- Kernel - mailweb.openeuler.org

[PATCH v2 1/1 openEuler-1.0-LTS] fork: add pid recover method for checkpoint and recover
by hejingxian 31 Dec '20

31 Dec '20

From: hejingxian <hejingxian(a)huawei.com> Date: Thu, 31 Dec 2020 20:18:56 +0800 Subject: [PATCH v2 1/1 openEuler-1.0-LTS] fork: add pid recover method for checkpoint and recover We record the pid of dump task in the reserved memory, and reserve the pids before init task start. In the recover process, set the fork_pid of the recovery task before call fork(). And then the fork_pid will be used to alloc pid. /proc/sys/kernel/ns_last_pid can also be used to fork child task with assigned pid. However, when there exist many tasks need to recover at the same time, we will fail to recover pids by using /proc/sys/kernel/ns_last_pid. Signed-off-by: Jingxian He <hejingxian(a)huawei.com> --- drivers/char/pin_memory.c | 24 +++++++++++++++++++++- include/linux/pin_mem.h | 6 ++++++ include/linux/sched.h | 9 +++++++++ init/init_task.c | 3 +++ kernel/pid.c | 33 ++++++++++++++++++++++++++++++ mm/Kconfig | 7 +++++++ mm/pin_mem.c | 51 +++++++++++++++++++++++++++++++++++++++++++++++ 7 files changed, 132 insertions(+), 1 deletion(-) diff --git a/drivers/char/pin_memory.c b/drivers/char/pin_memory.c index a0464e1..2c65de7 100644 --- a/drivers/char/pin_memory.c +++ b/drivers/char/pin_memory.c @@ -36,9 +36,12 @@ struct pin_mem_area_set { #define _SET_PIN_MEM_AREA 1 #define _CLEAR_PIN_MEM_AREA 2 #define _REMAP_PIN_MEM_AREA 3 +#define _SET_FORK_PID 4 + #define SET_PIN_MEM_AREA _IOW(PIN_MEM_MAGIC, _SET_PIN_MEM_AREA, struct pin_mem_area_set) #define CLEAR_PIN_MEM_AREA _IOW(PIN_MEM_MAGIC, _CLEAR_PIN_MEM_AREA, int) #define REMAP_PIN_MEM_AREA _IOW(PIN_MEM_MAGIC, _REMAP_PIN_MEM_AREA, int) +#define SET_FORK_PID _IOW(PIN_MEM_MAGIC, _SET_FORK_PID, int) static int set_pin_mem(struct pin_mem_area_set *pmas) { @@ -136,13 +139,29 @@ static int pin_mem_remap(unsigned long arg) return -EFAULT; } +static int set_fork_pid(unsigned long arg) +{ + int pid; + struct page_map_info * pmi = NULL; + void __user *buf = (void __user *)arg; + + if (!access_ok(buf, sizeof(int))) + goto fault; + if (copy_from_user(&pid, buf, sizeof(int))) + goto fault; + current->fork_pid = pid; + return 0; +fault: + return -EFAULT; +} + static long pin_memory_ioctl(struct file *file, unsigned cmd, unsigned long arg) { long ret = 0; if (_IOC_TYPE(cmd) != PIN_MEM_MAGIC) return -EINVAL; - if (_IOC_NR(cmd) > _REMAP_PIN_MEM_AREA) + if (_IOC_NR(cmd) > _SET_FORK_PID) return -EINVAL; switch (cmd) { @@ -155,6 +174,9 @@ static long pin_memory_ioctl(struct file *file, unsigned cmd, unsigned long arg) case REMAP_PIN_MEM_AREA: ret = pin_mem_remap(arg); break; + case SET_FORK_PID: + ret = set_fork_pid(arg); + break; default: return -EINVAL; } diff --git a/include/linux/pin_mem.h b/include/linux/pin_mem.h index 997ddb8..f47dab4 100644 --- a/include/linux/pin_mem.h +++ b/include/linux/pin_mem.h @@ -58,5 +58,11 @@ extern vm_fault_t do_anon_huge_page_remap(struct vm_area_struct *vma, unsigned l #endif extern void init_reserve_page_map(unsigned long map_addr, unsigned long map_size); +#ifdef CONFIG_PID_RECOVER +extern bool is_need_reserve_pids(void); +extern void reserve_pids(struct idr *idr, int pid_max); +extern void free_reserved_pid(struct idr *idr, int pid); +#endif + #endif /* CONFIG_PIN_MEMORY */ #endif /* _LINUX_PIN_MEMORY_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index ced2d7f..3d23e46 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1220,7 +1220,16 @@ struct task_struct { KABI_RESERVE(1) KABI_RESERVE(2) #endif +#ifdef CONFIG_PID_RECOVER +#ifndef __GENKSYMS__ + int fork_pid; + int reserve; +#else KABI_RESERVE(3) +#endif +#else + KABI_RESERVE(3) +#endif KABI_RESERVE(4) KABI_RESERVE(5) KABI_RESERVE(6) diff --git a/init/init_task.c b/init/init_task.c index 5aebe3b..0334ad8 100644 --- a/init/init_task.c +++ b/init/init_task.c @@ -179,6 +179,9 @@ struct task_struct init_task #ifdef CONFIG_SECURITY .security = NULL, #endif +#ifdef CONFIG_PID_RECOVER + .fork_pid = 0, +#endif }; EXPORT_SYMBOL(init_task); diff --git a/kernel/pid.c b/kernel/pid.c index bfdcd16..5d02591 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -41,6 +41,9 @@ #include <linux/sched/task.h> #include <linux/idr.h> #include <linux/kmemleak.h> +#ifdef CONFIG_PID_RECOVER +#include <linux/pin_mem.h> +#endif struct pid init_struct_pid = { .count = ATOMIC_INIT(1), @@ -185,6 +188,31 @@ struct pid *alloc_pid(struct pid_namespace *ns) if (idr_get_cursor(&tmp->idr) > RESERVED_PIDS) pid_min = RESERVED_PIDS; +#ifdef CONFIG_PID_RECOVER + if (!current->fork_pid) { + /* + * Store a null pointer so find_pid_ns does not find + * a partially initialized PID (see below). + */ + nr = idr_alloc_cyclic(&tmp->idr, NULL, pid_min, + tmp->pid_max, + GFP_ATOMIC); + } else { + /* Try to use fork_pid to alloc pid, and change to use the default way after fail. */ + test_pid_reserved_and_free(&tmp->idr, current->fork_pid); + pid_min = current->fork_pid; + current->fork_pid = 0; + nr = idr_alloc(&tmp->idr, NULL, pid_min, + tmp->pid_max, + GFP_ATOMIC); + pr_debug("Alloc pid by fork_pid ret: %d.", nr); + if (nr < 0) + nr = idr_alloc_cyclic(&tmp->idr, NULL, pid_min, + tmp->pid_max, + GFP_ATOMIC); + } +#else + /* * Store a null pointer so find_pid_ns does not find * a partially initialized PID (see below). @@ -192,6 +220,7 @@ struct pid *alloc_pid(struct pid_namespace *ns) nr = idr_alloc_cyclic(&tmp->idr, NULL, pid_min, tmp->pid_max, GFP_ATOMIC); +#endif spin_unlock_irq(&pidmap_lock); idr_preload_end(); @@ -500,6 +529,10 @@ void __init pid_idr_init(void) init_pid_ns.pid_cachep = KMEM_CACHE(pid, SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_ACCOUNT); +#ifdef CONFIG_PID_RECOVER + if (is_need_reserve_pids()) + reserve_pids(&init_pid_ns.idr, pid_max); +#endif hdr = register_sysctl_paths(pid_kern_path, pid_ctl_table); kmemleak_not_leak(hdr); diff --git a/mm/Kconfig b/mm/Kconfig index 31fa670..9ac20da 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -794,4 +794,11 @@ config PIN_MEMORY help Say y here to enable the pin memory feature for checkpoint and restore. + +config PID_RECOVER + bool "Support for pid recover" + depends on PIN_MEMORY + help + Say y here to enable the pid recover feature for checkpoint + and restore. endmenu diff --git a/mm/pin_mem.c b/mm/pin_mem.c index 89af938..2f77e86 100644 --- a/mm/pin_mem.c +++ b/mm/pin_mem.c @@ -688,4 +688,55 @@ vm_fault_t reserve_kernel_space_mem(unsigned long start_addr, unsigned int pages } EXPORT_SYMBOL_GPL(reserve_kernel_space_mem); +#ifdef CONFIG_PID_RECOVER +struct idr *reserve_idr; + +/* test if there exist pin memory tasks */ +bool is_need_reserve_pids(void) +{ + return (pin_pid_num > 0); +} + +void free_reserved_pid(struct idr *idr, int pid) +{ + unsigned int index; + struct page_map_info *pmi; + + if (!max_pin_pid_num || idr != reserve_idr) + return; + + for (index = 0; index < pin_pid_num; index++) { + pmi = &(user_space_reserve_start[index]); + if (pmi->pid == pid && pmi->pid_reserved) { + idr_remove(idr, pid); + return; + } + } +} + +/* reserve pids for check point tasks which pinned memory */ +void reserve_pids(struct idr *idr, int pid_max) +{ + int alloc_pid; + unsigned int index; + struct page_map_info *pmi; + + if (!max_pin_pid_num) + return; + reserve_idr = idr; + for (index = 0; index < pin_pid_num; index++) { + pmi = &(user_space_reserve_start[index]); + pmi->pid_reserved = true; + alloc_pid = idr_alloc(idr, NULL, pmi->pid, pid_max, GFP_ATOMIC); + if (alloc_pid != pmi->pid) { + if (alloc_pid > 0) + idr_remove(idr, alloc_pid); + pr_warn("Reserve pid (%d) fail, real pid is %d.\n", alloc_pid, pmi->pid); + pmi->pid_reserved = false; + continue; + } + } + return; +} +#endif /* CONFIG_PID_RECOVER */ #endif /* CONFIG_PIN_MEMORY */ -- 1.8.3.1

1 0

[PATCH v1 1/1 openEuler-1.0-LTS] fork: add pid recover method for checkpoint and recover
by hejingxian 31 Dec '20

31 Dec '20

From: hejingxian <hejingxian(a)huawei.com> Date: Thu, 10 Dec 2020 20:35:00 +0800 Subject: [PATCH v1 1/1 openEuler-1.0-LTS] fork: add pid recover method for checkpoint and recover We record the pid of dump task in the reserved memory, and reserve the pids before init task start. In the recover process, set the fork_pid of the recovery task before call fork(). And then the fork_pid will be used to alloc pid. /proc/sys/kernel/ns_last_pid can also be used to fork child task with assigned pid. However, when there exist many tasks need to recover at the same time, we will fail to recover pids by using /proc/sys/kernel/ns_last_pid. Signed-off-by: Jingxian He <hejingxian(a)huawei.com> --- drivers/char/pin_memory.c | 24 +++++++++++++++++++++- include/linux/pin_mem.h | 6 ++++++ include/linux/sched.h | 4 ++++ init/init_task.c | 3 +++ kernel/pid.c | 27 ++++++++++++++++++++++++- mm/Kconfig | 7 +++++++ mm/pin_mem.c | 51 +++++++++++++++++++++++++++++++++++++++++++++++ 7 files changed, 120 insertions(+), 2 deletions(-) diff --git a/drivers/char/pin_memory.c b/drivers/char/pin_memory.c index a0464e1..2c65de7 100644 --- a/drivers/char/pin_memory.c +++ b/drivers/char/pin_memory.c @@ -36,9 +36,12 @@ struct pin_mem_area_set { #define _SET_PIN_MEM_AREA 1 #define _CLEAR_PIN_MEM_AREA 2 #define _REMAP_PIN_MEM_AREA 3 +#define _SET_FORK_PID 4 + #define SET_PIN_MEM_AREA _IOW(PIN_MEM_MAGIC, _SET_PIN_MEM_AREA, struct pin_mem_area_set) #define CLEAR_PIN_MEM_AREA _IOW(PIN_MEM_MAGIC, _CLEAR_PIN_MEM_AREA, int) #define REMAP_PIN_MEM_AREA _IOW(PIN_MEM_MAGIC, _REMAP_PIN_MEM_AREA, int) +#define SET_FORK_PID _IOW(PIN_MEM_MAGIC, _SET_FORK_PID, int) static int set_pin_mem(struct pin_mem_area_set *pmas) { @@ -136,13 +139,29 @@ static int pin_mem_remap(unsigned long arg) return -EFAULT; } +static int set_fork_pid(unsigned long arg) +{ + int pid; + struct page_map_info * pmi = NULL; + void __user *buf = (void __user *)arg; + + if (!access_ok(buf, sizeof(int))) + goto fault; + if (copy_from_user(&pid, buf, sizeof(int))) + goto fault; + current->fork_pid = pid; + return 0; +fault: + return -EFAULT; +} + static long pin_memory_ioctl(struct file *file, unsigned cmd, unsigned long arg) { long ret = 0; if (_IOC_TYPE(cmd) != PIN_MEM_MAGIC) return -EINVAL; - if (_IOC_NR(cmd) > _REMAP_PIN_MEM_AREA) + if (_IOC_NR(cmd) > _SET_FORK_PID) return -EINVAL; switch (cmd) { @@ -155,6 +174,9 @@ static long pin_memory_ioctl(struct file *file, unsigned cmd, unsigned long arg) case REMAP_PIN_MEM_AREA: ret = pin_mem_remap(arg); break; + case SET_FORK_PID: + ret = set_fork_pid(arg); + break; default: return -EINVAL; } diff --git a/include/linux/pin_mem.h b/include/linux/pin_mem.h index 0ca44ac..4162043 100644 --- a/include/linux/pin_mem.h +++ b/include/linux/pin_mem.h @@ -58,5 +58,11 @@ extern vm_fault_t do_anon_huge_page_remap(struct vm_area_struct *vma, unsigned l #endif extern void init_reserve_page_map(unsigned long map_addr, unsigned long map_size); +#ifdef CONFIG_PID_RECOVER +extern bool is_need_reserve_pids(void); +extern void reserve_pids(struct idr *idr, int pid_max); +extern void free_reserved_pid(struct idr *idr, int pid); +#endif + #endif /* CONFIG_PIN_MEMORY */ #endif /* _LINUX_PIN_MEMORY_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 716ad1d..f6b1560 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1281,6 +1281,10 @@ struct task_struct { unsigned long prev_lowest_stack; #endif +#ifdef CONFIG_PID_RECOVER + int fork_pid; +#endif + /* * New fields for task_struct should be added above here, so that * they are included in the randomized portion of task_struct. diff --git a/init/init_task.c b/init/init_task.c index 9e5cbe5..40cd98f 100644 --- a/init/init_task.c +++ b/init/init_task.c @@ -181,6 +181,9 @@ struct task_struct init_task #ifdef CONFIG_SECURITY .security = NULL, #endif +#ifdef CONFIG_PID_RECOVER + .fork_pid = 0, +#endif }; EXPORT_SYMBOL(init_task); diff --git a/kernel/pid.c b/kernel/pid.c index 2278e24..dc93f7a 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -42,6 +42,9 @@ #include <linux/sched/signal.h> #include <linux/sched/task.h> #include <linux/idr.h> +#ifdef CONFIG_PID_RECOVER +#include <linux/pin_mem.h> +#endif struct pid init_struct_pid = { .count = REFCOUNT_INIT(1), @@ -208,8 +211,26 @@ struct pid *alloc_pid(struct pid_namespace *ns, pid_t *set_tid, idr_preload(GFP_KERNEL); spin_lock_irq(&pidmap_lock); - +#ifdef CONFIG_PID_RECOVER + if (current->fork_pid) { + int pid_min; + /* Try to free the reserved fork_pid, and then use it to alloc pid. */ + free_reserved_pid(&tmp->idr, current->fork_pid); + pid_min = current->fork_pid; + current->fork_pid = 0; + nr = idr_alloc(&tmp->idr, NULL, pid_min, + pid_min + 1, + GFP_ATOMIC); + /* + * If ENOSPC is returned it means that the PID is + * alreay in use. Return EEXIST in that case. + */ + if (nr == -ENOSPC) + nr = -EEXIST; + } else if (tid) { +#else if (tid) { +#endif nr = idr_alloc(&tmp->idr, NULL, tid, tid + 1, GFP_ATOMIC); /* @@ -577,4 +598,8 @@ void __init pid_idr_init(void) init_pid_ns.pid_cachep = KMEM_CACHE(pid, SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_ACCOUNT); +#ifdef CONFIG_PID_RECOVER + if (is_need_reserve_pids()) + reserve_pids(&init_pid_ns.idr, pid_max); +#endif } diff --git a/mm/Kconfig b/mm/Kconfig index c2dd088..847e8e4 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -745,4 +745,11 @@ config PIN_MEMORY help Say y here to enable the pin memory feature for checkpoint and restore. + +config PID_RECOVER + bool "Support for pid recover" + depends on PIN_MEMORY + help + Say y here to enable the pid recover feature for checkpoint + and restore. endmenu diff --git a/mm/pin_mem.c b/mm/pin_mem.c index ca3f23a..9426874 100644 --- a/mm/pin_mem.c +++ b/mm/pin_mem.c @@ -688,4 +688,55 @@ vm_fault_t reserve_kernel_space_mem(unsigned long start_addr, unsigned int pages } EXPORT_SYMBOL_GPL(reserve_kernel_space_mem); +#ifdef CONFIG_PID_RECOVER +struct idr *reserve_idr; + +/* test if there exist pin memory tasks */ +bool is_need_reserve_pids(void) +{ + return (pin_pid_num > 0); +} + +void free_reserved_pid(struct idr *idr, int pid) +{ + unsigned int index; + struct page_map_info *pmi; + + if (!max_pin_pid_num || idr != reserve_idr) + return; + + for (index = 0; index < pin_pid_num; index++) { + pmi = &(user_space_reserve_start[index]); + if (pmi->pid == pid && pmi->pid_reserved) { + idr_remove(idr, pid); + return; + } + } +} + +/* reserve pids for check point tasks which pinned memory */ +void reserve_pids(struct idr *idr, int pid_max) +{ + int alloc_pid; + unsigned int index; + struct page_map_info *pmi; + + if (!max_pin_pid_num) + return; + reserve_idr = idr; + for (index = 0; index < pin_pid_num; index++) { + pmi = &(user_space_reserve_start[index]); + pmi->pid_reserved = true; + alloc_pid = idr_alloc(idr, NULL, pmi->pid, pid_max, GFP_ATOMIC); + if (alloc_pid != pmi->pid) { + if (alloc_pid > 0) + idr_remove(idr, alloc_pid); + pr_warn("Reserve pid (%d) fail, real pid is %d.\n", alloc_pid, pmi->pid); + pmi->pid_reserved = false; + continue; + } + } + return; +} +#endif /* CONFIG_PID_RECOVER */ #endif /* CONFIG_PIN_MEMORY */ -- 1.8.3.1

2 1

[PATCH v1 0/1 openEuler-1.0-LTS] mm: add pin memory method for checkpoint add restore
by hejingxian 31 Dec '20

31 Dec '20

From: hejingxian <hejingxian(a)huawei.com> Date: Thu, 10 Dec 2020 20:31:15 +0800 Subject: [PATCH v1 0/1 openEuler-1.0-LTS] mm: add pin memory method for checkpoint add restore We can use the checkpoint and restore in userspace(criu) method to dump and restore tasks when updating the kernel. Currently, criu needs dump all memory data of tasks to files. When the memory size is very large(larger than 1G), the cost time of the dumping data will be very long(more than 1 min). We can pin the memory data of tasks and collect the corresponding physical pages mapping info in checkpoint process, and remap the physical pages to restore tasks after kernel is updated. The pin memory area info is saved in the reserve memblock named nvwa_res_first, which can keep usable in the kernel update process. The pin memory driver provides the following ioctl command for criu: 1) SET_PIN_MEM_AREA: set pin memory area, which can be remap to the restore task. 2) CLEAR_PIN_MEM_AREA: clear the pin memory area info, which enable user reset the pin data. 3) REMAP_PIN_MEM_AREA: remap the pages of the pin memory to the restore task. Signed-off-by: Jingxian He <hejingxian(a)huawei.com> --- arch/arm64/kernel/setup.c | 7 + arch/arm64/mm/init.c | 62 +++- drivers/char/Kconfig | 7 + drivers/char/Makefile | 1 + drivers/char/pin_memory.c | 198 +++++++++++++ include/linux/crash_core.h | 5 + include/linux/pin_mem.h | 62 ++++ kernel/crash_core.c | 11 + mm/Kconfig | 6 + mm/Makefile | 1 + mm/huge_memory.c | 61 ++++ mm/memory.c | 68 +++++ mm/pin_mem.c | 691 +++++++++++++++++++++++++++++++++++++++++++++ 13 files changed, 1179 insertions(+), 1 deletion(-) create mode 100644 drivers/char/pin_memory.c create mode 100644 include/linux/pin_mem.h create mode 100644 mm/pin_mem.c diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index 56f6645..40751ed 100644 --- a/arch/arm64/kernel/setup.c +++ b/arch/arm64/kernel/setup.c @@ -50,6 +50,9 @@ #include <asm/efi.h> #include <asm/xen/hypervisor.h> #include <asm/mmu_context.h> +#ifdef CONFIG_PIN_MEMORY +#include <linux/pin_memory.h> +#endif static int num_standard_resources; static struct resource *standard_resources; @@ -243,6 +246,10 @@ static void __init request_standard_resources(void) crashk_res.end <= res->end) request_resource(res, &crashk_res); #endif +#ifdef CONFIG_PIN_MEMORY + if (pin_memory_resource.end) + insert_resource(&iomem_resource, &pin_memory_resource); +#endif } } diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index b65dffd..dee3192 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -41,7 +41,9 @@ #include <linux/sizes.h> #include <asm/tlb.h> #include <asm/alternative.h> - +#ifdef CONFIG_PIN_MEMORY +#include <linux/pin_memory.h> +#endif #define ARM64_ZONE_DMA_BITS 30 /* @@ -68,6 +70,16 @@ phys_addr_t arm64_dma_phys_limit __ro_after_init; static phys_addr_t arm64_dma32_phys_limit __ro_after_init; +#ifdef CONFIG_PIN_MEMORY +struct resource pin_memory_resource = { + .name = "Pin memory maps", + .start = 0, + .end = 0, + .flags = IORESOURCE_MEM, + .desc = IORES_DESC_PIN_MEM_MAPS +}; +#endif + #ifdef CONFIG_KEXEC_CORE /* * reserve_crashkernel() - reserves memory for crash kernel @@ -129,6 +141,47 @@ static void __init reserve_crashkernel(void) } #endif /* CONFIG_KEXEC_CORE */ +#ifdef CONFIG_PIN_MEMORY +static void __init reserve_pin_memory_res(void) +{ + unsigned long long mem_start, mem_len; + int ret; + + ret = parse_pin_memory(boot_command_line, memblock_phys_mem_size(), + &mem_len, &mem_start); + if (ret || !mem_len) + return; + + mem_len = PAGE_ALIGN(mem_len); + + if (!memblock_is_region_memory(mem_start, mem_len)) { + pr_warn("cannot reserve for pin memory: region is not memory!\n"); + return; + } + + if (memblock_is_region_reserved(mem_start, mem_len)) { + pr_warn("cannot reserve for pin memory: region overlaps reserved memory!\n"); + return; + } + + if (!IS_ALIGNED(mem_start, SZ_2M)) { + pr_warn("cannot reserve for pin memory: base address is not 2MB aligned\n"); + return; + } + + memblock_reserve(mem_start, mem_len); + pr_debug("pin memory resource reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + mem_start, mem_start + mem_len, mem_len >> 20); + + pin_memory_resource.start = mem_start; + pin_memory_resource.end = mem_start + mem_len - 1; +} +#else +static void __init reserve_pin_memory_res(void) +{ +} +#endif /* CONFIG_PIN_MEMORY */ + #ifdef CONFIG_CRASH_DUMP static int __init early_init_dt_scan_elfcorehdr(unsigned long node, const char *uname, int depth, void *data) @@ -452,6 +505,8 @@ void __init arm64_memblock_init(void) reserve_crashkernel(); + reserve_pin_memory_res(); + reserve_elfcorehdr(); high_memory = __va(memblock_end_of_DRAM() - 1) + 1; @@ -573,6 +628,11 @@ void __init mem_init(void) /* this will put all unused low memory onto the freelists */ memblock_free_all(); +#ifdef CONFIG_PIN_MEMORY + /* pre alloc the pages for pin memory */ + init_reserve_page_map((unsigned long)pin_memory_resource.start, + (unsigned long)(pin_memory_resource.end - pin_memory_resource.start)); +#endif mem_init_print_info(NULL); /* diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index 26956c0..73af2f0 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -560,3 +560,10 @@ config RANDOM_TRUST_BOOTLOADER booloader is trustworthy so it will be added to the kernel's entropy pool. Otherwise, say N here so it will be regarded as device input that only mixes the entropy pool. + +config PIN_MEMORY_DEV + bool "/dev/pinmem character device" + depends PIN_MEMORY + default n + help + pin memory driver diff --git a/drivers/char/Makefile b/drivers/char/Makefile index 7c5ea6f..1941642 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -52,3 +52,4 @@ js-rtc-y = rtc.o obj-$(CONFIG_XILLYBUS) += xillybus/ obj-$(CONFIG_POWERNV_OP_PANEL) += powernv-op-panel.o obj-$(CONFIG_ADI) += adi.o +obj-$(CONFIG_PIN_MEMORY_DEV) += pin_memory.o diff --git a/drivers/char/pin_memory.c b/drivers/char/pin_memory.c new file mode 100644 index 00000000..a0464e1 --- /dev/null +++ b/drivers/char/pin_memory.c @@ -0,0 +1,198 @@ +/* + * Copyright @ Huawei Technologies Co., Ltd. 2020-2020. ALL rights reserved. + * Description: Euler pin memory driver + */ +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/kprobes.h> +#include <linux/spinlock.h> +#include <linux/workqueue.h> +#include <linux/sched.h> +#include <linux/mm.h> +#include <linux/init.h> +#include <linux/miscdevice.h> +#include <linux/fs.h> +#include <linux/mm_types.h> +#include <asm/processor.h> +#include <uapi/asm-generic/ioctl.h> +#include <uapi/asm-generic/mman-common.h> +#include <uapi/asm/setup.h> +#include <linux/pin_mem.h> +#include <linux/sched/mm.h> + +#define MAX_PIN_MEM_AREA_NUM 16 +struct _pin_mem_area { + unsigned long virt_start; + unsigned long virt_end; +}; + +struct pin_mem_area_set { + unsigned int pid; + unsigned int area_num; + struct _pin_mem_area mem_area[MAX_PIN_MEM_AREA_NUM]; +}; + +#define PIN_MEM_MAGIC 0x59 +#define _SET_PIN_MEM_AREA 1 +#define _CLEAR_PIN_MEM_AREA 2 +#define _REMAP_PIN_MEM_AREA 3 +#define SET_PIN_MEM_AREA _IOW(PIN_MEM_MAGIC, _SET_PIN_MEM_AREA, struct pin_mem_area_set) +#define CLEAR_PIN_MEM_AREA _IOW(PIN_MEM_MAGIC, _CLEAR_PIN_MEM_AREA, int) +#define REMAP_PIN_MEM_AREA _IOW(PIN_MEM_MAGIC, _REMAP_PIN_MEM_AREA, int) + +static int set_pin_mem(struct pin_mem_area_set *pmas) +{ + int i; + int ret = 0; + struct _pin_mem_area *pma; + struct mm_struct *mm; + struct task_struct *task; + struct pid *pid_s; + + pid_s = find_get_pid(pmas->pid); + if (!pid_s) { + pr_warn("Get pid struct fail:%d.\n", pmas->pid); + goto fail; + } + rcu_read_lock(); + task = pid_task(pid_s, PIDTYPE_PID); + if (!task) { + pr_warn("Get task struct fail:%d.\n", pmas->pid); + goto fail; + } + mm = get_task_mm(task); + for (i = 0; i < pmas->area_num; i++) { + pma = &(pmas->mem_area[i]); + ret = pin_mem_area(task, mm, pma->virt_start, pma->virt_end); + if (ret) { + mmput(mm); + goto fail; + } + } + mmput(mm); + rcu_read_unlock(); + return ret; + +fail: + rcu_read_unlock(); + return -EFAULT; +} + +static int set_pin_mem_area(unsigned long arg) +{ + struct pin_mem_area_set pmas; + void __user *buf = (void __user *)arg; + + if (!access_ok(buf, sizeof(pmas))) + return -EFAULT; + if (copy_from_user(&pmas, buf, sizeof(pmas))) + return -EINVAL; + if (pmas.area_num > MAX_PIN_MEM_AREA_NUM) { + pr_warn("Input area_num is too large.\n"); + return -EINVAL; + } + + return set_pin_mem(&pmas); +} + +static int pin_mem_remap(unsigned long arg) +{ + int pid; + struct task_struct *task; + struct mm_struct *mm; + vm_fault_t ret; + void __user *buf = (void __user *)arg; + struct pid *pid_s; + + if (!access_ok(buf, sizeof(int))) + return -EINVAL; + if (copy_from_user(&pid, buf, sizeof(int))) + return -EINVAL; + + pid_s = find_get_pid(pid); + if (!pid_s) { + pr_warn("Get pid struct fail:%d.\n", pid); + return -EINVAL; + } + rcu_read_lock(); + task = pid_task(pid_s, PIDTYPE_PID); + if (!task) { + pr_warn("Get task struct fail:%d.\n", pid); + goto fault; + } + mm = get_task_mm(task); + ret = do_mem_remap(pid, mm); + if (ret) { + pr_warn("Handle pin memory remap fail.\n"); + mmput(mm); + goto fault; + } + mmput(mm); + rcu_read_unlock(); + return 0; + +fault: + rcu_read_unlock(); + return -EFAULT; +} + +static long pin_memory_ioctl(struct file *file, unsigned cmd, unsigned long arg) +{ + long ret = 0; + + if (_IOC_TYPE(cmd) != PIN_MEM_MAGIC) + return -EINVAL; + if (_IOC_NR(cmd) > _REMAP_PIN_MEM_AREA) + return -EINVAL; + + switch (cmd) { + case SET_PIN_MEM_AREA: + ret = set_pin_mem_area(arg); + break; + case CLEAR_PIN_MEM_AREA: + clear_pin_memory_record(); + break; + case REMAP_PIN_MEM_AREA: + ret = pin_mem_remap(arg); + break; + default: + return -EINVAL; + } + return ret; +} + +static const struct file_operations pin_memory_fops = { + .owner = THIS_MODULE, + .unlocked_ioctl = pin_memory_ioctl, + .compat_ioctl = pin_memory_ioctl, +}; + +static struct miscdevice pin_memory_miscdev = { + .minor = MISC_DYNAMIC_MINOR, + .name = "pinmem", + .fops = &pin_memory_fops, +}; + +static int pin_memory_init(void) +{ + int err = misc_register(&pin_memory_miscdev); + if (!err) { + pr_info("pin_memory init\n"); + } else { + pr_warn("pin_memory init failed!\n"); + } + return err; +} + +static void pin_memory_exit(void) +{ + misc_deregister(&pin_memory_miscdev); + pr_info("pin_memory ko exists!\n"); +} + +module_init(pin_memory_init); +module_exit(pin_memory_exit); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Euler"); +MODULE_DESCRIPTION("pin memory"); diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h index 525510a..5baf40d 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -75,4 +75,9 @@ int parse_crashkernel_high(char *cmdline, unsigned long long system_ram, int parse_crashkernel_low(char *cmdline, unsigned long long system_ram, unsigned long long *crash_size, unsigned long long *crash_base); +#ifdef CONFIG_PIN_MEMORY +int __init parse_pin_memory(char *cmdline, unsigned long long system_ram, + unsigned long long *pin_size, unsigned long long *pin_base); +#endif + #endif /* LINUX_CRASH_CORE_H */ diff --git a/include/linux/pin_mem.h b/include/linux/pin_mem.h new file mode 100644 index 00000000..0ca44ac --- /dev/null +++ b/include/linux/pin_mem.h @@ -0,0 +1,62 @@ +/* + * Copyright (C) 2020. Huawei Technologies Co., Ltd. All rights reserved. + * Provide the pin memory method for check point and restore task. + */ +#ifndef _LINUX_PIN_MEMORY_H +#define _LINUX_PIN_MEMORY_H + +#ifdef CONFIG_PIN_MEMORY +#include <linux/errno.h> +#include <linux/kabi.h> +#include <linux/mm_types.h> +#include <linux/err.h> +#ifdef CONFIG_ARM64 +#include <linux/ioport.h> +#endif + +#define PAGE_BUDDY_MAPCOUNT_VALUE (~PG_buddy) + +#define COLLECT_PAGES_FINISH 0 +#define COLLECT_PAGES_NEED_CONTINUE 1 +#define COLLECT_PAGES_FAIL -1 + +#define COMPOUND_PAD_MASK 0xffffffff +#define COMPOUND_PAD_START 0x88 +#define COMPOUND_PAD_DELTA 0x40 +#define LIST_POISON4 0xdead000000000400 + +#define next_pme(pme) ((unsigned long *)(pme + 1) + pme->nr_pages) + +struct page_map_entry { + unsigned long virt_addr; + unsigned int nr_pages; + unsigned int is_huge_page; + unsigned long phy_addr_array[0]; +}; + +struct page_map_info { + int pid; + int pid_reserved; + unsigned int entry_num; + struct page_map_entry *pme; +}; + +extern struct page_map_info *get_page_map_info(int pid); +extern struct page_map_info *create_page_map_info(int pid); +extern vm_fault_t do_mem_remap(int pid, struct mm_struct *mm); +extern vm_fault_t do_anon_page_remap(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmd, struct page *page); +extern void clear_pin_memory_record(void); +extern int pin_mem_area(struct task_struct *task, struct mm_struct *mm, + unsigned long start_addr, unsigned long end_addr); +extern vm_fault_t do_anon_huge_page_remap(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmd, struct page *page); + +/* reserve space for pin memory*/ +#ifdef CONFIG_ARM64 +extern struct resource pin_memory_resource; +#endif +extern void init_reserve_page_map(unsigned long map_addr, unsigned long map_size); + +#endif /* CONFIG_PIN_MEMORY */ +#endif /* _LINUX_PIN_MEMORY_H */ diff --git a/kernel/crash_core.c b/kernel/crash_core.c index 9f1557b..7512696 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -292,6 +292,17 @@ int __init parse_crashkernel_low(char *cmdline, "crashkernel=", suffix_tbl[SUFFIX_LOW]); } +#ifdef CONFIG_PIN_MEMORY +int __init parse_pin_memory(char *cmdline, + unsigned long long system_ram, + unsigned long long *pin_size, + unsigned long long *pin_base) +{ + return __parse_crashkernel(cmdline, system_ram, pin_size, pin_base, + "pinmemory=", NULL); +} +#endif + Elf_Word *append_elf_note(Elf_Word *buf, char *name, unsigned int type, void *data, size_t data_len) { diff --git a/mm/Kconfig b/mm/Kconfig index ab80933..c2dd088 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -739,4 +739,10 @@ config ARCH_HAS_HUGEPD config MAPPING_DIRTY_HELPERS bool +config PIN_MEMORY + bool "Support for pin memory" + depends on CHECKPOINT_RESTORE + help + Say y here to enable the pin memory feature for checkpoint + and restore. endmenu diff --git a/mm/Makefile b/mm/Makefile index 1937cc2..7e1984e 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -108,3 +108,4 @@ obj-$(CONFIG_ZONE_DEVICE) += memremap.o obj-$(CONFIG_HMM_MIRROR) += hmm.o obj-$(CONFIG_MEMFD_CREATE) += memfd.o obj-$(CONFIG_MAPPING_DIRTY_HELPERS) += mapping_dirty_helpers.o +obj-$(CONFIG_PIN_MEMORY) += pin_mem.o diff --git a/mm/huge_memory.c b/mm/huge_memory.c index a880932..93dc582 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -3083,4 +3083,65 @@ void remove_migration_pmd(struct page_vma_mapped_walk *pvmw, struct page *new) mlock_vma_page(new); update_mmu_cache_pmd(vma, address, pvmw->pmd); } + +#ifdef CONFIG_PIN_MEMORY +vm_fault_t do_anon_huge_page_remap(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmd, struct page *page) +{ + gfp_t gfp; + pgtable_t pgtable; + spinlock_t *ptl; + pmd_t entry; + vm_fault_t ret = 0; + struct mem_cgroup *memcg; + + if (unlikely(anon_vma_prepare(vma))) + return VM_FAULT_OOM; + if (unlikely(khugepaged_enter(vma, vma->vm_flags))) + return VM_FAULT_OOM; + gfp = alloc_hugepage_direct_gfpmask(vma); + prep_transhuge_page(page); + if (mem_cgroup_try_charge_delay(page, vma->vm_mm, gfp, &memcg, true)) { + put_page(page); + count_vm_event(THP_FAULT_FALLBACK); + return VM_FAULT_FALLBACK; + } + pgtable = pte_alloc_one(vma->vm_mm, address); + if (unlikely(!pgtable)) { + ret = VM_FAULT_OOM; + goto release; + } + __SetPageUptodate(page); + ptl = pmd_lock(vma->vm_mm, pmd); + if (unlikely(!pmd_none(*pmd))) { + goto unlock_release; + } else { + ret = check_stable_address_space(vma->vm_mm); + if (ret) + goto unlock_release; + entry = mk_huge_pmd(page, vma->vm_page_prot); + entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma); + page_add_new_anon_rmap(page, vma, address, true); + mem_cgroup_commit_charge(page, memcg, false, true); + lru_cache_add_active_or_unevictable(page, vma); + pgtable_trans_huge_deposit(vma->vm_mm, pmd, pgtable); + set_pmd_at(vma->vm_mm, address, pmd, entry); + add_mm_counter(vma->vm_mm, MM_ANONPAGES, HPAGE_PMD_NR); + mm_inc_nr_ptes(vma->vm_mm); + spin_unlock(ptl); + count_vm_event(THP_FAULT_ALLOC); + } + + return 0; +unlock_release: + spin_unlock(ptl); +release: + if (pgtable) + pte_free(vma->vm_mm, pgtable); + mem_cgroup_cancel_charge(page, memcg, true); + put_page(page); + return ret; +} +#endif + #endif diff --git a/mm/memory.c b/mm/memory.c index 45442d9..dd416fd 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -4799,4 +4799,72 @@ void ptlock_free(struct page *page) { kmem_cache_free(page_ptl_cachep, page->ptl); } + +#ifdef CONFIG_PIN_MEMORY +vm_fault_t do_anon_page_remap(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmd, struct page *page) +{ + struct mem_cgroup *memcg; + pte_t entry; + spinlock_t *ptl; + pte_t *pte; + vm_fault_t ret = 0; + + if (pte_alloc(vma->vm_mm, pmd, address)) + return VM_FAULT_OOM; + + /* See the comment in pte_alloc_one_map() */ + if (unlikely(pmd_trans_unstable(pmd))) + return 0; + + /* Allocate our own private page. */ + if (unlikely(anon_vma_prepare(vma))) + goto oom; + + if (mem_cgroup_try_charge_delay(page, vma->vm_mm, GFP_KERNEL, &memcg, + false)) + goto oom_free_page; + + /* + * The memory barrier inside __SetPageUptodate makes sure that + * preceeding stores to the page contents become visible before + * the set_pte_at() write. + */ + __SetPageUptodate(page); + + entry = mk_pte(page, vma->vm_page_prot); + if (vma->vm_flags & VM_WRITE) + entry = pte_mkwrite(pte_mkdirty(entry)); + pte = pte_offset_map_lock(vma->vm_mm, pmd, address, + &ptl); + if (!pte_none(*pte)) { + ret = VM_FAULT_FALLBACK; + goto release; + } + + ret = check_stable_address_space(vma->vm_mm); + if (ret) + goto release; + inc_mm_counter_fast(vma->vm_mm, MM_ANONPAGES); + page_add_new_anon_rmap(page, vma, address, false); + mem_cgroup_commit_charge(page, memcg, false, false); + lru_cache_add_active_or_unevictable(page, vma); + + set_pte_at(vma->vm_mm, address, pte, entry); + /* No need to invalidate - it was non-present before */ + update_mmu_cache(vma, address, pte); +unlock: + pte_unmap_unlock(pte, ptl); + return ret; +release: + mem_cgroup_cancel_charge(page, memcg, false); + put_page(page); + goto unlock; +oom_free_page: + put_page(page); +oom: + return VM_FAULT_OOM; +} +#endif + #endif diff --git a/mm/pin_mem.c b/mm/pin_mem.c new file mode 100644 index 00000000..ca3f23a --- /dev/null +++ b/mm/pin_mem.c @@ -0,0 +1,691 @@ +/* + * Copyright (C) 2020. Huawei Technologies Co., Ltd. All rights reserved. + * Provide the pin memory method for check point and restore task. + */ +#ifdef CONFIG_PIN_MEMORY +#include <linux/fs.h> +#include <linux/init.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/slab.h> +#include <linux/time.h> +#include <linux/sched/cputime.h> +#include <linux/tick.h> +#include <asm/uaccess.h> +#include <linux/mm.h> +#include <linux/pin_mem.h> +#include <linux/idr.h> +#include <linux/page-isolation.h> +#include <linux/sched/mm.h> +#include <linux/ctype.h> + +#define MAX_PIN_PID_NUM 128 +static DEFINE_SPINLOCK(page_map_entry_lock); + +unsigned int pin_pid_num; +static unsigned int *pin_pid_num_addr; +static unsigned long __page_map_entry_start; +static unsigned long page_map_entry_end; +static struct page_map_info *user_space_reserve_start; +static struct page_map_entry *page_map_entry_start; +unsigned int max_pin_pid_num __read_mostly; + +static int __init setup_max_pin_pid_num(char *str) +{ + int ret = 1; + + if (!str) + goto out; + + ret = kstrtouint(str, 10, &max_pin_pid_num); +out: + if (ret) { + pr_warn("Unable to parse max pin pid num.\n"); + } else { + if (max_pin_pid_num > MAX_PIN_PID_NUM) { + max_pin_pid_num = 0; + pr_warn("Input max_pin_pid_num is too large.\n"); + } + } + return ret; +} +early_param("max_pin_pid_num", setup_max_pin_pid_num); + +struct page_map_info *create_page_map_info(int pid) +{ + struct page_map_info *new; + + if (!user_space_reserve_start) + return NULL; + + if (pin_pid_num >= max_pin_pid_num) { + pr_warn("Pin pid num too large than max_pin_pid_num, fail create: %d!", pid); + return NULL; + } + new = (struct page_map_info *)(user_space_reserve_start + pin_pid_num); + new->pid = pid; + new->pme = NULL; + new->entry_num = 0; + new->pid_reserved = false; + (*pin_pid_num_addr)++; + pin_pid_num++; + return new; +} +EXPORT_SYMBOL_GPL(create_page_map_info); + +struct page_map_info *get_page_map_info(int pid) +{ + int i; + + if (!user_space_reserve_start) + return NULL; + + for (i = 0; i < pin_pid_num; i++) { + if (user_space_reserve_start[i].pid == pid) { + return &(user_space_reserve_start[i]); + } + } + return NULL; +} +EXPORT_SYMBOL_GPL(get_page_map_info); + +static struct page *find_head_page(struct page *page) +{ + struct page *p = page; + + while (!PageBuddy(p)) { + if (PageLRU(p)) + return NULL; + p--; + } + return p; +} + +static void spilt_page_area_left(struct zone *zone, struct free_area *area, struct page *page, + unsigned long size, int order) +{ + unsigned long cur_size = 1 << order; + unsigned long total_size = 0; + struct page *tmp; + unsigned long tmp_size = size; + + while (size && cur_size > size) { + cur_size >>= 1; + order--; + area--; + if (cur_size <= size) { + list_add(&page[total_size].lru, &area->free_list[MIGRATE_MOVABLE]); + atomic_set(&(page[total_size]._mapcount), PAGE_BUDDY_MAPCOUNT_VALUE); + set_page_private(&page[total_size], order); + set_pageblock_migratetype(&page[total_size], MIGRATE_MOVABLE); + area->nr_free++; + total_size += cur_size; + size -= cur_size; + } + } +} + +static void spilt_page_area_right(struct zone *zone, struct free_area *area, struct page *page, + unsigned long size, int order) +{ + unsigned long cur_size = 1 << order; + struct page *right_page, *head_page; + unsigned long tmp_size = size; + + right_page = page + size; + while (size && cur_size > size) { + cur_size >>= 1; + order--; + area--; + if (cur_size <= size) { + head_page = right_page - cur_size; + list_add(&head_page->lru, &area->free_list[MIGRATE_MOVABLE]); + atomic_set(&(head_page->_mapcount), PAGE_BUDDY_MAPCOUNT_VALUE); + set_page_private(head_page, order); + set_pageblock_migratetype(head_page, MIGRATE_MOVABLE); + area->nr_free++; + size -= cur_size; + right_page = head_page; + } + } +} + +void reserve_page_from_buddy(unsigned long nr_pages, struct page *page) +{ + unsigned int current_order; + struct page *page_end; + struct free_area *area; + struct zone *zone; + struct page *head_page; + + head_page = find_head_page(page); + if (!head_page) { + pr_warn("Find page head fail."); + return; + } + current_order = head_page->private; + page_end = head_page + (1 << current_order); + zone = page_zone(head_page); + area = &(zone->free_area[current_order]); + list_del(&head_page->lru); + atomic_set(&head_page->_mapcount, -1); + set_page_private(head_page, 0); + area->nr_free--; + if (head_page != page) + spilt_page_area_left(zone, area, head_page, + (unsigned long)(page - head_page), current_order); + page = page + nr_pages; + if (page < page_end) { + spilt_page_area_right(zone, area, page, + (unsigned long)(page_end - page), current_order); + } else if (page > page_end) { + pr_warn("Find page end smaller than page."); + } +} + +static inline void reserve_user_normal_pages(struct page *page) +{ + if (!atomic_read(&page->_refcount)) { + atomic_inc(&page->_refcount); + reserve_page_from_buddy(1, page); + } else { + pr_warn("Page %pK refcount %d large than zero, no need reserve.\n", + page, page->_refcount.counter); + } +} + +static void init_huge_pmd_pages(struct page *head_page) +{ + int i = 0; + struct page *page = head_page; + unsigned long *temp; + unsigned long compound_pad = COMPOUND_PAD_START; + + __set_bit(PG_head, &page->flags); + __set_bit(PG_active, &page->flags); + atomic_set(&page->_refcount, 1); + page++; + i++; + page->compound_head = (unsigned long)head_page + 1; + page->_compound_pad_2 = (unsigned long)head_page & COMPOUND_PAD_MASK; + temp = (unsigned long *)(&(page->_compound_pad_2)); + temp[1] = LIST_POISON4; + page->compound_dtor = HUGETLB_PAGE_DTOR + 1; + page->compound_order = HPAGE_PMD_ORDER; + page++; + i++; + page->compound_head = (unsigned long)head_page + 1; + page->_compound_pad_2 = (unsigned long)head_page + compound_pad; + i++; + INIT_LIST_HEAD(&(page->deferred_list)); + for (; i < HPAGE_PMD_NR; i++) { + page = head_page + i; + page->compound_head = (unsigned long)head_page + 1; + compound_pad += COMPOUND_PAD_DELTA; + page->_compound_pad_2 = (unsigned long)head_page + compound_pad; + temp = (unsigned long *)(&(page->_compound_pad_2)); + temp[1] = LIST_POISON4; + } +} + +static void reserve_user_huge_pmd_pages(struct page *page) +{ + struct page *head_page; + + if (!atomic_read(&page->_refcount)) { + atomic_inc(&page->_refcount); + head_page = find_head_page(page); + reserve_page_from_buddy((1 << HPAGE_PMD_ORDER), page); + init_huge_pmd_pages(page); + } else { + pr_warn("Page %pK refcount %d large than zero, no need reserve.\n", + page, page->_refcount.counter); + } +} + +static void reserve_user_space_map_pages(void) +{ + struct page_map_info *pmi; + struct page_map_entry *pme; + unsigned int i, j, index; + struct page *page; + unsigned long flags; + unsigned long page_size; + int err = 0; + unsigned long phy_addr; + + if (!user_space_reserve_start) + return; + spin_lock_irqsave(&page_map_entry_lock, flags); + for (index = 0; index < pin_pid_num; index++) { + pmi = &(user_space_reserve_start[index]); + pme = pmi->pme; + + for (i = 0; i < pmi->entry_num; i++) { + err = 0; + for (j = 0; j < pme->nr_pages; j++) { + phy_addr = pme->phy_addr_array[j]; + if (!phy_addr) + continue; + page = phys_to_page(phy_addr); + if (atomic_read(&page->_refcount)) { + pme->phy_addr_array[j] = 0; + page_size = pme->is_huge_page ? HPAGE_PMD_SIZE : PAGE_SIZE; + continue; + } + if (!pme->is_huge_page) { + reserve_user_normal_pages(page); + } else { + reserve_user_huge_pmd_pages(page); + } + } + pme = (struct page_map_entry *)next_pme(pme); + if (err) + err_phy_num++; + } + page_size = pme->is_huge_page ? HPAGE_PMD_SIZE : PAGE_SIZE; + } + spin_unlock(&page_map_entry_lock); +} + + +/* The whole page map entry collect process must be Sequentially. + The user_space_reserve_start points to the first page map info for + the first dump task. And the page_map_entry_start points to + the first page map entry of the first dump vma. */ +static void init_page_map_info(unsigned int *map_addr) +{ + unsigned long map_len = pin_memory_resource.end - pin_memory_resource.start; + + if (user_space_reserve_start || !max_pin_pid_num) + return; + pin_pid_num = *map_addr; + pin_pid_num_addr = map_addr; + user_space_reserve_start = + (struct kup_page_map_info *)(map_addr + 1); + page_map_entry_start = + (struct page_map_entry *)(user_space_reserve_start + max_pin_pid_num); + page_map_entry_end = (unsigned long)map_addr + map_len; + if (pin_pid_num > 0) + reserve_user_space_map_pages(); +} + +int collect_pmd_huge_pages(struct task_struct *task, + unsigned long start_addr, unsigned long end_addr, struct page_map_entry *pme) +{ + long res; + int index = 0; + unsigned long start = start_addr; + struct page *temp_page; + + while (start < end_addr) { + temp_page = NULL; + res = get_user_pages_remote(task, task->mm, start, 1, + FOLL_TOUCH|FOLL_GET, &temp_page, NULL, NULL); + if (!res) { + pr_warn("Get huge page for addr(%lx) fail.", start); + return COLLECT_PAGES_FAIL; + } + if (PageHead(temp_page)) { + start += HPAGE_PMD_SIZE; + pme->phy_addr_array[index] = page_to_phys(temp_page); + index++; + } else { + pme->nr_pages = index; + atomic_dec(&((temp_page)->_refcount)); + return COLLECT_PAGES_NEED_CONTINUE; + } + } + pme->nr_pages = index; + return COLLECT_PAGES_FINISH; +} + +int collect_normal_pages(struct task_struct *task, + unsigned long start_addr, unsigned long end_addr, struct page_map_entry *pme) +{ + int res; + unsigned long next; + unsigned long i, nr_pages; + struct page *tmp_page; + unsigned long *phy_addr_array = pme->phy_addr_array; + struct page **page_array = (struct page **)pme->phy_addr_array; + + next = (start_addr & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE; + next = (next > end_addr) ? end_addr : next; + pme->nr_pages = 0; + while (start_addr < next) { + nr_pages = (next - start_addr) / PAGE_SIZE; + res = get_user_pages_remote(task, task->mm, start_addr, 1, + FOLL_TOUCH|FOLL_GET, &tmp_page, NULL, NULL); + if (!res) { + pr_warn("Get user pages of %lx fail.\n", start_addr); + return COLLECT_PAGES_FAIL; + } + if (PageHead(tmp_page)) { + atomic_dec(&(tmp_page->_refcount)); + return COLLECT_PAGES_NEED_CONTINUE; + } + atomic_dec(&(tmp_page->_refcount)); + if (PageTail(tmp_page)) { + start_addr = next; + pme->virt_addr = start_addr; + next = (next + HPAGE_PMD_SIZE) > end_addr ? end_addr : (next + HPAGE_PMD_SIZE); + continue; + } + res = get_user_pages_remote(task, task->mm, start_addr, nr_pages, + FOLL_TOUCH|FOLL_GET, page_array, NULL, NULL); + if (!res) { + pr_warn("Get user pages of %lx fail.\n", start_addr); + return COLLECT_PAGES_FAIL; + } + for (i = 0; i < nr_pages; i++) { + phy_addr_array[i] = page_to_phys(page_array[i]); + } + pme->nr_pages += nr_pages; + page_array += nr_pages; + phy_addr_array += nr_pages; + start_addr = next; + next = (next + HPAGE_PMD_SIZE) > end_addr ? end_addr : (next + HPAGE_PMD_SIZE); + } + return COLLECT_PAGES_FINISH; +} + +/* Users make sure that the pin memory belongs to anonymous vma. */ +int pin_mem_area(struct task_struct *task, struct mm_struct *mm, + unsigned long start_addr, unsigned long end_addr) +{ + int pid, ret; + int is_huge_page = false; + unsigned int page_size; + unsigned long nr_pages, flags; + struct page_map_entry *pme; + struct page_map_info *pmi; + struct vm_area_struct *vma; + unsigned long i; + struct page *tmp_page; + + if (!page_map_entry_start + || !task || !mm + || start_addr >= end_addr) + return -EFAULT; + + pid = task->pid; + spin_lock_irqsave(&page_map_entry_lock, flags); + nr_pages = ((end_addr - start_addr) / PAGE_SIZE); + if ((unsigned long)page_map_entry_start + nr_pages * sizeof(struct page *) + >= page_map_entry_end) { + pr_warn("Page map entry use up!\n"); + ret = -EFAULT; + goto finish; + } + vma = find_extend_vma(mm, start_addr); + if (!vma) { + pr_warn("Find no match vma!\n"); + ret = -EFAULT; + goto finish; + } + if (start_addr == (start_addr & HPAGE_PMD_MASK) && + transparent_hugepage_enabled(vma)) { + page_size = HPAGE_PMD_SIZE; + is_huge_page = true; + } else { + page_size = PAGE_SIZE; + } + pme = page_map_entry_start; + pme->virt_addr = start_addr; + pme->is_huge_page = is_huge_page; + memset(pme->phy_addr_array, 0, nr_pages * sizeof(unsigned long)); + down_write(&mm->mmap_sem); + if (!is_huge_page) { + ret = collect_normal_pages(task, start_addr, end_addr, pme); + if (!pme->nr_pages) { + if (ret == COLLECT_PAGES_FINISH) { + ret = 0; + up_write(&mm->mmap_sem); + goto finish; + } + pme->is_huge_page = true; + page_size = HPAGE_PMD_SIZE; + ret = collect_pmd_huge_pages(task, pme->virt_addr, end_addr, pme); + } + } else { + ret = collect_pmd_huge_pages(task, start_addr, end_addr, pme); + if (!pme->nr_pages) { + if (ret == COLLECT_PAGES_FINISH) { + ret = 0; + up_write(&mm->mmap_sem); + goto finish; + } + pme->is_huge_page = false; + page_size = PAGE_SIZE; + ret = collect_normal_pages(task, pme->virt_addr, end_addr, pme); + } + } + up_write(&mm->mmap_sem); + if (ret == COLLECT_PAGES_FAIL) { + ret = -EFAULT; + goto finish; + } + + /* check for zero pages */ + for (i = 0; i < pme->nr_pages; i++) { + tmp_page = phys_to_page(pme->phy_addr_array[i]); + if (!pme->is_huge_page) { + if (page_to_pfn(tmp_page) == my_zero_pfn(pme->virt_addr + i * PAGE_SIZE)) + pme->phy_addr_array[i] = 0; + } else if (is_huge_zero_page(tmp_page)) + pme->phy_addr_array[i] = 0; + } + + page_map_entry_start = (struct page_map_entry *)(next_pme(pme)); + pmi = get_page_map_info(pid); + if (!pmi) + pmi = create_page_map_info(pid); + if (!pmi) { + pr_warn("Create page map info fail for pid: %d!\n", pid); + ret = -EFAULT; + goto finish; + } + if (!pmi->pme) + pmi->pme = pme; + pmi->entry_num++; + spin_unlock_irqrestore(&page_map_entry_lock, flags); + if (ret == COLLECT_PAGES_NEED_CONTINUE) { + ret = pin_mem_area(task, mm, pme->virt_addr + pme->nr_pages * page_size, end_addr); + } + return ret; +finish: + spin_unlock_irqrestore(&page_map_entry_lock, flags); + return ret; +} +EXPORT_SYMBOL_GPL(pin_mem_area); + +vm_fault_t remap_normal_pages(struct mm_struct *mm, struct vm_area_struct *vma, + struct page_map_entry *pme) +{ + int ret; + unsigned int j; + pgd_t *pgd; + p4d_t *p4d; + pmd_t *pmd; + pud_t *pud; + struct page *page; + unsigned long address; + unsigned long phy_addr; + + for (j = 0; j < pme->nr_pages; j++) { + address = pme->virt_addr + j * PAGE_SIZE; + phy_addr = pme->phy_addr_array[j]; + if (!phy_addr) + continue; + page = phys_to_page(phy_addr); + if (page->flags & (1 << PG_reserved)) + page->flags -= (1 << PG_reserved); + if (page_to_pfn(page) == my_zero_pfn(address)) { + pme->phy_addr_array[j] = 0; + continue; + } + page->mapping = NULL; + pgd = pgd_offset(mm, address); + p4d = p4d_alloc(mm, pgd, address); + if (!p4d) + return VM_FAULT_OOM; + pud = pud_alloc(mm, p4d, address); + if (!pud) + return VM_FAULT_OOM; + pmd = pmd_alloc(mm, pud, address); + if (!pmd) + return VM_FAULT_OOM; + ret = do_anon_page_remap(vma, address, pmd, page); + if (ret == VM_FAULT_OOM) + return ret; + } + return 0; +} + +vm_fault_t remap_huge_pmd_pages(struct mm_struct *mm, struct vm_area_struct *vma, + struct page_map_entry *pme) +{ + int ret; + unsigned int j; + pgd_t *pgd; + p4d_t *p4d; + pmd_t *pmd; + pud_t *pud; + struct page *page; + unsigned long address; + unsigned long phy_addr; + + for (j = 0; j < pme->nr_pages; j++) { + address = pme->virt_addr + j * HPAGE_PMD_SIZE; + phy_addr = pme->phy_addr_array[j]; + if (!phy_addr) + continue; + page = phys_to_page(phy_addr); + if (page->flags & (1 << PG_reserved)) + page->flags -= (1 << PG_reserved); + if (is_huge_zero_page(page)) { + pme->phy_addr_array[j] = 0; + continue; + } + pgd = pgd_offset(mm, address); + p4d = p4d_alloc(mm, pgd, address); + if (!p4d) + return VM_FAULT_OOM; + pud = pud_alloc(mm, p4d, address); + if (!pud) + return VM_FAULT_OOM; + pmd = pmd_alloc(mm, pud, address); + if (!pmd) + return VM_FAULT_OOM; + ret = do_anon_huge_page_remap(vma, address, pmd, page); + if (ret == VM_FAULT_OOM) + return ret; + } + return 0; +} + +vm_fault_t do_mem_remap(int pid, struct mm_struct *mm) +{ + unsigned int i = 0; + vm_fault_t ret = 0; + struct vm_area_struct *vma; + struct page_map_info *pmi; + struct page_map_entry *pme; + + pmi = get_page_map_info(pid); + if (!pmi) + return -EFAULT; + down_write(&mm->mmap_sem); + pme = pmi->pme; + vma = mm->mmap; + while ((i < pmi->entry_num) && (vma != NULL)) { + if (pme->virt_addr >= vma->vm_start && pme->virt_addr < vma->vm_end) { + i++; + if (!vma_is_anonymous(vma)) { + pme = (struct page_map_entry *)(next_pme(pme)); + continue; + } + if (!pme->is_huge_page) { + ret = remap_normal_pages(mm, vma, pme); + if (ret < 0) + goto out; + } else { + ret = remap_huge_pmd_pages(mm, vma, pme); + if (ret < 0) + goto out; + } + pme = (struct page_map_entry *)(next_pme(pme)); + } else { + vma = vma->vm_next; + } + } +out: + up_write(&mm->mmap_sem); + return ret; +} +EXPORT_SYMBOL_GPL(do_mem_remap); + +#if defined(CONFIG_ARM64) +void init_reserve_page_map(unsigned long map_addr, unsigned long map_size) +{ + void *addr; + + if (!map_addr || !map_size) + return; + addr = phys_to_virt(map_addr); + init_page_map_info((unsigned int *)addr); +} +#else +void init_reserve_page_map(unsigned long map_addr, unsigned long map_size) +{ +} +#endif + +/* Clear all pin memory record. */ +void clear_pin_memory_record(void) +{ + if (pin_pid_num_addr) { + *pin_pid_num_addr = 0; + pin_pid_num = 0; + page_map_entry_start = (struct page_map_entry *)__page_map_entry_start; + } + if (kernel_space_reserve_start && kernel_pin_space_size > 0) { + *(unsigned long *)kernel_space_reserve_start = 0; + } +} +EXPORT_SYMBOL_GPL(clear_pin_memory_record); + +vm_fault_t reserve_kernel_space_mem(unsigned long start_addr, unsigned int pages) +{ + unsigned long i; + unsigned long entry_num; + struct page_map_entry *pme, *pme_start; + + + entry_num = *(unsigned long *)kernel_space_reserve_start; + pme_start = (struct page_map_entry *)(kernel_space_reserve_start + sizeof(entry_num)); + pme = pme_start; + spin_lock(&page_map_entry_lock); + for (i = 0; i < entry_num; i++) { + if (start_addr == pme->virt_addr) { + spin_unlock(&page_map_entry_lock); + return 0; + } + pme = pme + 1; + } + if ((unsigned long)(pme_start + entry_num) >= kernel_space_reserve_end) { + spin_unlock(&page_map_entry_lock); + return VM_FAULT_OOM; + } + pme = pme_start + entry_num; + pme->virt_addr = start_addr; + pme->nr_pages = pages; + pme->is_huge_page = false; + *(unsigned long *)kernel_space_reserve_start = entry_num + 1; + spin_unlock(&page_map_entry_lock); + return 0; +} +EXPORT_SYMBOL_GPL(reserve_kernel_space_mem); + +#endif /* CONFIG_PIN_MEMORY */ -- 1.8.3.1

1 0

[PATCH] mmap: fix a compiling error on mips
by Zhengyuan Liu 31 Dec '20

31 Dec '20

hulk inclusion category: bugfix bugzilla: NA CVE: NA MAP_PA32BIT was defined in uapi/asm-generic/mman.h, that was not automatically included by mm/mmap.c when building on mips platform, and result in following compiling error: mm/mmap.c: In function ‘do_mmap’: mm/mmap.c:1450:14: error: ‘MAP_PA32BIT’ undeclared (first use in this function); did you mean ‘MAP_32BIT’? if (flags & MAP_PA32BIT) ^~~~~~~~~~~ MAP_32BIT mm/mmap.c:1450:14: note: each undeclared identifier is reported only once for each function it appears in make[1]: *** [scripts/Makefile.build:303: mm/mmap.o] Error 1 Fixes: e138253f2b82 ("svm: add support for allocing memory which is within 4G physical address in svm_mmap") Signed-off-by: Zhengyuan Liu <liuzhengyuan(a)tj.kylinos.cn> --- arch/mips/include/uapi/asm/mman.h | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/mips/include/uapi/asm/mman.h b/arch/mips/include/uapi/asm/mman.h index 3035ca499cd8..0247a3c705af 100644 --- a/arch/mips/include/uapi/asm/mman.h +++ b/arch/mips/include/uapi/asm/mman.h @@ -51,6 +51,7 @@ #define MAP_STACK 0x40000 /* give out an address that is best suited for process/thread stacks */ #define MAP_HUGETLB 0x80000 /* create a huge page mapping */ #define MAP_FIXED_NOREPLACE 0x100000 /* MAP_FIXED which doesn't unmap underlying mapping */ +#define MAP_PA32BIT 0x400000 /* * Flags for msync -- 2.17.1

4 4

[PATCH v2 OLK-5.10 00/40] Backprot lts patches from Linux-5.10.3
by Chen Jun 30 Dec '20

30 Dec '20

*** BLURB HERE *** Alexey Kardashevskiy (1): serial_core: Check for port state when tty is in error state Anant Thazhemadam (2): fs: quota: fix array-index-out-of-bounds bug by passing correct argument to vfs_cleanup_quota_inode() nl80211: validate key indexes for cfg80211_registered_device Andi Kleen (1): x86/split-lock: Avoid returning with interrupts enabled Antti Palosaari (1): media: msi2500: assign SPI bus number dynamically Arnd Bergmann (1): scsi: megaraid_sas: Check user-provided offsets Artem Labazov (1): exfat: Avoid allocating upcase table using kcalloc() Chao Yu (1): f2fs: fix to seek incorrect data offset in inline data file Dae R. Jeong (1): md: fix a warning caused by a race between concurrent md_ioctl()s Dan Carpenter (1): usb: mtu3: fix memory corruption in mtu3_debugfs_regset() Eric Biggers (6): fscrypt: remove kernel-internal constants from UAPI header fscrypt: add fscrypt_is_nokey_name() ubifs: prevent creating duplicate encrypted filenames ext4: prevent creating duplicate encrypted filenames f2fs: prevent creating duplicate encrypted filenames crypto: af_alg - avoid undefined behavior accessing salg_name Fabio Estevam (1): usb: chipidea: ci_hdrc_imx: Pass DISABLE_DEVICE_STREAMING flag to imx6ul Jack Pham (1): usb: gadget: f_fs: Re-use SS descriptors for SuperSpeedPlus Jack Qiu (1): f2fs: init dirty_secmap incorrectly Jan Kara (1): quota: Sanity-check quota file headers on load Johan Hovold (1): USB: serial: option: add interface-number sanity check to flag handling Julian Sax (1): HID: i2c-hid: add Vero K147 to descriptor override Krzysztof Kozlowski (3): ARM: dts: exynos: fix roles of USB 3.0 ports on Odroid XU ARM: dts: exynos: fix USB 3.0 VBUS control and over-current pins on Exynos5410 ARM: dts: exynos: fix USB 3.0 pins supply being turned off on Odroid XU Mao Jinlong (1): coresight: tmc-etr: Check if page is valid before dma_map_page() Nicolin Chen (1): soc/tegra: fuse: Fix index bug in get_process_id Peilin Ye (1): Bluetooth: Fix slab-out-of-bounds read in hci_le_direct_adv_report_evt() Sai Prakash Ranjan (3): coresight: tmc-etf: Fix NULL ptr dereference in tmc_enable_etf_sink_perf() coresight: etb10: Fix possible NULL ptr dereference in etb_enable_perf() coresight: etm4x: Skip setting LPOVERRIDE bit for qcom, skip-power-up Suzuki K Poulose (5): coresight: tmc-etr: Fix barrier packet insertion for perf buffer coresight: etm4x: Fix accesses to TRCVMIDCTLR1 coresight: etm4x: Fix accesses to TRCCIDCTLR1 coresight: etm4x: Fix accesses to TRCPROCSELR coresight: etm4x: Handle TRCVIPCSSCTLR accesses Thierry Reding (1): net: ipconfig: Avoid spurious blank lines in boot log Will McVicker (2): USB: gadget: f_midi: setup SuperSpeed Plus descriptors USB: gadget: f_rndis: fix bitrate for SuperSpeed and above taehyun.cho (1): USB: gadget: f_acm: add support for SuperSpeed Plus arch/arm/boot/dts/exynos5410-odroidxu.dts | 6 ++- arch/arm/boot/dts/exynos5410-pinctrl.dtsi | 28 ++++++++++ arch/arm/boot/dts/exynos5410.dtsi | 4 ++ arch/x86/kernel/traps.c | 3 +- crypto/af_alg.c | 10 ++-- drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c | 8 +++ drivers/hwtracing/coresight/coresight-etb10.c | 4 +- .../coresight/coresight-etm4x-core.c | 41 +++++++++------ drivers/hwtracing/coresight/coresight-priv.h | 2 + .../hwtracing/coresight/coresight-tmc-etf.c | 4 +- .../hwtracing/coresight/coresight-tmc-etr.c | 4 +- drivers/md/md.c | 7 ++- drivers/media/usb/msi2500/msi2500.c | 2 +- drivers/scsi/megaraid/megaraid_sas_base.c | 16 ++++-- drivers/soc/tegra/fuse/speedo-tegra210.c | 2 +- drivers/tty/serial/serial_core.c | 4 ++ drivers/usb/chipidea/ci_hdrc_imx.c | 3 +- drivers/usb/gadget/function/f_acm.c | 2 +- drivers/usb/gadget/function/f_fs.c | 5 +- drivers/usb/gadget/function/f_midi.c | 6 +++ drivers/usb/gadget/function/f_rndis.c | 4 +- drivers/usb/mtu3/mtu3_debugfs.c | 2 +- drivers/usb/serial/option.c | 23 ++++++++- fs/crypto/fscrypt_private.h | 9 ++-- fs/crypto/hooks.c | 5 +- fs/crypto/keyring.c | 2 +- fs/crypto/keysetup.c | 4 +- fs/crypto/policy.c | 5 +- fs/exfat/nls.c | 6 +-- fs/ext4/namei.c | 3 ++ fs/f2fs/f2fs.h | 2 + fs/f2fs/file.c | 11 ++-- fs/f2fs/segment.c | 2 +- fs/quota/dquot.c | 2 +- fs/quota/quota_v2.c | 19 +++++++ fs/ubifs/dir.c | 17 +++++-- include/linux/fscrypt.h | 34 +++++++++++++ include/uapi/linux/fscrypt.h | 5 +- include/uapi/linux/if_alg.h | 16 ++++++ net/bluetooth/hci_event.c | 12 ++--- net/ipv4/ipconfig.c | 14 +++-- net/wireless/core.h | 2 + net/wireless/nl80211.c | 7 +-- net/wireless/util.c | 51 ++++++++++++++++--- 44 files changed, 332 insertions(+), 86 deletions(-) -- 2.25.0

2 41

[PATCH OLK-5.10 v2 0/2] some bugfix patches for kdump
by Chen Zhou 30 Dec '20

30 Dec '20

Some bugfix patches for kdump. Hanjun Guo (1): arm64: smp: wait 50 seconds for other CPUs to stop for kexec Yingliang Yang (1): arm64: kdump: fix timer interrupts panic, could not boot capture_kernel arch/arm64/kernel/machine_kexec.c | 16 ++++++++++++++++ arch/arm64/kernel/smp.c | 4 ++-- 2 files changed, 18 insertions(+), 2 deletions(-) -- 2.20.1

2 3

[PATCH OLK-5.10 v3 0/4] kvm: debugfs: support vcpu_stat debugfs
by Jiajun Chen 29 Dec '20

29 Dec '20

From: chenjiajun <chenjiajun8(a)huawei.com> virt inclusion category: feature bugzilla: 46853 CVE: NA Export kvm vcpu info in debugfs entry /sys/kernel/debug/kvm/vcpu_stat, this entry contains kvm exits and cputime items now. And then, we can get all kvm vcpu info through this debugfs instead of one entry per kvm exits items, which will greatly facilitates the collection of performance status. Currently, only support x86 and aarch64. -- ChangeList: v3: modify kzfree in patches v2: modify architecture-dependent st_max to avoid compile err in other arches v1: kvm: debugfs: support vcpu_stat debugfs chenjiajun (4): kvm: debugfs: Export vcpu stat via debugfs kvm: debugfs: export remaining aarch64 kvm exit reasons to debugfs kvm: debugfs: aarch64 export cpu time related items to debugfs kvm: debugfs: Export x86 kvm exits to vcpu_stat arch/arm64/include/asm/kvm_host.h | 21 ++++ arch/arm64/kvm/arm.c | 23 +++++ arch/arm64/kvm/guest.c | 33 +++++++ arch/arm64/kvm/handle_exit.c | 8 ++ arch/arm64/kvm/hyp/include/hyp/switch.h | 1 + arch/arm64/kvm/mmu.c | 1 + arch/arm64/kvm/sys_regs.c | 11 +++ arch/x86/include/asm/kvm_host.h | 13 +++ arch/x86/kvm/vmx/vmx.c | 10 ++ arch/x86/kvm/x86.c | 70 +++++++++++++ include/linux/kvm_host.h | 19 ++++ virt/kvm/kvm_main.c | 125 ++++++++++++++++++++++++ 12 files changed, 335 insertions(+) -- 2.29.GIT

3 6

[PATCH OLK-5.10 00/40] Backprot lts patches from Linux-5.10.3
by Chen Jun 28 Dec '20

28 Dec '20

*** BLURB HERE *** Alexey Kardashevskiy (1): serial_core: Check for port state when tty is in error state Anant Thazhemadam (2): fs: quota: fix array-index-out-of-bounds bug by passing correct argument to vfs_cleanup_quota_inode() nl80211: validate key indexes for cfg80211_registered_device Andi Kleen (1): x86/split-lock: Avoid returning with interrupts enabled Antti Palosaari (1): media: msi2500: assign SPI bus number dynamically Arnd Bergmann (1): scsi: megaraid_sas: Check user-provided offsets Artem Labazov (1): exfat: Avoid allocating upcase table using kcalloc() Chao Yu (1): f2fs: fix to seek incorrect data offset in inline data file Dae R. Jeong (1): md: fix a warning caused by a race between concurrent md_ioctl()s Dan Carpenter (1): usb: mtu3: fix memory corruption in mtu3_debugfs_regset() Eric Biggers (6): fscrypt: remove kernel-internal constants from UAPI header fscrypt: add fscrypt_is_nokey_name() ubifs: prevent creating duplicate encrypted filenames ext4: prevent creating duplicate encrypted filenames f2fs: prevent creating duplicate encrypted filenames crypto: af_alg - avoid undefined behavior accessing salg_name Fabio Estevam (1): usb: chipidea: ci_hdrc_imx: Pass DISABLE_DEVICE_STREAMING flag to imx6ul Jack Pham (1): usb: gadget: f_fs: Re-use SS descriptors for SuperSpeedPlus Jack Qiu (1): f2fs: init dirty_secmap incorrectly Jan Kara (1): quota: Sanity-check quota file headers on load Johan Hovold (1): USB: serial: option: add interface-number sanity check to flag handling Julian Sax (1): HID: i2c-hid: add Vero K147 to descriptor override Krzysztof Kozlowski (3): ARM: dts: exynos: fix roles of USB 3.0 ports on Odroid XU ARM: dts: exynos: fix USB 3.0 VBUS control and over-current pins on Exynos5410 ARM: dts: exynos: fix USB 3.0 pins supply being turned off on Odroid XU Mao Jinlong (1): coresight: tmc-etr: Check if page is valid before dma_map_page() Nicolin Chen (1): soc/tegra: fuse: Fix index bug in get_process_id Peilin Ye (1): Bluetooth: Fix slab-out-of-bounds read in hci_le_direct_adv_report_evt() Sai Prakash Ranjan (3): coresight: tmc-etf: Fix NULL ptr dereference in tmc_enable_etf_sink_perf() coresight: etb10: Fix possible NULL ptr dereference in etb_enable_perf() coresight: etm4x: Skip setting LPOVERRIDE bit for qcom, skip-power-up Suzuki K Poulose (5): coresight: tmc-etr: Fix barrier packet insertion for perf buffer coresight: etm4x: Fix accesses to TRCVMIDCTLR1 coresight: etm4x: Fix accesses to TRCCIDCTLR1 coresight: etm4x: Fix accesses to TRCPROCSELR coresight: etm4x: Handle TRCVIPCSSCTLR accesses Thierry Reding (1): net: ipconfig: Avoid spurious blank lines in boot log Will McVicker (2): USB: gadget: f_midi: setup SuperSpeed Plus descriptors USB: gadget: f_rndis: fix bitrate for SuperSpeed and above taehyun.cho (1): USB: gadget: f_acm: add support for SuperSpeed Plus arch/arm/boot/dts/exynos5410-odroidxu.dts | 6 ++- arch/arm/boot/dts/exynos5410-pinctrl.dtsi | 28 ++++++++++ arch/arm/boot/dts/exynos5410.dtsi | 4 ++ arch/x86/kernel/traps.c | 3 +- crypto/af_alg.c | 10 ++-- drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c | 8 +++ drivers/hwtracing/coresight/coresight-etb10.c | 4 +- .../coresight/coresight-etm4x-core.c | 41 +++++++++------ drivers/hwtracing/coresight/coresight-priv.h | 2 + .../hwtracing/coresight/coresight-tmc-etf.c | 4 +- .../hwtracing/coresight/coresight-tmc-etr.c | 4 +- drivers/md/md.c | 7 ++- drivers/media/usb/msi2500/msi2500.c | 2 +- drivers/scsi/megaraid/megaraid_sas_base.c | 16 ++++-- drivers/soc/tegra/fuse/speedo-tegra210.c | 2 +- drivers/tty/serial/serial_core.c | 4 ++ drivers/usb/chipidea/ci_hdrc_imx.c | 3 +- drivers/usb/gadget/function/f_acm.c | 2 +- drivers/usb/gadget/function/f_fs.c | 5 +- drivers/usb/gadget/function/f_midi.c | 6 +++ drivers/usb/gadget/function/f_rndis.c | 4 +- drivers/usb/mtu3/mtu3_debugfs.c | 2 +- drivers/usb/serial/option.c | 23 ++++++++- fs/crypto/fscrypt_private.h | 9 ++-- fs/crypto/hooks.c | 5 +- fs/crypto/keyring.c | 2 +- fs/crypto/keysetup.c | 4 +- fs/crypto/policy.c | 5 +- fs/exfat/nls.c | 6 +-- fs/ext4/namei.c | 3 ++ fs/f2fs/f2fs.h | 2 + fs/f2fs/file.c | 11 ++-- fs/f2fs/segment.c | 2 +- fs/quota/dquot.c | 2 +- fs/quota/quota_v2.c | 19 +++++++ fs/ubifs/dir.c | 17 +++++-- include/linux/fscrypt.h | 34 +++++++++++++ include/uapi/linux/fscrypt.h | 5 +- include/uapi/linux/if_alg.h | 16 ++++++ net/bluetooth/hci_event.c | 12 ++--- net/ipv4/ipconfig.c | 14 +++-- net/wireless/core.h | 2 + net/wireless/nl80211.c | 7 +-- net/wireless/util.c | 51 ++++++++++++++++--- 44 files changed, 332 insertions(+), 86 deletions(-) -- 2.25.0

2 22

[PATCH OLK-5.10 0/2] some bugfix patches for kdump
by Chen Zhou 28 Dec '20

28 Dec '20

some bugfix patches for kdump. Hanjun Guo (1): [Huawei] arm64: smp: wait 50 seconds for other CPUs to stop for kexec Yingliang Yang (1): [Huawei] arm64: kdump: fix timer interrupts panic, could not boot capture_kernel arch/arm64/kernel/machine_kexec.c | 16 ++++++++++++++++ arch/arm64/kernel/smp.c | 4 ++-- 2 files changed, 18 insertions(+), 2 deletions(-) -- 2.20.1

1 2

[PATCH kernel-4.19 1/4] modpost: add read_text_file() and get_line() helpers
by Yang Yingliang 28 Dec '20

28 Dec '20

From: Masahiro Yamada <masahiroy(a)kernel.org> mainline inclusion from mainline-5.8 commit ac5100f54329676469688d1b5415cd8d6428c909 category: bugfix bugzilla: 46847 CVE: NA ----------------------------------------------- modpost uses grab_file() to open a file, but it is not suitable for a text file because the mmap'ed file is not terminated by null byte. Actually, I see some issues for the use of grab_file(). The new helper, read_text_file() loads the whole file content into a malloc'ed buffer, and appends a null byte. Then, get_line() reads each line. To handle text files, I intend to replace as follows: grab_file() -> read_text_file() get_new_line() -> get_line() Signed-off-by: Masahiro Yamada <masahiroy(a)kernel.org> Signed-off-by: Wang Wensheng <wangwensheng4(a)huawei.com> Reviewed-by: Jian Cheng <cj.chengjian(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- scripts/mod/modpost.c | 49 +++++++++++++++++++++++++++++++++++++++++++ scripts/mod/modpost.h | 2 ++ 2 files changed, 51 insertions(+) diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c index 91a80036c05d..99334521909b 100644 --- a/scripts/mod/modpost.c +++ b/scripts/mod/modpost.c @@ -118,6 +118,55 @@ void *do_nofail(void *ptr, const char *expr) return ptr; } +char *read_text_file(const char *filename) +{ + struct stat st; + size_t nbytes; + int fd; + char *buf; + + fd = open(filename, O_RDONLY); + if (fd < 0) { + perror(filename); + exit(1); + } + + if (fstat(fd, &st) < 0) { + perror(filename); + exit(1); + } + + buf = NOFAIL(malloc(st.st_size + 1)); + + nbytes = st.st_size; + + while (nbytes) { + ssize_t bytes_read; + + bytes_read = read(fd, buf, nbytes); + if (bytes_read < 0) { + perror(filename); + exit(1); + } + + nbytes -= bytes_read; + } + buf[st.st_size] = '\0'; + + close(fd); + + return buf; +} + +char *get_line(char **stringp) +{ + /* do not return the unwanted extra line at EOF */ + if (*stringp && **stringp == '\0') + return NULL; + + return strsep(stringp, "\n"); +} + /* A list of all modules we processed */ static struct module *modules; diff --git a/scripts/mod/modpost.h b/scripts/mod/modpost.h index 8453d6ac2f77..ff809119f82b 100644 --- a/scripts/mod/modpost.h +++ b/scripts/mod/modpost.h @@ -185,6 +185,8 @@ void maybe_frob_rcs_version(const char *modfilename, void get_src_version(const char *modname, char sum[], unsigned sumlen); /* from modpost.c */ +char *read_text_file(const char *filename); +char *get_line(char **stringp); void *grab_file(const char *filename, unsigned long *size); char* get_next_line(unsigned long *pos, void *file, unsigned long size); void release_file(void *file, unsigned long size); -- 2.25.1

1 3

[PATCH OLK-5.10 0/6] Support for printing dirty pages
by Chen Jun 28 Dec '20

28 Dec '20

*** BLURB HERE *** yu kuai (6): fs/dirty_pages: dump the number of dirty pages for each inode fs/dirty_pages: fix index out of bounds and integer overflow fs/dirty_pages: fix wrong 'buff_num' after invalid input fs/dirty_pages: Adjust position of some code to improve the code fs/dirty_pages: fix kernel panic in concurrency mode fs/dirty_pages: remove set but not used variable 'm' fs/Makefile | 1 + fs/dirty_pages.c | 466 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 467 insertions(+) create mode 100644 fs/dirty_pages.c -- 2.25.0

2 7

[PATCH OLK-5.10 00/16] Backport v5.10.2 LTS
by Chen Jun 23 Dec '20

23 Dec '20

*** BLURB HERE *** Alan Stern (1): USB: legotower: fix logical error in recent commit Alexander Sverdlin (1): serial: 8250_omap: Avoid FIFO corruption caused by MDR1 access Bui Quang Minh (1): USB: dummy-hcd: Fix uninitialized array use in init() Hans de Goede (1): xhci-pci: Allow host runtime PM as default for Intel Alpine Ridge LP Li Jun (1): xhci: Give USB2 ports time to enter U3 in bus suspend Mika Westerberg (1): xhci-pci: Allow host runtime PM as default for Intel Maple Ridge xHCI Oliver Neukum (2): USB: add RESET_RESUME quirk for Snapscan 1212 USB: UAS: introduce a quirk to set no_write_same Peilin Ye (1): ptrace: Prevent kernel-infoleak in ptrace_get_syscall_info() Steven Rostedt (VMware) (2): ktest.pl: If size of log is too big to email, email error message ktest.pl: Fix the logic for truncating the size of the log file for email Takashi Iwai (3): ALSA: usb-audio: Fix potential out-of-bounds shift ALSA: usb-audio: Fix control 'access overflow' errors from chmap ALSA: pcm: oss: Fix potential out-of-bounds shift Tejas Joglekar (1): usb: xhci: Set quirk for XHCI_SG_TRB_CACHE_SIZE_QUIRK Thomas Gleixner (1): USB: sisusbvga: Make console support depend on BROKEN .../admin-guide/kernel-parameters.txt | 1 + drivers/tty/serial/8250/8250_omap.c | 5 ----- drivers/usb/core/quirks.c | 3 +++ drivers/usb/gadget/udc/dummy_hcd.c | 2 +- drivers/usb/host/xhci-hub.c | 4 ++++ drivers/usb/host/xhci-pci.c | 6 +++++- drivers/usb/host/xhci-plat.c | 3 +++ drivers/usb/host/xhci.h | 1 + drivers/usb/misc/legousbtower.c | 2 +- drivers/usb/misc/sisusbvga/Kconfig | 2 +- drivers/usb/storage/uas.c | 3 +++ drivers/usb/storage/unusual_uas.h | 7 +++++-- drivers/usb/storage/usb.c | 3 +++ include/linux/usb_usual.h | 2 ++ include/uapi/linux/ptrace.h | 3 ++- sound/core/oss/pcm_oss.c | 6 +++++- sound/usb/format.c | 2 ++ sound/usb/stream.c | 6 +++--- tools/testing/ktest/ktest.pl | 20 +++++++++++-------- 19 files changed, 57 insertions(+), 24 deletions(-) -- 2.25.0

2 17

Re: [PATCH OLK-5.10 0/6] Support for printing dirty pages
by Xiexiuqi 23 Dec '20

23 Dec '20

For this series, Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> -----邮件原件----- 发件人: chenjun (AM) 发送时间: 2020年12月23日 10:03 收件人: kernel(a)openeuler.org 抄送: Xiangrui (Euler) <rui.xiang(a)huawei.com>; Guohanjun (Hanjun Guo) <guohanjun(a)huawei.com>; Xiexiuqi <xiexiuqi(a)huawei.com> 主题: [PATCH OLK-5.10 0/6] Support for printing dirty pages *** BLURB HERE *** yu kuai (6): fs/dirty_pages: dump the number of dirty pages for each inode fs/dirty_pages: fix index out of bounds and integer overflow fs/dirty_pages: fix wrong 'buff_num' after invalid input fs/dirty_pages: Adjust position of some code to improve the code fs/dirty_pages: fix kernel panic in concurrency mode fs/dirty_pages: remove set but not used variable 'm' fs/Makefile | 1 + fs/dirty_pages.c | 466 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 467 insertions(+) create mode 100644 fs/dirty_pages.c -- 2.25.0

1 0

[PATCH kernel-4.19 01/24] net: hns3: modify location of one print information
by Yang Yingliang 23 Dec '20

23 Dec '20

From: Junxin Chen <chenjunxin1(a)huawei.com> driver inclusion category: cleanup bugzilla: NA CVE: NA ---------------------------- This patch modifies location of one print information. Signed-off-by: Junxin Chen <chenjunxin1(a)huawei.com> Signed-off-by: Guangbin Huang <huangguangbin2(a)huawei.com> Reviewed-by: Peng Li <lipeng321(a)huawei.com> Signed-off-by: Yonglong Liu <liuyonglong(a)huawei.com> Reviewed-by: Yongxin Li <liyongxin1(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 082625de7a30..2c1c0efad80e 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -3454,10 +3454,10 @@ static void hclge_do_reset(struct hclge_dev *hdev) switch (hdev->reset_type) { case HNAE3_GLOBAL_RESET: + dev_info(&pdev->dev, "global reset requested\n"); val = hclge_read_dev(&hdev->hw, HCLGE_GLOBAL_RESET_REG); hnae3_set_bit(val, HCLGE_GLOBAL_RESET_BIT, 1); hclge_write_dev(&hdev->hw, HCLGE_GLOBAL_RESET_REG, val); - dev_info(&pdev->dev, "global reset requested\n"); break; case HNAE3_FUNC_RESET: dev_info(&pdev->dev, "PF reset requested\n"); -- 2.25.1

1 23

[PATCH kernel-4.19 1/2] Page flags: Add reserved page flags for future extension
by Yang Yingliang 23 Dec '20

23 Dec '20

From: Xiangyou Xie <xiexiangyou(a)huawei.com> euleros inclusion category: feature bugzilla: 46840 CVE: NA ------------------------------------------------- Add reserved page flags for internal extension. For the new page flags which backported from kernel upstream, place them behind the the reserved page flags. Signed-off-by: Xiangyou Xie <xiexiangyou(a)huawei.com> Reviewed-by: Zenghui Yu <yuzenghui(a)huawei.com> Reviewed-by: Jing Xiangfeng <jingxiangfeng(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/linux/page-flags.h | 8 ++++++++ include/trace/events/mmflags.h | 4 +++- 2 files changed, 11 insertions(+), 1 deletion(-) diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 17eeaf76259b..0d395970a988 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -103,6 +103,14 @@ enum pageflags { PG_idle, #endif PG_percpu_ref, + + /* Add reserved page flags for internal extension. For the new page + * flags which backported from kernel upstream, please place them + * behind the the reserved page flags. + */ + PG_reserve_pgflag_0, + PG_reserve_pgflag_1, + __NR_PAGEFLAGS, /* Filesystems */ diff --git a/include/trace/events/mmflags.h b/include/trace/events/mmflags.h index 894305dad41e..d8965ccb9c4b 100644 --- a/include/trace/events/mmflags.h +++ b/include/trace/events/mmflags.h @@ -106,7 +106,9 @@ IF_HAVE_PG_UNCACHED(PG_uncached, "uncached" ) \ IF_HAVE_PG_HWPOISON(PG_hwpoison, "hwpoison" ) \ IF_HAVE_PG_IDLE(PG_young, "young" ) \ IF_HAVE_PG_IDLE(PG_idle, "idle"), \ - {1UL << PG_percpu_ref, "percpu_ref" } + {1UL << PG_percpu_ref, "percpu_ref" }, \ + {1UL << PG_reserve_pgflag_0, "reserve_pgflag_0"}, \ + {1UL << PG_reserve_pgflag_1, "reserve_pgflag_1"} #define show_page_flags(flags) \ (flags) ? __print_flags(flags, "|", \ -- 2.25.1

1 1

[PATCH 01/10] genirq, sched/isolation: Isolate from handling managed interrupts
by Yang Yingliang 23 Dec '20

23 Dec '20

From: Ming Lei <ming.lei(a)redhat.com> mainline inclusion from mainline-v5.6-rc1 commit 11ea68f553e244851d15793a7fa33a97c46d8271 category: bugfix bugzilla: 45956 CVE: NA -------------------------------- The affinity of managed interrupts is completely handled in the kernel and cannot be changed via the /proc/irq/* interfaces from user space. As the kernel tries to spread out interrupts evenly accross CPUs on x86 to prevent vector exhaustion, it can happen that a managed interrupt whose affinity mask contains both isolated and housekeeping CPUs is routed to an isolated CPU. As a consequence IO submitted on a housekeeping CPU causes interrupts on the isolated CPU. Add a new sub-parameter 'managed_irq' for 'isolcpus' and the corresponding logic in the interrupt affinity selection code. The subparameter indicates to the interrupt affinity selection logic that it should try to avoid the above scenario. This isolation is best effort and only effective if the automatically assigned interrupt mask of a device queue contains isolated and housekeeping CPUs. If housekeeping CPUs are online then such interrupts are directed to the housekeeping CPU so that IO submitted on the housekeeping CPU cannot disturb the isolated CPU. If a queue's affinity mask contains only isolated CPUs then this parameter has no effect on the interrupt routing decision, though interrupts are only happening when tasks running on those isolated CPUs submit IO. IO submitted on housekeeping CPUs has no influence on those queues. If the affinity mask contains both housekeeping and isolated CPUs, but none of the contained housekeeping CPUs is online, then the interrupt is also routed to an isolated CPU. Interrupts are only delivered when one of the isolated CPUs in the affinity mask submits IO. If one of the contained housekeeping CPUs comes online, the CPU hotplug logic migrates the interrupt automatically back to the upcoming housekeeping CPU. Depending on the type of interrupt controller, this can require that at least one interrupt is delivered to the isolated CPU in order to complete the migration. [ tglx: Removed unused parameter, added and edited comments/documentation and rephrased the changelog so it contains more details. ] Signed-off-by: Ming Lei <ming.lei(a)redhat.com> Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Link: https://lore.kernel.org/r/20200120091625.17912-1-ming.lei@redhat.com Signed-off-by: Liu Chao <liuchao173(a)huawei.com> Reviewed-by: Jian Cheng <cj.chengjian(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- .../admin-guide/kernel-parameters.txt | 26 +++++++++++- include/linux/sched/isolation.h | 7 ++++ kernel/irq/cpuhotplug.c | 21 +++++++++- kernel/irq/manage.c | 41 ++++++++++++++++++- kernel/sched/isolation.c | 12 ++++++ 5 files changed, 102 insertions(+), 5 deletions(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 60c06c7d243d..a0e84095a1a5 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -1911,9 +1911,31 @@ <cpu number> begins at 0 and the maximum value is "number of CPUs in system - 1". - The format of <cpu-list> is described above. - + managed_irq + + Isolate from being targeted by managed interrupts + which have an interrupt mask containing isolated + CPUs. The affinity of managed interrupts is + handled by the kernel and cannot be changed via + the /proc/irq/* interfaces. + + This isolation is best effort and only effective + if the automatically assigned interrupt mask of a + device queue contains isolated and housekeeping + CPUs. If housekeeping CPUs are online then such + interrupts are directed to the housekeeping CPU + so that IO submitted on the housekeeping CPU + cannot disturb the isolated CPU. + + If a queue's affinity mask contains only isolated + CPUs then this parameter has no effect on the + interrupt routing decision, though interrupts are + only delivered when tasks running on those + isolated CPUs submit IO. IO submitted on + housekeeping CPUs has no influence on those + queues. + The format of <cpu-list> is described above. iucv= [HW,NET] diff --git a/include/linux/sched/isolation.h b/include/linux/sched/isolation.h index 4a6582c27dea..461e1f6ab3eb 100644 --- a/include/linux/sched/isolation.h +++ b/include/linux/sched/isolation.h @@ -13,12 +13,14 @@ enum hk_flags { HK_FLAG_TICK = (1 << 4), HK_FLAG_DOMAIN = (1 << 5), HK_FLAG_WQ = (1 << 6), + HK_FLAG_MANAGED_IRQ = (1 << 7), }; #ifdef CONFIG_CPU_ISOLATION DECLARE_STATIC_KEY_FALSE(housekeeping_overriden); extern int housekeeping_any_cpu(enum hk_flags flags); extern const struct cpumask *housekeeping_cpumask(enum hk_flags flags); +extern bool housekeeping_enabled(enum hk_flags flags); extern void housekeeping_affine(struct task_struct *t, enum hk_flags flags); extern bool housekeeping_test_cpu(int cpu, enum hk_flags flags); extern void __init housekeeping_init(void); @@ -35,6 +37,11 @@ static inline const struct cpumask *housekeeping_cpumask(enum hk_flags flags) return cpu_possible_mask; } +static inline bool housekeeping_enabled(enum hk_flags flags) +{ + return false; +} + static inline void housekeeping_affine(struct task_struct *t, enum hk_flags flags) { } static inline void housekeeping_init(void) { } diff --git a/kernel/irq/cpuhotplug.c b/kernel/irq/cpuhotplug.c index 6c7ca2e983a5..02236b13b359 100644 --- a/kernel/irq/cpuhotplug.c +++ b/kernel/irq/cpuhotplug.c @@ -12,6 +12,7 @@ #include <linux/interrupt.h> #include <linux/ratelimit.h> #include <linux/irq.h> +#include <linux/sched/isolation.h> #include "internals.h" @@ -171,6 +172,20 @@ void irq_migrate_all_off_this_cpu(void) } } +static bool hk_should_isolate(struct irq_data *data, unsigned int cpu) +{ + const struct cpumask *hk_mask; + + if (!housekeeping_enabled(HK_FLAG_MANAGED_IRQ)) + return false; + + hk_mask = housekeeping_cpumask(HK_FLAG_MANAGED_IRQ); + if (cpumask_subset(irq_data_get_effective_affinity_mask(data), hk_mask)) + return false; + + return cpumask_test_cpu(cpu, hk_mask); +} + static void irq_restore_affinity_of_irq(struct irq_desc *desc, unsigned int cpu) { struct irq_data *data = irq_desc_get_irq_data(desc); @@ -188,9 +203,11 @@ static void irq_restore_affinity_of_irq(struct irq_desc *desc, unsigned int cpu) /* * If the interrupt can only be directed to a single target * CPU then it is already assigned to a CPU in the affinity - * mask. No point in trying to move it around. + * mask. No point in trying to move it around unless the + * isolation mechanism requests to move it to an upcoming + * housekeeping CPU. */ - if (!irqd_is_single_target(data)) + if (!irqd_is_single_target(data) || hk_should_isolate(data, cpu)) irq_set_affinity_locked(data, affinity, false); } diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index a8c66acee82a..34eda1e772fa 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -18,6 +18,7 @@ #include <linux/sched.h> #include <linux/sched/rt.h> #include <linux/sched/task.h> +#include <linux/sched/isolation.h> #include <uapi/linux/sched/types.h> #include <linux/task_work.h> @@ -227,7 +228,45 @@ int irq_do_set_affinity(struct irq_data *data, const struct cpumask *mask, if (!chip || !chip->irq_set_affinity) return -EINVAL; - ret = chip->irq_set_affinity(data, mask, force); + /* + * If this is a managed interrupt and housekeeping is enabled on + * it check whether the requested affinity mask intersects with + * a housekeeping CPU. If so, then remove the isolated CPUs from + * the mask and just keep the housekeeping CPU(s). This prevents + * the affinity setter from routing the interrupt to an isolated + * CPU to avoid that I/O submitted from a housekeeping CPU causes + * interrupts on an isolated one. + * + * If the masks do not intersect or include online CPU(s) then + * keep the requested mask. The isolated target CPUs are only + * receiving interrupts when the I/O operation was submitted + * directly from them. + * + * If all housekeeping CPUs in the affinity mask are offline, the + * interrupt will be migrated by the CPU hotplug code once a + * housekeeping CPU which belongs to the affinity mask comes + * online. + */ + if (irqd_affinity_is_managed(data) && + housekeeping_enabled(HK_FLAG_MANAGED_IRQ)) { + const struct cpumask *hk_mask, *prog_mask; + + static DEFINE_RAW_SPINLOCK(tmp_mask_lock); + static struct cpumask tmp_mask; + + hk_mask = housekeeping_cpumask(HK_FLAG_MANAGED_IRQ); + + raw_spin_lock(&tmp_mask_lock); + cpumask_and(&tmp_mask, mask, hk_mask); + if (!cpumask_intersects(&tmp_mask, cpu_online_mask)) + prog_mask = mask; + else + prog_mask = &tmp_mask; + ret = chip->irq_set_affinity(data, prog_mask, force); + raw_spin_unlock(&tmp_mask_lock); + } else { + ret = chip->irq_set_affinity(data, mask, force); + } switch (ret) { case IRQ_SET_MASK_OK: case IRQ_SET_MASK_OK_DONE: diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c index e6802181900f..6e85049d68cf 100644 --- a/kernel/sched/isolation.c +++ b/kernel/sched/isolation.c @@ -13,6 +13,12 @@ EXPORT_SYMBOL_GPL(housekeeping_overriden); static cpumask_var_t housekeeping_mask; static unsigned int housekeeping_flags; +bool housekeeping_enabled(enum hk_flags flags) +{ + return !!(housekeeping_flags & flags); +} +EXPORT_SYMBOL_GPL(housekeeping_enabled); + int housekeeping_any_cpu(enum hk_flags flags) { if (static_branch_unlikely(&housekeeping_overriden)) @@ -140,6 +146,12 @@ static int __init housekeeping_isolcpus_setup(char *str) continue; } + if (!strncmp(str, "managed_irq,", 12)) { + str += 12; + flags |= HK_FLAG_MANAGED_IRQ; + continue; + } + pr_warn("isolcpus: Error, unknown flag\n"); return 0; } -- 2.25.1

1 9

[PATCH 01/10] genirq, sched/isolation: Isolate from handling managed interrupts
by Yang Yingliang 23 Dec '20

23 Dec '20

From: Ming Lei <ming.lei(a)redhat.com> mainline inclusion from mainline-v5.6-rc1 commit 11ea68f553e244851d15793a7fa33a97c46d8271 category: bugfix bugzilla: 45956 CVE: NA -------------------------------- The affinity of managed interrupts is completely handled in the kernel and cannot be changed via the /proc/irq/* interfaces from user space. As the kernel tries to spread out interrupts evenly accross CPUs on x86 to prevent vector exhaustion, it can happen that a managed interrupt whose affinity mask contains both isolated and housekeeping CPUs is routed to an isolated CPU. As a consequence IO submitted on a housekeeping CPU causes interrupts on the isolated CPU. Add a new sub-parameter 'managed_irq' for 'isolcpus' and the corresponding logic in the interrupt affinity selection code. The subparameter indicates to the interrupt affinity selection logic that it should try to avoid the above scenario. This isolation is best effort and only effective if the automatically assigned interrupt mask of a device queue contains isolated and housekeeping CPUs. If housekeeping CPUs are online then such interrupts are directed to the housekeeping CPU so that IO submitted on the housekeeping CPU cannot disturb the isolated CPU. If a queue's affinity mask contains only isolated CPUs then this parameter has no effect on the interrupt routing decision, though interrupts are only happening when tasks running on those isolated CPUs submit IO. IO submitted on housekeeping CPUs has no influence on those queues. If the affinity mask contains both housekeeping and isolated CPUs, but none of the contained housekeeping CPUs is online, then the interrupt is also routed to an isolated CPU. Interrupts are only delivered when one of the isolated CPUs in the affinity mask submits IO. If one of the contained housekeeping CPUs comes online, the CPU hotplug logic migrates the interrupt automatically back to the upcoming housekeeping CPU. Depending on the type of interrupt controller, this can require that at least one interrupt is delivered to the isolated CPU in order to complete the migration. [ tglx: Removed unused parameter, added and edited comments/documentation and rephrased the changelog so it contains more details. ] Signed-off-by: Ming Lei <ming.lei(a)redhat.com> Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Link: https://lore.kernel.org/r/20200120091625.17912-1-ming.lei@redhat.com Signed-off-by: Liu Chao <liuchao173(a)huawei.com> Reviewed-by: Jian Cheng <cj.chengjian(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- .../admin-guide/kernel-parameters.txt | 26 +++++++++++- include/linux/sched/isolation.h | 7 ++++ kernel/irq/cpuhotplug.c | 21 +++++++++- kernel/irq/manage.c | 41 ++++++++++++++++++- kernel/sched/isolation.c | 12 ++++++ 5 files changed, 102 insertions(+), 5 deletions(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 60c06c7d243d..a0e84095a1a5 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -1911,9 +1911,31 @@ <cpu number> begins at 0 and the maximum value is "number of CPUs in system - 1". - The format of <cpu-list> is described above. - + managed_irq + + Isolate from being targeted by managed interrupts + which have an interrupt mask containing isolated + CPUs. The affinity of managed interrupts is + handled by the kernel and cannot be changed via + the /proc/irq/* interfaces. + + This isolation is best effort and only effective + if the automatically assigned interrupt mask of a + device queue contains isolated and housekeeping + CPUs. If housekeeping CPUs are online then such + interrupts are directed to the housekeeping CPU + so that IO submitted on the housekeeping CPU + cannot disturb the isolated CPU. + + If a queue's affinity mask contains only isolated + CPUs then this parameter has no effect on the + interrupt routing decision, though interrupts are + only delivered when tasks running on those + isolated CPUs submit IO. IO submitted on + housekeeping CPUs has no influence on those + queues. + The format of <cpu-list> is described above. iucv= [HW,NET] diff --git a/include/linux/sched/isolation.h b/include/linux/sched/isolation.h index 4a6582c27dea..461e1f6ab3eb 100644 --- a/include/linux/sched/isolation.h +++ b/include/linux/sched/isolation.h @@ -13,12 +13,14 @@ enum hk_flags { HK_FLAG_TICK = (1 << 4), HK_FLAG_DOMAIN = (1 << 5), HK_FLAG_WQ = (1 << 6), + HK_FLAG_MANAGED_IRQ = (1 << 7), }; #ifdef CONFIG_CPU_ISOLATION DECLARE_STATIC_KEY_FALSE(housekeeping_overriden); extern int housekeeping_any_cpu(enum hk_flags flags); extern const struct cpumask *housekeeping_cpumask(enum hk_flags flags); +extern bool housekeeping_enabled(enum hk_flags flags); extern void housekeeping_affine(struct task_struct *t, enum hk_flags flags); extern bool housekeeping_test_cpu(int cpu, enum hk_flags flags); extern void __init housekeeping_init(void); @@ -35,6 +37,11 @@ static inline const struct cpumask *housekeeping_cpumask(enum hk_flags flags) return cpu_possible_mask; } +static inline bool housekeeping_enabled(enum hk_flags flags) +{ + return false; +} + static inline void housekeeping_affine(struct task_struct *t, enum hk_flags flags) { } static inline void housekeeping_init(void) { } diff --git a/kernel/irq/cpuhotplug.c b/kernel/irq/cpuhotplug.c index 6c7ca2e983a5..02236b13b359 100644 --- a/kernel/irq/cpuhotplug.c +++ b/kernel/irq/cpuhotplug.c @@ -12,6 +12,7 @@ #include <linux/interrupt.h> #include <linux/ratelimit.h> #include <linux/irq.h> +#include <linux/sched/isolation.h> #include "internals.h" @@ -171,6 +172,20 @@ void irq_migrate_all_off_this_cpu(void) } } +static bool hk_should_isolate(struct irq_data *data, unsigned int cpu) +{ + const struct cpumask *hk_mask; + + if (!housekeeping_enabled(HK_FLAG_MANAGED_IRQ)) + return false; + + hk_mask = housekeeping_cpumask(HK_FLAG_MANAGED_IRQ); + if (cpumask_subset(irq_data_get_effective_affinity_mask(data), hk_mask)) + return false; + + return cpumask_test_cpu(cpu, hk_mask); +} + static void irq_restore_affinity_of_irq(struct irq_desc *desc, unsigned int cpu) { struct irq_data *data = irq_desc_get_irq_data(desc); @@ -188,9 +203,11 @@ static void irq_restore_affinity_of_irq(struct irq_desc *desc, unsigned int cpu) /* * If the interrupt can only be directed to a single target * CPU then it is already assigned to a CPU in the affinity - * mask. No point in trying to move it around. + * mask. No point in trying to move it around unless the + * isolation mechanism requests to move it to an upcoming + * housekeeping CPU. */ - if (!irqd_is_single_target(data)) + if (!irqd_is_single_target(data) || hk_should_isolate(data, cpu)) irq_set_affinity_locked(data, affinity, false); } diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index a8c66acee82a..34eda1e772fa 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -18,6 +18,7 @@ #include <linux/sched.h> #include <linux/sched/rt.h> #include <linux/sched/task.h> +#include <linux/sched/isolation.h> #include <uapi/linux/sched/types.h> #include <linux/task_work.h> @@ -227,7 +228,45 @@ int irq_do_set_affinity(struct irq_data *data, const struct cpumask *mask, if (!chip || !chip->irq_set_affinity) return -EINVAL; - ret = chip->irq_set_affinity(data, mask, force); + /* + * If this is a managed interrupt and housekeeping is enabled on + * it check whether the requested affinity mask intersects with + * a housekeeping CPU. If so, then remove the isolated CPUs from + * the mask and just keep the housekeeping CPU(s). This prevents + * the affinity setter from routing the interrupt to an isolated + * CPU to avoid that I/O submitted from a housekeeping CPU causes + * interrupts on an isolated one. + * + * If the masks do not intersect or include online CPU(s) then + * keep the requested mask. The isolated target CPUs are only + * receiving interrupts when the I/O operation was submitted + * directly from them. + * + * If all housekeeping CPUs in the affinity mask are offline, the + * interrupt will be migrated by the CPU hotplug code once a + * housekeeping CPU which belongs to the affinity mask comes + * online. + */ + if (irqd_affinity_is_managed(data) && + housekeeping_enabled(HK_FLAG_MANAGED_IRQ)) { + const struct cpumask *hk_mask, *prog_mask; + + static DEFINE_RAW_SPINLOCK(tmp_mask_lock); + static struct cpumask tmp_mask; + + hk_mask = housekeeping_cpumask(HK_FLAG_MANAGED_IRQ); + + raw_spin_lock(&tmp_mask_lock); + cpumask_and(&tmp_mask, mask, hk_mask); + if (!cpumask_intersects(&tmp_mask, cpu_online_mask)) + prog_mask = mask; + else + prog_mask = &tmp_mask; + ret = chip->irq_set_affinity(data, prog_mask, force); + raw_spin_unlock(&tmp_mask_lock); + } else { + ret = chip->irq_set_affinity(data, mask, force); + } switch (ret) { case IRQ_SET_MASK_OK: case IRQ_SET_MASK_OK_DONE: diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c index e6802181900f..6e85049d68cf 100644 --- a/kernel/sched/isolation.c +++ b/kernel/sched/isolation.c @@ -13,6 +13,12 @@ EXPORT_SYMBOL_GPL(housekeeping_overriden); static cpumask_var_t housekeeping_mask; static unsigned int housekeeping_flags; +bool housekeeping_enabled(enum hk_flags flags) +{ + return !!(housekeeping_flags & flags); +} +EXPORT_SYMBOL_GPL(housekeeping_enabled); + int housekeeping_any_cpu(enum hk_flags flags) { if (static_branch_unlikely(&housekeeping_overriden)) @@ -140,6 +146,12 @@ static int __init housekeeping_isolcpus_setup(char *str) continue; } + if (!strncmp(str, "managed_irq,", 12)) { + str += 12; + flags |= HK_FLAG_MANAGED_IRQ; + continue; + } + pr_warn("isolcpus: Error, unknown flag\n"); return 0; } -- 2.25.1

1 2

[PATCH v3 OLK-5.10 00/26] KVM RISC-V Support
by MingWang Li 23 Dec '20

23 Dec '20

From: Mingwang Li <limingwang(a)huawei.com> euleros inclusion category: feature feature: initial KVM RISC-V support bugzilla: 46845 CVE: NA This series adds initial KVM RISC-V support. Currently, we are able to boot Linux on RV64/RV32 Guest with multiple VCPUs. This series can be found in riscv_kvm_v15 branch at: https//github.com/avpatel/linux.git Changes since v2 - fix build error Changes since v1 - follow guileline for submiting patches to openEuler - add the testing result in this issue Reference: https://gitee.com/openeuler/kernel/issues/I26X9V Alistair Francis (1): Revert "riscv: Use latest system call ABI" Anup Patel (19): RISC-V: Add fragmented config for debug options RISC-V: Enable CPU Hotplug in defconfigs RISC-V: Add Microchip PolarFire kconfig option RISC-V: Initial DTS for Microchip ICICLE board RISC-V: Enable drivers for Microchip PolarFire ICICLE board RISC-V: Add hypervisor extension related CSR defines RISC-V: Add initial skeletal KVM support RISC-V: KVM: Implement VCPU create, init and destroy functions RISC-V: KVM: Implement VCPU interrupts and requests handling RISC-V: KVM: Implement KVM_GET_ONE_REG/KVM_SET_ONE_REG ioctls RISC-V: KVM: Implement VCPU world-switch RISC-V: KVM: Handle MMIO exits for VCPU RISC-V: KVM: Handle WFI exits for VCPU RISC-V: KVM: Implement VMID allocator RISC-V: KVM: Implement stage2 page table programming RISC-V: KVM: Implement MMU notifiers RISC-V: KVM: Document RISC-V specific parts of KVM API RISC-V: KVM: Add MAINTAINERS entry RISC-V: Enable KVM for RV64 and RV32 Atish Patra (4): RISC-V: KVM: Add timer functionality RISC-V: KVM: FP lazy save/restore RISC-V: KVM: Implement ONE REG interface for FP registers RISC-V: KVM: Add SBI v0.1 support Mingwang Li (1): RISCV: KVM: fix bug in migration m34782 (1): Microchip Polarfire SoC Clock Driver Documentation/virt/kvm/api.rst | 193 +++- MAINTAINERS | 11 + arch/riscv/Kconfig | 2 + arch/riscv/Kconfig.socs | 7 + arch/riscv/Makefile | 2 + arch/riscv/boot/dts/Makefile | 1 + arch/riscv/boot/dts/microchip/Makefile | 2 + .../boot/dts/microchip/icicle-kit-es.dts | 307 +++++ arch/riscv/configs/defconfig | 30 +- arch/riscv/configs/extra_debug.config | 21 + arch/riscv/configs/rv32_defconfig | 26 +- arch/riscv/include/asm/csr.h | 89 ++ arch/riscv/include/asm/kvm_host.h | 278 +++++ arch/riscv/include/asm/kvm_types.h | 7 + arch/riscv/include/asm/kvm_vcpu_timer.h | 44 + arch/riscv/include/asm/pgtable-bits.h | 1 + arch/riscv/include/uapi/asm/kvm.h | 128 +++ arch/riscv/include/uapi/asm/unistd.h | 5 +- arch/riscv/kernel/asm-offsets.c | 156 +++ arch/riscv/kvm/Kconfig | 36 + arch/riscv/kvm/Makefile | 15 + arch/riscv/kvm/main.c | 118 ++ arch/riscv/kvm/mmu.c | 860 ++++++++++++++ arch/riscv/kvm/tlb.S | 74 ++ arch/riscv/kvm/vcpu.c | 1012 +++++++++++++++++ arch/riscv/kvm/vcpu_exit.c | 701 ++++++++++++ arch/riscv/kvm/vcpu_sbi.c | 173 +++ arch/riscv/kvm/vcpu_switch.S | 400 +++++++ arch/riscv/kvm/vcpu_timer.c | 225 ++++ arch/riscv/kvm/vm.c | 81 ++ arch/riscv/kvm/vmid.c | 120 ++ drivers/clk/Kconfig | 5 + drivers/clk/Makefile | 1 + drivers/clk/microchip/Makefile | 2 + drivers/clk/microchip/clk-pfsoc.c | 508 +++++++++ drivers/clocksource/timer-riscv.c | 8 + include/clocksource/timer-riscv.h | 16 + include/uapi/linux/kvm.h | 8 + 38 files changed, 5619 insertions(+), 54 deletions(-) create mode 100644 arch/riscv/boot/dts/microchip/Makefile create mode 100644 arch/riscv/boot/dts/microchip/icicle-kit-es.dts create mode 100644 arch/riscv/configs/extra_debug.config create mode 100644 arch/riscv/include/asm/kvm_host.h create mode 100644 arch/riscv/include/asm/kvm_types.h create mode 100644 arch/riscv/include/asm/kvm_vcpu_timer.h create mode 100644 arch/riscv/include/uapi/asm/kvm.h create mode 100644 arch/riscv/kvm/Kconfig create mode 100644 arch/riscv/kvm/Makefile create mode 100644 arch/riscv/kvm/main.c create mode 100644 arch/riscv/kvm/mmu.c create mode 100644 arch/riscv/kvm/tlb.S create mode 100644 arch/riscv/kvm/vcpu.c create mode 100644 arch/riscv/kvm/vcpu_exit.c create mode 100644 arch/riscv/kvm/vcpu_sbi.c create mode 100644 arch/riscv/kvm/vcpu_switch.S create mode 100644 arch/riscv/kvm/vcpu_timer.c create mode 100644 arch/riscv/kvm/vm.c create mode 100644 arch/riscv/kvm/vmid.c create mode 100644 drivers/clk/microchip/clk-pfsoc.c create mode 100644 include/clocksource/timer-riscv.h -- 2.19.1

3 28

[PATCH OLK-5.10 1/1] mm, page_alloc: avoid page_to_pfn() in move_freepages()
by Chen Jun 22 Dec '20

22 Dec '20

From: Kefeng Wang <wangkefeng.wang(a)huawei.com> hulk inclusion category: bugfix bugzilla: 46856 CVE: NA --------------------------- The start_pfn and end_pfn are already available in move_freepages_block(), there is no need to go back and forth between page and pfn in move_freepages and move_freepages_block, and pfn_valid_within() should validate pfn first before touching the page. Signed-off-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> --- mm/page_alloc.c | 28 +++++++++++++--------------- 1 file changed, 13 insertions(+), 15 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index eaa227a479e4..636afce90efa 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2361,19 +2361,21 @@ static inline struct page *__rmqueue_cma_fallback(struct zone *zone, * boundary. If alignment is required, use move_freepages_block() */ static int move_freepages(struct zone *zone, - struct page *start_page, struct page *end_page, + unsigned long start_pfn, unsigned long end_pfn, int migratetype, int *num_movable) { struct page *page; + unsigned long pfn; unsigned int order; int pages_moved = 0; - for (page = start_page; page <= end_page;) { - if (!pfn_valid_within(page_to_pfn(page))) { - page++; + for (pfn = start_pfn; pfn <= end_pfn;) { + if (!pfn_valid_within(pfn)) { + pfn++; continue; } + page = pfn_to_page(pfn); if (!PageBuddy(page)) { /* * We assume that pages that could be isolated for @@ -2383,8 +2385,7 @@ static int move_freepages(struct zone *zone, if (num_movable && (PageLRU(page) || __PageMovable(page))) (*num_movable)++; - - page++; + pfn++; continue; } @@ -2394,7 +2395,7 @@ static int move_freepages(struct zone *zone, order = buddy_order(page); move_to_free_list(page, zone, order, migratetype); - page += 1 << order; + pfn += 1 << order; pages_moved += 1 << order; } @@ -2404,25 +2405,22 @@ static int move_freepages(struct zone *zone, int move_freepages_block(struct zone *zone, struct page *page, int migratetype, int *num_movable) { - unsigned long start_pfn, end_pfn; - struct page *start_page, *end_page; + unsigned long start_pfn, end_pfn, pfn; if (num_movable) *num_movable = 0; - start_pfn = page_to_pfn(page); - start_pfn = start_pfn & ~(pageblock_nr_pages-1); - start_page = pfn_to_page(start_pfn); - end_page = start_page + pageblock_nr_pages - 1; + pfn = page_to_pfn(page); + start_pfn = pfn & ~(pageblock_nr_pages - 1); end_pfn = start_pfn + pageblock_nr_pages - 1; /* Do not cross zone boundaries */ if (!zone_spans_pfn(zone, start_pfn)) - start_page = page; + start_pfn = pfn; if (!zone_spans_pfn(zone, end_pfn)) return 0; - return move_freepages(zone, start_page, end_page, migratetype, + return move_freepages(zone, start_pfn, end_pfn, migratetype, num_movable); } -- 2.25.0

1 0

[PATCH OLK-5.10 00/24] Support ILP32 for ARM64
by Chen Jun 22 Dec '20

22 Dec '20

*** BLURB HERE *** Andrew Pinski (3): arm64: rename COMPAT to AARCH32_EL0 arm64: uapi: set __BITS_PER_LONG correctly for ILP32 and LP64 arm64:ilp32: add ARM64_ILP32 to Kconfig Dave Martin (1): arm64: signal: Make parse_user_sigframe() independent of rt_sigframe layout James Morse (1): ptrace: Add compat PTRACE_{G,S}ETSIGMASK handlers Philipp Tomsich (1): arm64:ilp32: add vdso-ilp32 and use for signal return Xiongfeng Wang (4): arm64: rename functions that reference compat term arm64: secomp: fix the secure computing mode 1 syscall check for ilp32 ilp32: avoid clearing upper 32 bits of syscall return value for ilp32 ilp32: skip ARM erratum 1418040 for ilp32 application Yury Norov (14): thread: move thread bits accessors to separated file arm64: ilp32: add documentation on the ILP32 ABI for ARM64 arm64: introduce is_a32_compat_{task,thread} for AArch32 compat arm64: ilp32: add is_ilp32_compat_{task,thread} and TIF_32BIT_AARCH64 arm64: introduce AUDIT_ARCH_AARCH64ILP32 for ilp32 arm64: introduce binfmt_elf32.c arm64: change compat_elf_hwcap and compat_elf_hwcap2 prefix to a32 arm64: ilp32: introduce binfmt_ilp32.c arm64: ilp32: share aarch32 syscall handlers arm64: ilp32: introduce syscall table for ILP32 arm64: signal: share lp64 signal structures and routines to ilp32 arm64: signal32: move ilp32 and aarch32 common code to separated file arm64: ilp32: introduce ilp32-specific sigframe and ucontext arm64: ptrace: handle ptrace_request differently for aarch32 and ilp32 Documentation/arm64/ilp32.txt | 52 +++ arch/arm64/Kconfig | 20 +- arch/arm64/Makefile | 3 + arch/arm64/include/asm/arch_timer.h | 4 +- arch/arm64/include/asm/compat.h | 19 +- arch/arm64/include/asm/elf.h | 37 +- arch/arm64/include/asm/fpsimd.h | 2 +- arch/arm64/include/asm/ftrace.h | 2 +- arch/arm64/include/asm/hwcap.h | 8 +- arch/arm64/include/asm/is_compat.h | 78 ++++ arch/arm64/include/asm/processor.h | 15 +- arch/arm64/include/asm/ptrace.h | 10 +- arch/arm64/include/asm/seccomp.h | 32 +- arch/arm64/include/asm/signal32.h | 31 +- arch/arm64/include/asm/signal32_common.h | 13 + arch/arm64/include/asm/signal_common.h | 303 +++++++++++++++ arch/arm64/include/asm/signal_ilp32.h | 23 ++ arch/arm64/include/asm/syscall.h | 17 +- arch/arm64/include/asm/thread_info.h | 4 +- arch/arm64/include/asm/unistd.h | 5 + arch/arm64/include/asm/vdso.h | 6 + arch/arm64/include/uapi/asm/bitsperlong.h | 9 +- arch/arm64/include/uapi/asm/unistd.h | 15 +- arch/arm64/kernel/Makefile | 8 +- arch/arm64/kernel/armv8_deprecated.c | 6 +- arch/arm64/kernel/asm-offsets.c | 13 +- arch/arm64/kernel/binfmt_elf32.c | 34 ++ arch/arm64/kernel/binfmt_ilp32.c | 88 +++++ arch/arm64/kernel/cpufeature.c | 30 +- arch/arm64/kernel/cpuinfo.c | 18 +- arch/arm64/kernel/debug-monitors.c | 4 +- arch/arm64/kernel/entry-common.c | 4 +- arch/arm64/kernel/entry.S | 6 +- arch/arm64/kernel/head.S | 2 +- arch/arm64/kernel/hw_breakpoint.c | 8 +- arch/arm64/kernel/perf_callchain.c | 28 +- arch/arm64/kernel/perf_regs.c | 4 +- arch/arm64/kernel/process.c | 15 +- arch/arm64/kernel/proton-pack.c | 2 +- arch/arm64/kernel/ptrace.c | 38 +- arch/arm64/kernel/signal.c | 348 ++++-------------- arch/arm64/kernel/signal32.c | 97 ++--- arch/arm64/kernel/signal32_common.c | 37 ++ arch/arm64/kernel/signal_ilp32.c | 67 ++++ arch/arm64/kernel/sys32.c | 104 +----- arch/arm64/kernel/sys32_common.c | 106 ++++++ arch/arm64/kernel/sys_compat.c | 12 +- arch/arm64/kernel/sys_ilp32.c | 82 +++++ arch/arm64/kernel/syscall.c | 39 +- arch/arm64/kernel/traps.c | 5 +- arch/arm64/kernel/vdso-ilp32/.gitignore | 2 + arch/arm64/kernel/vdso-ilp32/Makefile | 108 ++++++ arch/arm64/kernel/vdso-ilp32/vdso-ilp32.S | 22 ++ arch/arm64/kernel/vdso-ilp32/vdso-ilp32.lds.S | 88 +++++ arch/arm64/kernel/vdso.c | 50 ++- include/linux/sched.h | 1 + include/linux/thread_bits.h | 87 +++++ include/linux/thread_info.h | 75 +--- include/uapi/linux/audit.h | 1 + kernel/ptrace.c | 52 ++- 60 files changed, 1673 insertions(+), 726 deletions(-) create mode 100644 Documentation/arm64/ilp32.txt create mode 100644 arch/arm64/include/asm/is_compat.h create mode 100644 arch/arm64/include/asm/signal32_common.h create mode 100644 arch/arm64/include/asm/signal_common.h create mode 100644 arch/arm64/include/asm/signal_ilp32.h create mode 100644 arch/arm64/kernel/binfmt_elf32.c create mode 100644 arch/arm64/kernel/binfmt_ilp32.c create mode 100644 arch/arm64/kernel/signal32_common.c create mode 100644 arch/arm64/kernel/signal_ilp32.c create mode 100644 arch/arm64/kernel/sys32_common.c create mode 100644 arch/arm64/kernel/sys_ilp32.c create mode 100644 arch/arm64/kernel/vdso-ilp32/.gitignore create mode 100644 arch/arm64/kernel/vdso-ilp32/Makefile create mode 100644 arch/arm64/kernel/vdso-ilp32/vdso-ilp32.S create mode 100644 arch/arm64/kernel/vdso-ilp32/vdso-ilp32.lds.S create mode 100644 include/linux/thread_bits.h -- 2.25.0

2 25

[PATCH kernel-4.19 01/25] ascend: share_pool: make the function share_k2u_to_spg work
by Yang Yingliang 22 Dec '20

22 Dec '20

From: Ding Tianhong <dingtianhong(a)huawei.com> ascend inclusion category: feature bugzilla: NA CVE: NA ------------------------------------------------- The share_k2u_to_spg was designed to translate the kernel virtual address to user level address and share to the whole group, but tht function use the uncorrect file to mmap the memory region, it was not really work for whole group, and will consume more memory, fix this problem to use the correct hugepage file and make share to the whole group for each task. Signed-off-by: Ding Tianhong <dingtianhong(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- mm/share_pool.c | 117 +++++++++++++++++++++++++++++++++++------------- 1 file changed, 87 insertions(+), 30 deletions(-) diff --git a/mm/share_pool.c b/mm/share_pool.c index f8dfbe80968b..a83254200e33 100644 --- a/mm/share_pool.c +++ b/mm/share_pool.c @@ -267,7 +267,10 @@ static int spa_dec_usage(enum spa_type type, unsigned long size, bool is_dvpp) } static unsigned long sp_mmap(struct mm_struct *mm, struct file *file, - struct sp_area *spa, unsigned long *populate); + struct sp_area *spa, unsigned long *populate); +static void sp_munmap(struct mm_struct *mm, unsigned long addr, unsigned long size); +static unsigned long sp_remap_kva_to_vma(unsigned long kva, struct sp_area *spa, + struct mm_struct *mm); static void free_sp_group_id(unsigned int spg_id) { @@ -596,6 +599,15 @@ int sp_group_add_task(int pid, int spg_id) atomic_inc(&spa->use_count); spin_unlock(&sp_area_lock); + if (spa->type == SPA_TYPE_K2SPG && spa->kva) { + addr = sp_remap_kva_to_vma(spa->kva, spa, mm); + if (IS_ERR_VALUE(addr)) + pr_warn("share pool: task add group remap k2u failed, ret %ld\n", addr); + + spin_lock(&sp_area_lock); + continue; + } + down_write(&mm->mmap_sem); addr = sp_mmap(mm, file, spa, &populate); if (IS_ERR_VALUE(addr)) { @@ -611,9 +623,11 @@ int sp_group_add_task(int pid, int spg_id) if (populate) { ret = do_mm_populate(mm, spa->va_start, populate, 0); if (ret) { - if (printk_ratelimit()) + if (printk_ratelimit()) { pr_warn("share pool: task add group failed when mm populate " - "failed (potential no enough memory): %d\n", ret); + "failed (potential no enough memory): %d " + "spa flag is %d\n", ret, spa->type); + } sp_munmap_task_areas(mm, spa->link.next); spin_lock(&sp_area_lock); break; @@ -1480,12 +1494,16 @@ static unsigned long sp_remap_kva_to_vma(unsigned long kva, struct sp_area *spa, int hsize_log = MAP_HUGE_2MB >> MAP_HUGE_SHIFT; unsigned long addr, buf, offset; - if (spa->is_hugepage) { - file = hugetlb_file_setup(HUGETLB_ANON_FILE, spa_size(spa), VM_NORESERVE, - &user, HUGETLB_ANONHUGE_INODE, hsize_log); - if (IS_ERR(file)) { - pr_err("share pool: file setup for k2u hugepage failed %ld\n", PTR_ERR(file)); - return PTR_ERR(file); + if (spg_valid(spa->spg)) { + file = spa_file(spa); + } else { + if (spa->is_hugepage) { + file = hugetlb_file_setup(HUGETLB_ANON_FILE, spa_size(spa), VM_NORESERVE, + &user, HUGETLB_ANONHUGE_INODE, hsize_log); + if (IS_ERR(file)) { + pr_err("share pool: file setup for k2u hugepage failed %ld\n", PTR_ERR(file)); + return PTR_ERR(file); + } } } @@ -1510,7 +1528,8 @@ static unsigned long sp_remap_kva_to_vma(unsigned long kva, struct sp_area *spa, ret = remap_vmalloc_hugepage_range(vma, (void *)kva, 0); if (ret) { do_munmap(mm, ret_addr, spa_size(spa), NULL); - pr_err("share pool: remap vmalloc hugepage failed, ret %d\n", ret); + pr_err("share pool: remap vmalloc hugepage failed, " + "ret %d, kva is %lx\n", ret, kva); ret_addr = ret; goto put_mm; } @@ -1538,7 +1557,7 @@ static unsigned long sp_remap_kva_to_vma(unsigned long kva, struct sp_area *spa, up_write(&mm->mmap_sem); mmput(mm); put_file: - if (file) + if (!spa->spg && file) fput(file); return ret_addr; @@ -1615,6 +1634,35 @@ static void *sp_make_share_kva_to_spg(unsigned long kva, struct sp_area *spa, return p; } +static bool vmalloc_area_set_flag(struct sp_area *spa, unsigned long kva, unsigned long flags) +{ + struct vm_struct *area; + + area = find_vm_area((void *)kva); + if (area) { + area->flags |= flags; + spa->kva = kva; + return true; + } + + return false; +} + +static bool vmalloc_area_clr_flag(struct sp_area *spa, unsigned long kva, unsigned long flags) +{ + struct vm_struct *area; + + spa->kva = 0; + + area = find_vm_area((void *)kva); + if (area) { + area->flags &= ~flags; + return true; + } + + return false; +} + /** * Share kernel memory to a specified process or sp_group * @kva: the VA of shared kernel memory @@ -1638,7 +1686,6 @@ void *sp_make_share_k2u(unsigned long kva, unsigned long size, unsigned long size_aligned; unsigned int page_size = PAGE_SIZE; int ret; - struct vm_struct *area; if (sp_flags & ~SP_DVPP) { if (printk_ratelimit()) @@ -1679,8 +1726,13 @@ void *sp_make_share_k2u(unsigned long kva, unsigned long size, PTR_ERR(spa)); return spa; } + + if (!vmalloc_area_set_flag(spa, kva_aligned, VM_SHAREPOOL)) { + pr_err("%s: the kva %ld is not valid\n", __func__, kva_aligned); + goto out; + } + uva = sp_make_share_kva_to_task(kva_aligned, spa, pid); - mutex_unlock(&sp_mutex); } else if (spg_valid(spg)) { /* k2u to group */ if (spg_id != SPG_ID_DEFAULT && spg_id != spg->id) { @@ -1699,26 +1751,31 @@ void *sp_make_share_k2u(unsigned long kva, unsigned long size, return spa; } + if (!vmalloc_area_set_flag(spa, kva_aligned, VM_SHAREPOOL)) { + pr_err("%s: the kva %ld is not valid\n", __func__, kva_aligned); + goto out; + } + uva = sp_make_share_kva_to_spg(kva_aligned, spa, spg); - mutex_unlock(&sp_mutex); } else { mutex_unlock(&sp_mutex); pr_err("share pool: failed to make k2u\n"); return NULL; } - if (!IS_ERR(uva)) + if (!IS_ERR(uva)) { uva = uva + (kva - kva_aligned); + } else { + /* associate vma and spa */ + if (!vmalloc_area_clr_flag(spa, kva_aligned, VM_SHAREPOOL)) + pr_warn("share pool: %s: the kva %ld is not valid \n", + __func__, kva_aligned); + } +out: __sp_area_drop(spa); + mutex_unlock(&sp_mutex); - if (!IS_ERR(uva)) { - /* associate vma and spa */ - area = find_vm_area((void *)kva); - if (area) - area->flags |= VM_SHAREPOOL; - spa->kva = kva; - } sp_dump_stack(); return uva; @@ -1990,7 +2047,6 @@ static int sp_unshare_uva(unsigned long uva, unsigned long size, int pid, int sp unsigned long uva_aligned; unsigned long size_aligned; unsigned int page_size; - struct vm_struct *area; mutex_lock(&sp_mutex); /* @@ -2061,7 +2117,7 @@ static int sp_unshare_uva(unsigned long uva, unsigned long size, int pid, int sp if (printk_ratelimit()) pr_info("share pool: no need to unshare uva(to task), " "target process mm is exiting\n"); - goto out_drop_area; + goto out_clr_flag; } if (spa->mm != mm) { @@ -2095,7 +2151,7 @@ static int sp_unshare_uva(unsigned long uva, unsigned long size, int pid, int sp if (printk_ratelimit()) pr_info("share pool: no need to unshare uva(to group), " "spa doesn't belong to a sp group or group is dead\n"); - goto out_drop_area; + goto out_clr_flag; } /* alway allow kthread and dvpp channel destroy procedure */ @@ -2112,11 +2168,12 @@ static int sp_unshare_uva(unsigned long uva, unsigned long size, int pid, int sp sp_dump_stack(); -out_drop_area: +out_clr_flag: /* deassociate vma and spa */ - area = find_vm_area((void *)spa->kva); - if (area) - area->flags &= ~VM_SHAREPOOL; + if (!vmalloc_area_clr_flag(spa, spa->kva, VM_SHAREPOOL)) + pr_warn("share pool: %s: the spa->kva %ld is not valid\n", __func__, spa->kva); + +out_drop_area: __sp_area_drop(spa); out_unlock: mutex_unlock(&sp_mutex); @@ -2162,7 +2219,7 @@ static int sp_unshare_kva(unsigned long kva, unsigned long size) if (page) put_page(page); else - pr_err("share pool: vmalloc %pK to page/hugepage failed\n", + pr_warn("share pool: vmalloc %pK to page/hugepage failed\n", (void *)addr); } -- 2.25.1

1 24

[PATCH v2 OLK-5.10 00/26] KVM RISC-V Support
by MingWang Li 22 Dec '20

22 Dec '20

From: Mingwang Li <limingwang(a)huawei.com> euleros inclusion category: feature feature: initial KVM RISC-V support bugzilla: 46845 CVE: NA This series adds initial KVM RISC-V support. Currently, we are able to boot Linux on RV64/RV32 Guest with multiple VCPUs. This series can be found in riscv_kvm_v15 branch at: https//github.com/avpatel/linux.git Changes since v1 - follow guileline for submiting patches to openEuler - add the testing result in this issue Reference: https://gitee.com/openeuler/kernel/issues/I26X9V Alistair Francis (1): Revert "riscv: Use latest system call ABI" Anup Patel (19): RISC-V: Add fragmented config for debug options RISC-V: Enable CPU Hotplug in defconfigs RISC-V: Add Microchip PolarFire kconfig option RISC-V: Initial DTS for Microchip ICICLE board RISC-V: Enable drivers for Microchip PolarFire ICICLE board RISC-V: Add hypervisor extension related CSR defines RISC-V: Add initial skeletal KVM support RISC-V: KVM: Implement VCPU create, init and destroy functions RISC-V: KVM: Implement VCPU interrupts and requests handling RISC-V: KVM: Implement KVM_GET_ONE_REG/KVM_SET_ONE_REG ioctls RISC-V: KVM: Implement VCPU world-switch RISC-V: KVM: Handle MMIO exits for VCPU RISC-V: KVM: Handle WFI exits for VCPU RISC-V: KVM: Implement VMID allocator RISC-V: KVM: Implement stage2 page table programming RISC-V: KVM: Implement MMU notifiers RISC-V: KVM: Document RISC-V specific parts of KVM API RISC-V: KVM: Add MAINTAINERS entry RISC-V: Enable KVM for RV64 and RV32 Atish Patra (4): RISC-V: KVM: Add timer functionality RISC-V: KVM: FP lazy save/restore RISC-V: KVM: Implement ONE REG interface for FP registers RISC-V: KVM: Add SBI v0.1 support Mingwang Li (1): RISCV: KVM: fix bug in migration m34782 (1): Microchip Polarfire SoC Clock Driver Documentation/virt/kvm/api.rst | 193 +++- MAINTAINERS | 11 + arch/riscv/Kconfig | 2 + arch/riscv/Kconfig.socs | 7 + arch/riscv/Makefile | 2 + arch/riscv/boot/dts/Makefile | 1 + arch/riscv/boot/dts/microchip/Makefile | 2 + .../boot/dts/microchip/icicle-kit-es.dts | 307 +++++ arch/riscv/configs/defconfig | 30 +- arch/riscv/configs/extra_debug.config | 21 + arch/riscv/configs/rv32_defconfig | 26 +- arch/riscv/include/asm/csr.h | 89 ++ arch/riscv/include/asm/kvm_host.h | 278 +++++ arch/riscv/include/asm/kvm_types.h | 7 + arch/riscv/include/asm/kvm_vcpu_timer.h | 44 + arch/riscv/include/asm/pgtable-bits.h | 1 + arch/riscv/include/uapi/asm/kvm.h | 128 +++ arch/riscv/include/uapi/asm/unistd.h | 5 +- arch/riscv/kernel/asm-offsets.c | 156 +++ arch/riscv/kvm/Kconfig | 36 + arch/riscv/kvm/Makefile | 15 + arch/riscv/kvm/main.c | 118 ++ arch/riscv/kvm/mmu.c | 860 ++++++++++++++ arch/riscv/kvm/tlb.S | 74 ++ arch/riscv/kvm/vcpu.c | 1012 +++++++++++++++++ arch/riscv/kvm/vcpu_exit.c | 701 ++++++++++++ arch/riscv/kvm/vcpu_sbi.c | 173 +++ arch/riscv/kvm/vcpu_switch.S | 400 +++++++ arch/riscv/kvm/vcpu_timer.c | 225 ++++ arch/riscv/kvm/vm.c | 81 ++ arch/riscv/kvm/vmid.c | 120 ++ drivers/clk/Kconfig | 5 + drivers/clk/Makefile | 1 + drivers/clk/microchip/Makefile | 2 + drivers/clk/microchip/clk-pfsoc.c | 508 +++++++++ drivers/clocksource/timer-riscv.c | 8 + include/clocksource/timer-riscv.h | 16 + include/uapi/linux/kvm.h | 8 + 38 files changed, 5619 insertions(+), 54 deletions(-) create mode 100644 arch/riscv/boot/dts/microchip/Makefile create mode 100644 arch/riscv/boot/dts/microchip/icicle-kit-es.dts create mode 100644 arch/riscv/configs/extra_debug.config create mode 100644 arch/riscv/include/asm/kvm_host.h create mode 100644 arch/riscv/include/asm/kvm_types.h create mode 100644 arch/riscv/include/asm/kvm_vcpu_timer.h create mode 100644 arch/riscv/include/uapi/asm/kvm.h create mode 100644 arch/riscv/kvm/Kconfig create mode 100644 arch/riscv/kvm/Makefile create mode 100644 arch/riscv/kvm/main.c create mode 100644 arch/riscv/kvm/mmu.c create mode 100644 arch/riscv/kvm/tlb.S create mode 100644 arch/riscv/kvm/vcpu.c create mode 100644 arch/riscv/kvm/vcpu_exit.c create mode 100644 arch/riscv/kvm/vcpu_sbi.c create mode 100644 arch/riscv/kvm/vcpu_switch.S create mode 100644 arch/riscv/kvm/vcpu_timer.c create mode 100644 arch/riscv/kvm/vm.c create mode 100644 arch/riscv/kvm/vmid.c create mode 100644 drivers/clk/microchip/clk-pfsoc.c create mode 100644 include/clocksource/timer-riscv.h -- 2.19.1

4 30

[PATCH OLK-5.10 v3 0/4] kvm: debugfs: support vcpu_stat debugfs
by Jiajun Chen 22 Dec '20

22 Dec '20

From: chenjiajun <chenjiajun8(a)huawei.com> virt inclusion category: feature bugzilla: 46853 CVE: NA Export kvm vcpu info in debugfs entry /sys/kernel/debug/kvm/vcpu_stat, this entry contains kvm exits and cputime items now. And then, we can get all kvm vcpu info through this debugfs instead of one entry per kvm exits items, which will greatly facilitates the collection of performance status. Currently, only support x86 and aarch64. -- ChangeList: v3: modify kzfree in patches v2: modify architecture-dependent st_max to avoid compile err in other arches v1: kvm: debugfs: support vcpu_stat debugfs chenjiajun (4): kvm: debugfs: Export vcpu stat via debugfs kvm: debugfs: export remaining aarch64 kvm exit reasons to debugfs kvm: debugfs: aarch64 export cpu time related items to debugfs kvm: debugfs: Export x86 kvm exits to vcpu_stat arch/arm64/include/asm/kvm_host.h | 21 ++++ arch/arm64/kvm/arm.c | 23 +++++ arch/arm64/kvm/guest.c | 33 +++++++ arch/arm64/kvm/handle_exit.c | 8 ++ arch/arm64/kvm/hyp/include/hyp/switch.h | 1 + arch/arm64/kvm/mmu.c | 1 + arch/arm64/kvm/sys_regs.c | 11 +++ arch/x86/include/asm/kvm_host.h | 13 +++ arch/x86/kvm/vmx/vmx.c | 10 ++ arch/x86/kvm/x86.c | 70 +++++++++++++ include/linux/kvm_host.h | 19 ++++ virt/kvm/kvm_main.c | 125 ++++++++++++++++++++++++ 12 files changed, 335 insertions(+) -- 2.29.GIT

1 4

[PATCH OLK-5.10 0/4] kvm: debugfs: support vcpu_stat debugfs
by Jiajun Chen 22 Dec '20

22 Dec '20

From: chenjiajun <chenjiajun8(a)huawei.com> virt inclusion category: feature bugzilla: 46853 CVE: NA Export kvm vcpu info in debugfs entry /sys/kernel/debug/kvm/vcpu_stat, this entry contains kvm exits and cputime items now. And then, we can get all kvm vcpu info through this debugfs instead of one entry per kvm exits items, which will greatly facilitates the collection of performance status. Currently, only support x86 and aarch64. -- ChangeList: v3: modify kzfree in patches v2: modify architecture-dependent st_max to avoid compile err in other arches v1: kvm: debugfs: support vcpu_stat debugfs chenjiajun (4): kvm: debugfs: Export vcpu stat via debugfs kvm: debugfs: export remaining aarch64 kvm exit reasons to debugfs kvm: debugfs: aarch64 export cpu time related items to debugfs kvm: debugfs: Export x86 kvm exits to vcpu_stat arch/arm64/include/asm/kvm_host.h | 21 ++++ arch/arm64/kvm/arm.c | 23 +++++ arch/arm64/kvm/guest.c | 33 +++++++ arch/arm64/kvm/handle_exit.c | 8 ++ arch/arm64/kvm/hyp/include/hyp/switch.h | 1 + arch/arm64/kvm/mmu.c | 1 + arch/arm64/kvm/sys_regs.c | 11 +++ arch/x86/include/asm/kvm_host.h | 13 +++ arch/x86/kvm/vmx/vmx.c | 10 ++ arch/x86/kvm/x86.c | 70 +++++++++++++ include/linux/kvm_host.h | 19 ++++ virt/kvm/kvm_main.c | 125 ++++++++++++++++++++++++ 12 files changed, 335 insertions(+) -- 2.29.GIT

2 5

[PATCH OLK-5.10 0/3] Support MBIGEN
by Chen Jun 22 Dec '20

22 Dec '20

*** BLURB HERE *** Yang Yingliang (3): irqchip/mbigen: rename register marcros irqchip/mbigen: add support for a MBIGEN generating SPIs dt-bindings/irqchip/mbigen: add example of MBIGEN generate SPIs .../hisilicon,mbigen-v2.txt | 17 +++++++- drivers/irqchip/irq-mbigen.c | 39 +++++++++++++------ 2 files changed, 44 insertions(+), 12 deletions(-) -- 2.25.0

2 4

[PATCH OLK-5.10 0/4] kvm: debugfs: support vcpu_stat debugfs
by Jiajun Chen 22 Dec '20

22 Dec '20

From: chenjiajun <chenjiajun8(a)huawei.com> virt inclusion category: feature bugzilla: 46853 CVE: NA Export kvm vcpu info in debugfs entry /sys/kernel/debug/kvm/vcpu_stat, this entry contains kvm exits and cputime items now. And then, we can get all kvm vcpu info through this debugfs instead of one entry per kvm exits items, which will greatly facilitates the collection of performance status. Currently, only support x86 and aarch64. -- ChangeList: v2: modify architecture-dependent st_max to avoid compile err in other arches v1: kvm: debugfs: support vcpu_stat debugfs chenjiajun (4): kvm: debugfs: Export vcpu stat via debugfs kvm: debugfs: export remaining aarch64 kvm exit reasons to debugfs kvm: debugfs: aarch64 export cpu time related items to debugfs kvm: debugfs: Export x86 kvm exits to vcpu_stat arch/arm64/include/asm/kvm_host.h | 21 ++++ arch/arm64/kvm/arm.c | 23 +++++ arch/arm64/kvm/guest.c | 33 +++++++ arch/arm64/kvm/handle_exit.c | 8 ++ arch/arm64/kvm/hyp/include/hyp/switch.h | 1 + arch/arm64/kvm/mmu.c | 1 + arch/arm64/kvm/sys_regs.c | 11 +++ arch/x86/include/asm/kvm_host.h | 13 +++ arch/x86/kvm/vmx/vmx.c | 10 ++ arch/x86/kvm/x86.c | 70 +++++++++++++ include/linux/kvm_host.h | 19 ++++ virt/kvm/kvm_main.c | 124 ++++++++++++++++++++++++ 12 files changed, 334 insertions(+) -- 2.29.GIT

1 4

[PATCH OLK-5.10 1/1] PCI: Add MCFG quirks for some Hisilicon Chip host controllers
by Chen Jun 22 Dec '20

22 Dec '20

From: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> euler inclusion category: bugfix bugzilla: 46851 CVE: NA ------------------------------------------------- The PCIe controller in some Hisilicon Chip is not completely ECAM-compliant. Part of its PCIe cores do not support ECAM. Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: yangerkun <yangerkun(a)huawei.com> --- drivers/acpi/pci_mcfg.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/acpi/pci_mcfg.c b/drivers/acpi/pci_mcfg.c index 95f23acd5b80..3f310cfc9e7d 100644 --- a/drivers/acpi/pci_mcfg.c +++ b/drivers/acpi/pci_mcfg.c @@ -77,6 +77,10 @@ static struct mcfg_fixup mcfg_quirks[] = { HISI_QUAD_DOM("HIP07 ", 4, &hisi_pcie_ops), HISI_QUAD_DOM("HIP07 ", 8, &hisi_pcie_ops), HISI_QUAD_DOM("HIP07 ", 12, &hisi_pcie_ops), + HISI_QUAD_DOM("HIP12 ", 0x20, &hisi_pcie_ops), + HISI_QUAD_DOM("HIP12 ", 0x24, &hisi_pcie_ops), + HISI_QUAD_DOM("HIP12 ", 0x28, &hisi_pcie_ops), + HISI_QUAD_DOM("HIP12 ", 0x2c, &hisi_pcie_ops), #define THUNDER_PEM_RES(addr, node) \ DEFINE_RES_MEM((addr) + ((u64) (node) << 44), 0x39 * SZ_16M) -- 2.25.0

1 0

[PATCH OLK-5.10 0/2] Export some interfaces about cache
by Chen Jun 22 Dec '20

22 Dec '20

*** BLURB HERE *** Hanjun Guo (2): arm64: cache: Add flush_dcache_area() for module use arm64: cache: Export and add cache invalidation and clean ABIs for module use arch/arm64/include/asm/cacheflush.h | 51 +++++++++++++++++++++++++++++ arch/arm64/mm/flush.c | 3 ++ 2 files changed, 54 insertions(+) -- 2.25.0

1 2

[PATCH OLK-5.10 1/1] security: restrict init parameters by configuration
by Chen Jun 22 Dec '20

22 Dec '20

From: Xiangyu Lu <luxiangyu(a)huawei.com> euler inclusion category: bugfix bugzilla: 46850 CVE: NA --------------------------------- Linux kernel allow to specify a single-user mode, or specify the init process by init parameter, which could bypass the login authentication mechanisms, direct access to root identify. Close init kernel boot parameters through CONFIG_SECURITY_BOOT_INIT. Signed-off-by: Xiangyu Lu <luxiangyu(a)huawei.com> Reviewed-by: Wang Kai <morgan.wang(a)huawei.com> Signed-off-by: Weilong Chen <chenweilong(a)huawei.com> [hj: backport from hulk-3.10 for security enhancement] Signed-off-by: Hanjun Guo <hanjun.guo(a)linaro.org> Signed-off-by: gaobo <gaobo794(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: zhangyi (F) <yi.zhang(a)huawei.com> --- init/main.c | 2 ++ security/Kconfig | 6 ++++++ 2 files changed, 8 insertions(+) diff --git a/init/main.c b/init/main.c index 32b2a8affafd..1e3c3371ea5d 100644 --- a/init/main.c +++ b/init/main.c @@ -572,6 +572,7 @@ static int __init unknown_bootoption(char *param, char *val, return 0; } +#ifndef CONFIG_SECURITY_BOOT_INIT static int __init init_setup(char *str) { unsigned int i; @@ -600,6 +601,7 @@ static int __init rdinit_setup(char *str) return 1; } __setup("rdinit=", rdinit_setup); +#endif #ifndef CONFIG_SMP static const unsigned int setup_max_cpus = NR_CPUS; diff --git a/security/Kconfig b/security/Kconfig index 7561f6f99f1d..a178816f61ed 100644 --- a/security/Kconfig +++ b/security/Kconfig @@ -291,5 +291,11 @@ config LSM source "security/Kconfig.hardening" +config SECURITY_BOOT_INIT + bool "Disable init & rdinit parameters in cmdline" + default n + help + No support init and rdinit parameters in cmdline + endmenu -- 2.25.0

1 0

[PATCH OLK-5.10 0/2] config: add initial openeuler_defconfig for arm64 & x86
by Xie XiuQi 21 Dec '20

21 Dec '20

Add initial configs for arm64 & x86 platform. Use openEuler-20.03's config as base, re-generate it on 5.10 kernel. The major changes: - use 52 bit VA & PA - enable CONFIG_PSI - support more kunpeng drivers on arm64 Xie XiuQi (2): config: add initial openeuler_defconfig for arm64 config: add initial openeuler_defconfig for x86 arch/arm64/configs/openeuler_defconfig | 7068 ++++++++++++++++++++ arch/x86/configs/openeuler_defconfig | 8271 ++++++++++++++++++++++++ 2 files changed, 15339 insertions(+) create mode 100644 arch/arm64/configs/openeuler_defconfig create mode 100644 arch/x86/configs/openeuler_defconfig -- 2.20.1

1 2

[PATCH OLK-5.10 1/1] ARM: mm: non-LPAE systems HugeTLB support for hulk
by Chen Jun 21 Dec '20

21 Dec '20

From: Linxu Fang <fanglinxu(a)huawei.com> maillist inclusion category: feature bugzilla: 46792 CVE: NA Reference: https://lists.linaro.org/pipermail/linaro-kernel/2013-October/008031.html ---------------------------------------- It's a improved version of Steve Capper's RFC patch, see: https://lists.linaro.org/pipermail/linaro-kernel/2013-October/008031.html Signed-off-by: Linxu Fang <fanglinxu(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Jing Xiangfeng <jingxiangfeng(a)huawei.com> --- arch/arm/Kconfig | 2 +- arch/arm/include/asm/hugetlb-2level.h | 148 +++++++++++++++++++ arch/arm/include/asm/hugetlb.h | 6 +- arch/arm/include/asm/pgtable-2level-hwdef.h | 2 + arch/arm/include/asm/pgtable-2level.h | 154 +++++++++++++++++++- arch/arm/include/asm/pgtable-3level.h | 2 +- arch/arm/include/asm/pgtable.h | 14 ++ arch/arm/kernel/head.S | 4 + arch/arm/mm/fault.c | 26 ++-- arch/arm/mm/fsr-2level.c | 4 +- arch/arm/mm/hugetlbpage.c | 2 +- arch/arm/mm/mmu.c | 25 ++++ 12 files changed, 369 insertions(+), 20 deletions(-) create mode 100644 arch/arm/include/asm/hugetlb-2level.h diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 002e0cf025f5..3efce886d7a0 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -1545,7 +1545,7 @@ config HW_PERF_EVENTS config SYS_SUPPORTS_HUGETLBFS def_bool y - depends on ARM_LPAE + depends on ARM_LPAE || (!CPU_USE_DOMAINS && !MEMORY_FAILURE && !CPU_SW_DOMAIN_PAN) config HAVE_ARCH_TRANSPARENT_HUGEPAGE def_bool y diff --git a/arch/arm/include/asm/hugetlb-2level.h b/arch/arm/include/asm/hugetlb-2level.h new file mode 100644 index 000000000000..ce746dae2497 --- /dev/null +++ b/arch/arm/include/asm/hugetlb-2level.h @@ -0,0 +1,148 @@ +/* + * arch/arm/include/asm/hugetlb-2level.h + * + * Copyright (C) 2014 Linaro Ltd. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#ifndef _ASM_ARM_HUGETLB_2LEVEL_H +#define _ASM_ARM_HUGETLB_2LEVEL_H + +#define __HAVE_ARCH_HUGE_PTEP_GET +static inline pte_t huge_ptep_get(pte_t *ptep) +{ + pmd_t pmd = *((pmd_t *)ptep); + pte_t retval; + + if (!pmd_val(pmd)) + return __pte(0); + + retval = __pte((pteval_t) (pmd_val(pmd) & HPAGE_MASK) + | arm_hugepteprotval); + + if (pmd_exec(pmd)) + retval = pte_mkexec(retval); + else + retval = pte_mknexec(retval); + + if (pmd_young(pmd)) + retval = pte_mkyoung(retval); + else + retval = pte_mkold(retval); + + if (pmd_dirty(pmd)) + retval = pte_mkdirty(retval); + else + retval = pte_mkclean(retval); + + if (pmd_write(pmd)) + retval = pte_mkwrite(retval); + else + retval = pte_wrprotect(retval); + + if (pmd & PMD_SECT_BUFFERABLE) + retval |= PMD_SECT_BUFFERABLE; + else + retval &= ~PMD_SECT_BUFFERABLE; + + if (pmd & PMD_SECT_CACHEABLE) + retval |= PMD_SECT_CACHEABLE; + else + retval &= ~PMD_SECT_CACHEABLE; + + if (pmd & PMD_SECT_TEX(1)) + retval |= L_PTE_MT_DEV_SHARED; + else + retval &= ~L_PTE_MT_DEV_SHARED; + + if (pmd & PMD_SECT_S) + retval |= L_PTE_SHARED; + else + retval &= ~(L_PTE_SHARED); + + if (pmd_protnone(pmd)) + retval = pte_mkprotnone(retval); + else + retval = pte_rmprotnone(retval); + + return retval; +} + +#define __HAVE_ARCH_HUGE_SET_HUGE_PTE_AT +static inline void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pte) +{ + pmdval_t pmdval = (pmdval_t) pte_val(pte); + pmd_t *pmdp = (pmd_t *) ptep; + + /* take the target address bits from the pte only */ + pmdval &= HPAGE_MASK; + + /* + * now use pmd_modify to translate the permission bits from the pte + * and set the memory type information. + */ + pmdval = pmd_val(pmd_modify(__pmd(pmdval), __pgprot(pte_val(pte)))); + + __sync_icache_dcache(pte); + + set_pmd_at(mm, addr, pmdp, __pmd(pmdval)); +} + +static inline pte_t pte_mkhuge(pte_t pte) { return pte; } + +#define __HAVE_ARCH_HUGE_PTEP_CLEAR_FLUSH +static inline void huge_ptep_clear_flush(struct vm_area_struct *vma, + unsigned long addr, pte_t *ptep) +{ + pmd_t *pmdp = (pmd_t *)ptep; + + pmd_clear(pmdp); + flush_tlb_range(vma, addr, addr + HPAGE_SIZE); +} + +#define __HAVE_ARCH_HUGE_PTEP_SET_WRPROTECT +static inline void huge_ptep_set_wrprotect(struct mm_struct *mm, + unsigned long addr, pte_t *ptep) +{ + pmd_t *pmdp = (pmd_t *) ptep; + + set_pmd_at(mm, addr, pmdp, pmd_wrprotect(*pmdp)); +} + +#define __HAVE_ARCH_HUGE_PTEP_GET_AND_CLEAR +static inline pte_t huge_ptep_get_and_clear(struct mm_struct *mm, + unsigned long addr, pte_t *ptep) +{ + pmd_t *pmdp = (pmd_t *)ptep; + pte_t pte = huge_ptep_get(ptep); + + pmd_clear(pmdp); + + return pte; +} + +#define __HAVE_ARCH_HUGE_PTEP_SET_ACCESS_FLAGS +static inline int huge_ptep_set_access_flags(struct vm_area_struct *vma, + unsigned long addr, pte_t *ptep, + pte_t pte, int dirty) +{ + int changed = !pte_same(huge_ptep_get(ptep), pte); + + if (changed) { + set_huge_pte_at(vma->vm_mm, addr, ptep, pte); + flush_tlb_range(vma, addr, addr + HPAGE_SIZE); + } + + return changed; +} + +#endif /* _ASM_ARM_HUGETLB_2LEVEL_H */ diff --git a/arch/arm/include/asm/hugetlb.h b/arch/arm/include/asm/hugetlb.h index d02d6ca88e92..44fd1a5f3a32 100644 --- a/arch/arm/include/asm/hugetlb.h +++ b/arch/arm/include/asm/hugetlb.h @@ -11,7 +11,11 @@ #define _ASM_ARM_HUGETLB_H #include <asm/page.h> -#include <asm/hugetlb-3level.h> +#ifdef CONFIG_ARM_LPAE + #include <asm/hugetlb-3level.h> +#else +#include <asm/hugetlb-2level.h> +#endif #include <asm-generic/hugetlb.h> static inline void arch_clear_hugepage_flags(struct page *page) diff --git a/arch/arm/include/asm/pgtable-2level-hwdef.h b/arch/arm/include/asm/pgtable-2level-hwdef.h index 556937e1790e..b3ea5d00b5c7 100644 --- a/arch/arm/include/asm/pgtable-2level-hwdef.h +++ b/arch/arm/include/asm/pgtable-2level-hwdef.h @@ -20,7 +20,9 @@ #define PMD_PXNTABLE (_AT(pmdval_t, 1) << 2) /* v7 */ #define PMD_BIT4 (_AT(pmdval_t, 1) << 4) #define PMD_DOMAIN(x) (_AT(pmdval_t, (x)) << 5) +#ifndef PMD_DOMAIN_MASK #define PMD_DOMAIN_MASK PMD_DOMAIN(0x0f) +#endif #define PMD_PROTECTION (_AT(pmdval_t, 1) << 9) /* v5 */ /* * - section diff --git a/arch/arm/include/asm/pgtable-2level.h b/arch/arm/include/asm/pgtable-2level.h index baf7d0204eb5..000feaf88248 100644 --- a/arch/arm/include/asm/pgtable-2level.h +++ b/arch/arm/include/asm/pgtable-2level.h @@ -7,6 +7,8 @@ #ifndef _ASM_PGTABLE_2LEVEL_H #define _ASM_PGTABLE_2LEVEL_H +#include <asm/tlbflush.h> + #define __PAGETABLE_PMD_FOLDED 1 /* @@ -185,13 +187,27 @@ #define pud_clear(pudp) do { } while (0) #define set_pud(pud,pudp) do { } while (0) +static inline int pmd_large(pmd_t pmd) +{ + if ((pmd_val(pmd) & PMD_TYPE_MASK) == PMD_TYPE_FAULT) + return pmd_val(pmd); + + return ((pmd_val(pmd) & PMD_TYPE_MASK) == PMD_TYPE_SECT); +} + +static inline int pte_huge(pte_t pte) +{ + pmd_t pmd = (pmd_t)pte; + + return pmd_large(pmd); +} + static inline pmd_t *pmd_offset(pud_t *pud, unsigned long addr) { return (pmd_t *)pud; } #define pmd_offset pmd_offset -#define pmd_large(pmd) (pmd_val(pmd) & 2) #define pmd_leaf(pmd) (pmd_val(pmd) & 2) #define pmd_bad(pmd) (pmd_val(pmd) & 2) #define pmd_present(pmd) (pmd_val(pmd)) @@ -215,6 +231,142 @@ static inline pmd_t *pmd_offset(pud_t *pud, unsigned long addr) #define set_pte_ext(ptep,pte,ext) cpu_set_pte_ext(ptep,pte,ext) +/* + * now follows some of the definitions to allow huge page support, we can't put + * these in the hugetlb source files as they are also required for transparent + * hugepage support. + */ + +#define HPAGE_SHIFT PMD_SHIFT +#define HPAGE_SIZE (_AC(1, UL) << HPAGE_SHIFT) +#define HPAGE_MASK (~(HPAGE_SIZE - 1)) +#define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) + +#define HUGE_LINUX_PTE_COUNT (PAGE_OFFSET >> HPAGE_SHIFT) +#define HUGE_LINUX_PTE_SIZE (HUGE_LINUX_PTE_COUNT * sizeof(pte_t *)) +#define HUGE_LINUX_PTE_INDEX(addr) (addr >> HPAGE_SHIFT) + +/* + * We re-purpose the following domain bits in the section descriptor + */ +#ifndef PMD_DOMAIN_MASK +#define PMD_DOMAIN_MASK (_AT(pmdval_t, 0xF) << 5) +#endif +#define PMD_DSECT_PROT_NONE (_AT(pmdval_t, 1) << 5) +#define PMD_DSECT_DIRTY (_AT(pmdval_t, 1) << 6) +#define PMD_DSECT_AF (_AT(pmdval_t, 1) << 7) + +#define PMD_BIT_FUNC(fn, op) \ +static inline pmd_t pmd_##fn(pmd_t pmd) { pmd_val(pmd) op; return pmd; } + +extern pmdval_t arm_hugepmdprotval; +extern pteval_t arm_hugepteprotval; + +#define pmd_mkhuge(pmd) (__pmd((pmd_val(pmd) & ~PMD_TYPE_MASK) | PMD_TYPE_SECT)) + +PMD_BIT_FUNC(mkold, &= ~PMD_DSECT_AF); +PMD_BIT_FUNC(mkdirty, |= PMD_DSECT_DIRTY); +PMD_BIT_FUNC(mkclean, &= ~PMD_DSECT_DIRTY); +PMD_BIT_FUNC(mkyoung, |= PMD_DSECT_AF); +PMD_BIT_FUNC(mkwrite, |= PMD_SECT_AP_WRITE); +PMD_BIT_FUNC(wrprotect, &= ~PMD_SECT_AP_WRITE); +PMD_BIT_FUNC(mknotpresent, &= ~PMD_TYPE_MASK); +PMD_BIT_FUNC(mkexec, &= ~PMD_SECT_XN); +PMD_BIT_FUNC(mknexec, |= PMD_SECT_XN); +PMD_BIT_FUNC(mkprotnone, |= PMD_DSECT_PROT_NONE); +PMD_BIT_FUNC(rmprotnone, &= ~PMD_DSECT_PROT_NONE); + +#ifdef CONFIG_NUMA_BALANCING +#define pmd_protnone(pmd) (pmd_val(pmd) & PMD_DSECT_PROT_NONE) +#else +static inline int pmd_protnone(pmd_t pmd); +#endif +#define pmd_young(pmd) (pmd_val(pmd) & PMD_DSECT_AF) +#define pmd_write(pmd) (pmd_val(pmd) & PMD_SECT_AP_WRITE) +#define pmd_exec(pmd) (!(pmd_val(pmd) & PMD_SECT_XN)) +#define pmd_dirty(pmd) (pmd_val(pmd) & PMD_DSECT_DIRTY) + +#define __HAVE_ARCH_PMD_WRITE + +static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp, pmd_t pmd) +{ + /* + * we can sometimes be passed a pmd pointing to a level 2 descriptor + * from collapse_huge_page. + */ + if ((pmd_val(pmd) & PMD_TYPE_MASK) == PMD_TYPE_TABLE) { + pmdp[0] = __pmd(pmd_val(pmd)); + pmdp[1] = __pmd(pmd_val(pmd) + 256 * sizeof(pte_t)); + } else { + if (pmd_protnone(pmd)) + pmd_val(pmd) &= ~PMD_TYPE_MASK; + else + pmd_val(pmd) |= PMD_TYPE_SECT; + + pmdp[0] = __pmd(pmd_val(pmd)); + pmdp[1] = __pmd(pmd_val(pmd) + SECTION_SIZE); + } + + flush_pmd_entry(pmdp); +} + +#define pmd_modify(pmd, prot) \ +({ \ + pmd_t pmdret = __pmd((pmd_val(pmd) \ + & (PMD_MASK | PMD_DOMAIN_MASK)) \ + | arm_hugepmdprotval); \ + pgprot_t inprot = prot; \ + pte_t newprot = __pte(pgprot_val(inprot)); \ + \ + if (pte_dirty(newprot)) \ + pmdret = pmd_mkdirty(pmdret); \ + else \ + pmdret = pmd_mkclean(pmdret); \ + \ + if (pte_exec(newprot)) \ + pmdret = pmd_mkexec(pmdret); \ + else \ + pmdret = pmd_mknexec(pmdret); \ + \ + if (pte_write(newprot)) \ + pmdret = pmd_mkwrite(pmdret); \ + else \ + pmdret = pmd_wrprotect(pmdret); \ + \ + if (pte_young(newprot)) \ + pmdret = pmd_mkyoung(pmdret); \ + else \ + pmdret = pmd_mkold(pmdret); \ + \ + if (pte_protnone(newprot)) \ + pmdret = pmd_mkprotnone(pmdret); \ + else \ + pmdret = pmd_rmprotnone(pmdret); \ + \ + if (pte_val(newprot) & PMD_SECT_BUFFERABLE) \ + pmdret |= PMD_SECT_BUFFERABLE; \ + else \ + pmdret &= ~PMD_SECT_BUFFERABLE; \ + \ + if (pte_val(newprot) & PMD_SECT_CACHEABLE) \ + pmdret |= PMD_SECT_CACHEABLE; \ + else \ + pmdret &= ~PMD_SECT_CACHEABLE; \ + \ + if (pte_val(newprot) & L_PTE_MT_DEV_SHARED) \ + pmdret |= PMD_SECT_TEX(1); \ + else \ + pmdret &= ~(PMD_SECT_TEX(1)); \ + \ + if (pte_val(newprot) & L_PTE_SHARED) \ + pmdret |= PMD_SECT_S; \ + else \ + pmdret &= ~PMD_SECT_S; \ + \ + pmdret; \ +}) + /* * We don't have huge page support for short descriptors, for the moment * define empty stubs for use by pin_page_for_write. diff --git a/arch/arm/include/asm/pgtable-3level.h b/arch/arm/include/asm/pgtable-3level.h index 2b85d175e999..af53b40d1627 100644 --- a/arch/arm/include/asm/pgtable-3level.h +++ b/arch/arm/include/asm/pgtable-3level.h @@ -55,7 +55,7 @@ * Hugetlb definitions. */ #define HPAGE_SHIFT PMD_SHIFT -#define HPAGE_SIZE (_AC(1, UL) << HPAGE_SHIFT) +#define HPAGE_SIZE (_AC(1, ULL) << HPAGE_SHIFT) #define HPAGE_MASK (~(HPAGE_SIZE - 1)) #define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) diff --git a/arch/arm/include/asm/pgtable.h b/arch/arm/include/asm/pgtable.h index c02f24400369..e3a6ebcfe415 100644 --- a/arch/arm/include/asm/pgtable.h +++ b/arch/arm/include/asm/pgtable.h @@ -196,6 +196,10 @@ static inline pte_t *pmd_page_vaddr(pmd_t pmd) #define pte_young(pte) (pte_isset((pte), L_PTE_YOUNG)) #define pte_exec(pte) (pte_isclear((pte), L_PTE_XN)) +#ifdef CONFIG_NUMA_BALANCING +#define pte_protnone(pte) (pte_val(pte) & L_PTE_NONE) +#endif + #define pte_valid_user(pte) \ (pte_valid(pte) && pte_isset((pte), L_PTE_USER) && pte_young(pte)) @@ -274,6 +278,16 @@ static inline pte_t pte_mknexec(pte_t pte) return set_pte_bit(pte, __pgprot(L_PTE_XN)); } +static inline pte_t pte_mkprotnone(pte_t pte) +{ + return set_pte_bit(pte, __pgprot(L_PTE_NONE)); +} + +static inline pte_t pte_rmprotnone(pte_t pte) +{ + return clear_pte_bit(pte, __pgprot(L_PTE_NONE)); +} + static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) { const pteval_t mask = L_PTE_XN | L_PTE_RDONLY | L_PTE_USER | diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S index f8904227e7fd..dfc95659bb89 100644 --- a/arch/arm/kernel/head.S +++ b/arch/arm/kernel/head.S @@ -460,8 +460,12 @@ __enable_mmu: #endif #ifdef CONFIG_ARM_LPAE mcrr p15, 0, r4, r5, c2 @ load TTBR0 +#else +#ifdef CONFIG_SYS_SUPPORTS_HUGETLBFS + ldr r5, =0x55555555 #else mov r5, #DACR_INIT +#endif mcr p15, 0, r5, c3, c0, 0 @ load domain access register mcr p15, 0, r4, c2, c0, 0 @ load page table pointer #endif diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c index efa402025031..9985bab5ee66 100644 --- a/arch/arm/mm/fault.c +++ b/arch/arm/mm/fault.c @@ -26,6 +26,10 @@ #ifdef CONFIG_MMU +#if defined(CONFIG_SYS_SUPPORTS_HUGETLBFS) && !defined(CONFIG_ARM_LPAE) +#include <linux/hugetlb.h> +#endif + /* * This is useful to dump out the page tables associated with * 'addr' in mm 'mm'. @@ -76,6 +80,15 @@ void show_pte(const char *lvl, struct mm_struct *mm, unsigned long addr) break; if (pmd_bad(*pmd)) { +#if !defined(CONFIG_ARM_LPAE) && defined(CONFIG_HUGETLBFS) + if (pte_huge((pte_t)*pgd)) { + pte_t huge_pte = huge_ptep_get((pte_t *)pgd); + + pr_alert("[%08lx] *ppgd=%08llx", addr, + (long long)pmd_val(huge_pte)); + break; + } +#endif pr_cont("(bad)"); break; } @@ -462,19 +475,6 @@ do_translation_fault(unsigned long addr, unsigned int fsr, } #endif /* CONFIG_MMU */ -/* - * Some section permission faults need to be handled gracefully. - * They can happen due to a __{get,put}_user during an oops. - */ -#ifndef CONFIG_ARM_LPAE -static int -do_sect_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs) -{ - do_bad_area(addr, fsr, regs); - return 0; -} -#endif /* CONFIG_ARM_LPAE */ - /* * This abort handler always returns "fault". */ diff --git a/arch/arm/mm/fsr-2level.c b/arch/arm/mm/fsr-2level.c index f2be95197265..f22e33ab0dc6 100644 --- a/arch/arm/mm/fsr-2level.c +++ b/arch/arm/mm/fsr-2level.c @@ -17,7 +17,7 @@ static struct fsr_info fsr_info[] = { { do_bad, SIGBUS, 0, "external abort on non-linefetch" }, { do_bad, SIGSEGV, SEGV_ACCERR, "page domain fault" }, { do_bad, SIGBUS, 0, "external abort on translation" }, - { do_sect_fault, SIGSEGV, SEGV_ACCERR, "section permission fault" }, + { do_page_fault, SIGSEGV, SEGV_ACCERR, "section permission fault" }, { do_bad, SIGBUS, 0, "external abort on translation" }, { do_page_fault, SIGSEGV, SEGV_ACCERR, "page permission fault" }, /* @@ -57,7 +57,7 @@ static struct fsr_info ifsr_info[] = { { do_bad, SIGBUS, 0, "unknown 10" }, { do_bad, SIGSEGV, SEGV_ACCERR, "page domain fault" }, { do_bad, SIGBUS, 0, "external abort on translation" }, - { do_sect_fault, SIGSEGV, SEGV_ACCERR, "section permission fault" }, + { do_page_fault, SIGSEGV, SEGV_ACCERR, "section permission fault" }, { do_bad, SIGBUS, 0, "external abort on translation" }, { do_page_fault, SIGSEGV, SEGV_ACCERR, "page permission fault" }, { do_bad, SIGBUS, 0, "unknown 16" }, diff --git a/arch/arm/mm/hugetlbpage.c b/arch/arm/mm/hugetlbpage.c index dd7a0277c5c0..5ede7ef5e531 100644 --- a/arch/arm/mm/hugetlbpage.c +++ b/arch/arm/mm/hugetlbpage.c @@ -30,5 +30,5 @@ int pud_huge(pud_t pud) int pmd_huge(pmd_t pmd) { - return pmd_val(pmd) && !(pmd_val(pmd) & PMD_TABLE_BIT); + return pmd_large(pmd); } diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c index ab69250a86bc..911ce69f04a0 100644 --- a/arch/arm/mm/mmu.c +++ b/arch/arm/mm/mmu.c @@ -402,6 +402,20 @@ void __set_fixmap(enum fixed_addresses idx, phys_addr_t phys, pgprot_t prot) local_flush_tlb_kernel_range(vaddr, vaddr + PAGE_SIZE); } +/* + * If the system supports huge pages and we are running with short descriptors, + * then compute the pmd and linux pte prot values for a huge page. + * + * These values are used by both the HugeTLB and THP code. + */ +#if defined(CONFIG_SYS_SUPPORTS_HUGETLBFS) && !defined(CONFIG_ARM_LPAE) +pmdval_t arm_hugepmdprotval; +EXPORT_SYMBOL(arm_hugepmdprotval); + +pteval_t arm_hugepteprotval; +EXPORT_SYMBOL(arm_hugepteprotval); +#endif + /* * Adjust the PMD section entries according to the CPU in use. */ @@ -667,6 +681,17 @@ static void __init build_mem_type_table(void) if (t->prot_sect) t->prot_sect |= PMD_DOMAIN(t->domain); } + +#if defined(CONFIG_SYS_SUPPORTS_HUGETLBFS) && !defined(CONFIG_ARM_LPAE) + /* + * we assume all huge pages are user pages and that hardware access + * flag updates are disabled (i.e. SCTLR.AFE == 0b). + */ + arm_hugepteprotval = mem_types[MT_MEMORY_RWX].prot_pte | L_PTE_USER | L_PTE_VALID; + + arm_hugepmdprotval = mem_types[MT_MEMORY_RWX].prot_sect | PMD_SECT_AP_READ + | PMD_SECT_nG; +#endif } #ifdef CONFIG_ARM_DMA_MEM_BUFFERABLE -- 2.25.0

2 1

[PATCH OLK-5.10 1/1] ARM: mm: non-LPAE systems HugeTLB support for hulk
by Chen Jun 21 Dec '20

21 Dec '20

From: f00440829 <fanglinxu(a)huawei.com> euler inclusion category: feature bugzilla: 46792 CVE: NA ---------------------------------------- It's a improved version of Steve Capper's RFC patch, see: https://lists.linaro.org/pipermail/linaro-kernel/2013-October/008031.html Signed-off-by: Linxu Fang <fanglinxu(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Jing Xiangfeng <jingxiangfeng(a)huawei.com> Signed-off-by: Chen Jun <chenjun102(a)huawei.com> --- arch/arm/Kconfig | 2 +- arch/arm/include/asm/hugetlb-2level.h | 148 +++++++++++++++++++ arch/arm/include/asm/hugetlb.h | 6 +- arch/arm/include/asm/pgtable-2level-hwdef.h | 2 + arch/arm/include/asm/pgtable-2level.h | 154 +++++++++++++++++++- arch/arm/include/asm/pgtable-3level.h | 2 +- arch/arm/include/asm/pgtable.h | 14 ++ arch/arm/kernel/head.S | 4 + arch/arm/mm/fault.c | 26 ++-- arch/arm/mm/fsr-2level.c | 4 +- arch/arm/mm/hugetlbpage.c | 2 +- arch/arm/mm/mmu.c | 25 ++++ 12 files changed, 369 insertions(+), 20 deletions(-) create mode 100644 arch/arm/include/asm/hugetlb-2level.h diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 002e0cf025f5..3efce886d7a0 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -1545,7 +1545,7 @@ config HW_PERF_EVENTS config SYS_SUPPORTS_HUGETLBFS def_bool y - depends on ARM_LPAE + depends on ARM_LPAE || (!CPU_USE_DOMAINS && !MEMORY_FAILURE && !CPU_SW_DOMAIN_PAN) config HAVE_ARCH_TRANSPARENT_HUGEPAGE def_bool y diff --git a/arch/arm/include/asm/hugetlb-2level.h b/arch/arm/include/asm/hugetlb-2level.h new file mode 100644 index 000000000000..ce746dae2497 --- /dev/null +++ b/arch/arm/include/asm/hugetlb-2level.h @@ -0,0 +1,148 @@ +/* + * arch/arm/include/asm/hugetlb-2level.h + * + * Copyright (C) 2014 Linaro Ltd. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#ifndef _ASM_ARM_HUGETLB_2LEVEL_H +#define _ASM_ARM_HUGETLB_2LEVEL_H + +#define __HAVE_ARCH_HUGE_PTEP_GET +static inline pte_t huge_ptep_get(pte_t *ptep) +{ + pmd_t pmd = *((pmd_t *)ptep); + pte_t retval; + + if (!pmd_val(pmd)) + return __pte(0); + + retval = __pte((pteval_t) (pmd_val(pmd) & HPAGE_MASK) + | arm_hugepteprotval); + + if (pmd_exec(pmd)) + retval = pte_mkexec(retval); + else + retval = pte_mknexec(retval); + + if (pmd_young(pmd)) + retval = pte_mkyoung(retval); + else + retval = pte_mkold(retval); + + if (pmd_dirty(pmd)) + retval = pte_mkdirty(retval); + else + retval = pte_mkclean(retval); + + if (pmd_write(pmd)) + retval = pte_mkwrite(retval); + else + retval = pte_wrprotect(retval); + + if (pmd & PMD_SECT_BUFFERABLE) + retval |= PMD_SECT_BUFFERABLE; + else + retval &= ~PMD_SECT_BUFFERABLE; + + if (pmd & PMD_SECT_CACHEABLE) + retval |= PMD_SECT_CACHEABLE; + else + retval &= ~PMD_SECT_CACHEABLE; + + if (pmd & PMD_SECT_TEX(1)) + retval |= L_PTE_MT_DEV_SHARED; + else + retval &= ~L_PTE_MT_DEV_SHARED; + + if (pmd & PMD_SECT_S) + retval |= L_PTE_SHARED; + else + retval &= ~(L_PTE_SHARED); + + if (pmd_protnone(pmd)) + retval = pte_mkprotnone(retval); + else + retval = pte_rmprotnone(retval); + + return retval; +} + +#define __HAVE_ARCH_HUGE_SET_HUGE_PTE_AT +static inline void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, + pte_t *ptep, pte_t pte) +{ + pmdval_t pmdval = (pmdval_t) pte_val(pte); + pmd_t *pmdp = (pmd_t *) ptep; + + /* take the target address bits from the pte only */ + pmdval &= HPAGE_MASK; + + /* + * now use pmd_modify to translate the permission bits from the pte + * and set the memory type information. + */ + pmdval = pmd_val(pmd_modify(__pmd(pmdval), __pgprot(pte_val(pte)))); + + __sync_icache_dcache(pte); + + set_pmd_at(mm, addr, pmdp, __pmd(pmdval)); +} + +static inline pte_t pte_mkhuge(pte_t pte) { return pte; } + +#define __HAVE_ARCH_HUGE_PTEP_CLEAR_FLUSH +static inline void huge_ptep_clear_flush(struct vm_area_struct *vma, + unsigned long addr, pte_t *ptep) +{ + pmd_t *pmdp = (pmd_t *)ptep; + + pmd_clear(pmdp); + flush_tlb_range(vma, addr, addr + HPAGE_SIZE); +} + +#define __HAVE_ARCH_HUGE_PTEP_SET_WRPROTECT +static inline void huge_ptep_set_wrprotect(struct mm_struct *mm, + unsigned long addr, pte_t *ptep) +{ + pmd_t *pmdp = (pmd_t *) ptep; + + set_pmd_at(mm, addr, pmdp, pmd_wrprotect(*pmdp)); +} + +#define __HAVE_ARCH_HUGE_PTEP_GET_AND_CLEAR +static inline pte_t huge_ptep_get_and_clear(struct mm_struct *mm, + unsigned long addr, pte_t *ptep) +{ + pmd_t *pmdp = (pmd_t *)ptep; + pte_t pte = huge_ptep_get(ptep); + + pmd_clear(pmdp); + + return pte; +} + +#define __HAVE_ARCH_HUGE_PTEP_SET_ACCESS_FLAGS +static inline int huge_ptep_set_access_flags(struct vm_area_struct *vma, + unsigned long addr, pte_t *ptep, + pte_t pte, int dirty) +{ + int changed = !pte_same(huge_ptep_get(ptep), pte); + + if (changed) { + set_huge_pte_at(vma->vm_mm, addr, ptep, pte); + flush_tlb_range(vma, addr, addr + HPAGE_SIZE); + } + + return changed; +} + +#endif /* _ASM_ARM_HUGETLB_2LEVEL_H */ diff --git a/arch/arm/include/asm/hugetlb.h b/arch/arm/include/asm/hugetlb.h index d02d6ca88e92..44fd1a5f3a32 100644 --- a/arch/arm/include/asm/hugetlb.h +++ b/arch/arm/include/asm/hugetlb.h @@ -11,7 +11,11 @@ #define _ASM_ARM_HUGETLB_H #include <asm/page.h> -#include <asm/hugetlb-3level.h> +#ifdef CONFIG_ARM_LPAE + #include <asm/hugetlb-3level.h> +#else +#include <asm/hugetlb-2level.h> +#endif #include <asm-generic/hugetlb.h> static inline void arch_clear_hugepage_flags(struct page *page) diff --git a/arch/arm/include/asm/pgtable-2level-hwdef.h b/arch/arm/include/asm/pgtable-2level-hwdef.h index 556937e1790e..b3ea5d00b5c7 100644 --- a/arch/arm/include/asm/pgtable-2level-hwdef.h +++ b/arch/arm/include/asm/pgtable-2level-hwdef.h @@ -20,7 +20,9 @@ #define PMD_PXNTABLE (_AT(pmdval_t, 1) << 2) /* v7 */ #define PMD_BIT4 (_AT(pmdval_t, 1) << 4) #define PMD_DOMAIN(x) (_AT(pmdval_t, (x)) << 5) +#ifndef PMD_DOMAIN_MASK #define PMD_DOMAIN_MASK PMD_DOMAIN(0x0f) +#endif #define PMD_PROTECTION (_AT(pmdval_t, 1) << 9) /* v5 */ /* * - section diff --git a/arch/arm/include/asm/pgtable-2level.h b/arch/arm/include/asm/pgtable-2level.h index baf7d0204eb5..000feaf88248 100644 --- a/arch/arm/include/asm/pgtable-2level.h +++ b/arch/arm/include/asm/pgtable-2level.h @@ -7,6 +7,8 @@ #ifndef _ASM_PGTABLE_2LEVEL_H #define _ASM_PGTABLE_2LEVEL_H +#include <asm/tlbflush.h> + #define __PAGETABLE_PMD_FOLDED 1 /* @@ -185,13 +187,27 @@ #define pud_clear(pudp) do { } while (0) #define set_pud(pud,pudp) do { } while (0) +static inline int pmd_large(pmd_t pmd) +{ + if ((pmd_val(pmd) & PMD_TYPE_MASK) == PMD_TYPE_FAULT) + return pmd_val(pmd); + + return ((pmd_val(pmd) & PMD_TYPE_MASK) == PMD_TYPE_SECT); +} + +static inline int pte_huge(pte_t pte) +{ + pmd_t pmd = (pmd_t)pte; + + return pmd_large(pmd); +} + static inline pmd_t *pmd_offset(pud_t *pud, unsigned long addr) { return (pmd_t *)pud; } #define pmd_offset pmd_offset -#define pmd_large(pmd) (pmd_val(pmd) & 2) #define pmd_leaf(pmd) (pmd_val(pmd) & 2) #define pmd_bad(pmd) (pmd_val(pmd) & 2) #define pmd_present(pmd) (pmd_val(pmd)) @@ -215,6 +231,142 @@ static inline pmd_t *pmd_offset(pud_t *pud, unsigned long addr) #define set_pte_ext(ptep,pte,ext) cpu_set_pte_ext(ptep,pte,ext) +/* + * now follows some of the definitions to allow huge page support, we can't put + * these in the hugetlb source files as they are also required for transparent + * hugepage support. + */ + +#define HPAGE_SHIFT PMD_SHIFT +#define HPAGE_SIZE (_AC(1, UL) << HPAGE_SHIFT) +#define HPAGE_MASK (~(HPAGE_SIZE - 1)) +#define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) + +#define HUGE_LINUX_PTE_COUNT (PAGE_OFFSET >> HPAGE_SHIFT) +#define HUGE_LINUX_PTE_SIZE (HUGE_LINUX_PTE_COUNT * sizeof(pte_t *)) +#define HUGE_LINUX_PTE_INDEX(addr) (addr >> HPAGE_SHIFT) + +/* + * We re-purpose the following domain bits in the section descriptor + */ +#ifndef PMD_DOMAIN_MASK +#define PMD_DOMAIN_MASK (_AT(pmdval_t, 0xF) << 5) +#endif +#define PMD_DSECT_PROT_NONE (_AT(pmdval_t, 1) << 5) +#define PMD_DSECT_DIRTY (_AT(pmdval_t, 1) << 6) +#define PMD_DSECT_AF (_AT(pmdval_t, 1) << 7) + +#define PMD_BIT_FUNC(fn, op) \ +static inline pmd_t pmd_##fn(pmd_t pmd) { pmd_val(pmd) op; return pmd; } + +extern pmdval_t arm_hugepmdprotval; +extern pteval_t arm_hugepteprotval; + +#define pmd_mkhuge(pmd) (__pmd((pmd_val(pmd) & ~PMD_TYPE_MASK) | PMD_TYPE_SECT)) + +PMD_BIT_FUNC(mkold, &= ~PMD_DSECT_AF); +PMD_BIT_FUNC(mkdirty, |= PMD_DSECT_DIRTY); +PMD_BIT_FUNC(mkclean, &= ~PMD_DSECT_DIRTY); +PMD_BIT_FUNC(mkyoung, |= PMD_DSECT_AF); +PMD_BIT_FUNC(mkwrite, |= PMD_SECT_AP_WRITE); +PMD_BIT_FUNC(wrprotect, &= ~PMD_SECT_AP_WRITE); +PMD_BIT_FUNC(mknotpresent, &= ~PMD_TYPE_MASK); +PMD_BIT_FUNC(mkexec, &= ~PMD_SECT_XN); +PMD_BIT_FUNC(mknexec, |= PMD_SECT_XN); +PMD_BIT_FUNC(mkprotnone, |= PMD_DSECT_PROT_NONE); +PMD_BIT_FUNC(rmprotnone, &= ~PMD_DSECT_PROT_NONE); + +#ifdef CONFIG_NUMA_BALANCING +#define pmd_protnone(pmd) (pmd_val(pmd) & PMD_DSECT_PROT_NONE) +#else +static inline int pmd_protnone(pmd_t pmd); +#endif +#define pmd_young(pmd) (pmd_val(pmd) & PMD_DSECT_AF) +#define pmd_write(pmd) (pmd_val(pmd) & PMD_SECT_AP_WRITE) +#define pmd_exec(pmd) (!(pmd_val(pmd) & PMD_SECT_XN)) +#define pmd_dirty(pmd) (pmd_val(pmd) & PMD_DSECT_DIRTY) + +#define __HAVE_ARCH_PMD_WRITE + +static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp, pmd_t pmd) +{ + /* + * we can sometimes be passed a pmd pointing to a level 2 descriptor + * from collapse_huge_page. + */ + if ((pmd_val(pmd) & PMD_TYPE_MASK) == PMD_TYPE_TABLE) { + pmdp[0] = __pmd(pmd_val(pmd)); + pmdp[1] = __pmd(pmd_val(pmd) + 256 * sizeof(pte_t)); + } else { + if (pmd_protnone(pmd)) + pmd_val(pmd) &= ~PMD_TYPE_MASK; + else + pmd_val(pmd) |= PMD_TYPE_SECT; + + pmdp[0] = __pmd(pmd_val(pmd)); + pmdp[1] = __pmd(pmd_val(pmd) + SECTION_SIZE); + } + + flush_pmd_entry(pmdp); +} + +#define pmd_modify(pmd, prot) \ +({ \ + pmd_t pmdret = __pmd((pmd_val(pmd) \ + & (PMD_MASK | PMD_DOMAIN_MASK)) \ + | arm_hugepmdprotval); \ + pgprot_t inprot = prot; \ + pte_t newprot = __pte(pgprot_val(inprot)); \ + \ + if (pte_dirty(newprot)) \ + pmdret = pmd_mkdirty(pmdret); \ + else \ + pmdret = pmd_mkclean(pmdret); \ + \ + if (pte_exec(newprot)) \ + pmdret = pmd_mkexec(pmdret); \ + else \ + pmdret = pmd_mknexec(pmdret); \ + \ + if (pte_write(newprot)) \ + pmdret = pmd_mkwrite(pmdret); \ + else \ + pmdret = pmd_wrprotect(pmdret); \ + \ + if (pte_young(newprot)) \ + pmdret = pmd_mkyoung(pmdret); \ + else \ + pmdret = pmd_mkold(pmdret); \ + \ + if (pte_protnone(newprot)) \ + pmdret = pmd_mkprotnone(pmdret); \ + else \ + pmdret = pmd_rmprotnone(pmdret); \ + \ + if (pte_val(newprot) & PMD_SECT_BUFFERABLE) \ + pmdret |= PMD_SECT_BUFFERABLE; \ + else \ + pmdret &= ~PMD_SECT_BUFFERABLE; \ + \ + if (pte_val(newprot) & PMD_SECT_CACHEABLE) \ + pmdret |= PMD_SECT_CACHEABLE; \ + else \ + pmdret &= ~PMD_SECT_CACHEABLE; \ + \ + if (pte_val(newprot) & L_PTE_MT_DEV_SHARED) \ + pmdret |= PMD_SECT_TEX(1); \ + else \ + pmdret &= ~(PMD_SECT_TEX(1)); \ + \ + if (pte_val(newprot) & L_PTE_SHARED) \ + pmdret |= PMD_SECT_S; \ + else \ + pmdret &= ~PMD_SECT_S; \ + \ + pmdret; \ +}) + /* * We don't have huge page support for short descriptors, for the moment * define empty stubs for use by pin_page_for_write. diff --git a/arch/arm/include/asm/pgtable-3level.h b/arch/arm/include/asm/pgtable-3level.h index 2b85d175e999..af53b40d1627 100644 --- a/arch/arm/include/asm/pgtable-3level.h +++ b/arch/arm/include/asm/pgtable-3level.h @@ -55,7 +55,7 @@ * Hugetlb definitions. */ #define HPAGE_SHIFT PMD_SHIFT -#define HPAGE_SIZE (_AC(1, UL) << HPAGE_SHIFT) +#define HPAGE_SIZE (_AC(1, ULL) << HPAGE_SHIFT) #define HPAGE_MASK (~(HPAGE_SIZE - 1)) #define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) diff --git a/arch/arm/include/asm/pgtable.h b/arch/arm/include/asm/pgtable.h index c02f24400369..e3a6ebcfe415 100644 --- a/arch/arm/include/asm/pgtable.h +++ b/arch/arm/include/asm/pgtable.h @@ -196,6 +196,10 @@ static inline pte_t *pmd_page_vaddr(pmd_t pmd) #define pte_young(pte) (pte_isset((pte), L_PTE_YOUNG)) #define pte_exec(pte) (pte_isclear((pte), L_PTE_XN)) +#ifdef CONFIG_NUMA_BALANCING +#define pte_protnone(pte) (pte_val(pte) & L_PTE_NONE) +#endif + #define pte_valid_user(pte) \ (pte_valid(pte) && pte_isset((pte), L_PTE_USER) && pte_young(pte)) @@ -274,6 +278,16 @@ static inline pte_t pte_mknexec(pte_t pte) return set_pte_bit(pte, __pgprot(L_PTE_XN)); } +static inline pte_t pte_mkprotnone(pte_t pte) +{ + return set_pte_bit(pte, __pgprot(L_PTE_NONE)); +} + +static inline pte_t pte_rmprotnone(pte_t pte) +{ + return clear_pte_bit(pte, __pgprot(L_PTE_NONE)); +} + static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) { const pteval_t mask = L_PTE_XN | L_PTE_RDONLY | L_PTE_USER | diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S index f8904227e7fd..dfc95659bb89 100644 --- a/arch/arm/kernel/head.S +++ b/arch/arm/kernel/head.S @@ -460,8 +460,12 @@ __enable_mmu: #endif #ifdef CONFIG_ARM_LPAE mcrr p15, 0, r4, r5, c2 @ load TTBR0 +#else +#ifdef CONFIG_SYS_SUPPORTS_HUGETLBFS + ldr r5, =0x55555555 #else mov r5, #DACR_INIT +#endif mcr p15, 0, r5, c3, c0, 0 @ load domain access register mcr p15, 0, r4, c2, c0, 0 @ load page table pointer #endif diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c index efa402025031..9985bab5ee66 100644 --- a/arch/arm/mm/fault.c +++ b/arch/arm/mm/fault.c @@ -26,6 +26,10 @@ #ifdef CONFIG_MMU +#if defined(CONFIG_SYS_SUPPORTS_HUGETLBFS) && !defined(CONFIG_ARM_LPAE) +#include <linux/hugetlb.h> +#endif + /* * This is useful to dump out the page tables associated with * 'addr' in mm 'mm'. @@ -76,6 +80,15 @@ void show_pte(const char *lvl, struct mm_struct *mm, unsigned long addr) break; if (pmd_bad(*pmd)) { +#if !defined(CONFIG_ARM_LPAE) && defined(CONFIG_HUGETLBFS) + if (pte_huge((pte_t)*pgd)) { + pte_t huge_pte = huge_ptep_get((pte_t *)pgd); + + pr_alert("[%08lx] *ppgd=%08llx", addr, + (long long)pmd_val(huge_pte)); + break; + } +#endif pr_cont("(bad)"); break; } @@ -462,19 +475,6 @@ do_translation_fault(unsigned long addr, unsigned int fsr, } #endif /* CONFIG_MMU */ -/* - * Some section permission faults need to be handled gracefully. - * They can happen due to a __{get,put}_user during an oops. - */ -#ifndef CONFIG_ARM_LPAE -static int -do_sect_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs) -{ - do_bad_area(addr, fsr, regs); - return 0; -} -#endif /* CONFIG_ARM_LPAE */ - /* * This abort handler always returns "fault". */ diff --git a/arch/arm/mm/fsr-2level.c b/arch/arm/mm/fsr-2level.c index f2be95197265..f22e33ab0dc6 100644 --- a/arch/arm/mm/fsr-2level.c +++ b/arch/arm/mm/fsr-2level.c @@ -17,7 +17,7 @@ static struct fsr_info fsr_info[] = { { do_bad, SIGBUS, 0, "external abort on non-linefetch" }, { do_bad, SIGSEGV, SEGV_ACCERR, "page domain fault" }, { do_bad, SIGBUS, 0, "external abort on translation" }, - { do_sect_fault, SIGSEGV, SEGV_ACCERR, "section permission fault" }, + { do_page_fault, SIGSEGV, SEGV_ACCERR, "section permission fault" }, { do_bad, SIGBUS, 0, "external abort on translation" }, { do_page_fault, SIGSEGV, SEGV_ACCERR, "page permission fault" }, /* @@ -57,7 +57,7 @@ static struct fsr_info ifsr_info[] = { { do_bad, SIGBUS, 0, "unknown 10" }, { do_bad, SIGSEGV, SEGV_ACCERR, "page domain fault" }, { do_bad, SIGBUS, 0, "external abort on translation" }, - { do_sect_fault, SIGSEGV, SEGV_ACCERR, "section permission fault" }, + { do_page_fault, SIGSEGV, SEGV_ACCERR, "section permission fault" }, { do_bad, SIGBUS, 0, "external abort on translation" }, { do_page_fault, SIGSEGV, SEGV_ACCERR, "page permission fault" }, { do_bad, SIGBUS, 0, "unknown 16" }, diff --git a/arch/arm/mm/hugetlbpage.c b/arch/arm/mm/hugetlbpage.c index dd7a0277c5c0..5ede7ef5e531 100644 --- a/arch/arm/mm/hugetlbpage.c +++ b/arch/arm/mm/hugetlbpage.c @@ -30,5 +30,5 @@ int pud_huge(pud_t pud) int pmd_huge(pmd_t pmd) { - return pmd_val(pmd) && !(pmd_val(pmd) & PMD_TABLE_BIT); + return pmd_large(pmd); } diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c index ab69250a86bc..911ce69f04a0 100644 --- a/arch/arm/mm/mmu.c +++ b/arch/arm/mm/mmu.c @@ -402,6 +402,20 @@ void __set_fixmap(enum fixed_addresses idx, phys_addr_t phys, pgprot_t prot) local_flush_tlb_kernel_range(vaddr, vaddr + PAGE_SIZE); } +/* + * If the system supports huge pages and we are running with short descriptors, + * then compute the pmd and linux pte prot values for a huge page. + * + * These values are used by both the HugeTLB and THP code. + */ +#if defined(CONFIG_SYS_SUPPORTS_HUGETLBFS) && !defined(CONFIG_ARM_LPAE) +pmdval_t arm_hugepmdprotval; +EXPORT_SYMBOL(arm_hugepmdprotval); + +pteval_t arm_hugepteprotval; +EXPORT_SYMBOL(arm_hugepteprotval); +#endif + /* * Adjust the PMD section entries according to the CPU in use. */ @@ -667,6 +681,17 @@ static void __init build_mem_type_table(void) if (t->prot_sect) t->prot_sect |= PMD_DOMAIN(t->domain); } + +#if defined(CONFIG_SYS_SUPPORTS_HUGETLBFS) && !defined(CONFIG_ARM_LPAE) + /* + * we assume all huge pages are user pages and that hardware access + * flag updates are disabled (i.e. SCTLR.AFE == 0b). + */ + arm_hugepteprotval = mem_types[MT_MEMORY_RWX].prot_pte | L_PTE_USER | L_PTE_VALID; + + arm_hugepmdprotval = mem_types[MT_MEMORY_RWX].prot_sect | PMD_SECT_AP_READ + | PMD_SECT_nG; +#endif } #ifdef CONFIG_ARM_DMA_MEM_BUFFERABLE -- 2.25.0

1 0

[PATCH v3 OLK-5.10 0/2] Backport 5.10.1 LTS
by Chen Jun 21 Dec '20

21 Dec '20

v2->v3: remove the patch "Linux-5.10.1.patch" v1->v2: modifty the commit message: linux-5.10.1 -> stable-5.10.1 *** BLURB HERE *** Greg Kroah-Hartman (2): Revert "md: change mddev 'chunk_sectors' from int to unsigned" Revert "dm raid: fix discard limits for raid1 and raid10" drivers/md/dm-raid.c | 12 +++++------- drivers/md/md.h | 4 ++-- 2 files changed, 7 insertions(+), 9 deletions(-) -- 2.25.0

1 2

[PATCH openEuler-1.0-LTS] defconfig: update the defconfigs to support NVDIMM
by Xie XiuQi 21 Dec '20

21 Dec '20

From: Hanjun Guo <guohanjun(a)huawei.com> hulk inclusion category: feature feature: ACPI NVDIMM Firmware Interface Table (NFIT) bugzilla: 46783 CVE: NA Reference: https://gitee.com/open_euler/dashboard/issues?id=I26KWQ Backport this patch from kernel-4.19 to openEuler 20.03 LTS SP1. 564e83bbdcae defconfig: update the defconfigs to support NVDIMM --------------------------- openEuler wants NVDIMM support, update the defconfig. Signed-off-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Signed-off-by: Xie XiuQi <xiexiuqi(a)huawei.com> --- arch/arm64/configs/hulk_defconfig | 2 +- arch/arm64/configs/openeuler_defconfig | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/configs/hulk_defconfig b/arch/arm64/configs/hulk_defconfig index 20c7c527dc33..090ff6c93730 100644 --- a/arch/arm64/configs/hulk_defconfig +++ b/arch/arm64/configs/hulk_defconfig @@ -635,7 +635,7 @@ CONFIG_ACPI_HED=y # CONFIG_ACPI_CUSTOM_METHOD is not set # CONFIG_ACPI_BGRT is not set CONFIG_ACPI_REDUCED_HARDWARE_ONLY=y -# CONFIG_ACPI_NFIT is not set +CONFIG_ACPI_NFIT=m CONFIG_HAVE_ACPI_APEI=y CONFIG_ACPI_APEI=y CONFIG_ACPI_APEI_GHES=y diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index c68f4c9ee65c..5561751ad6eb 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -627,7 +627,7 @@ CONFIG_ACPI_HED=y # CONFIG_ACPI_CUSTOM_METHOD is not set # CONFIG_ACPI_BGRT is not set CONFIG_ACPI_REDUCED_HARDWARE_ONLY=y -# CONFIG_ACPI_NFIT is not set +CONFIG_ACPI_NFIT=m CONFIG_HAVE_ACPI_APEI=y CONFIG_ACPI_APEI=y CONFIG_ACPI_APEI_GHES=y -- 2.20.1

1 0

TCM(Nationz Tech) on SPI driver for openEuler
by Li ZhiGang 21 Dec '20

21 Dec '20

hi openEuler kernel team: 我是麒麟软件共性支撑部李志刚, 公司安排我将 TCM on SPI 的驱动移植到您的社区源码; 代码我已经核外编译验证过, 下一步请你们指导我将其放入你们社区(请告诉我最好放入哪个目录). 不好意思我没有社区开发经验, 请您告诉我如何开始好吧? 譬如是不需要我注册个账户, 最好有个简介性的文档. TCM 是国民技术公司的类米国 TPM 芯片/产品, 用于安全和可信计算等. 麒麟软件共性支撑部李志刚 2020-12-18

2 2

[PATCH OLK-5.10 00/26] KVM RISC-V Support
by MingWang Li 19 Dec '20

19 Dec '20

From: Mingwang Li <limingwang(a)huawei.com> euleros inclusion category: feature bugzilla: NA CVE: NA This series adds initial KVM RISC-V support. Currently, we are able to boot Linux on RV64/RV32 Guest with multiple VCPUs. This series can be found in riscv_kvm_v15 branch at: https//github.com/avpatel/linux.git Link: https://gitee.com/openeuler/kernel/issues/I26X9V Alistair Francis (1): Revert "riscv: Use latest system call ABI" Anup Patel (19): RISC-V: Add fragmented config for debug options RISC-V: Enable CPU Hotplug in defconfigs RISC-V: Add Microchip PolarFire kconfig option RISC-V: Initial DTS for Microchip ICICLE board RISC-V: Enable drivers for Microchip PolarFire ICICLE board RISC-V: Add hypervisor extension related CSR defines RISC-V: Add initial skeletal KVM support RISC-V: KVM: Implement VCPU create, init and destroy functions RISC-V: KVM: Implement VCPU interrupts and requests handling RISC-V: KVM: Implement KVM_GET_ONE_REG/KVM_SET_ONE_REG ioctls RISC-V: KVM: Implement VCPU world-switch RISC-V: KVM: Handle MMIO exits for VCPU RISC-V: KVM: Handle WFI exits for VCPU RISC-V: KVM: Implement VMID allocator RISC-V: KVM: Implement stage2 page table programming RISC-V: KVM: Implement MMU notifiers RISC-V: KVM: Document RISC-V specific parts of KVM API RISC-V: KVM: Add MAINTAINERS entry RISC-V: Enable KVM for RV64 and RV32 Atish Patra (4): RISC-V: KVM: Add timer functionality RISC-V: KVM: FP lazy save/restore RISC-V: KVM: Implement ONE REG interface for FP registers RISC-V: KVM: Add SBI v0.1 support Mingwang Li (1): RISCV: KVM: fix bug in migration m34782 (1): Microchip Polarfire SoC Clock Driver Documentation/virt/kvm/api.rst | 193 +++- MAINTAINERS | 11 + arch/riscv/Kconfig | 2 + arch/riscv/Kconfig.socs | 7 + arch/riscv/Makefile | 2 + arch/riscv/boot/dts/Makefile | 1 + arch/riscv/boot/dts/microchip/Makefile | 2 + .../boot/dts/microchip/icicle-kit-es.dts | 307 +++++ arch/riscv/configs/defconfig | 30 +- arch/riscv/configs/extra_debug.config | 21 + arch/riscv/configs/rv32_defconfig | 26 +- arch/riscv/include/asm/csr.h | 89 ++ arch/riscv/include/asm/kvm_host.h | 278 +++++ arch/riscv/include/asm/kvm_types.h | 7 + arch/riscv/include/asm/kvm_vcpu_timer.h | 44 + arch/riscv/include/asm/pgtable-bits.h | 1 + arch/riscv/include/uapi/asm/kvm.h | 128 +++ arch/riscv/include/uapi/asm/unistd.h | 5 +- arch/riscv/kernel/asm-offsets.c | 156 +++ arch/riscv/kvm/Kconfig | 36 + arch/riscv/kvm/Makefile | 15 + arch/riscv/kvm/main.c | 118 ++ arch/riscv/kvm/mmu.c | 860 ++++++++++++++ arch/riscv/kvm/tlb.S | 74 ++ arch/riscv/kvm/vcpu.c | 1012 +++++++++++++++++ arch/riscv/kvm/vcpu_exit.c | 701 ++++++++++++ arch/riscv/kvm/vcpu_sbi.c | 173 +++ arch/riscv/kvm/vcpu_switch.S | 400 +++++++ arch/riscv/kvm/vcpu_timer.c | 225 ++++ arch/riscv/kvm/vm.c | 81 ++ arch/riscv/kvm/vmid.c | 120 ++ drivers/clk/Kconfig | 5 + drivers/clk/Makefile | 1 + drivers/clk/microchip/Makefile | 2 + drivers/clk/microchip/clk-pfsoc.c | 508 +++++++++ drivers/clocksource/timer-riscv.c | 8 + include/clocksource/timer-riscv.h | 16 + include/uapi/linux/kvm.h | 8 + 38 files changed, 5619 insertions(+), 54 deletions(-) create mode 100644 arch/riscv/boot/dts/microchip/Makefile create mode 100644 arch/riscv/boot/dts/microchip/icicle-kit-es.dts create mode 100644 arch/riscv/configs/extra_debug.config create mode 100644 arch/riscv/include/asm/kvm_host.h create mode 100644 arch/riscv/include/asm/kvm_types.h create mode 100644 arch/riscv/include/asm/kvm_vcpu_timer.h create mode 100644 arch/riscv/include/uapi/asm/kvm.h create mode 100644 arch/riscv/kvm/Kconfig create mode 100644 arch/riscv/kvm/Makefile create mode 100644 arch/riscv/kvm/main.c create mode 100644 arch/riscv/kvm/mmu.c create mode 100644 arch/riscv/kvm/tlb.S create mode 100644 arch/riscv/kvm/vcpu.c create mode 100644 arch/riscv/kvm/vcpu_exit.c create mode 100644 arch/riscv/kvm/vcpu_sbi.c create mode 100644 arch/riscv/kvm/vcpu_switch.S create mode 100644 arch/riscv/kvm/vcpu_timer.c create mode 100644 arch/riscv/kvm/vm.c create mode 100644 arch/riscv/kvm/vmid.c create mode 100644 drivers/clk/microchip/clk-pfsoc.c create mode 100644 include/clocksource/timer-riscv.h -- 2.19.1

2 27

[PATCH 1/4] net/hinic: Set default features when probe netdev
by Yang Yingliang 17 Dec '20

17 Dec '20

From: Chiqijun <chiqijun(a)huawei.com> driver inclusion category: bugfix bugzilla: 4472 ----------------------------------------------------------------------- In the x86 system, the hinic device turns off LRO by default, and the LRO state will not be set when the driver is loaded. If the LRO of the chip is turned on before the driver is loaded, it will cause the chip to still turn on LRO after reloading the driver, but the LRO that the user sees is turned off. Therefore, the driver is forced to set the default features of the device when probe. Signed-off-by: Chiqijun <chiqijun(a)huawei.com> Reviewed-by: Wangxiaoyun <cloud.wangxiaoyun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/net/ethernet/huawei/hinic/hinic_main.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/huawei/hinic/hinic_main.c b/drivers/net/ethernet/huawei/hinic/hinic_main.c index cf7b6ceef060..26cd2929f60a 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_main.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c @@ -1491,7 +1491,7 @@ static int set_feature_lro(struct hinic_nic_dev *nic_dev, static int set_features(struct hinic_nic_dev *nic_dev, netdev_features_t pre_features, - netdev_features_t features, bool force_change) + netdev_features_t features) { netdev_features_t failed_features = 0; u32 err; @@ -1518,7 +1518,7 @@ static int hinic_set_features(struct net_device *netdev, struct hinic_nic_dev *nic_dev = netdev_priv(netdev); return set_features(nic_dev, nic_dev->netdev->features, - features, false); + features); } static netdev_features_t hinic_fix_features(struct net_device *netdev, @@ -1563,7 +1563,8 @@ static int hinic_set_default_hw_feature(struct hinic_nic_dev *nic_dev) } /* enable all hw features in netdev->features */ - return set_features(nic_dev, 0, nic_dev->netdev->features, true); + return set_features(nic_dev, ~nic_dev->netdev->features, + nic_dev->netdev->features); } static int hinic_setup_tc_mqprio(struct net_device *dev, -- 2.25.1

1 3

[PATCH 1/2] ext4: Protect superblock modifications with a buffer lock
by Yang Yingliang 17 Dec '20

17 Dec '20

From: Jan Kara <jack(a)suse.cz> hulk inclusion category: bugfix bugzilla: 46758 CVE: NA --------------------------- Protect all superblock modifications (including checksum computation) with a superblock buffer lock. That way we are sure computed checksum matches current superblock contents (a mismatch could cause checksum failures in nojournal mode or if an unjournalled superblock update races with a journalled one). Also we avoid modifying superblock contents while it is being written out (which can cause DIF/DIX failures if we are running in nojournal mode). Signed-off-by: Jan Kara <jack(a)suse.cz> [backport the 10th patch: https://www.spinics.net/lists/linux-ext4/msg75423.html drop the other lock_buffer besides operation for orphan] Signed-off-by: yangerkun <yangerkun(a)huawei.com> Reviewed-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/ext4/ext4_jbd2.c | 1 - fs/ext4/file.c | 1 + fs/ext4/inode.c | 1 + fs/ext4/namei.c | 6 ++++++ fs/ext4/resize.c | 4 ++++ fs/ext4/xattr.c | 1 + 6 files changed, 13 insertions(+), 1 deletion(-) diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c index a589b7f79558..f9ac7dfd93bf 100644 --- a/fs/ext4/ext4_jbd2.c +++ b/fs/ext4/ext4_jbd2.c @@ -361,7 +361,6 @@ int __ext4_handle_dirty_super(const char *where, unsigned int line, struct buffer_head *bh = EXT4_SB(sb)->s_sbh; int err = 0; - ext4_superblock_csum_set(sb); if (ext4_handle_valid(handle)) { err = jbd2_journal_dirty_metadata(handle, bh); if (err) diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 52d155b4e733..1703871fa2d0 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -434,6 +434,7 @@ static int ext4_sample_last_mounted(struct super_block *sb, goto out_journal; strlcpy(sbi->s_es->s_last_mounted, cp, sizeof(sbi->s_es->s_last_mounted)); + ext4_superblock_csum_set(sb); ext4_handle_dirty_super(handle, sb); out_journal: ext4_journal_stop(handle); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 8d7e2cd9ae37..6ee89cd7a408 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -5398,6 +5398,7 @@ static int ext4_do_update_inode(handle_t *handle, if (err) goto out_brelse; ext4_set_feature_large_file(sb); + ext4_superblock_csum_set(sb); ext4_handle_sync(handle); err = ext4_handle_dirty_super(handle, sb); } diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index ffc3695eb153..762eb6913240 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -2889,7 +2889,10 @@ int ext4_orphan_add(handle_t *handle, struct inode *inode) (le32_to_cpu(sbi->s_es->s_inodes_count))) { /* Insert this inode at the head of the on-disk orphan list */ NEXT_ORPHAN(inode) = le32_to_cpu(sbi->s_es->s_last_orphan); + lock_buffer(sbi->s_sbh); sbi->s_es->s_last_orphan = cpu_to_le32(inode->i_ino); + ext4_superblock_csum_set(sb); + unlock_buffer(sbi->s_sbh); dirty = true; } list_add(&EXT4_I(inode)->i_orphan, &sbi->s_orphan); @@ -2972,7 +2975,10 @@ int ext4_orphan_del(handle_t *handle, struct inode *inode) mutex_unlock(&sbi->s_orphan_lock); goto out_brelse; } + lock_buffer(sbi->s_sbh); sbi->s_es->s_last_orphan = cpu_to_le32(ino_next); + ext4_superblock_csum_set(inode->i_sb); + unlock_buffer(sbi->s_sbh); mutex_unlock(&sbi->s_orphan_lock); err = ext4_handle_dirty_super(handle, inode->i_sb); } else { diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 6a0c5c880354..c2e007d836e4 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -901,6 +901,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, ext4_kvfree_array_rcu(o_group_desc); le16_add_cpu(&es->s_reserved_gdt_blocks, -1); + ext4_superblock_csum_set(sb); err = ext4_handle_dirty_super(handle, sb); if (err) ext4_std_error(sb, err); @@ -1423,6 +1424,7 @@ static void ext4_update_super(struct super_block *sb, * active. */ ext4_r_blocks_count_set(es, ext4_r_blocks_count(es) + reserved_blocks); + ext4_superblock_csum_set(sb); /* Update the free space counts */ percpu_counter_add(&sbi->s_freeclusters_counter, @@ -1721,6 +1723,7 @@ static int ext4_group_extend_no_check(struct super_block *sb, ext4_blocks_count_set(es, o_blocks_count + add); ext4_free_blocks_count_set(es, ext4_free_blocks_count(es) + add); + ext4_superblock_csum_set(sb); ext4_debug("freeing blocks %llu through %llu\n", o_blocks_count, o_blocks_count + add); /* We add the blocks to the bitmap and set the group need init bit */ @@ -1882,6 +1885,7 @@ static int ext4_convert_meta_bg(struct super_block *sb, struct inode *inode) ext4_set_feature_meta_bg(sb); sbi->s_es->s_first_meta_bg = cpu_to_le32(num_desc_blocks(sb, sbi->s_groups_count)); + ext4_superblock_csum_set(sb); err = ext4_handle_dirty_super(handle, sb); if (err) { diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 24cf730ba6b0..ae029dccebc1 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -791,6 +791,7 @@ static void ext4_xattr_update_super_block(handle_t *handle, BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "get_write_access"); if (ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh) == 0) { ext4_set_feature_xattr(sb); + ext4_superblock_csum_set(sb); ext4_handle_dirty_super(handle, sb); } } -- 2.25.1

1 1

[PATCH 01/15] arm64: KVM: Remove some extra semicolon in kvm_target_cpu
by Yang Yingliang 17 Dec '20

17 Dec '20

From: zhong jiang <zhongjiang(a)huawei.com> mainline inclusion from mainline-4.20 commit f0725345e3e127032376e4fcb6b0fc893237fcef category: bugfix bugzilla: NA CVE: NA ------------------------------------------------- There are some extra semicolon in kvm_target_cpu, remove it. Signed-off-by: zhong jiang <zhongjiang(a)huawei.com> Signed-off-by: Marc Zyngier <marc.zyngier(a)arm.com> Signed-off-by: Xiangyou Xie <xiexiangyou(a)huawei.com> Reviewed-by: Ying Fang <fangying1(a)huawei.com> Reviewed-by: Zenghui Yu <yuzenghui(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/kvm/guest.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm64/kvm/guest.c b/arch/arm64/kvm/guest.c index ff9908c5d496..ff00e1c8023a 100644 --- a/arch/arm64/kvm/guest.c +++ b/arch/arm64/kvm/guest.c @@ -432,15 +432,15 @@ int __attribute_const__ kvm_target_cpu(void) return KVM_ARM_TARGET_CORTEX_A53; case ARM_CPU_PART_CORTEX_A57: return KVM_ARM_TARGET_CORTEX_A57; - }; + } break; case ARM_CPU_IMP_APM: switch (part_number) { case APM_CPU_PART_POTENZA: return KVM_ARM_TARGET_XGENE_POTENZA; - }; + } break; - }; + } /* Return a default generic target */ return KVM_ARM_TARGET_GENERIC_V8; -- 2.25.1

1 14

[PATCH] mm: thp: make the THP mapcount atomic against __split_huge_pmd_locked()
by Yang Yingliang 17 Dec '20

17 Dec '20

From: Andrea Arcangeli <aarcange(a)redhat.com> stable inclusion from linux-4.19.129 commit 453d8a481b127edfa7fb76af1bc7586b7a63bdd2 CVE: CVE-2020-29368 -------------------------------- commit c444eb564fb16645c172d550359cb3d75fe8a040 upstream. Write protect anon page faults require an accurate mapcount to decide if to break the COW or not. This is implemented in the THP path with reuse_swap_page() -> page_trans_huge_map_swapcount()/page_trans_huge_mapcount(). If the COW triggers while the other processes sharing the page are under a huge pmd split, to do an accurate reading, we must ensure the mapcount isn't computed while it's being transferred from the head page to the tail pages. reuse_swap_cache() already runs serialized by the page lock, so it's enough to add the page lock around __split_huge_pmd_locked too, in order to add the missing serialization. Note: the commit in "Fixes" is just to facilitate the backporting, because the code before such commit didn't try to do an accurate THP mapcount calculation and it instead used the page_count() to decide if to COW or not. Both the page_count and the pin_count are THP-wide refcounts, so they're inaccurate if used in reuse_swap_page(). Reverting such commit (besides the unrelated fix to the local anon_vma assignment) would have also opened the window for memory corruption side effects to certain workloads as documented in such commit header. Signed-off-by: Andrea Arcangeli <aarcange(a)redhat.com> Suggested-by: Jann Horn <jannh(a)google.com> Reported-by: Jann Horn <jannh(a)google.com> Acked-by: Kirill A. Shutemov <kirill.shutemov(a)linux.intel.com> Fixes: 6d0a07edd17c ("mm: thp: calculate the mapcount correctly for THP pages during WP faults") Cc: stable(a)vger.kernel.org Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- mm/huge_memory.c | 31 ++++++++++++++++++++++++++++--- 1 file changed, 28 insertions(+), 3 deletions(-) diff --git a/mm/huge_memory.c b/mm/huge_memory.c index b60cbf77e902..816a7fd3c6ff 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -2286,6 +2286,8 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, spinlock_t *ptl; struct mm_struct *mm = vma->vm_mm; unsigned long haddr = address & HPAGE_PMD_MASK; + bool was_locked = false; + pmd_t _pmd; mmu_notifier_invalidate_range_start(mm, haddr, haddr + HPAGE_PMD_SIZE); ptl = pmd_lock(mm, pmd); @@ -2295,11 +2297,32 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, * pmd against. Otherwise we can end up replacing wrong page. */ VM_BUG_ON(freeze && !page); - if (page && page != pmd_page(*pmd)) - goto out; + if (page) { + VM_WARN_ON_ONCE(!PageLocked(page)); + was_locked = true; + if (page != pmd_page(*pmd)) + goto out; + } +repeat: if (pmd_trans_huge(*pmd)) { - page = pmd_page(*pmd); + if (!page) { + page = pmd_page(*pmd); + if (unlikely(!trylock_page(page))) { + get_page(page); + _pmd = *pmd; + spin_unlock(ptl); + lock_page(page); + spin_lock(ptl); + if (unlikely(!pmd_same(*pmd, _pmd))) { + unlock_page(page); + put_page(page); + page = NULL; + goto repeat; + } + put_page(page); + } + } if (PageMlocked(page)) clear_page_mlock(page); } else if (!(pmd_devmap(*pmd) || is_pmd_migration_entry(*pmd))) @@ -2307,6 +2330,8 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, __split_huge_pmd_locked(vma, pmd, haddr, freeze); out: spin_unlock(ptl); + if (!was_locked && page) + unlock_page(page); /* * No need to double call mmu_notifier->invalidate_range() callback. * They are 3 cases to consider inside __split_huge_pmd_locked(): -- 2.25.1

1 0

[PATCH] romfs: fix uninitialized memory leak in romfs_dev_read()
by Yang Yingliang 17 Dec '20

17 Dec '20

From: Jann Horn <jannh(a)google.com> stable inclusion from linux-4.19.142 commit 9660983738399465fd0e3b1977a61bbd29b2e5be CVE: CVE-2020-29371 -------------------------------- commit bcf85fcedfdd17911982a3e3564fcfec7b01eebd upstream. romfs has a superblock field that limits the size of the filesystem; data beyond that limit is never accessed. romfs_dev_read() fetches a caller-supplied number of bytes from the backing device. It returns 0 on success or an error code on failure; therefore, its API can't represent short reads, it's all-or-nothing. However, when romfs_dev_read() detects that the requested operation would cross the filesystem size limit, it currently silently truncates the requested number of bytes. This e.g. means that when the content of a file with size 0x1000 starts one byte before the filesystem size limit, ->readpage() will only fill a single byte of the supplied page while leaving the rest uninitialized, leaking that uninitialized memory to userspace. Fix it by returning an error code instead of truncating the read when the requested read operation would go beyond the end of the filesystem. Fixes: da4458bda237 ("NOMMU: Make it possible for RomFS to use MTD devices directly") Signed-off-by: Jann Horn <jannh(a)google.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Reviewed-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Cc: David Howells <dhowells(a)redhat.com> Cc: <stable(a)vger.kernel.org> Link: http://lkml.kernel.org/r/20200818013202.2246365-1-jannh@google.com Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/romfs/storage.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/fs/romfs/storage.c b/fs/romfs/storage.c index f86f51f99ace..1dcadd22b440 100644 --- a/fs/romfs/storage.c +++ b/fs/romfs/storage.c @@ -221,10 +221,8 @@ int romfs_dev_read(struct super_block *sb, unsigned long pos, size_t limit; limit = romfs_maxsize(sb); - if (pos >= limit) + if (pos >= limit || buflen > limit - pos) return -EIO; - if (buflen > limit - pos) - buflen = limit - pos; #ifdef CONFIG_ROMFS_ON_MTD if (sb->s_mtd) -- 2.25.1

1 0

[PATCH v2 OLK-5.10 0/3] Backport 5.10.1 LTS
by Chen Jun 17 Dec '20

17 Dec '20

v1->v2 modifty the commit message: linux-5.10.1 -> stable-5.10.1 *** BLURB HERE *** Greg Kroah-Hartman (3): Revert "md: change mddev 'chunk_sectors' from int to unsigned" Revert "dm raid: fix discard limits for raid1 and raid10" Linux 5.10.1 Makefile | 2 +- drivers/md/dm-raid.c | 12 +++++------- drivers/md/md.h | 4 ++-- 3 files changed, 8 insertions(+), 10 deletions(-) -- 2.25.0

2 6

[OLK-5.10 0/3] Backport 5.10.1 LTS
by Chen Jun 16 Dec '20

16 Dec '20

*** BLURB HERE *** Greg Kroah-Hartman (3): Revert "md: change mddev 'chunk_sectors' from int to unsigned" Revert "dm raid: fix discard limits for raid1 and raid10" Linux 5.10.1 Makefile | 2 +- drivers/md/dm-raid.c | 12 +++++------- drivers/md/md.h | 4 ++-- 3 files changed, 8 insertions(+), 10 deletions(-) -- 2.25.0

2 6

为什么你一定要参加 openEuler Summit 的 TC、Maintainer、SIG 组工作会议
by public openEuler 16 Dec '20

16 Dec '20

嗨，各位 openEuler 社区的开发者：很高兴能够向你们汇报 openEuler Summit 的会议进度。对于各位来说，本周最重要的事情是 TC、Maintainer和SIG 组开放工作会议的议程已经上线了。 *为什么你一定要参与 TC、Maintainer、SIG组开放工作会议？* 1. 你可以清楚的了解 openEuler 社区的运作模式 2. 你可以和 TC、Maintainer、SIG 的成员共同探讨openEuler 社区的下一个版本的计划和技术路线 3. 可以跟社区里的技术大佬进行面对面交流。 4. 更好的规划自己在社区中的成长路线 *会议时间和房间号：* TC 开放工作会议 Maintainer 开放工作会议 SIG 组开放工作会议时间 12/24 13:00-14:30 12/24 15:00-17:00 12/25 13:00-15:30 房间号悦府6和悦府7 悦府6和悦府7 大宴会厅B *TC **开放工作会议议程：* 1. 讨论 openEuler 支持 App Stream 机制 2. 申请成立智能运维 SIG 3. 讨论从 Gitee issue 整体切换到bugzilla 4. 讨论2021年 TC 合作的开展方式 5. …… *Maintainer **开放工作会议议程* 1. 开场 & 颁发 Maintainer 纪念徽章 2. openEuler 社区运作分析 3. 我在 openEuler 社区担任 Maintainer 的那些事儿 4. 自由问答 5. 总结 *参与本次 SIG 组开放工作会议的 SIG 组：* 1. sig-DDE 2. sig-UKUI 3. sig-Ha 4. sig-arrch32 5. A-Tune 6. iSulad & Container 7. sig-ai-bitdata 8. sig-security-facility 9. sig-confidential-computing 10. Compiler 11. Doc 12. Infrastructure 13. security-committee 14. sig-Compatibility-Infra 15. sig-QA 16. sig-release-management 17. Kernel 18. Virt *openEuler Summit **专属伴手礼：* [image: image.png] *openEuler **卫衣* [image: image.png] *openEuler Maintainer **纪念徽章* *欢迎通过openEuler 社区开发者专属的报名渠道报名：* https://etherpad.openeuler.org/p/openEuler-Summit-2020 openEuler Summit 官网： https://openeuler.org/zh/interaction/summit-list/

1 0

[PATCH 1/2] arm64: arch_timer: only do cntvct workaround on VDSO path on D05
by Yang Yingliang 16 Dec '20

16 Dec '20

hulk inclusion category: bugfix bugzilla: NA CVE: NA -------------------------------- On other board do do cntvct workaround on VDSO path may cause unexpected error, so we do this only on D05. Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/clocksource/arm_arch_timer.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/clocksource/arm_arch_timer.c b/drivers/clocksource/arm_arch_timer.c index 443079810300..50030326d27b 100644 --- a/drivers/clocksource/arm_arch_timer.c +++ b/drivers/clocksource/arm_arch_timer.c @@ -551,9 +551,14 @@ void arch_timer_enable_workaround(const struct arch_timer_erratum_workaround *wa * change both the default value and the vdso itself. */ if (wa->read_cntvct_el0) { - clocksource_counter.archdata.vdso_direct = true; - vdso_default = true; - vdso_fix = true; + if (wa->read_cntvct_el0 == hisi_161010101_read_cntvct_el0) { + clocksource_counter.archdata.vdso_direct = true; + vdso_default = true; + vdso_fix = true; + } else { + clocksource_counter.archdata.vdso_direct = false; + vdso_default = false; + } } } -- 2.25.1

1 1

[PATCH] scsi: libiscsi: Fix cmds hung when sd_shutdown
by Yang Yingliang 15 Dec '20

15 Dec '20

From: Wu Bo <wubo40(a)huawei.com> hulk inclusion category: bugfix bugzilla: NA CVE: NA https://gitee.com/src-openeuler/kernel/issues/I28N9J --------------------------- For some reason, during reboot the system, iscsi.service failed to logout all sessions. kernel will hang forever on its sd_sync_cache() logic, after issuing the SYNCHRONIZE_CACHE cmd to all still existent paths. [ 1044.098991] reboot: Mddev shutdown finished. [ 1044.099311] reboot: Usermodehelper disable finished. [ 1050.611244] connection2:0: ping timeout of 5 secs expired, recv timeout 5, last rx 4295152378, last ping 4295153633, now 4295154944 [ 1348.599676] Call trace: [ 1348.599887] __switch_to+0xe8/0x150 [ 1348.600113] __schedule+0x33c/0xa08 [ 1348.600372] schedule+0x2c/0x88 [ 1348.600567] schedule_timeout+0x184/0x3a8 [ 1348.600820] io_schedule_timeout+0x28/0x48 [ 1348.601089] wait_for_common_io.constprop.2+0x168/0x258 [ 1348.601425] wait_for_completion_io_timeout+0x28/0x38 [ 1348.601762] blk_execute_rq+0x98/0xd8 [ 1348.602006] __scsi_execute+0xe0/0x1e8 [ 1348.602262] sd_sync_cache+0xd0/0x220 [sd_mod] [ 1348.602551] sd_shutdown+0x6c/0xf8 [sd_mod] [ 1348.602826] device_shutdown+0x13c/0x250 [ 1348.603078] kernel_restart_prepare+0x5c/0x68 [ 1348.603400] kernel_restart+0x20/0x98 [ 1348.603683] __se_sys_reboot+0x214/0x260 [ 1348.603987] __arm64_sys_reboot+0x24/0x30 [ 1348.604300] el0_svc_common+0x80/0x1b8 [ 1348.604590] el0_svc_handler+0x78/0xe0 [ 1348.604877] el0_svc+0x10/0x260 d754941225 (scsi: libiscsi: Allow sd_shutdown on bad transport) Once solved this problem. The iscsi_eh_cmd_timed_out() function add system_state judgment, and will return BLK_EH_DONE and mark the result as DID_NO_CONNECT when system_state is not SYSTEM_RUNNING, To tell upper layers that the command was handled during the transport layer error handler helper. The scsi Mid Layer timeout handler function(scsi_times_out) will be abort the cmd if the scsi LLD timeout handler return BLK_EH_DONE. if abort cmd failed, will enter scsi EH logic. Scsi EH will do reset target logic, if reset target failed, Will call iscsi_eh_session_reset() function to drop the session. The iscsi_eh_session_reset function will wait for a relogin, session termination from userspace, or a recovery/replacement timeout. But at this time, the app iscsid has exited, and the session was marked as ISCSI_STATE_FAILED, So the SCSI EH process will never be scheduled back again. PID: 9123 TASK: ffff80020c1b4d80 CPU: 3 COMMAND: "scsi_eh_2" #0 [ffff00008632bb70] __switch_to at ffff000080088738 #1 [ffff00008632bb90] __schedule at ffff000080a00480 #2 [ffff00008632bc20] schedule at ffff000080a00b58 #3 [ffff00008632bc30] iscsi_eh_session_reset at ffff000000d1ab9c [libiscsi] #4 [ffff00008632bcb0] iscsi_eh_recover_target at ffff000000d1d1fc [libiscsi] #5 [ffff00008632bd00] scsi_try_target_reset at ffff0000806f0bac #6 [ffff00008632bd30] scsi_eh_ready_devs at ffff0000806f2724 #7 [ffff00008632bde0] scsi_error_handler at ffff0000806f41d4 #8 [ffff00008632be70] kthread at ffff000080119ae0 Reported-by: Tianxiong Lu <lutianxiong(a)huawei.com> Signed-off-by: Wu Bo <wubo40(a)huawei.com> Signed-off-by: Yu Kuai <yukuai3(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/scsi/libiscsi.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index ec356e035c5f..35d603db33bb 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -2441,7 +2441,17 @@ int iscsi_eh_session_reset(struct scsi_cmnd *sc) mutex_lock(&session->eh_mutex); spin_lock_bh(&session->frwd_lock); - if (session->state == ISCSI_STATE_TERMINATE) { + + /* + * During shutdown, if session is prematurely disconnected, recovery + * won't happen and there will be hung cmds. To solve this case, all + * cmds would be enter scsi EH, but the EH path will wait for + * wait_event_interruptible() to complete until the state of the + * session is ISCSI_STATE_TERMINATE, ISCSI_STATE_LOGGED_IN or + * ISCSI_STATE_RECOVERY_FAILED. + */ + if (session->state == ISCSI_STATE_TERMINATE || + unlikely(system_state != SYSTEM_RUNNING)) { failed: ISCSI_DBG_EH(session, "failing session reset: Could not log back into " -- 2.25.1

1 0

[PATCH] scsi: iscsi: Perform connection failure entirely in kernel space
by Yang Yingliang 15 Dec '20

15 Dec '20

From: Bharath Ravi <rbharath(a)google.com> mainline inclusion from mainline-v5.7-rc1 commit 0ab710458da113a71c461c4df27e7f1353d9f864 category: bugfix bugzilla: NA CVE: NA https://gitee.com/src-openeuler/kernel/issues/I28N9J -------------------------------- Connection failure processing depends on a daemon being present to (at least) stop the connection and start recovery. This is a problem on a multipath scenario, where if the daemon failed for whatever reason, the SCSI path is never marked as down, multipath won't perform the failover and IO to the device will be forever waiting for that connection to come back. This patch performs the connection failure entirely inside the kernel. This way, the failover can happen and pending IO can continue even if the daemon is dead. Once the daemon comes alive again, it can execute recovery procedures if applicable. Cc: Mike Christie <mchristi(a)redhat.com> Cc: Lee Duncan <LDuncan(a)suse.com> Cc: Bart Van Assche <bvanassche(a)acm.org> Link: https://lore.kernel.org/r/20200125061925.191601-1-krisman@collabora.com Co-developed-by: Dave Clausen <dclausen(a)google.com> Co-developed-by: Nick Black <nlb(a)google.com> Co-developed-by: Vaibhav Nagarnaik <vnagarnaik(a)google.com> Co-developed-by: Anatol Pomazau <anatol(a)google.com> Co-developed-by: Tahsin Erdogan <tahsin(a)google.com> Co-developed-by: Frank Mayhar <fmayhar(a)google.com> Co-developed-by: Junho Ryu <jayr(a)google.com> Co-developed-by: Khazhismel Kumykov <khazhy(a)google.com> Reviewed-by: Reviewed-by: Khazhismel Kumykov <khazhy(a)google.com> Co-developed-by: Gabriel Krisman Bertazi <krisman(a)collabora.com> Reviewed-by: Lee Duncan <lduncan(a)suse.com> Signed-off-by: Bharath Ravi <rbharath(a)google.com> Signed-off-by: Dave Clausen <dclausen(a)google.com> Signed-off-by: Nick Black <nlb(a)google.com> Signed-off-by: Vaibhav Nagarnaik <vnagarnaik(a)google.com> Signed-off-by: Anatol Pomazau <anatol(a)google.com> Signed-off-by: Tahsin Erdogan <tahsin(a)google.com> Signed-off-by: Frank Mayhar <fmayhar(a)google.com> Signed-off-by: Junho Ryu <jayr(a)google.com> Signed-off-by: Khazhismel Kumykov <khazhy(a)google.com> Signed-off-by: Gabriel Krisman Bertazi <krisman(a)collabora.com> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/scsi/scsi_transport_iscsi.c | 68 +++++++++++++++++++++++++++++ include/scsi/scsi_transport_iscsi.h | 1 + 2 files changed, 69 insertions(+) diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 698347301198..e1ac96b8940b 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -81,6 +81,12 @@ struct iscsi_internal { struct transport_container session_cont; }; +/* Worker to perform connection failure on unresponsive connections + * completely in kernel space. + */ +static void stop_conn_work_fn(struct work_struct *work); +static DECLARE_WORK(stop_conn_work, stop_conn_work_fn); + static atomic_t iscsi_session_nr; /* sysfs session id for next new session */ static struct workqueue_struct *iscsi_eh_timer_workq; @@ -1609,6 +1615,7 @@ static DEFINE_MUTEX(rx_queue_mutex); static LIST_HEAD(sesslist); static DEFINE_SPINLOCK(sesslock); static LIST_HEAD(connlist); +static LIST_HEAD(connlist_err); static DEFINE_SPINLOCK(connlock); static uint32_t iscsi_conn_get_sid(struct iscsi_cls_conn *conn) @@ -2247,6 +2254,7 @@ iscsi_create_conn(struct iscsi_cls_session *session, int dd_size, uint32_t cid) mutex_init(&conn->ep_mutex); INIT_LIST_HEAD(&conn->conn_list); + INIT_LIST_HEAD(&conn->conn_list_err); conn->transport = transport; conn->cid = cid; @@ -2293,6 +2301,7 @@ int iscsi_destroy_conn(struct iscsi_cls_conn *conn) spin_lock_irqsave(&connlock, flags); list_del(&conn->conn_list); + list_del(&conn->conn_list_err); spin_unlock_irqrestore(&connlock, flags); transport_unregister_device(&conn->dev); @@ -2407,6 +2416,51 @@ int iscsi_offload_mesg(struct Scsi_Host *shost, } EXPORT_SYMBOL_GPL(iscsi_offload_mesg); +static void stop_conn_work_fn(struct work_struct *work) +{ + struct iscsi_cls_conn *conn, *tmp; + unsigned long flags; + LIST_HEAD(recovery_list); + + spin_lock_irqsave(&connlock, flags); + if (list_empty(&connlist_err)) { + spin_unlock_irqrestore(&connlock, flags); + return; + } + list_splice_init(&connlist_err, &recovery_list); + spin_unlock_irqrestore(&connlock, flags); + + list_for_each_entry_safe(conn, tmp, &recovery_list, conn_list_err) { + uint32_t sid = iscsi_conn_get_sid(conn); + struct iscsi_cls_session *session; + + mutex_lock(&rx_queue_mutex); + + session = iscsi_session_lookup(sid); + if (session) { + if (system_state != SYSTEM_RUNNING) { + session->recovery_tmo = 0; + conn->transport->stop_conn(conn, + STOP_CONN_TERM); + } else { + conn->transport->stop_conn(conn, + STOP_CONN_RECOVER); + } + } + + list_del_init(&conn->conn_list_err); + + mutex_unlock(&rx_queue_mutex); + + /* we don't want to hold rx_queue_mutex for too long, + * for instance if many conns failed at the same time, + * since this stall other iscsi maintenance operations. + * Give other users a chance to proceed. + */ + cond_resched(); + } +} + void iscsi_conn_error_event(struct iscsi_cls_conn *conn, enum iscsi_err error) { struct nlmsghdr *nlh; @@ -2414,6 +2468,12 @@ void iscsi_conn_error_event(struct iscsi_cls_conn *conn, enum iscsi_err error) struct iscsi_uevent *ev; struct iscsi_internal *priv; int len = nlmsg_total_size(sizeof(*ev)); + unsigned long flags; + + spin_lock_irqsave(&connlock, flags); + list_add(&conn->conn_list_err, &connlist_err); + spin_unlock_irqrestore(&connlock, flags); + queue_work(system_unbound_wq, &stop_conn_work); priv = iscsi_if_transport_lookup(conn->transport); if (!priv) @@ -2743,11 +2803,19 @@ static int iscsi_if_destroy_conn(struct iscsi_transport *transport, struct iscsi_uevent *ev) { struct iscsi_cls_conn *conn; + unsigned long flags; conn = iscsi_conn_lookup(ev->u.d_conn.sid, ev->u.d_conn.cid); if (!conn) return -EINVAL; + spin_lock_irqsave(&connlock, flags); + if (!list_empty(&conn->conn_list_err)) { + spin_unlock_irqrestore(&connlock, flags); + return -EAGAIN; + } + spin_unlock_irqrestore(&connlock, flags); + ISCSI_DBG_TRANS_CONN(conn, "Destroying transport conn\n"); if (transport->destroy_conn) transport->destroy_conn(conn); diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h index b266d2a3bcb1..7de0bf6a9b60 100644 --- a/include/scsi/scsi_transport_iscsi.h +++ b/include/scsi/scsi_transport_iscsi.h @@ -203,6 +203,7 @@ extern void iscsi_ping_comp_event(uint32_t host_no, struct iscsi_cls_conn { struct list_head conn_list; /* item in connlist */ + struct list_head conn_list_err; /* item in connlist_err */ void *dd_data; /* LLD private data */ struct iscsi_transport *transport; uint32_t cid; /* connection id */ -- 2.25.1

1 0

[PATCH 00/38] backport stable-4.19.163
by Yang Yingliang 14 Dec '20

14 Dec '20

Bjørn Mork (1): USB: serial: option: fix Quectel BG96 matching Bob Peterson (1): gfs2: check for empty rgrp tree in gfs2_ri_update Christian Eggers (3): i2c: imx: Don't generate STOP condition if arbitration has been lost i2c: imx: Fix reset of I2SR_IAL flag i2c: imx: Check for I2SR_IAL after every byte Florian Westphal (1): netfilter: nf_tables: avoid false-postive lockdep splat Giacinto Cifelli (1): USB: serial: option: add support for Thales Cinterion EXS82 Greg Kroah-Hartman (1): Linux 4.19.163 Hans de Goede (2): pinctrl: baytrail: Replace WARN with dev_info_once when setting direct-irq pin to output pinctrl: baytrail: Fix pin being driven low for a while on gpiod_get(..., GPIOD_OUT_HIGH) Jakub Kicinski (1): Revert "geneve: pull IP header before ECN decapsulation" Jan-Niklas Burfeind (1): USB: serial: ch341: add new Product ID for CH341A Jian-Hong Pan (1): ALSA: hda/realtek: Enable headset of ASUS UX482EG & B9400CEA with ALC294 Johan Hovold (2): USB: serial: kl5kusb105: fix memleak on open USB: serial: ch341: sort device-id entries Kailang Yang (1): ALSA: hda/realtek - Add new codec supported for ALC897 Lukas Wunner (3): spi: Introduce device-managed SPI controller allocation spi: bcm-qspi: Fix use-after-free on unbind spi: bcm2835: Fix use-after-free on unbind Luo Meng (1): Input: i8042 - fix error return code in i8042_setup_aux() Masami Hiramatsu (2): x86/uprobes: Do not use prefixes.nbytes when looping over prefixes.bytes x86/insn-eval: Use new for_each_insn_prefix() macro to loop over prefixes bytes Mike Snitzer (2): dm: remove invalid sparse __acquires and __releases annotations dm writecache: remove BUG() and fail gracefully instead Mikulas Patocka (1): dm writecache: fix the maximum number of arguments Naveen N. Rao (1): ftrace: Fix updating FTRACE_FL_TRAMP Paulo Alcantara (1): cifs: fix potential use-after-free in cifs_echo_request() Peter Ujfalusi (1): spi: bcm2835: Release the DMA channel if probe fails after dma_init Qian Cai (1): mm/swapfile: do not sleep with a spin lock held Steven Rostedt (VMware) (1): tracing: Fix userstacktrace option for instances Suganath Prabu S (1): scsi: mpt3sas: Fix ioctl timeout Suravee Suthikulpanit (1): iommu/amd: Set DTE[IntTabLen] to represent 512 IRTEs Takashi Iwai (2): ALSA: hda/realtek: Add mute LED quirk to yet another HP x360 model ALSA: hda/generic: Add option to enforce preferred_dacs pairs Vamsi Krishna Samavedam (1): usb: gadget: f_fs: Use local copy of descriptors for userspace copy Vincent Palatin (1): USB: serial: option: add Fibocom NL668 variants Yang Shi (1): mm: list_lru: set shrinker map bit when child nr_items is not zero Zhihao Cheng (1): i2c: qup: Fix error return code in qup_i2c_bam_schedule_desc() Makefile | 2 +- arch/x86/include/asm/insn.h | 15 +++++ arch/x86/kernel/uprobes.c | 10 ++-- arch/x86/lib/insn-eval.c | 5 +- drivers/i2c/busses/i2c-imx.c | 42 +++++++++++--- drivers/i2c/busses/i2c-qup.c | 3 +- drivers/input/serio/i8042.c | 3 +- drivers/iommu/amd_iommu_types.h | 2 +- drivers/md/dm-writecache.c | 4 +- drivers/md/dm.c | 2 - drivers/net/geneve.c | 20 ++----- drivers/pinctrl/intel/pinctrl-baytrail.c | 67 ++++++++++++++++++----- drivers/scsi/mpt3sas/mpt3sas_ctl.c | 2 +- drivers/spi/spi-bcm-qspi.c | 34 ++++-------- drivers/spi/spi-bcm2835.c | 22 +++----- drivers/spi/spi.c | 58 +++++++++++++++++++- drivers/usb/gadget/function/f_fs.c | 6 +- drivers/usb/serial/ch341.c | 5 +- drivers/usb/serial/kl5kusb105.c | 10 ++-- drivers/usb/serial/option.c | 10 ++-- fs/cifs/connect.c | 2 + fs/gfs2/rgrp.c | 4 ++ include/linux/spi/spi.h | 19 +++++++ kernel/trace/ftrace.c | 22 +++++++- kernel/trace/trace.c | 7 ++- kernel/trace/trace.h | 6 +- mm/list_lru.c | 10 ++-- mm/swapfile.c | 4 +- net/netfilter/nf_tables_api.c | 3 +- sound/pci/hda/hda_generic.c | 12 ++-- sound/pci/hda/hda_generic.h | 1 + sound/pci/hda/patch_realtek.c | 6 ++ tools/objtool/arch/x86/include/asm/insn.h | 15 +++++ tools/perf/util/intel-pt-decoder/insn.h | 15 +++++ 34 files changed, 329 insertions(+), 119 deletions(-) -- 2.25.1

1 38

[PATCH 1/5] scsi: hisi_sas: Delete down() when handle Block-IO
by Yang Yingliang 14 Dec '20

14 Dec '20

From: Yang Xingui <yangxingui(a)huawei.com> driver inclusion category: bugfix bugzilla: NA CVE: NA Call down() at RCU Read-Side Critical Secrion sometimes may caugh a RCU preempt Calltrace. [ 3506.800288] rcu: INFO: rcu_preempt detected stalls on CPUs/tasks: [ 3506.806357] rcu: Tasks blocked on level-1 rcu_node (CPUs 0-15): P22895/1:b..l [ 3506.813556] (detected by 108, t=162805 jiffies, g=124209, q=59192378) [ 3506.820056] kworker/0:1H D 0 22895 2 0x00000228 [ 3506.825524] Workqueue: kblockd blk_mq_run_work_fn [ 3506.830207] Call trace: [ 3506.832644] __switch_to+0xf0/0x140 [ 3506.836118] __schedule+0x890/0xdc8 [ 3506.839590] schedule+0x80/0x120 [ 3506.842804] schedule_timeout+0x22c/0x328 [ 3506.846796] __down+0x74/0xc8 [ 3506.849751] down+0x54/0x70 [ 3506.852539] hisi_sas_task_exec.isra.25+0x840/0xf28 [hisi_sas_main] [ 3506.858780] hisi_sas_queue_command+0x40/0x84 [hisi_sas_main] [ 3506.864502] sas_queuecommand+0x144/0x1c8 [libsas] [ 3506.869271] scsi_queue_rq+0x72c/0xa20 [ 3506.873005] blk_mq_dispatch_rq_list+0x110/0x680 [ 3506.877601] blk_mq_do_dispatch_sched+0x108/0x148 [ 3506.882285] __blk_mq_sched_dispatch_requests+0x120/0x1b8 [ 3506.887659] blk_mq_sched_dispatch_requests+0x40/0x78 [ 3506.892688] __blk_mq_run_hw_queue+0xb8/0x138 [ 3506.897026] blk_mq_run_work_fn+0x28/0x38 Actually it's no need for keeping IO stopping here to wait for controller reset complete. We can directly return error code. Signed-off-by: Yang Xingui <yangxingui(a)huawei.com> Reviewed-by: Zhu Xiongxiong <zhuxiongxiong(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/scsi/hisi_sas/hisi_sas_main.c | 14 ++------------ 1 file changed, 2 insertions(+), 12 deletions(-) diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c index 124aa26fa4b8..0128c547e4c8 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_main.c +++ b/drivers/scsi/hisi_sas/hisi_sas_main.c @@ -611,19 +611,9 @@ static int hisi_sas_task_exec(struct sas_task *task, gfp_t gfp_flags, hisi_hba = dev_to_hisi_hba(device); dev = hisi_hba->dev; - if (unlikely(test_bit(HISI_SAS_REJECT_CMD_BIT, &hisi_hba->flags))) { - /* - * For IOs from upper layer, it may already disable preempt - * in the IO path, if disable preempt again in down(), - * function schedule() will report schedule_bug(), so check - * preemptible() before goto down(). - */ - if (!preemptible()) - return -EINVAL; + if (unlikely(test_bit(HISI_SAS_REJECT_CMD_BIT, &hisi_hba->flags))) + return -EINVAL; - down(&hisi_hba->sem); - up(&hisi_hba->sem); - } /* protect task_prep and start_delivery sequence */ rc = hisi_sas_task_prep(task, is_tmf, tmf, &pass, &dq); -- 2.25.1

1 4

[PATCH v2] arm64/mpam: Avoid compilation error of enable_acpi redefinition
by Wang ShaoBo 11 Dec '20

11 Dec '20

hulk inclusion category: bugfix bugzilla: 46772 CVE: NA Reference: https://gitee.com/openeuler/kernel/issues/I28HIY ------------------------------------------------- enable_acpi is declaration as a enumerated type in mpam, where it's also defined as a static inline function in acpi.h, this following compilation error will be triggered if we use their two like this. In file included from ./include/acpi/acpi_io.h:7:0, from ./include/linux/acpi.h:47, from ./arch/arm64/include/asm/machine_t.h:26, from arch/arm64/kernel/process.c:64: ./arch/arm64/include/asm/acpi.h:82:20: error: ‘enable_acpi’ redeclared as different kind of symbol static inline void enable_acpi(void) ^~~~~~~~~~~ In file included from arch/arm64/kernel/process.c:61:0: ./arch/arm64/include/asm/mpam_sched.h:46:2: note: previous definition of ‘enable_acpi’ was here enable_acpi, We should rename enable_acpi properly for mpam. Reported-by: Zhengyuan Liu <liuzhengyuan(a)tj.kylinos.cn> Signed-off-by: Wang ShaoBo <bobo.shaobowang(a)huawei.com> --- arch/arm64/include/asm/mpam_sched.h | 6 +++--- arch/arm64/kernel/mpam.c | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/arm64/include/asm/mpam_sched.h b/arch/arm64/include/asm/mpam_sched.h index 350296157087a..ddc7ce3e1a9e1 100644 --- a/arch/arm64/include/asm/mpam_sched.h +++ b/arch/arm64/include/asm/mpam_sched.h @@ -41,9 +41,9 @@ static inline void mpam_sched_in(void) } enum mpam_enable_type { - enable_denied = 0, - enable_default, - enable_acpi, + MPAM_ENABLE_DENIED = 0, + MPAM_ENABLE_DEFAULT, + MPAM_ENABLE_ACPI, }; extern enum mpam_enable_type __read_mostly mpam_enabled; diff --git a/arch/arm64/kernel/mpam.c b/arch/arm64/kernel/mpam.c index 3f89cec8e2fb1..96f4d013a9034 100644 --- a/arch/arm64/kernel/mpam.c +++ b/arch/arm64/kernel/mpam.c @@ -252,7 +252,7 @@ int __init mpam_force_init(void) { int ret; - if (mpam_enabled != enable_default) + if (mpam_enabled != MPAM_ENABLE_DEFAULT) return 0; ret = mpam_nodes_discovery_start(); @@ -1380,9 +1380,9 @@ enum mpam_enable_type __read_mostly mpam_enabled; static int __init mpam_setup(char *str) { if (!strcmp(str, "=acpi")) - mpam_enabled = enable_acpi; + mpam_enabled = MPAM_ENABLE_ACPI; else - mpam_enabled = enable_default; + mpam_enabled = MPAM_ENABLE_DEFAULT; return 1; } __setup("mpam", mpam_setup); -- 2.17.1

1 0

【会议纪要】 Re: openEuler kernel sig meeting
by Xie XiuQi 11 Dec '20

11 Dec '20

会议纪要：与会人：Zheng Zhenpeng, xiexiuqi， dukaitian, myeuler, jiaoff, Liu Yuan, Wenlong.Ding, 潘晨博 1. 网迅万兆网卡驱动合入 openEuler 内核讨论 - 郑振鹏、刘源 > > 网迅万兆网卡是国内具有完全知识产权的万兆网络控制器，相关资料请访问官方网站： > 网卡成品及资料 > https://www.net-swift.com/a/383.html > 网卡芯片 > https://www.net-swift.com/a/275.html > > 之前我们已向EulerOS发布版本提供万兆网卡驱动rpm包，此次希望能加深合作， > 将驱动源码推入openeuler内核。 > 根据开源规则（遵从GPLV2），我厂已在官网上提供源码下载 1）openEuler 欢迎网讯网卡驱动合入社区，使 openEuler 支持更丰富的设备。 2）kernel sig 同意网讯万兆网卡驱动合入 openEuler，后续驱动可以按照开源社区的规范，准备补丁； - 建议合入openEuler 20.03 LTS SP2 （kernel-4.19)，如果时间赶不上的话，与 Release sig 讨论是否可以合入后续 update 版本； - 该驱动代码量约 5w+ 行，首次合入时，不要求把所有开发历史作为补丁发出来，可以按照模块或者功能做适当的补丁拆分； 3）建议加入到兼容性sig 组，兼容性 sig 与网讯网卡确认的认证过程 4）与 Release sig 对齐 Release 节奏 5）另有针对网讯网卡针对 dpdk 的适配，也欢迎参与 openEuler，贡献社区。 On 2020/12/10 9:32, zhenpengzheng(a)net-swift.com wrote: > 收到，我会按时参加。 > > ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ > > **************************************************************** > > 郑振鹏(zheng zhenpeng) > > 北京网迅科技有限公司杭州分公司软件工程师 > > 浙江省杭州市西湖区文三路478号华星时代广场A座507室 310013 > > Beijing WangXun Technology Co., Ltd. Software Engineer. > > Room A507, HuaXing Times Square, No.478 West Wensan Road. > > West Lake District, Hangzhou City, 310013 ZHEJIANG, P.R.CHINA. > > > > Office: +86(0571)89807901-8014 > > Mobile: +86-13656681762 > > E-Mail: z <mailto:jianwang@trustnetic.com>henpengzheng(a)net-swift.com > > **************************************************************** > > > *发件人：* Xie XiuQi <mailto:xiexiuqi@huawei.com> > *发送时间：* 2020-12-09 20:09 > *收件人：* kernel(a)openeuler.org <mailto:kernel@openeuler.org> > *抄送：* tc(a)openeuler.org <mailto:tc@openeuler.org>; dev(a)openeuler.org <mailto:dev@openeuler.org>; release(a)openeuler.org <mailto:release@openeuler.org>; 郑振鹏 <mailto:zhenpengzheng@net-swift.com>; Liuyuan (Compatibility, Cloud Infrastructure Service Product Dept.) <mailto:liuyuan36@huawei.com> > *主题：* Re: [Release] Re: 【议题申报】 openEuler kernel sig meeting > 议题： > > 1）网迅万兆网卡驱动合入 openEuler 内核讨论 - 郑振鹏、刘源 > > 网迅万兆网卡是国内具有完全知识产权的万兆网络控制器，相关资料请访问官方网站： > 网卡成品及资料 > https://www.net-swift.com/a/383.html > 网卡芯片 > https://www.net-swift.com/a/275.html > > 之前我们已向EulerOS发布版本提供万兆网卡驱动rpm包，此次希望能加深合作，将驱动源码推入openeuler内核。 > 根据开源规则（遵从GPLV2），我厂已在官网上提供源码下载 > > > 欢迎继续申报议题 > > On 2020/12/2 19:52, Xie XiuQi wrote: > > > > > > On 2020/12/2 19:26, openEuler conference wrote: > >> 您好！ > >> > >> openEuler Kernel SIG 邀请您参加 2020-12-11 10:00 召开的ZOOM会议 > >> > >> 会议主题：openEuler kernel sig meeting > >> > >> 会议链接：https://zoom.us/j/95008051399?pwd=RVRzNk0zSVUxdmVaaWZjT1d4UEhVQT09 > >> > >> 更多资讯尽在：https://openeuler.org/zh/ > >> > >> > >> > >> Hello! > >> > >> openEuler Kernel SIG invites you to attend the ZOOM conference will be held at 2020-12-11 10:00, > >> > >> The subject of the conference is openEuler kernel sig meeting, > >> > >> You can join the meeting at https://zoom.us/j/95008051399?pwd=RVRzNk0zSVUxdmVaaWZjT1d4UEhVQT09. > >> > >> More information <https://openeuler.org/zh/> > >> > >> > >> _______________________________________________ > >> Release mailing list -- release(a)openeuler.org > >> To unsubscribe send an email to release-leave(a)openeuler.org > >> > > _______________________________________________ > > Release mailing list -- release(a)openeuler.org > > To unsubscribe send an email to release-leave(a)openeuler.org > > > >

1 0

[PATCH] arm64/mpam: Avoid compilation error of enable_acpi redefinition
by Wang ShaoBo 10 Dec '20

10 Dec '20

hulk inclusion category: bugfix bugzilla: 46772 CVE: NA Reference: https://gitee.com/openeuler/kernel/issues/I28HIY ------------------------------------------------- enable_acpi is declaration as a enumerated type in mpam, where it's also defined as a static inline function in acpi.h, this following compilation error will be triggered if we use their two like this. In file included from ./include/acpi/acpi_io.h:7:0, from ./include/linux/acpi.h:47, from ./arch/arm64/include/asm/machine_t.h:26, from arch/arm64/kernel/process.c:64: ./arch/arm64/include/asm/acpi.h:82:20: error: ‘enable_acpi’ redeclared as different kind of symbol static inline void enable_acpi(void) ^~~~~~~~~~~ In file included from arch/arm64/kernel/process.c:61:0: ./arch/arm64/include/asm/mpam_sched.h:46:2: note: previous definition of ‘enable_acpi’ was here enable_acpi, We should rename enable_acpi properly for mpam. Signed-off-by: Wang ShaoBo <bobo.shaobowang(a)huawei.com> --- arch/arm64/include/asm/mpam_sched.h | 6 +++--- arch/arm64/kernel/mpam.c | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/arm64/include/asm/mpam_sched.h b/arch/arm64/include/asm/mpam_sched.h index 350296157087a..ddc7ce3e1a9e1 100644 --- a/arch/arm64/include/asm/mpam_sched.h +++ b/arch/arm64/include/asm/mpam_sched.h @@ -41,9 +41,9 @@ static inline void mpam_sched_in(void) } enum mpam_enable_type { - enable_denied = 0, - enable_default, - enable_acpi, + MPAM_ENABLE_DENIED = 0, + MPAM_ENABLE_DEFAULT, + MPAM_ENABLE_ACPI, }; extern enum mpam_enable_type __read_mostly mpam_enabled; diff --git a/arch/arm64/kernel/mpam.c b/arch/arm64/kernel/mpam.c index 3f89cec8e2fb1..96f4d013a9034 100644 --- a/arch/arm64/kernel/mpam.c +++ b/arch/arm64/kernel/mpam.c @@ -252,7 +252,7 @@ int __init mpam_force_init(void) { int ret; - if (mpam_enabled != enable_default) + if (mpam_enabled != MPAM_ENABLE_DEFAULT) return 0; ret = mpam_nodes_discovery_start(); @@ -1380,9 +1380,9 @@ enum mpam_enable_type __read_mostly mpam_enabled; static int __init mpam_setup(char *str) { if (!strcmp(str, "=acpi")) - mpam_enabled = enable_acpi; + mpam_enabled = MPAM_ENABLE_ACPI; else - mpam_enabled = enable_default; + mpam_enabled = MPAM_ENABLE_DEFAULT; return 1; } __setup("mpam", mpam_setup); -- 2.17.1

2 1

[PATCH 00/32] backport stable-4.19.162
by Yang Yingliang 10 Dec '20

10 Dec '20

Alexander Duyck (1): tcp: Set INET_ECN_xmit configuration in tcp_reinit_congestion_control Anmol Karn (1): rose: Fix Null pointer dereference in rose_send_frame() Antoine Tenart (2): netfilter: bridge: reset skb->pkt_type after NF_INET_POST_ROUTING traversal net: ip6_gre: set dev->hard_header_len when using header_ops Dan Carpenter (2): net/x25: prevent a couple of overflows chelsio/chtls: fix a double free in chtls_setkey() Eran Ben Elisha (1): net/mlx5: Fix wrong address reclaim when command interface is down Eric Dumazet (1): geneve: pull IP header before ECN decapsulation Greg Kroah-Hartman (1): Linux 4.19.162 Guillaume Nault (1): ipv4: Fix tos mask in inet_rtm_getroute() Hector Martin (1): ALSA: usb-audio: US16x08: fix value count for level meters Jamie Iles (1): bonding: wait for sysfs kobject destruction before freeing struct slave Jens Axboe (1): tun: honor IOCB_NOWAIT flag Julian Wiedmann (1): net/af_iucv: set correct sk_protocol for child sockets Krzysztof Kozlowski (1): dt-bindings: net: correct interrupt flags in examples Lijun Pan (1): ibmvnic: fix call_netdevice_notifiers in do_reset Maxim Mikityanskiy (1): net/tls: Protect from calling tls_dev_del for TLS RX twice Po-Hsu Lin (1): Input: i8042 - add ByteSpeed touchpad to noloop table Sanjay Govind (1): Input: xpad - support Ardwiino Controllers Shiraz Saleem (1): RDMA/i40iw: Address an mmap handler exploit in i40iw Sylwester Dziedziuch (1): i40e: Fix removing driver while bare-metal VFs pass traffic Thomas Falcon (2): ibmvnic: Ensure that SCRQ entry reads are correctly ordered ibmvnic: Fix TX completion error handling Vadim Fedorenko (1): net/tls: missing received data after fast remote close Vasily Averin (1): tracing: Remove WARN_ON in start_thread() Vinay Kumar Yadav (1): chelsio/chtls: fix panic during unload reload chtls Wang Hai (2): ipv6: addrlabel: fix possible memory leak in ip6addrlbl_net_init net: mvpp2: Fix error return code in mvpp2_open() Willem de Bruijn (1): sock: set sk_err to ee_errno on dequeue from errq Yves-Alexis Perez (1): usbnet: ipheth: fix connectivity with iOS 14 Zhang Changzhong (2): cxgb3: fix error return code in t3_sge_alloc_qset() net: pasemi: fix error return code in pasemi_mac_open() .../devicetree/bindings/net/nfc/nxp-nci.txt | 2 +- .../devicetree/bindings/net/nfc/pn544.txt | 2 +- Makefile | 2 +- drivers/crypto/chelsio/chtls/chtls_cm.c | 1 + drivers/crypto/chelsio/chtls/chtls_hw.c | 1 + drivers/infiniband/hw/i40iw/i40iw_main.c | 5 -- drivers/infiniband/hw/i40iw/i40iw_verbs.c | 36 +++-------- drivers/input/joystick/xpad.c | 2 + drivers/input/serio/i8042-x86ia64io.h | 4 ++ drivers/net/bonding/bond_main.c | 61 +++++++++++++------ drivers/net/bonding/bond_sysfs_slave.c | 18 +----- drivers/net/ethernet/chelsio/cxgb3/sge.c | 1 + drivers/net/ethernet/ibm/ibmvnic.c | 26 ++++++-- drivers/net/ethernet/intel/i40e/i40e.h | 1 + drivers/net/ethernet/intel/i40e/i40e_main.c | 22 ++++--- .../ethernet/intel/i40e/i40e_virtchnl_pf.c | 26 ++++---- .../net/ethernet/marvell/mvpp2/mvpp2_main.c | 1 + .../ethernet/mellanox/mlx5/core/pagealloc.c | 21 ++++++- drivers/net/ethernet/pasemi/pasemi_mac.c | 8 ++- drivers/net/geneve.c | 20 ++++-- drivers/net/tun.c | 14 ++++- drivers/net/usb/ipheth.c | 2 +- include/net/bonding.h | 8 +++ include/net/tls.h | 6 ++ kernel/trace/trace_hwlat.c | 2 +- net/bridge/br_netfilter_hooks.c | 7 ++- net/core/skbuff.c | 2 +- net/ipv4/route.c | 7 ++- net/ipv4/tcp_cong.c | 5 ++ net/ipv6/addrlabel.c | 26 +++++--- net/ipv6/ip6_gre.c | 16 ++++- net/iucv/af_iucv.c | 4 +- net/rose/rose_loopback.c | 17 ++++-- net/tls/tls_device.c | 5 +- net/tls/tls_sw.c | 6 ++ net/x25/af_x25.c | 6 +- sound/usb/mixer_us16x08.c | 2 +- 37 files changed, 260 insertions(+), 135 deletions(-) -- 2.25.1

1 32

[PATCH] fanotify: fix merging marks masks with FAN_ONDIR
by Yang Yingliang 10 Dec '20

10 Dec '20

From: Amir Goldstein <amir73il(a)gmail.com> mainline inclusion from mainline-5.7-rc1 commit 55bf882c7f13dda8bbe624040c6d5b4fbb812d16 category: bugfix bugzilla: 46764 CVE: NA https://gitee.com/openeuler/kernel/issues/I27PTB?from=project-issue --------------------------- Change the logic of FAN_ONDIR in two ways that are similar to the logic of FAN_EVENT_ON_CHILD, that was fixed in commit 54a307ba8d3c ("fanotify: fix logic of events on child"): 1. The flag is meaningless in ignore mask 2. The flag refers only to events in the mask of the mark where it is set This is what the fanotify_mark.2 man page says about FAN_ONDIR: "Without this flag, only events for files are created." It doesn't say anything about setting this flag in ignore mask to stop getting events on directories nor can I think of any setup where this capability would be useful. Currently, when marks masks are merged, the FAN_ONDIR flag set in one mark affects the events that are set in another mark's mask and this behavior causes unexpected results. For example, a user adds a mark on a directory with mask FAN_ATTRIB | FAN_ONDIR and a mount mark with mask FAN_OPEN (without FAN_ONDIR). An opendir() of that directory (which is inside that mount) generates a FAN_OPEN event even though neither of the marks requested to get open events on directories. Link: https://lore.kernel.org/r/20200319151022.31456-10-amir73il@gmail.com Signed-off-by: Amir Goldstein <amir73il(a)gmail.com> Signed-off-by: Jan Kara <jack(a)suse.cz> Conflicts: fs/notify/fanotify/fanotify.c Signed-off-by: yangerkun <yangerkun(a)huawei.com> Reviewed-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/notify/fanotify/fanotify.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c index ca3405f73264..cdf0dc9ddbbb 100644 --- a/fs/notify/fanotify/fanotify.c +++ b/fs/notify/fanotify/fanotify.c @@ -117,6 +117,12 @@ static bool fanotify_should_send_event(struct fsnotify_iter_info *iter_info, /* Apply ignore mask regardless of ISDIR and ON_CHILD flags */ marks_ignored_mask |= mark->ignored_mask; + /* + * If the event is on dir and this mark doesn't care about + * events on dir, don't send it! + */ + if (d_is_dir(path->dentry) && !(mark->mask & FS_ISDIR)) + continue; /* * If the event is for a child and this mark doesn't care about @@ -130,10 +136,6 @@ static bool fanotify_should_send_event(struct fsnotify_iter_info *iter_info, marks_mask |= mark->mask; } - if (d_is_dir(path->dentry) && - !(marks_mask & FS_ISDIR & ~marks_ignored_mask)) - return false; - if (event_mask & FAN_ALL_OUTGOING_EVENTS & marks_mask & ~marks_ignored_mask) return true; -- 2.25.1

1 0

openEuler kernel sig meeting
by openEuler conference 09 Dec '20

09 Dec '20

3 3

[PATCH 01/17] share_pool: Fix series of bugs
by Yang Yingliang 08 Dec '20

08 Dec '20

From: Tang Yizhou <tangyizhou(a)huawei.com> ascend inclusion category: bugfix bugzilla: NA CVE: NA ------------------------------------------------- 1. Error handling of sp_free(). 2. When sp_alloc(..., SP_HUGEPAGE, ...) rollbacks to normal page, we need to call vfs_fallocate() otherwise memory leaks until sp group is dead. 3. When sp_alloc(..., SP_HUGEPAGE, ...) rollbacks to normal page, we need to clear SP_HUGEPAGE bit in sp_flags then spa_stat interface will show the spa as a normal page spa correctly. 4. Add the reference count of a spg in find_or_alloc_sp_group() due to closer relationship. Signed-off-by: Tang Yizhou <tangyizhou(a)huawei.com> Reviewed-by: Ding Tianhong <dingtianhong(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- mm/share_pool.c | 38 +++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-) diff --git a/mm/share_pool.c b/mm/share_pool.c index d39c2c3d728c..a71f6fb214ce 100644 --- a/mm/share_pool.c +++ b/mm/share_pool.c @@ -358,7 +358,7 @@ static struct sp_group *find_or_alloc_sp_group(int spg_id) spg->hugepage_failures = 0; spg->dvpp_multi_spaces = false; spg->owner = current->group_leader; - atomic_set(&spg->use_count, 0); + atomic_set(&spg->use_count, 1); INIT_LIST_HEAD(&spg->procs); INIT_LIST_HEAD(&spg->spa_list); @@ -391,6 +391,10 @@ static struct sp_group *find_or_alloc_sp_group(int spg_id) ret = PTR_ERR(spg->file_hugetlb); goto out_fput; } + } else { + if (!spg_valid(spg)) + return ERR_PTR(-ENODEV); + atomic_inc(&spg->use_count); } return spg; @@ -540,12 +544,6 @@ int sp_group_add_task(int pid, int spg_id) goto out_put_task; } - if (!spg_valid(spg)) { - ret = -ENODEV; - goto out_put_task; - } - atomic_inc(&spg->use_count); - /* access control permission check */ if (sysctl_ac_mode == AC_SINGLE_OWNER) { if (spg->owner != current->group_leader) { @@ -1102,6 +1100,7 @@ int sp_free(unsigned long addr) if (printk_ratelimit()) pr_err("share pool: sp free failed, addr %pK is not from sp_alloc\n", (void *)addr); + goto drop_spa; } if (!spg_valid(spa->spg)) @@ -1312,31 +1311,32 @@ void *sp_alloc(unsigned long size, unsigned long sp_flags, int spg_id) __sp_free(spg, sp_addr, size_aligned, list_next_entry(mm, sp_node)); + if (printk_ratelimit()) + pr_warn("share pool: allocation failed due to mm populate failed" + "(potential no enough memory when -12): %d\n", ret); + p = ERR_PTR(ret); + + mode = FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE; + offset = sp_addr - MMAP_SHARE_POOL_START; + + ret = vfs_fallocate(spa_file(spa), mode, offset, spa_size(spa)); + if (ret) + pr_err("share pool: sp alloc normal page fallocate failed %d\n", ret); + if (file == spg->file_hugetlb) { spg->hugepage_failures++; /* fallback to small pages */ if (!(sp_flags & SP_HUGEPAGE_ONLY)) { file = spg->file; - spa->is_hugepage = false; size_aligned = ALIGN(size, PAGE_SIZE); + sp_flags &= ~SP_HUGEPAGE; __sp_area_drop(spa); mmput(mm); goto try_again; } } - if (printk_ratelimit()) - pr_warn("share pool: allocation failed due to mm populate failed" - "(potential no enough memory when -12): %d\n", ret); - p = ERR_PTR(ret); - - mode = FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE; - offset = sp_addr - MMAP_SHARE_POOL_START; - ret = vfs_fallocate(spa_file(spa), mode, offset, spa_size(spa)); - if (ret) - pr_err("share pool: fallocate failed %d\n", ret); - mmput(mm); break; } -- 2.25.1

1 16

[PATCH 1/7] ascend: share_pool: Use sharepool_no_page to alloc hugepage
by Yang Yingliang 08 Dec '20

08 Dec '20

From: Weilong Chen <chenweilong(a)huawei.com> ascend inclusion category: feature bugzilla: NA CVE: NA ------------------------------------------------- Sharepool applies for a dedicated interface for large pages, which optimizes the efficiency of memory application Reviewed-by: Ding Tianhong <dingtianhong(a)huawei.com> Signed-off-by: Weilong Chen <chenweilong(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/linux/hugetlb.h | 4 ++ include/linux/share_pool.h | 5 +++ mm/hugetlb.c | 22 +++++++--- mm/share_pool.c | 90 ++++++++++++++++++++++++++++++++++++++ 4 files changed, 115 insertions(+), 6 deletions(-) diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index bd658f44e133..be649e5ba322 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -645,4 +645,8 @@ static inline spinlock_t *huge_pte_lock(struct hstate *h, return ptl; } +#ifdef CONFIG_ASCEND_SHARE_POOL +pte_t make_huge_pte(struct vm_area_struct *vma, struct page *page, int writable); +#endif + #endif /* _LINUX_HUGETLB_H */ diff --git a/include/linux/share_pool.h b/include/linux/share_pool.h index 2557ef138122..47d8579c23ec 100644 --- a/include/linux/share_pool.h +++ b/include/linux/share_pool.h @@ -220,6 +220,11 @@ static inline void sp_dump_stack(void) dump_stack(); } +vm_fault_t sharepool_no_page(struct mm_struct *mm, + struct vm_area_struct *vma, + struct address_space *mapping, pgoff_t idx, + unsigned long address, pte_t *ptep, unsigned int flags); + #else static inline int sp_group_add_task(int pid, int spg_id) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 9d2035632aed..a68e7e1afd82 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -3352,8 +3352,13 @@ const struct vm_operations_struct hugetlb_vm_ops = { .pagesize = hugetlb_vm_op_pagesize, }; +#ifdef CONFIG_ASCEND_SHARE_POOL +pte_t make_huge_pte(struct vm_area_struct *vma, struct page *page, + int writable) +#else static pte_t make_huge_pte(struct vm_area_struct *vma, struct page *page, int writable) +#endif { pte_t entry; @@ -3370,6 +3375,9 @@ static pte_t make_huge_pte(struct vm_area_struct *vma, struct page *page, return entry; } +#ifdef CONFIG_ASCEND_SHARE_POOL +EXPORT_SYMBOL(make_huge_pte); +#endif static void set_huge_ptep_writable(struct vm_area_struct *vma, unsigned long address, pte_t *ptep) @@ -3962,12 +3970,6 @@ static vm_fault_t hugetlb_no_page(struct mm_struct *mm, } page = alloc_huge_page(vma, haddr, 0); - if (IS_ERR(page) && sp_check_vm_share_pool(vma->vm_flags)) { - page = alloc_huge_page_node(hstate_file(vma->vm_file), - numa_mem_id()); - if (!page) - page = ERR_PTR(-ENOMEM); - } if (IS_ERR(page)) { /* * Returning error will result in faulting task being @@ -4155,7 +4157,15 @@ vm_fault_t hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, entry = huge_ptep_get(ptep); if (huge_pte_none(entry)) { +#ifdef CONFIG_ASCEND_SHARE_POOL + if (sp_check_vm_share_pool(vma->vm_flags)) { + ret = sharepool_no_page(mm, vma, mapping, idx, address, ptep, flags); + } else { + ret = hugetlb_no_page(mm, vma, mapping, idx, address, ptep, flags); + } +#else ret = hugetlb_no_page(mm, vma, mapping, idx, address, ptep, flags); +#endif goto out_mutex; } diff --git a/mm/share_pool.c b/mm/share_pool.c index 23195a76c2c6..286e74f99360 100644 --- a/mm/share_pool.c +++ b/mm/share_pool.c @@ -41,6 +41,8 @@ #include <linux/idr.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> +#include <linux/rmap.h> +#include <linux/hugetlb.h> /* access control mode macros */ #define AC_NONE 0 @@ -2399,6 +2401,94 @@ static int spa_stat_show(struct seq_file *seq, void *offset) return 0; } +vm_fault_t sharepool_no_page(struct mm_struct *mm, + struct vm_area_struct *vma, + struct address_space *mapping, pgoff_t idx, + unsigned long address, pte_t *ptep, unsigned int flags) +{ + struct hstate *h = hstate_vma(vma); + vm_fault_t ret = VM_FAULT_SIGBUS; + unsigned long size; + struct page *page; + pte_t new_pte; + spinlock_t *ptl; + unsigned long haddr = address & huge_page_mask(h); + bool new_page = false; + int err; + +retry: + page = find_lock_page(mapping, idx); + if (!page) { + size = i_size_read(mapping->host) >> huge_page_shift(h); + if (idx >= size) + goto out; + + page = alloc_huge_page(vma, haddr, 0); + if (IS_ERR(page)) { + page = alloc_huge_page_node(hstate_file(vma->vm_file), + numa_mem_id()); + if (!page) + page = ERR_PTR(-ENOMEM); + } + if (IS_ERR(page)) { + ptl = huge_pte_lock(h, mm, ptep); + if (!huge_pte_none(huge_ptep_get(ptep))) { + ret = 0; + spin_unlock(ptl); + goto out; + } + spin_unlock(ptl); + ret = vmf_error(PTR_ERR(page)); + goto out; + } + __SetPageUptodate(page); + new_page = true; + + /* sharepool pages are all shared */ + err = huge_add_to_page_cache(page, mapping, idx); + if (err) { + put_page(page); + if (err == -EEXIST) + goto retry; + goto out; + } + } + + + ptl = huge_pte_lock(h, mm, ptep); + size = i_size_read(mapping->host) >> huge_page_shift(h); + if (idx >= size) + goto backout; + + ret = 0; + if (!huge_pte_none(huge_ptep_get(ptep))) + goto backout; + + page_dup_rmap(page, true); + new_pte = make_huge_pte(vma, page, ((vma->vm_flags & VM_WRITE) + && (vma->vm_flags & VM_SHARED))); + set_huge_pte_at(mm, haddr, ptep, new_pte); + + hugetlb_count_add(pages_per_huge_page(h), mm); + + spin_unlock(ptl); + + if (new_page) { + SetPagePrivate(&page[1]); + } + + unlock_page(page); +out: + return ret; + +backout: + spin_unlock(ptl); + unlock_page(page); + put_page(page); + goto out; +} +EXPORT_SYMBOL(sharepool_no_page); + /* * Called by proc_root_init() to initialize the /proc/sharepool subtree */ -- 2.25.1

1 6

Compilation error on openeuler 2003 LTS
by Zhengyuan Liu 08 Dec '20

08 Dec '20

Building the latest 2003 LTS can trigger following compilation error: In file included from ./include/acpi/acpi_io.h:7:0, from ./include/linux/acpi.h:47, from ./arch/arm64/include/asm/machine_t.h:26, from arch/arm64/kernel/process.c:64: ./arch/arm64/include/asm/acpi.h:82:20: error: ‘enable_acpi’ redeclared as different kind of symbol static inline void enable_acpi(void) ^~~~~~~~~~~ In file included from arch/arm64/kernel/process.c:61:0: ./arch/arm64/include/asm/mpam_sched.h:46:2: note: previous definition of ‘enable_acpi’ was here enable_acpi, The redefinition was introduced by commit 7e66740ad725 ("MPAM / ACPI: Refactoring MPAM init process and set MPAM ACPI as entrance")

2 2

openEuler Summit 报名渠道开启，我们为社区内的开发者准备了很多专属环节
by public openEuler 07 Dec '20

07 Dec '20

各位社区内的开发者： openEuler Summit 所有议程 <https://openeuler.org/zh/interaction/summit-list/> 本周已经全部上线。本次峰会设置了 Linux 操作系统、云与云原生、虚拟化、基础软件、开源基础设施、安全与可信六大专场。一些通过 Call for Speaker 提交的议题已经入选。欢迎大家快来围观有没有你感兴趣的议题。 [image: image.png] 针对社区的开发者，我们准备了很多专属环节： 1. 大会为社区开发者准备了专属报名通道，在现场有专属的签到通道和定制的伴手礼。 2. 大会为Maintainer 准备了一份十分具有荣誉感和纪念意义的徽章。大会现场还准备了惊喜环节：大会主题演讲时会通过报名时你填写的 Gitee ID 和Gitee 后台的数据演示一个非常炫酷的Live Demo。我们诚挚地邀请社区开发者到现场参会。社区开发者专属报名通道：https://etherpad.openeuler.org/p/openEuler-Summit-2020 P.S：通过该链接报名的人员可以参加 Day1和 Day2 两天会议。 openEuler 也将迎来自己的一周岁生日，社区准备了一个“我为 openEuler 代言”的百人视频合集，诚挚地邀请所有热爱openEuler 的人来拍摄一段视频，来纪念openEuler这朝气蓬勃的第一年。 “我为 openEuler 代言”的视频拍摄需求：自拍形式，对着镜头回答以下问题： 1. 我想要的操作系统是****的， 2. 我在**时候参与/认识/加入/知道的openEuler（社区） 3. 我在openEuler贡献了**** 4. 我是***，我为openEuler代言使用场景：您拍摄的视频素材将用于 openEuler Summit 2020 前期的会议宣传和开场视频中。拍摄指导： 1. 背景任您喜欢 2. 人物面对镜头，位于画面中央，取半身构图，横向拍摄16:9 的画面比例 3. 最好选择白炽灯，白色光源 4. 尽量由另一个人辅助，用手机后置摄像头拍摄，不要用前置，拍摄前擦拭摄像头 5. 读口播时，声音越大越好 6. 尽量保证每一句话都是完整通顺的，中间重来的部分可以忽略 7. 视频输出，用电脑将手机视频导出，打压缩包传输即可（不要用微信上传，会压缩画质） 8. 拍摄完毕后请将视频上传到百度网盘，将分享链接期限设置为永久，然后发给 public(a)openeuler.io 9. 视频收集截止日期：2020年12月11日23:59分

1 0

[PATCH 00/58] backport stable-4.19.161
by Yang Yingliang 03 Dec '20

03 Dec '20

Alan Stern (2): USB: core: Change %pK for __user pointers to %px USB: core: Fix regression in Hercules audio card Anand K Mistry (1): x86/speculation: Fix prctl() when spectre_v2_user={seccomp,prctl},ibpb Ard Biesheuvel (1): efivarfs: revert "fix memory leak in efivarfs_create()" Benjamin Berg (1): platform/x86: thinkpad_acpi: Send tablet mode switch at wakeup time Brian Masney (1): x86/xen: don't unbind uninitialized lock_kicker_irq Chris Ye (1): HID: add HID_QUIRK_INCREMENT_USAGE_ON_DUPLICATE for Gamevice devices Cong Wang (1): netfilter: clear skb->next in NF_HOOK_LIST() Dexuan Cui (1): video: hyperv_fb: Fix the cache type when mapping the VRAM Filipe Manana (1): btrfs: fix lockdep splat when reading qgroup config on mount Frank Yang (1): HID: cypress: Support Varmilo Keyboards' media hotkeys Greg Kroah-Hartman (1): Linux 4.19.161 Hans de Goede (2): Input: i8042 - allow insmod to succeed on devices without an i8042 controller HID: Add Logitech Dinovo Edge battery quirk Hauke Mehrtens (1): wireless: Use linux/stddef.h instead of stddef.h Igor Lubashev (1): perf event: Check ref_reloc_sym before using it Jens Axboe (1): proc: don't allow async path resolution of /proc/self components Jiri Kosina (1): HID: add support for Sega Saturn Johannes Thumshirn (1): btrfs: don't access possibly stale fs_info data for printing duplicate device Julian Wiedmann (1): s390/qeth: fix tear down of async TX buffers Kai Vehmanen (1): ALSA: hda/hdmi: fix incorrect locking in hdmi_pcm_close Kaixu Xia (1): platform/x86: toshiba_acpi: Fix the wrong variable assignment Krzysztof Kozlowski (1): nfc: s3fwrn5: use signed integer for parsing GPIO numbers Laurent Pinchart (1): xtensa: uaccess: Add missing __user to strncpy_from_user() prototype Lee Duncan (1): scsi: libiscsi: Fix NOP race condition Lijun Pan (2): ibmvnic: fix NULL pointer dereference in reset_sub_crq_queues ibmvnic: fix NULL pointer dereference in ibmvic_reset_crq Lyude Paul (1): drm/atomic_helper: Stop modesets on unregistered connectors harder Marc Ferland (1): dmaengine: xilinx_dma: use readl_poll_timeout_atomic variant Marc Kleine-Budde (3): ARM: dts: dra76x: m_can: fix order of clocks can: gs_usb: fix endianess problem with candleLight firmware can: m_can: fix nominal bitiming tseg2 min for version >= 3.1 Marc Zyngier (1): phy: tegra: xusb: Fix dangling pointer on probe failure Masami Hiramatsu (1): perf probe: Fix to die_entrypc() returns error correctly Michael Chan (1): bnxt_en: Release PCI regions when DMA mask setup fails during probe. Mike Christie (1): scsi: target: iscsi: Fix cmd abort fabric stop race Minwoo Im (1): nvme: free sq/cq dbbuf pointers when dbbuf set fails Pablo Ceballos (1): HID: hid-sensor-hub: Fix issue with devices with no report ID Paolo Bonzini (2): KVM: x86: handle !lapic_in_kernel case in kvm_cpu_*_extint KVM: x86: Fix split-irqchip vs interrupt injection window request Raju Rangoju (1): cxgb4: fix the panic caused by non smac rewrite Rui Miguel Silva (1): optee: add writeback to valid memory type Sami Tolvanen (1): perf/x86: fix sysfs type mismatches Shay Agroskin (1): net: ena: set initial DMA width to avoid intel iommu issue Stanley Chu (1): scsi: ufs: Fix race between shutdown and runtime resume flow Sugar Zhang (1): dmaengine: pl330: _prep_dma_memcpy: Fix wrong burst size Taehee Yoo (1): batman-adv: set .owner to THIS_MODULE Will Deacon (2): arm64: pgtable: Fix pte_accessible() arm64: pgtable: Ensure dirty bit is preserved across pte_wrprotect() Xiaochen Shen (2): x86/resctrl: Remove superfluous kernfs_get() calls to prevent refcount leak x86/resctrl: Add necessary kernfs_put() calls to prevent refcount leak Xiongfeng Wang (1): IB/mthca: fix return value of error branch in mthca_init_cq() Zenghui Yu (1): KVM: arm64: vgic-v3: Drop the reporting of GICR_TYPER.Last for userspace Zhang Changzhong (2): bnxt_en: fix error return code in bnxt_init_one() bnxt_en: fix error return code in bnxt_init_board() Zhang Qilong (2): usb: gadget: f_midi: Fix memleak in f_midi_alloc usb: gadget: Fix memleak in gadgetfs_fill_super penghao (1): USB: quirks: Add USB_QUIRK_DISCONNECT_SUSPEND quirk for Lenovo A630Z TIO built-in usb-audio card Makefile | 2 +- arch/arm/boot/dts/dra76x.dtsi | 4 +- arch/arm64/include/asm/pgtable.h | 34 ++--- arch/x86/events/intel/cstate.c | 6 +- arch/x86/events/intel/rapl.c | 14 +- arch/x86/events/intel/uncore.c | 4 +- arch/x86/events/intel/uncore.h | 12 +- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kernel/cpu/bugs.c | 4 +- arch/x86/kernel/cpu/intel_rdt_rdtgroup.c | 65 ++++----- arch/x86/kvm/irq.c | 85 +++++------- arch/x86/kvm/lapic.c | 2 +- arch/x86/kvm/x86.c | 18 +-- arch/x86/xen/spinlock.c | 12 +- arch/xtensa/include/asm/uaccess.h | 2 +- drivers/dma/pl330.c | 2 +- drivers/dma/xilinx/xilinx_dma.c | 4 +- drivers/gpu/drm/drm_atomic.c | 21 --- drivers/gpu/drm/drm_atomic_helper.c | 21 ++- drivers/gpu/drm/drm_connector.c | 11 +- drivers/gpu/drm/i915/intel_dp_mst.c | 8 +- drivers/hid/hid-cypress.c | 44 +++++- drivers/hid/hid-ids.h | 8 ++ drivers/hid/hid-input.c | 3 + drivers/hid/hid-quirks.c | 5 + drivers/hid/hid-sensor-hub.c | 3 +- drivers/infiniband/hw/mthca/mthca_cq.c | 10 +- drivers/input/serio/i8042.c | 12 +- drivers/net/can/m_can/m_can.c | 2 +- drivers/net/can/usb/gs_usb.c | 131 ++++++++++-------- drivers/net/ethernet/amazon/ena/ena_netdev.c | 17 ++- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 4 +- .../net/ethernet/chelsio/cxgb4/cxgb4_filter.c | 3 +- drivers/net/ethernet/ibm/ibmvnic.c | 6 + drivers/nfc/s3fwrn5/i2c.c | 4 +- drivers/nvme/host/pci.c | 15 ++ drivers/phy/tegra/xusb.c | 1 + drivers/platform/x86/thinkpad_acpi.c | 1 + drivers/platform/x86/toshiba_acpi.c | 3 +- drivers/s390/net/qeth_core_main.c | 6 - drivers/scsi/libiscsi.c | 23 +-- drivers/scsi/ufs/ufshcd.c | 6 +- drivers/target/iscsi/iscsi_target.c | 17 ++- drivers/tee/optee/call.c | 3 +- drivers/usb/core/devio.c | 14 +- drivers/usb/core/quirks.c | 10 ++ drivers/usb/gadget/function/f_midi.c | 10 +- drivers/usb/gadget/legacy/inode.c | 3 + drivers/video/fbdev/hyperv_fb.c | 7 +- fs/btrfs/qgroup.c | 2 +- fs/btrfs/volumes.c | 8 +- fs/efivarfs/inode.c | 2 + fs/efivarfs/super.c | 1 - fs/proc/self.c | 7 + include/drm/drm_connector.h | 71 +++++++++- include/linux/netfilter.h | 2 +- include/scsi/libiscsi.h | 3 + include/uapi/linux/wireless.h | 6 +- net/batman-adv/log.c | 1 + sound/pci/hda/patch_hdmi.c | 20 +-- tools/perf/util/dwarf-aux.c | 8 ++ tools/perf/util/event.c | 7 +- virt/kvm/arm/vgic/vgic-mmio-v3.c | 22 ++- 63 files changed, 540 insertions(+), 323 deletions(-) -- 2.25.1

1 49

[PATCH 00/87] backport patches from linux-4.19.y for openEuler-1.0-LTS
by Yang Yingliang 02 Dec '20

02 Dec '20

Al Viro (1): don't dump the threads that had been already exiting when zapped. Aleksandr Nogikh (1): netem: fix zero division in tabledist Andy Shevchenko (2): device property: Keep secondary firmware node secondary by type device property: Don't clear secondary pointer for shared primary firmware node Arjun Roy (1): tcp: Prevent low rmem stalls with SO_RCVLOWAT. Arnaldo Carvalho de Melo (1): perf scripting python: Avoid declaring function pointers with a visibility attribute Boris Protopopov (1): Convert trailing spaces and periods in path components Brian Foster (1): xfs: flush new eof page on truncate to avoid post-eof corruption Chen Zhou (1): selinux: Fix error return code in sel_ib_pkey_sid_slow() Christoph Hellwig (2): nbd: fix a block_device refcount leak in nbd_release xfs: fix a missing unlock on error in xfs_fs_map_blocks Chunyan Zhang (1): tick/common: Touch watchdog in tick_unfreeze() on all CPUs Dan Carpenter (1): futex: Don't enable IRQs unconditionally in put_pi_state() Darrick J. Wong (12): xfs: fix realtime bitmap/summary file truncation when growing rt volume xfs: don't free rt blocks when we're doing a REMAP bunmapi call xfs: set xefi_discard when creating a deferred agfl free log intent item xfs: fix scrub flagging rtinherit even if there is no rt device xfs: fix flags argument to rmap lookup when converting shared file rmaps xfs: set the unwritten bit in rmap lookup flags in xchk_bmap_get_rmapextents xfs: fix rmap key and record comparison functions xfs: fix brainos in the refcount scrubber's rmap fragment processor vfs: remove lockdep bogosity in __sb_start_write xfs: fix the minrecs logic when dealing with inode root child blocks xfs: strengthen rmap record flags checking xfs: revert "xfs: fix rmap key and record comparison functions" Dinghao Liu (1): ext4: fix error handling code in add_new_gdb Dongli Zhang (1): page_frag: Recover from memory pressure Douglas Gilbert (1): sgl_alloc_order: fix memory leak Eddy Wu (1): fork: fix copy_process(CLONE_PARENT) race with the exiting ->real_parent Eric Biggers (1): ext4: fix leaking sysfs kobject after failed mount Gabriel Krisman Bertazi (2): blk-cgroup: Fix memleak on error path blk-cgroup: Pre-allocate tree node on blkg_conf_prep George Spelvin (1): random32: make prandom_u32() output unpredictable Gerald Schaefer (1): mm/userfaultfd: do not access vma->vm_mm after calling handle_userfault() Heiner Kallweit (1): net: bridge: add missing counters to ndo_get_stats64 callback Jan Kara (2): ext4: Detect already used quota file early ext4: fix bogus warning in ext4_update_dx_flag() Jason A. Donenfeld (1): netfilter: use actual socket sk rather than skb sk when routing harder Jiri Olsa (2): perf python scripting: Fix printable strings in python3 scripts perf tools: Add missing swap for ino_generation Jonathan Cameron (1): ACPI: Add out of bounds and numa_off protections to pxm_to_node() Joseph Qi (1): ext4: unlock xattr_sem properly in ext4_inline_data_truncate() Kaixu Xia (1): ext4: correctly report "not supported" for {usr, grp}jquota when !CONFIG_QUOTA Lang Dai (1): uio: free uio id after uio file node is freed Lee Jones (1): Fonts: Replace discarded const qualifier Leo Yan (1): perf lock: Don't free "lock_seq_stat" if read_count isn't zero Luo Meng (2): ext4: fix invalid inode checksum fail_function: Remove a redundant mutex unlock Mao Wenan (1): net: Update window_clamp if SOCK_RCVBUF is set Marc Zyngier (1): arm64: Run ARCH_WORKAROUND_1 enabling code on all CPUs Mateusz Nosek (1): futex: Fix incorrect should_fail_futex() handling Matteo Croce (2): Revert "kernel/reboot.c: convert simple_strtoul to kstrtoint" reboot: fix overflow parsing reboot cpu number Michael Schaller (1): efivarfs: Replace invalid slashes with exclamation marks in dentries. Mike Galbraith (1): futex: Handle transient "ownerless" rtmutex state correctly Miklos Szeredi (1): fuse: fix page dereference after free Ming Lei (1): scsi: core: Don't start concurrent async scan on same host Nicholas Piggin (1): mm: fix exec activate_mm vs TLB shootdown and lazy tlb switching race Oleg Nesterov (1): ptrace: fix task_join_group_stop() for the case when current is traced Oliver Herms (1): IPv6: Set SIT tunnel hard_header_len to zero Peter Zijlstra (2): serial: pl011: Fix lockdep splat when handling magic-sysrq interrupt perf: Fix get_recursion_context() Qiujun Huang (2): ring-buffer: Return 0 on success from ring_buffer_resize() tracing: Fix out of bounds write in get_trace_buf Ronnie Sahlberg (1): cifs: handle -EINTR in cifs_setattr Ryan Sharpelletti (1): tcp: only postpone PROBE_RTT if RTT is < current min_rtt estimate Shijie Luo (1): mm: mempolicy: fix potential pte_unmap_unlock pte error Shin'ichiro Kawasaki (1): uio: Fix use-after-free in uio_unregister_device() Stefano Brivio (1): netfilter: ipset: Update byte and packet counters regardless of whether they match Steven Rostedt (VMware) (3): ring-buffer: Fix recursion protection transitions between interrupt context ftrace: Fix recursion check for NMI test ftrace: Handle tracing when switching between context Tyler Hicks (1): tpm: efi: Don't create binary_bios_measurements file for an empty log Valentin Schneider (1): arm64: topology: Stop using MPIDR for topology information Vamshi K Sthambamkadi (1): efivarfs: fix memory leak in efivarfs_create() Wang Hai (2): devlink: Add missing genlmsg_cancel() in devlink_nl_sb_port_pool_fill() inet_diag: Fix error path to cancel the meseage in inet_req_diag_fill() Wengang Wang (1): ocfs2: initialize ip_next_orphan Will Deacon (1): arm64: psci: Avoid printing in cpu_psci_cpu_die() Xiubo Li (1): nbd: make the config put is called before the notifying the waiter Yi-Hung Wei (1): ip_tunnels: Set tunnel option flag when tunnel metadata is present Yicong Yang (1): libfs: fix error cast of negative value in simple_attr_write() Yunsheng Lin (1): net: sch_generic: fix the missing new qdisc assignment bug Zeng Tao (1): time: Prevent undefined behaviour in timespec64_to_ns() Zhengyuan Liu (1): arm64/mm: return cpu_all_mask when node is NUMA_NO_NODE Zqiang (1): kthread_worker: prevent queuing delayed work from timer_fn when it is being canceled zhuoliang zhang (1): net: xfrm: fix a race condition during allocing spi arch/Kconfig | 7 + arch/arm64/include/asm/numa.h | 3 + arch/arm64/kernel/cpu_errata.c | 8 + arch/arm64/kernel/psci.c | 5 +- arch/arm64/kernel/topology.c | 43 +- arch/arm64/mm/numa.c | 6 +- block/blk-cgroup.c | 15 +- drivers/acpi/numa.c | 2 +- drivers/base/core.c | 4 +- drivers/block/nbd.c | 3 +- drivers/char/random.c | 1 - drivers/char/tpm/eventlog/efi.c | 5 + drivers/net/geneve.c | 3 +- drivers/scsi/scsi_scan.c | 7 +- drivers/tty/serial/amba-pl011.c | 11 +- drivers/uio/uio.c | 12 +- fs/cifs/cifs_unicode.c | 8 +- fs/cifs/inode.c | 13 +- fs/efivarfs/super.c | 4 + fs/exec.c | 15 +- fs/ext4/ext4.h | 3 +- fs/ext4/inline.c | 1 + fs/ext4/inode.c | 11 +- fs/ext4/resize.c | 4 +- fs/ext4/super.c | 10 +- fs/fuse/dev.c | 28 +- fs/libfs.c | 6 +- fs/ocfs2/super.c | 1 + fs/super.c | 33 +- fs/xfs/libxfs/xfs_alloc.c | 1 + fs/xfs/libxfs/xfs_bmap.c | 19 +- fs/xfs/libxfs/xfs_bmap.h | 2 +- fs/xfs/libxfs/xfs_rmap.c | 2 +- fs/xfs/scrub/bmap.c | 10 +- fs/xfs/scrub/btree.c | 45 +- fs/xfs/scrub/inode.c | 3 +- fs/xfs/scrub/refcount.c | 8 +- fs/xfs/xfs_iops.c | 10 + fs/xfs/xfs_pnfs.c | 2 +- fs/xfs/xfs_rtalloc.c | 10 +- include/linux/netfilter_ipv4.h | 2 +- include/linux/netfilter_ipv6.h | 2 +- include/linux/prandom.h | 36 +- include/linux/time64.h | 4 + include/net/ip_tunnels.h | 7 +- kernel/events/internal.h | 2 +- kernel/exit.c | 5 +- kernel/fail_function.c | 5 +- kernel/fork.c | 10 +- kernel/futex.c | 25 +- kernel/kthread.c | 3 +- kernel/reboot.c | 28 +- kernel/signal.c | 19 +- kernel/time/itimer.c | 4 - kernel/time/tick-common.c | 2 + kernel/time/timer.c | 7 - kernel/trace/ring_buffer.c | 66 ++- kernel/trace/trace.c | 2 +- kernel/trace/trace.h | 26 +- kernel/trace/trace_selftest.c | 9 +- lib/fonts/font_10x18.c | 2 +- lib/fonts/font_6x10.c | 2 +- lib/fonts/font_6x11.c | 2 +- lib/fonts/font_7x14.c | 2 +- lib/fonts/font_8x16.c | 2 +- lib/fonts/font_8x8.c | 2 +- lib/fonts/font_acorn_8x8.c | 2 +- lib/fonts/font_mini_4x6.c | 2 +- lib/fonts/font_pearl_8x8.c | 2 +- lib/fonts/font_sun12x22.c | 2 +- lib/fonts/font_sun8x16.c | 2 +- lib/random32.c | 462 +++++++++++------- lib/scatterlist.c | 2 +- mm/huge_memory.c | 9 +- mm/mempolicy.c | 6 +- mm/page_alloc.c | 5 + net/bridge/br_device.c | 1 + net/core/devlink.c | 6 +- net/ipv4/inet_diag.c | 4 +- net/ipv4/netfilter.c | 12 +- net/ipv4/netfilter/ipt_SYNPROXY.c | 2 +- net/ipv4/netfilter/iptable_mangle.c | 2 +- net/ipv4/netfilter/nf_nat_l3proto_ipv4.c | 2 +- net/ipv4/netfilter/nf_reject_ipv4.c | 2 +- net/ipv4/netfilter/nft_chain_route_ipv4.c | 2 +- net/ipv4/syncookies.c | 9 +- net/ipv4/tcp.c | 2 + net/ipv4/tcp_bbr.c | 2 +- net/ipv4/tcp_input.c | 3 +- net/ipv6/netfilter.c | 6 +- net/ipv6/netfilter/ip6table_mangle.c | 2 +- net/ipv6/netfilter/nf_nat_l3proto_ipv6.c | 2 +- net/ipv6/netfilter/nft_chain_route_ipv6.c | 2 +- net/ipv6/sit.c | 2 - net/ipv6/syncookies.c | 10 +- net/netfilter/ipset/ip_set_core.c | 3 +- net/netfilter/ipvs/ip_vs_core.c | 4 +- net/sched/sch_generic.c | 3 + net/sched/sch_netem.c | 9 +- net/xfrm/xfrm_state.c | 8 +- security/selinux/ibpkey.c | 4 +- tools/perf/builtin-lock.c | 2 +- tools/perf/util/print_binary.c | 2 +- .../scripting-engines/trace-event-python.c | 7 +- tools/perf/util/session.c | 1 + 105 files changed, 809 insertions(+), 461 deletions(-) -- 2.25.1

1 87

Re: openeuler patch tag咨询
by Xie XiuQi 02 Dec '20

02 Dec '20

Hi, On 2020/12/2 11:12, Zhengyuan Liu wrote: > > > On 2020/12/2 上午10:59, Yang Yingliang wrote: >> >> On 2020/12/2 10:47, Zhengyuan Liu wrote: >>> Hi, Xiuqi >>> >>> 最近在　backport OpenEuler内核补丁时对于补丁来源有个疑问，每个Patch　来源类别都在 >>> commit log　里面标注了，如 mainline inclusion, stable inclusion, 这两个比较好理解．但是像： >>> hulk >>> ascend >>> driver >>> euleros >>> 这几个我查阅openeuler开发文档也没找到说明，　所以想请问下你们这几个是怎么做区分的． >>> 这个开发者在向openeuler社区提补丁时也用得到，方便的话麻烦你们更新到文档里去．　这个确实是存在一些问题，这个 tag 的划分，既没有覆盖全，划分的标准也不完全一致。当前是给的比较主观的一些 tag，既有按领域的，也有按来源的,也有按子系统的。总之目的是为了方便识别和统计。下面表格是我准备发到社区讨论的，也请大家帮忙看看。 | 补丁 tag | 说明 | | ------------------- | ----------------------------------- | | mainline inclusion | 主线 backport | | stable inclusion | 上游 stable 分支的补丁 | | maillist inclusion | 发到社区，但是还没进主线的补丁 | | hulk inclusion | openEuler kernel 团队开发的补丁 | | virt inclusion | 虚拟化相关的补丁 | | ascend inclusion | 昇腾相关的补丁或特性 | | kunpeng inclusion | 鲲鹏相关的补丁或特性 | | dist inclusion | 从其他发行版移植的补丁 | | openEuler inclusion | 其他发送到openEuler社区的非上游补丁 | >> >> 非主线的补丁会用到上面的分类，只要补丁是来自主线的统一用mainline >> >> ascend是支持昇腾芯片的一些特有补丁 >> >> driver是驱动补丁 >> >> euleros和hulk是一类，主要是一些未进主线的或者待进主线的通用补丁 > > 感谢回复． > > driver相当于从hulk一类单独挑出来的是吧．从现有的分类来看，好像也没有严格的区分． > > >> >> >>> >>> 谢谢． >>> _______________________________________________ >>> Kernel mailing list -- kernel(a)openeuler.org >>> To unsubscribe send an email to kernel-leave(a)openeuler.org >> > >

1 0

openeuler patch tag咨询
by Zhengyuan Liu 02 Dec '20

02 Dec '20

Hi, Xiuqi 最近在　backport OpenEuler内核补丁时对于补丁来源有个疑问，每个Patch　来源类别都在 commit log　里面标注了，如 mainline inclusion, stable inclusion, 这两个比较好理解．但是像： hulk ascend driver euleros 这几个我查阅openeuler开发文档也没找到说明，　所以想请问下你们这几个是怎么做区分的．这个开发者在向openeuler社区提补丁时也用得到，方便的话麻烦你们更新到文档里去．　谢谢．

3 3

bonding: add an option to specify a delay between peer notifications
by liaichun 02 Dec '20

02 Dec '20

From: Vincent Bernat <vincent(a)bernat.ch> mainline inclusion from mainline-5.5-rc5 commit 07a4ddec3ce9b0a533b5f90f582f1057390d5e63 category: bugfix issue: I27SUX bugzilla: NA ------------------------------ Currently, gratuitous ARP/ND packets are sent every `miimon' milliseconds. This commit allows a user to specify a custom delay through a new option, `peer_notif_delay'. Like for `updelay' and `downdelay', this delay should be a multiple of `miimon' to avoid managing an additional work queue. The configuration logic is copied from `updelay' and `downdelay'. However, the default value cannot be set using a module parameter: Netlink or sysfs should be used to configure this feature. When setting `miimon' to 100 and `peer_notif_delay' to 500, we can observe the 500 ms delay is respected: 20:30:19.354693 ARP, Request who-has 203.0.113.10 tell 203.0.113.10, length 28 20:30:19.874892 ARP, Request who-has 203.0.113.10 tell 203.0.113.10, length 28 20:30:20.394919 ARP, Request who-has 203.0.113.10 tell 203.0.113.10, length 28 20:30:20.914963 ARP, Request who-has 203.0.113.10 tell 203.0.113.10, length 28 In bond_mii_monitor(), I have tried to keep the lock logic readable. The change is due to the fact we cannot rely on a notification to lower the value of `bond->send_peer_notif' as `NETDEV_NOTIFY_PEERS' is only triggered once every N times, while we need to decrement the counter each time. iproute2 also needs to be updated to be able to specify this new attribute through `ip link'. Signed-off-by: Vincent Bernat <vincent(a)bernat.ch> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: wangli <wangli221(a)huawei.com> Signed-off-by: Aichun Li <liaichun(a)huawei.com> --- drivers/net/bonding/bond_main.c | 31 +-- drivers/net/bonding/bond_netlink.c | 14 ++ drivers/net/bonding/bond_options.c | 75 +++++--- drivers/net/bonding/bond_procfs.c | 3 +- drivers/net/bonding/bond_sysfs.c | 13 ++ include/net/bond_options.h | 1 + include/net/bonding.h | 3 +- include/uapi/linux/if_link.h | 1 + tools/include/uapi/linux/if_link.h | 1 + 9 files changed, 102 insertions(+), 40 deletions(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index a59333b..b1f3520 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -786,6 +786,8 @@ static bool bond_should_notify_peers(struct bonding *bond) slave ? slave->dev->name : "NULL"); if (!slave || !bond->send_peer_notif || + (bond->send_peer_notif % + max(1, bond->params.peer_notif_delay) != 0) || !netif_carrier_ok(bond->dev) || test_bit(__LINK_STATE_LINKWATCH_PENDING, &slave->dev->state)) return false; @@ -878,15 +880,18 @@ void bond_change_active_slave(struct bonding *bond, struct slave *new_active) if (netif_running(bond->dev)) { bond->send_peer_notif = - bond->params.num_peer_notif; + bond->params.num_peer_notif * + max(1, bond->params.peer_notif_delay); should_notify_peers = bond_should_notify_peers(bond); } call_netdevice_notifiers(NETDEV_BONDING_FAILOVER, bond->dev); - if (should_notify_peers) + if (should_notify_peers) { + bond->send_peer_notif--; call_netdevice_notifiers(NETDEV_NOTIFY_PEERS, bond->dev); + } } } @@ -2290,6 +2295,7 @@ static void bond_mii_monitor(struct work_struct *work) struct bonding *bond = container_of(work, struct bonding, mii_work.work); bool should_notify_peers = false; + bool commit; unsigned long delay; struct slave *slave; struct list_head *iter; @@ -2302,10 +2308,18 @@ static void bond_mii_monitor(struct work_struct *work) rcu_read_lock(); should_notify_peers = bond_should_notify_peers(bond); - - if (bond_miimon_inspect(bond)) { + commit = !!bond_miimon_inspect(bond); + if (bond->send_peer_notif) { + rcu_read_unlock(); + if (rtnl_trylock()) { + bond->send_peer_notif--; + rtnl_unlock(); + } + } else { rcu_read_unlock(); + } + if (commit) { /* Race avoidance with bond_close cancel of workqueue */ if (!rtnl_trylock()) { delay = 1; @@ -2319,9 +2333,7 @@ static void bond_mii_monitor(struct work_struct *work) bond_miimon_commit(bond); rtnl_unlock(); /* might sleep, hold no other locks */ - } else - rcu_read_unlock(); - + } re_arm: if (bond->params.miimon) queue_delayed_work(bond->wq, &bond->mii_work, delay); @@ -3093,10 +3105,6 @@ static int bond_master_netdev_event(unsigned long event, case NETDEV_REGISTER: bond_create_proc_entry(event_bond); break; - case NETDEV_NOTIFY_PEERS: - if (event_bond->send_peer_notif) - event_bond->send_peer_notif--; - break; default: break; } @@ -4743,6 +4751,7 @@ static int bond_check_params(struct bond_params *params) params->arp_all_targets = arp_all_targets_value; params->updelay = updelay; params->downdelay = downdelay; + params->peer_notif_delay = 0; params->use_carrier = use_carrier; params->lacp_fast = lacp_fast; params->primary[0] = 0; diff --git a/drivers/net/bonding/bond_netlink.c b/drivers/net/bonding/bond_netlink.c index fbcd8a7..ae8e4e4 100644 --- a/drivers/net/bonding/bond_netlink.c +++ b/drivers/net/bonding/bond_netlink.c @@ -112,6 +112,7 @@ static int bond_fill_slave_info(struct sk_buff *skb, [IFLA_BOND_AD_ACTOR_SYSTEM] = { .type = NLA_BINARY, .len = ETH_ALEN }, [IFLA_BOND_TLB_DYNAMIC_LB] = { .type = NLA_U8 }, + [IFLA_BOND_PEER_NOTIF_DELAY] = { .type = NLA_U32 }, }; static const struct nla_policy bond_slave_policy[IFLA_BOND_SLAVE_MAX + 1] = { @@ -219,6 +220,14 @@ static int bond_changelink(struct net_device *bond_dev, struct nlattr *tb[], if (err) return err; } + if (data[IFLA_BOND_PEER_NOTIF_DELAY]) { + int delay = nla_get_u32(data[IFLA_BOND_PEER_NOTIF_DELAY]); + + bond_opt_initval(&newval, delay); + err = __bond_opt_set(bond, BOND_OPT_PEER_NOTIF_DELAY, &newval); + if (err) + return err; + } if (data[IFLA_BOND_USE_CARRIER]) { int use_carrier = nla_get_u8(data[IFLA_BOND_USE_CARRIER]); @@ -497,6 +506,7 @@ static size_t bond_get_size(const struct net_device *bond_dev) nla_total_size(sizeof(u16)) + /* IFLA_BOND_AD_USER_PORT_KEY */ nla_total_size(ETH_ALEN) + /* IFLA_BOND_AD_ACTOR_SYSTEM */ nla_total_size(sizeof(u8)) + /* IFLA_BOND_TLB_DYNAMIC_LB */ + nla_total_size(sizeof(u32)) + /* IFLA_BOND_PEER_NOTIF_DELAY */ 0; } @@ -539,6 +549,10 @@ static int bond_fill_info(struct sk_buff *skb, bond->params.downdelay * bond->params.miimon)) goto nla_put_failure; + if (nla_put_u32(skb, IFLA_BOND_PEER_NOTIF_DELAY, + bond->params.peer_notif_delay * bond->params.miimon)) + goto nla_put_failure; + if (nla_put_u8(skb, IFLA_BOND_USE_CARRIER, bond->params.use_carrier)) goto nla_put_failure; diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c index 80867bd..ceae69c 100644 --- a/drivers/net/bonding/bond_options.c +++ b/drivers/net/bonding/bond_options.c @@ -28,6 +28,8 @@ static int bond_option_updelay_set(struct bonding *bond, const struct bond_opt_value *newval); static int bond_option_downdelay_set(struct bonding *bond, const struct bond_opt_value *newval); +static int bond_option_peer_notif_delay_set(struct bonding *bond, + const struct bond_opt_value *newval); static int bond_option_use_carrier_set(struct bonding *bond, const struct bond_opt_value *newval); static int bond_option_arp_interval_set(struct bonding *bond, @@ -78,7 +80,10 @@ static int bond_option_ad_actor_system_set(struct bonding *bond, const struct bond_opt_value *newval); static int bond_option_ad_user_port_key_set(struct bonding *bond, const struct bond_opt_value *newval); - +static int _bond_option_delay_set(struct bonding *bond, + const struct bond_opt_value *newval, + const char *name, + int *target); static const struct bond_opt_value bond_mode_tbl[] = { { "balance-rr", BOND_MODE_ROUNDROBIN, BOND_VALFLAG_DEFAULT}, @@ -428,6 +433,13 @@ static int bond_option_ad_user_port_key_set(struct bonding *bond, .desc = "Number of peer notifications to send on failover event", .values = bond_num_peer_notif_tbl, .set = bond_option_num_peer_notif_set + }, + [BOND_OPT_PEER_NOTIF_DELAY] = { + .id = BOND_OPT_PEER_NOTIF_DELAY, + .name = "peer_notif_delay", + .desc = "Delay between each peer notification on failover event, in milliseconds", + .values = bond_intmax_tbl, + .set = bond_option_peer_notif_delay_set } }; @@ -850,6 +862,9 @@ static int bond_option_miimon_set(struct bonding *bond, if (bond->params.downdelay) netdev_dbg(bond->dev, "Note: Updating downdelay (to %d) since it is a multiple of the miimon value\n", bond->params.downdelay * bond->params.miimon); + if (bond->params.peer_notif_delay) + netdev_dbg(bond->dev, "Note: Updating peer_notif_delay (to %d) since it is a multiple of the miimon value\n", + bond->params.peer_notif_delay * bond->params.miimon); if (newval->value && bond->params.arp_interval) { netdev_dbg(bond->dev, "MII monitoring cannot be used with ARP monitoring - disabling ARP monitoring...\n"); bond->params.arp_interval = 0; @@ -873,52 +888,58 @@ static int bond_option_miimon_set(struct bonding *bond, return 0; } -/* Set up and down delays. These must be multiples of the - * MII monitoring value, and are stored internally as the multiplier. - * Thus, we must translate to MS for the real world. +/* Set up, down and peer notification delays. These must be multiples + * of the MII monitoring value, and are stored internally as the + * multiplier.Thus, we must translate to MS for the real world. */ -static int bond_option_updelay_set(struct bonding *bond, - const struct bond_opt_value *newval) +static int _bond_option_delay_set(struct bonding *bond, + const struct bond_opt_value *newval, + const char *name, + int *target) { int value = newval->value; if (!bond->params.miimon) { - netdev_err(bond->dev, "Unable to set up delay as MII monitoring is disabled\n"); + netdev_err(bond->dev, "Unable to set %s as MII monitoring is disabled\n", + name); return -EPERM; } if ((value % bond->params.miimon) != 0) { - netdev_warn(bond->dev, "up delay (%d) is not a multiple of miimon (%d), updelay rounded to %d ms\n", + netdev_warn(bond->dev, + "%s (%d) is not a multiple of miimon (%d), value rounded to %d ms\n", + name, value, bond->params.miimon, (value / bond->params.miimon) * bond->params.miimon); } - bond->params.updelay = value / bond->params.miimon; - netdev_dbg(bond->dev, "Setting up delay to %d\n", - bond->params.updelay * bond->params.miimon); + *target = value / bond->params.miimon; + netdev_dbg(bond->dev, "Setting %s to %d\n", + name, + *target * bond->params.miimon); return 0; } +static int bond_option_updelay_set(struct bonding *bond, + const struct bond_opt_value *newval) +{ + return _bond_option_delay_set(bond, newval, "up delay", + &bond->params.updelay); +} + static int bond_option_downdelay_set(struct bonding *bond, const struct bond_opt_value *newval) { - int value = newval->value; - - if (!bond->params.miimon) { - netdev_err(bond->dev, "Unable to set down delay as MII monitoring is disabled\n"); - return -EPERM; - } - if ((value % bond->params.miimon) != 0) { - netdev_warn(bond->dev, "down delay (%d) is not a multiple of miimon (%d), delay rounded to %d ms\n", - value, bond->params.miimon, - (value / bond->params.miimon) * - bond->params.miimon); - } - bond->params.downdelay = value / bond->params.miimon; - netdev_dbg(bond->dev, "Setting down delay to %d\n", - bond->params.downdelay * bond->params.miimon); + return _bond_option_delay_set(bond, newval, "down delay", + &bond->params.downdelay); +} - return 0; +static int bond_option_peer_notif_delay_set(struct bonding *bond, + const struct bond_opt_value *newval) +{ + return _bond_option_delay_set(bond, newval, + "peer notification delay", + &bond->params.peer_notif_delay); } static int bond_option_use_carrier_set(struct bonding *bond, diff --git a/drivers/net/bonding/bond_procfs.c b/drivers/net/bonding/bond_procfs.c index 9f7d83e..850dde4 100644 --- a/drivers/net/bonding/bond_procfs.c +++ b/drivers/net/bonding/bond_procfs.c @@ -104,7 +104,8 @@ static void bond_info_show_master(struct seq_file *seq) bond->params.updelay * bond->params.miimon); seq_printf(seq, "Down Delay (ms): %d\n", bond->params.downdelay * bond->params.miimon); - + seq_printf(seq, "Peer Notification Delay (ms): %d\n", + bond->params.peer_notif_delay * bond->params.miimon); /* ARP information */ if (bond->params.arp_interval > 0) { diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c index 3584725..0ecafe1 100644 --- a/drivers/net/bonding/bond_sysfs.c +++ b/drivers/net/bonding/bond_sysfs.c @@ -343,6 +343,18 @@ static ssize_t bonding_show_updelay(struct device *d, static DEVICE_ATTR(updelay, 0644, bonding_show_updelay, bonding_sysfs_store_option); +static ssize_t bonding_show_peer_notif_delay(struct device *d, + struct device_attribute *attr, + char *buf) +{ + struct bonding *bond = to_bond(d); + + return sprintf(buf, "%d\n", + bond->params.peer_notif_delay * bond->params.miimon); +} +static DEVICE_ATTR(peer_notif_delay, 0644, + bonding_show_peer_notif_delay, bonding_sysfs_store_option); + /* Show the LACP interval. */ static ssize_t bonding_show_lacp(struct device *d, struct device_attribute *attr, @@ -734,6 +746,7 @@ static DEVICE_ATTR(ad_user_port_key, 0644, &dev_attr_arp_ip_target.attr, &dev_attr_downdelay.attr, &dev_attr_updelay.attr, + &dev_attr_peer_notif_delay.attr, &dev_attr_lacp_rate.attr, &dev_attr_ad_select.attr, &dev_attr_xmit_hash_policy.attr, diff --git a/include/net/bond_options.h b/include/net/bond_options.h index d79d28f..6889a04 100644 --- a/include/net/bond_options.h +++ b/include/net/bond_options.h @@ -67,6 +67,7 @@ enum { BOND_OPT_AD_ACTOR_SYSTEM, BOND_OPT_AD_USER_PORT_KEY, BOND_OPT_NUM_PEER_NOTIF_ALIAS, + BOND_OPT_PEER_NOTIF_DELAY, BOND_OPT_LAST }; diff --git a/include/net/bonding.h b/include/net/bonding.h index 81166488..c241010 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -114,6 +114,7 @@ struct bond_params { int fail_over_mac; int updelay; int downdelay; + int peer_notif_delay; int lacp_fast; unsigned int min_links; int ad_select; @@ -205,7 +206,7 @@ struct bonding { */ spinlock_t mode_lock; spinlock_t stats_lock; - u8 send_peer_notif; + u64 send_peer_notif; u8 igmp_retrans; #ifdef CONFIG_PROC_FS struct proc_dir_entry *proc_entry; diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h index 9d8eac5..f9a0359 100644 --- a/include/uapi/linux/if_link.h +++ b/include/uapi/linux/if_link.h @@ -615,6 +615,7 @@ enum { IFLA_BOND_AD_USER_PORT_KEY, IFLA_BOND_AD_ACTOR_SYSTEM, IFLA_BOND_TLB_DYNAMIC_LB, + IFLA_BOND_PEER_NOTIF_DELAY, __IFLA_BOND_MAX, }; diff --git a/tools/include/uapi/linux/if_link.h b/tools/include/uapi/linux/if_link.h index 43391e2..f57af58 100644 --- a/tools/include/uapi/linux/if_link.h +++ b/tools/include/uapi/linux/if_link.h @@ -614,6 +614,7 @@ enum { IFLA_BOND_AD_USER_PORT_KEY, IFLA_BOND_AD_ACTOR_SYSTEM, IFLA_BOND_TLB_DYNAMIC_LB, + IFLA_BOND_PEER_NOTIF_DELAY, __IFLA_BOND_MAX, }; -- 1.8.3.1

1 0

[PATCH 1/6] share_pool: Add sp_area cache
by Yang Yingliang 02 Dec '20

02 Dec '20

From: Tang Yizhou <tangyizhou(a)huawei.com> ascend inclusion category: feature bugzilla: NA CVE: NA ------------------------------------------------- Provide a free area cache for the share pool VA allocator, based on the algorithm used by linux mainline commit 89699605fe7c. This reduces the number of rbtree operations and linear traversals over the share pool extents in order to find a free area, by starting off at the last point that a free area was found. The free area cache is reset if areas are freed behind it, or if we are searching for an another area (such as DVPP 16G area) than last time. So allocation patterns are not changed. After this patch, the search will start from where it left off, giving closer to an amortized O(1). Test environment: Huawei 1951 DC (8 CPU cores) with 21G memory, no load. Test method: A single thread process first call sp_alloc() to allocate a specified number of 2M hugepages, then we calculate the allocation time when sp_alloc() another 2M hugepage. The results are in microsecond. test 1, first sp_alloc() 256 2M-hugepage, total 512M test 2, first sp_alloc() 512 2M-hugepage, total 1G test 3, first sp_alloc() 1024 2M-hugepage, total 2G test 4, first sp_alloc() 1536 2M-hugepage, total 3G test 5, first sp_alloc() 2048 2M-hugepage, total 4G test 6, first sp_alloc() 4096 2M-hugepage, total 8G test 7, first sp_alloc() 6072 2M-hugepage, total 12G test 8, first sp_alloc() 8192 2M-hugepage, total 16G test1 test2 test3 test4 231 238 240 252 279 253 315 268 242 238 247 253 282 255 326 265 233 234 250 243 272 251 314 258 239 224 245 246 273 261 324 262 234 233 252 257 277 262 326 265 225 231 243 243 279 249 325 264 236 261 246 248 265 262 323 266 233 238 247 246 281 259 331 265 239 222 243 241 270 248 325 263 241 231 239 241 335 246 321 268 avg: 235.3 235 245.2 247 281.3 254.6 323 264.4 res: - - 9.49% 18.14% test5 test6 test7 test8 371 280 720 458 1001 629 1547 909 369 283 691 465 1005 718 1533 903 374 279 954 470 1003 680 1371 908 363 279 697 457 1004 923 1375 930 369 286 711 464 1016 683 1395 1083 382 280 967 491 1029 695 1413 1096 378 284 688 823 1008 689 1419 905 376 360 921 469 1285 696 1554 1085 374 287 896 485 1030 682 1381 902 380 276 706 545 1286 717 1606 1097 avg: 373.6 289.4 791.5 512.7 1066.7 717.5 1459.4 981.8 res: 22.54% 35.52% 32.74% 32.73% Suggested-by: Zefan Li <lizefan(a)huawei.com> Signed-off-by: Tang Yizhou <tangyizhou(a)huawei.com> Reviewed-by: Ding Tianhong <dingtianhong(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- mm/share_pool.c | 104 ++++++++++++++++++++++++++++++++++-------------- 1 file changed, 75 insertions(+), 29 deletions(-) diff --git a/mm/share_pool.c b/mm/share_pool.c index 24c5dd680451..b9d30d36b7c2 100644 --- a/mm/share_pool.c +++ b/mm/share_pool.c @@ -730,6 +730,11 @@ static void __insert_sp_area(struct sp_area *spa) rb_insert_color(&spa->rb_node, &sp_area_root); } +/* The sp_area cache globals are protected by sp_area_lock */ +static struct rb_node *free_sp_area_cache; +static unsigned long cached_hole_size; +static unsigned long cached_vstart; /* affected by SP_DVPP and sp_config_dvpp_range() */ + /* * Allocate a region of VA from the share pool. * @size - the size of VA to allocate @@ -741,7 +746,7 @@ static void __insert_sp_area(struct sp_area *spa) static struct sp_area *sp_alloc_area(unsigned long size, unsigned long flags, struct sp_group *spg, enum spa_type type) { - struct sp_area *spa, *err; + struct sp_area *spa, *first, *err; struct rb_node *n; unsigned long vstart = MMAP_SHARE_POOL_START; unsigned long vend = MMAP_SHARE_POOL_16G_START; @@ -763,8 +768,6 @@ static struct sp_area *sp_alloc_area(unsigned long size, unsigned long flags, } } - addr = vstart; - spa = kmalloc(sizeof(struct sp_area), GFP_KERNEL); if (unlikely(!spa)) { if (printk_ratelimit()) @@ -774,45 +777,75 @@ static struct sp_area *sp_alloc_area(unsigned long size, unsigned long flags, spin_lock(&sp_area_lock); - n = sp_area_root.rb_node; - if (n) { - struct sp_area *first = NULL; + /* + * Invalidate cache if we have more permissive parameters. + * cached_hole_size notes the largest hole noticed _below_ + * the sp_area cached in free_sp_area_cache: if size fits + * into that hole, we want to scan from vstart to reuse + * the hole instead of allocating above free_sp_area_cache. + * Note that sp_free_area may update free_sp_area_cache + * without updating cached_hole_size. + */ + if (!free_sp_area_cache || size_align < cached_hole_size || + vstart != cached_vstart) { + cached_hole_size = 0; + free_sp_area_cache = NULL; + } + + /* record if we encounter less permissive parameters */ + cached_vstart = vstart; + + /* find starting point for our search */ + if (free_sp_area_cache) { + first = rb_entry(free_sp_area_cache, struct sp_area, rb_node); + addr = first->va_end; + if (addr + size_align < addr) { + err = ERR_PTR(-EOVERFLOW); + goto error; + } + } else { + addr = vstart; + if (addr + size_align < addr) { + err = ERR_PTR(-EOVERFLOW); + goto error; + } + + n = sp_area_root.rb_node; + first = NULL; - do { + while (n) { struct sp_area *tmp; tmp = rb_entry(n, struct sp_area, rb_node); if (tmp->va_end >= addr) { - if (!first && tmp->va_start < addr + size_align) - first = tmp; - n = n->rb_left; - } else { first = tmp; + if (tmp->va_start <= addr) + break; + n = n->rb_left; + } else n = n->rb_right; - } - } while (n); + } if (!first) goto found; + } - if (first->va_end < addr) { - n = rb_next(&first->rb_node); - if (n) - first = rb_entry(n, struct sp_area, rb_node); - else - goto found; + /* from the starting point, traverse areas until a suitable hole is found */ + while (addr + size_align > first->va_start && addr + size_align <= vend) { + if (addr + cached_hole_size < first->va_start) + cached_hole_size = first->va_start - addr; + addr = first->va_end; + if (addr + size_align < addr) { + err = ERR_PTR(-EOVERFLOW); + goto error; } - while (addr + size_align >= first->va_start && - addr + size_align <= vend) { - addr = first->va_end; - - n = rb_next(&first->rb_node); - if (n) - first = rb_entry(n, struct sp_area, rb_node); - else - goto found; - } + n = rb_next(&first->rb_node); + if (n) + first = rb_entry(n, struct sp_area, rb_node); + else + goto found; } + found: if (addr + size_align > vend) { err = ERR_PTR(-EOVERFLOW); @@ -833,6 +866,7 @@ static struct sp_area *sp_alloc_area(unsigned long size, unsigned long flags, } __insert_sp_area(spa); + free_sp_area_cache = &spa->rb_node; if (spa->spg) { atomic_inc(&spg->spa_num); atomic64_add(size, &spg->size); @@ -889,6 +923,18 @@ static void sp_free_area(struct sp_area *spa) { lockdep_assert_held(&sp_area_lock); + if (free_sp_area_cache) { + struct sp_area *cache; + cache = rb_entry(free_sp_area_cache, struct sp_area, rb_node); + if (spa->va_start <= cache->va_start) { + free_sp_area_cache = rb_prev(&spa->rb_node); + /* + * We don't try to update cached_hole_size, + * but it won't go very wrong. + */ + } + } + spa_dec_usage(spa->type, spa->real_size); /* won't fail */ if (spa->spg) { atomic_dec(&spa->spg->spa_num); -- 2.25.1

1 5

[PATCH 1/2] nvme: fix nvme_stop_queues cost long time error
by Yang Yingliang 02 Dec '20

02 Dec '20

From: Di Wang <wangdi44(a)huawei.com> driver inclusion category: bugfix bugzilla: NA Link: https://gitee.com/openeuler/kernel/issues/I1WGZE -------------------------------- It will cost a lot of time when synchronize_rcu() is called for every namspace in nvme_stop_queues(). Reviewed-by: Chao Leng <lengchao(a)huawei.com> Reviewed-by: Jike Cheng <chengjike.cheng(a)huawei.com> Signed-off-by: Di Wang <wangdi44(a)huawei.com> Signed-off-by: Lijie <lijie34(a)huawei.com> Acked-by: Hanjun Guo <guohanjun(a)huawei.com> Reviewed-by: Hou Tao <houtao1(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- block/blk-mq.c | 18 ++++++++++++++++++ drivers/nvme/host/core.c | 6 +++++- include/linux/blk-mq.h | 1 + 3 files changed, 24 insertions(+), 1 deletion(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index 7096f23b18da..41e92dbff821 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -242,6 +242,24 @@ void blk_mq_quiesce_queue(struct request_queue *q) } EXPORT_SYMBOL_GPL(blk_mq_quiesce_queue); +bool blk_mq_quiesce_queue_without_rcu(struct request_queue *q) +{ + struct blk_mq_hw_ctx *hctx; + unsigned int i; + bool rcu = false; + + blk_mq_quiesce_queue_nowait(q); + + queue_for_each_hw_ctx(q, hctx, i) { + if (hctx->flags & BLK_MQ_F_BLOCKING) + synchronize_srcu(hctx->srcu); + else + rcu = true; + } + return rcu; +} +EXPORT_SYMBOL_GPL(blk_mq_quiesce_queue_without_rcu); + /* * blk_mq_unquiesce_queue() - counterpart of blk_mq_quiesce_queue() * @q: request queue. diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index 8443c6c5c5af..c9faa824de26 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -3904,11 +3904,15 @@ EXPORT_SYMBOL_GPL(nvme_start_freeze); void nvme_stop_queues(struct nvme_ctrl *ctrl) { struct nvme_ns *ns; + bool rcu = false; down_read(&ctrl->namespaces_rwsem); list_for_each_entry(ns, &ctrl->namespaces, list) - blk_mq_quiesce_queue(ns->queue); + rcu = (blk_mq_quiesce_queue_without_rcu(ns->queue) || rcu); up_read(&ctrl->namespaces_rwsem); + + if (rcu) + synchronize_rcu(); } EXPORT_SYMBOL_GPL(nvme_stop_queues); diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index d26edab21d5c..730de600075b 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -309,6 +309,7 @@ void blk_mq_start_hw_queues(struct request_queue *q); void blk_mq_start_stopped_hw_queue(struct blk_mq_hw_ctx *hctx, bool async); void blk_mq_start_stopped_hw_queues(struct request_queue *q, bool async); void blk_mq_quiesce_queue(struct request_queue *q); +bool blk_mq_quiesce_queue_without_rcu(struct request_queue *q); void blk_mq_unquiesce_queue(struct request_queue *q); void blk_mq_delay_run_hw_queue(struct blk_mq_hw_ctx *hctx, unsigned long msecs); bool blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async); -- 2.25.1

1 1

Re: [Release] openEuler 20.03 LTS SP1 内核版本咨询
by Xie XiuQi 02 Dec '20

02 Dec '20

Hi 亚芬， On 2020/12/1 16:47, 方亚芬 wrote: > 内核组的老师们，大家好。我是 openEuler 社区树莓派 SIG 组的 maintainer 方亚芬。 > > > 针对即将发布的 openEuler 20.03 LTS SP1，有个问题想咨询一下。 > > 当前 openEuler 4.19 内核源码最新版本为 4.19.90-2011.6.0（https://gitee.com/openeuler/kernel/tree/4.19.90-2011.6.0/）… src-openeuler/kernel（https://gitee.com/src-openeuler/kernel）下三个分支更新版本不同，分别为： > 1. openEuler 20.03 LTS Next：4.19.90-2010.2.0 > 2. openEuler 20.03 LTS SP1：4.19.90-2011.4.0 > 3. openEuler 20.03 LTS：4.19.90-2011.6.0 > > > 针对该情况有以下两个问题： > 1. 之前树莓派镜像的内核版本一直根据 src-openeuler/kernel 下内核版本做更新，即将发布的 openEuler 20.03 LTS SP1 > 树莓派镜像采用内核版本要与 src-openeuler/kernel 的 openEuler 20.03 LTS SP1 分支内核版本一致？还是需要采用 openEuler 4.19 内核源码的最新版本（https://gitee.com/openeuler/kernel/tree/openEuler-1.0-LTS/，当前为 4.19.90-2011.6.0）？ 20.03 LTS SP1 和 20.03 LTS 的kernel同源，使用相同的内核分支，4.19.90-2011.4.0 和 4.19.90-2011.6.0 只是合入节奏的问题，实际上是同一个分支。对应的源码分支都是 openEuler-1.0-LTS. > > 2. openEuler 20.03 LTS SP1 正式发布前，内核会一直不定期更新，还是会有个具体的截止时间停止更新呢？ > SP1 包括之前的 20.03 LTS 都会陆续有少量的 cve 或 bugfix 补丁，SP1 正式发布之后，也会有陆续的更新，会持续到 SP1、SP0 生命周期结束。我理解你的诉求，你关心树莓派的内核 rebase 到openEuler具体哪个版本。我觉得这个影响应该不大，近期到 SP1 release，合入的补丁应该会很少。这个 SP1 的 release 计划， https://gitee.com/openeuler/release-management/blob/master/openEuler-20.03-… > ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ > 祝工作顺利，生活愉快。 > > ------------------------------------------ > 方亚芬 > 中国科学院软件研究所智能软件研究中心 > 邮箱：yafen(a)iscas.ac.cn <mailto:jiran@iscas.ac.cn> > 地址：北京市海淀区中关村南四街4号5楼 > > _______________________________________________ > Release mailing list -- release(a)openeuler.org > To unsubscribe send an email to release-leave(a)openeuler.org >

2 1

openEuler Summit 社区内报名通道开启，欢迎报名，诚挚的欢迎您为 openEuler 代言
by public＠openeuler.io 30 Nov '20

30 Nov '20

亲爱的 openEuler 社区开发者： openEuler Summit 是由 openEuler 社区举办的开发者交流会，首届线下 openEuler Summit 2020 将于 12 月 24-25 日在北京・望京凯悦酒店举办。openEuler 是一个面向未来的开源操作系统，依托于最具活力的 openEuler 社区，它的未来将不止于操作系统！ openEuler Summit 2020 将聚焦于软硬件协同、云边端协同的技术趋势，万物互联的多算力场景，挖掘技术创新与社区生态的价值。开发者、用户、社区贡献者、软件爱好者在 openEuler Summit 上汇聚成海，连接行业上下游，让全产业链共享操作系统创新价值。目前各大 Session 议题已经评审完毕，今日将会上线部分议程，大会报名通道开启，TC、Maintainer、SIG 工作组会议也在有条不紊地推进中。我们也将持续向大家同步会议进程，以便于社区成员了解 openEuler Summit 最新进展。 openEuler Summit 社区内开发者报名专属通道：https://etherpad.openeuler.org/p/openEuler-Summit-2020 在 openEuler 一周年之际，大会准备了一个 100个社区开发者为openEuler 代言的视频，诚挚地邀请您按照下方的要求拍摄一段视频，将 openEuler 操作系统推荐给更多的人。拍摄需求：自拍形式，对着镜头回答以下问题： 1. 我想要的操作系统是****的， 2. 我在**时候参与/认识/加入/知道的openEuler（社区） 3. 我在openEuler贡献了**** 4. 我是***，我为openEuler代言使用场景：您拍摄的视频素材将用于 openEuler Summit 2020 前期的会议宣传和开场视频中。拍摄指导： 1. 背景任您喜欢 2. 人物面对镜头，取半身构图，横向拍摄16:9 的画面比例 3. 最好选择白炽灯，白色光源 4. 尽量由另一个人辅助，用手机后置摄像头拍摄，不要用前置，拍摄前擦拭摄像头 5. 读口播时，声音越大越好 6. 尽量保证每一句话都是完整通顺的，中间重来的部分可以忽略 7. 视频输出，用电脑将手机视频导出，打压缩包传输即可（不要用微信上传，会压缩画质） 8. 拍摄完毕后请将视频上传到百度网盘，将分享链接期限设置为永久，然后发给 public(a)openeuler.io 9. 样片在附件，请您参考

1 0

[PATCH 1/3] Revert "refcount_t: Add ACQUIRE ordering on success for dec(sub)_and_test() variants"
by Yang Yingliang 28 Nov '20

28 Nov '20

hulk inclusion category: bugfix bugzilla: NA CVE: NA ---------------------------------------- This reverts commit 99294378e022ec2785f88247a7373767fd090e3a. 288e4521f0f6 ("x86/asm: 'Simplify' GEN_*_RMWcc() macros") need be apply before this commit to avoid compile error. Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- Documentation/core-api/refcount-vs-atomic.rst | 24 +++---------------- arch/x86/include/asm/refcount.h | 18 ++------------ lib/refcount.c | 18 ++++---------- 3 files changed, 10 insertions(+), 50 deletions(-) diff --git a/Documentation/core-api/refcount-vs-atomic.rst b/Documentation/core-api/refcount-vs-atomic.rst index 976e85adffe8..322851bada16 100644 --- a/Documentation/core-api/refcount-vs-atomic.rst +++ b/Documentation/core-api/refcount-vs-atomic.rst @@ -54,13 +54,6 @@ must propagate to all other CPUs before the release operation (A-cumulative property). This is implemented using :c:func:`smp_store_release`. -An ACQUIRE memory ordering guarantees that all post loads and -stores (all po-later instructions) on the same CPU are -completed after the acquire operation. It also guarantees that all -po-later stores on the same CPU must propagate to all other CPUs -after the acquire operation executes. This is implemented using -:c:func:`smp_acquire__after_ctrl_dep`. - A control dependency (on success) for refcounters guarantees that if a reference for an object was successfully obtained (reference counter increment or addition happened, function returned true), @@ -126,24 +119,13 @@ Memory ordering guarantees changes: result of obtaining pointer to the object! -case 5) - generic dec/sub decrement-based RMW ops that return a value ---------------------------------------------------------------------- +case 5) - decrement-based RMW ops that return a value +----------------------------------------------------- Function changes: * :c:func:`atomic_dec_and_test` --> :c:func:`refcount_dec_and_test` * :c:func:`atomic_sub_and_test` --> :c:func:`refcount_sub_and_test` - -Memory ordering guarantees changes: - - * fully ordered --> RELEASE ordering + ACQUIRE ordering on success - - -case 6) other decrement-based RMW ops that return a value ---------------------------------------------------------- - -Function changes: - * no atomic counterpart --> :c:func:`refcount_dec_if_one` * ``atomic_add_unless(&var, -1, 1)`` --> ``refcount_dec_not_one(&var)`` @@ -154,7 +136,7 @@ Memory ordering guarantees changes: .. note:: :c:func:`atomic_add_unless` only provides full order on success. -case 7) - lock-based RMW +case 6) - lock-based RMW ------------------------ Function changes: diff --git a/arch/x86/include/asm/refcount.h b/arch/x86/include/asm/refcount.h index 88061df7add6..19b90521954c 100644 --- a/arch/x86/include/asm/refcount.h +++ b/arch/x86/include/asm/refcount.h @@ -67,28 +67,14 @@ static __always_inline void refcount_dec(refcount_t *r) static __always_inline __must_check bool refcount_sub_and_test(unsigned int i, refcount_t *r) { - bool ret = GEN_BINARY_SUFFIXED_RMWcc(LOCK_PREFIX "subl", REFCOUNT_CHECK_LT_ZERO, + GEN_BINARY_SUFFIXED_RMWcc(LOCK_PREFIX "subl", REFCOUNT_CHECK_LT_ZERO, r->refs.counter, "er", i, "%0", e, "cx"); - - if (ret) { - smp_acquire__after_ctrl_dep(); - return true; - } - - return false; } static __always_inline __must_check bool refcount_dec_and_test(refcount_t *r) { - bool ret = GEN_UNARY_SUFFIXED_RMWcc(LOCK_PREFIX "decl", REFCOUNT_CHECK_LT_ZERO, + GEN_UNARY_SUFFIXED_RMWcc(LOCK_PREFIX "decl", REFCOUNT_CHECK_LT_ZERO, r->refs.counter, "%0", e, "cx"); - - if (ret) { - smp_acquire__after_ctrl_dep(); - return true; - } - - return false; } static __always_inline __must_check diff --git a/lib/refcount.c b/lib/refcount.c index 6e904af0fb3e..ebcf8cd49e05 100644 --- a/lib/refcount.c +++ b/lib/refcount.c @@ -33,9 +33,6 @@ * Note that the allocator is responsible for ordering things between free() * and alloc(). * - * The decrements dec_and_test() and sub_and_test() also provide acquire - * ordering on success. - * */ #include <linux/mutex.h> @@ -167,8 +164,8 @@ EXPORT_SYMBOL(refcount_inc_checked); * at UINT_MAX. * * Provides release memory ordering, such that prior loads and stores are done - * before, and provides an acquire ordering on success such that free() - * must come after. + * before, and provides a control dependency such that free() must come after. + * See the comment on top. * * Use of this function is not recommended for the normal reference counting * use case in which references are taken and released one at a time. In these @@ -193,12 +190,7 @@ bool refcount_sub_and_test_checked(unsigned int i, refcount_t *r) } while (!atomic_try_cmpxchg_release(&r->refs, &val, new)); - if (!new) { - smp_acquire__after_ctrl_dep(); - return true; - } - return false; - + return !new; } EXPORT_SYMBOL(refcount_sub_and_test_checked); @@ -210,8 +202,8 @@ EXPORT_SYMBOL(refcount_sub_and_test_checked); * decrement when saturated at UINT_MAX. * * Provides release memory ordering, such that prior loads and stores are done - * before, and provides an acquire ordering on success such that free() - * must come after. + * before, and provides a control dependency such that free() must come after. + * See the comment on top. * * Return: true if the resulting refcount is 0, false otherwise */ -- 2.25.1

1 2

[PATCH 01/11] mm: move nr_deactivate accounting to shrink_active_list()
by Yang Yingliang 28 Nov '20

28 Nov '20

From: Kirill Tkhai <ktkhai(a)virtuozzo.com> mainline inclusion from mainline-v5.2-rc1 commit 9851ac13592df77958ae7bac6ba39e71420c38ec category: bugfix bugzilla: 36232 CVE: NA ------------------------------------------------- We know which LRU is not active. [chris(a)chrisdown.name: fix build on !CONFIG_MEMCG] Link: http://lkml.kernel.org/r/20190322150513.GA22021@chrisdown.name Link: http://lkml.kernel.org/r/155290128498.31489.18250485448913338607.stgit@loca… Signed-off-by: Kirill Tkhai <ktkhai(a)virtuozzo.com> Signed-off-by: Chris Down <chris(a)chrisdown.name> Reviewed-by: Daniel Jordan <daniel.m.jordan(a)oracle.com> Cc: Michal Hocko <mhocko(a)kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Liu Shixin <liushixin2(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/linux/memcontrol.h | 6 ++++++ mm/vmscan.c | 10 ++++------ 2 files changed, 10 insertions(+), 6 deletions(-) diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index f6f51b14a332..029b4848ab6f 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -1109,6 +1109,12 @@ static inline void count_memcg_events(struct mem_cgroup *memcg, { } +static inline void __count_memcg_events(struct mem_cgroup *memcg, + enum vm_event_item idx, + unsigned long count) +{ +} + static inline void count_memcg_page_event(struct page *page, int idx) { diff --git a/mm/vmscan.c b/mm/vmscan.c index 38fcda38bf7b..92f2489efda8 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2077,12 +2077,6 @@ static unsigned move_active_pages_to_lru(struct lruvec *lruvec, } } - if (!is_active_lru(lru)) { - __count_vm_events(PGDEACTIVATE, nr_moved); - count_memcg_events(lruvec_memcg(lruvec), PGDEACTIVATE, - nr_moved); - } - return nr_moved; } @@ -2178,6 +2172,10 @@ static void shrink_active_list(unsigned long nr_to_scan, nr_activate = move_active_pages_to_lru(lruvec, &l_active, &l_hold, lru); nr_deactivate = move_active_pages_to_lru(lruvec, &l_inactive, &l_hold, lru - LRU_ACTIVE); + + __count_vm_events(PGDEACTIVATE, nr_deactivate); + __count_memcg_events(lruvec_memcg(lruvec), PGDEACTIVATE, nr_deactivate); + __mod_node_page_state(pgdat, NR_ISOLATED_ANON + file, -nr_taken); spin_unlock_irq(&pgdat->lru_lock); -- 2.25.1

1 10

[PATCH 01/13] bpf: remove duplicated include from syscall.c
by Yang Yingliang 28 Nov '20

28 Nov '20

From: YueHaibing <yuehaibing(a)huawei.com> mainline inclusion from mainline-v4.20-rc1 commit efbaec89c642cd1d4977fc6df9923697e1598d4e category: other bugzilla: 43460 CVE: NA --------------------------------------- Remove duplicated include. Signed-off-by: YueHaibing <yuehaibing(a)huawei.com> Signed-off-by: Daniel Borkmann <daniel(a)iogearbox.net> Signed-off-by: liuxin <liuxin264(a)huawei.com> Reviewed-by: Cheng Jian <cj.chengjian(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- kernel/bpf/syscall.c | 1 - 1 file changed, 1 deletion(-) diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index cb473159a55c..3e9ff9897c7c 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -30,7 +30,6 @@ #include <linux/cred.h> #include <linux/timekeeping.h> #include <linux/ctype.h> -#include <linux/btf.h> #include <linux/nospec.h> #define IS_FD_ARRAY(map) ((map)->map_type == BPF_MAP_TYPE_PROG_ARRAY || \ -- 2.25.1

1 12

[PATCH 00/91] backport stable-4.19.160
by Yang Yingliang 28 Nov '20

28 Nov '20

Aaron Lewis (1): selftests: kvm: Fix the segment descriptor layout to match the actual layout Alejandro Concepcion Rodriguez (1): can: dev: can_restart(): post buffer from the right context Anant Thazhemadam (2): can: af_can: prevent potential access of uninitialized member in can_rcv() can: af_can: prevent potential access of uninitialized member in canfd_rcv() Arvind Sankar (1): efi/x86: Free efi_pgd with free_pages() Aya Levin (1): net/mlx4_core: Fix init_hca fields offset Brian O'Keefe (1): staging: rtl8723bs: Add 024c:0627 to the list of SDIO device-ids Can Guo (1): scsi: ufs: Fix unbalanced scsi_block_reqs_cnt caused by ufshcd_hold() Chen Yu (1): x86/microcode/intel: Check patch signature before saving microcode for early loading Chen-Yu Tsai (4): Revert "arm: sun8i: orangepi-pc-plus: Set EMAC activity LEDs to active high" ARM: dts: sun8i: h3: orangepi-plus2e: Enable RGMII RX/TX delay on Ethernet PHY ARM: dts: sun8i: a83t: Enable both RGMII RX/TX delay on Ethernet PHY arm64: dts: allwinner: a64: bananapi-m64: Enable RGMII RX/TX delay on PHY Colin Ian King (1): can: peak_usb: fix potential integer overflow on shift of a int Dan Carpenter (2): Input: adxl34x - clean up a data type in adxl34x_probe() ALSA: firewire: Clean up a locking issue in copy_resp_to_buf() Darrick J. Wong (4): vfs: remove lockdep bogosity in __sb_start_write xfs: fix the minrecs logic when dealing with inode root child blocks xfs: strengthen rmap record flags checking xfs: revert "xfs: fix rmap key and record comparison functions" Dongli Zhang (1): page_frag: Recover from memory pressure Edwin Peer (1): bnxt_en: read EEPROM A2h address using page 0 Fabio Estevam (1): ARM: dts: imx50-evk: Fix the chip select 1 IOMUX Felix Fietkau (2): mac80211: minstrel: remove deferred sampling code mac80211: minstrel: fix tx status processing corner case Filip Moc (1): net: usb: qmi_wwan: Set DTR quirk for MR400 Florian Fainelli (1): net: Have netpoll bring-up DSA management interface Fugang Duan (1): tty: serial: imx: keep console clocks always on Gerald Schaefer (1): mm/userfaultfd: do not access vma->vm_mm after calling handle_userfault() Greg Kroah-Hartman (1): Linux 4.19.160 Hans de Goede (3): ACPI: button: Add DMI quirk for Medion Akoya E2228T iio: accel: kxcjk1013: Replace is_smo8500_device with an acpi_type enum iio: accel: kxcjk1013: Add support for KIOX010A ACPI DSM for setting tablet-mode Heiner Kallweit (1): net: bridge: add missing counters to ndo_get_stats64 callback Ido Schimmel (1): mlxsw: core: Use variable timeout for EMAD retries Jan Kara (1): ext4: fix bogus warning in ext4_update_dx_flag() Jernej Skrabec (3): arm64: dts: allwinner: a64: Pine64 Plus: Fix ethernet node arm64: dts: allwinner: h5: OrangePi PC2: Fix ethernet node ARM: dts: sun8i: r40: bananapi-m2-ultra: Fix ethernet node Jianqun Xu (1): pinctrl: rockchip: enable gpio pclk for rockchip_gpio_to_irq Jimmy Assarsson (1): can: kvaser_usb: kvaser_usb_hydra: Fix KCAN bittiming limits Joakim Tjernlund (1): ALSA: usb-audio: Add delay quirk for all Logitech USB devices Joel Stanley (2): net/ncsi: Fix netlink registration net: ftgmac100: Fix crash when removing driver Johannes Berg (1): mac80211: free sta in sta_info_insert_finish() on errors Leo Yan (1): perf lock: Don't free "lock_seq_stat" if read_count isn't zero Luo Meng (1): fail_function: Remove a redundant mutex unlock Marc Kleine-Budde (1): can: mcba_usb: mcba_usb_start_xmit(): first fill skb, then pass to can_put_echo_skb() Max Filippov (1): xtensa: disable preemption around cache alias management calls Michał Mirosław (3): regulator: fix memory leak with repeated set_machine_constraints() regulator: avoid resolve_supply() infinite recursion regulator: workaround self-referent regulators Mickaël Salaün (2): ptrace: Set PF_SUPERPRIV when checking capability seccomp: Set PF_SUPERPRIV when checking capability Necip Fazil Yildiran (1): Input: resistive-adc-touch - fix kconfig dependency on IIO_BUFFER Nenad Peric (1): arm64: dts: allwinner: h5: OrangePi Prime: Fix ethernet node Nishanth Menon (1): regulator: ti-abb: Fix array out of bound read access on the first transition Paul Moore (2): netlabel: fix our progress tracking in netlbl_unlabel_staticlist() netlabel: fix an uninitialized warning in netlbl_unlabel_staticlist() PeiSen Hou (1): ALSA: hda/realtek: Add some Clove SSID in the ALC293(ALC1220) Randy Dunlap (1): MIPS: export has_transparent_hugepage() for modules Ryan Sharpelletti (1): tcp: only postpone PROBE_RTT if RTT is < current min_rtt estimate Sean Nyekjaer (1): regulator: pfuze100: limit pfuze-support-disable-sw to pfuze{100,200} Sebastian Andrzej Siewior (1): atm: nicstar: Unmap DMA on send error Sergey Matyukevich (1): arm: dts: imx6qdl-udoo: fix rgmii phy-mode for ksz9031 phy Srinivasa Rao Mandadapu (1): ASoC: qcom: lpass-platform: Fix memory leak Stefan Haberland (1): s390/dasd: fix null pointer dereference for ERP requests Subash Abhinov Kasiviswanathan (1): net: qualcomm: rmnet: Fix incorrect receive packet handling during cleanup Sven Van Asbroeck (2): lan743x: fix issue causing intermittent kernel log warnings lan743x: prevent entire kernel HANG on open, for some platforms Takashi Iwai (1): ALSA: mixart: Fix mutex deadlock Takashi Sakamoto (1): ALSA: ctl: fix error path at adding user-defined element set Thomas Richter (1): s390/cpum_sf.c: fix file permission for cpum_sfb_size Tobias Waldekranz (1): net: dsa: mv88e6xxx: Avoid VTU corruption on 6097 Vamshi K Sthambamkadi (1): efivarfs: fix memory leak in efivarfs_create() Vladyslav Tarasiuk (1): net/mlx5: Disable QoS when min_rates on all VFs are zero Wang Hai (2): devlink: Add missing genlmsg_cancel() in devlink_nl_sb_port_pool_fill() inet_diag: Fix error path to cancel the meseage in inet_req_diag_fill() Will Deacon (1): arm64: psci: Avoid printing in cpu_psci_cpu_die() Wu Bo (1): can: m_can: m_can_handle_state_change(): fix state change Xie He (1): net: x25: Increase refcnt of "struct x25_neigh" in x25_rx_call_request Xin Long (1): sctp: change to hold/put transport for proto_unreach_timer Xiongfeng Wang (1): drm/sun4i: dw-hdmi: fix error return code in sun8i_dw_hdmi_bind() Yi-Hung Wei (1): ip_tunnels: Set tunnel option flag when tunnel metadata is present Yicong Yang (1): libfs: fix error cast of negative value in simple_attr_write() Zhang Changzhong (4): ah6: fix error return code in ah6_input() net: b44: fix error return code in b44_init_one() qed: fix error return code in qed_iwarp_ll2_start() qlcnic: fix error return code in qlcnic_83xx_restart_hw() Zhang Qilong (2): can: ti_hecc: Fix memleak in ti_hecc_probe MIPS: Alchemy: Fix memleak in alchemy_clk_setup_cpu Makefile | 2 +- arch/arm/boot/dts/imx50-evk.dts | 2 +- arch/arm/boot/dts/imx6qdl-udoo.dtsi | 2 +- arch/arm/boot/dts/sun8i-a83t-bananapi-m3.dts | 2 +- .../boot/dts/sun8i-a83t-cubietruck-plus.dts | 2 +- .../boot/dts/sun8i-h3-orangepi-pc-plus.dts | 5 -- .../arm/boot/dts/sun8i-h3-orangepi-plus2e.dts | 2 +- .../boot/dts/sun8i-r40-bananapi-m2-ultra.dts | 2 +- .../dts/allwinner/sun50i-a64-bananapi-m64.dts | 2 +- .../dts/allwinner/sun50i-a64-pine64-plus.dts | 2 +- .../dts/allwinner/sun50i-h5-orangepi-pc2.dts | 2 +- .../allwinner/sun50i-h5-orangepi-prime.dts | 2 +- arch/arm64/kernel/psci.c | 5 +- arch/mips/alchemy/common/clock.c | 9 ++- arch/mips/mm/tlb-r4k.c | 1 + arch/s390/kernel/perf_cpum_sf.c | 2 +- arch/x86/kernel/cpu/microcode/intel.c | 63 +++---------------- arch/x86/platform/efi/efi_64.c | 24 +++---- arch/xtensa/mm/cache.c | 14 +++++ drivers/acpi/button.c | 13 +++- drivers/atm/nicstar.c | 2 + drivers/gpu/drm/sun4i/sun8i_dw_hdmi.c | 1 + drivers/iio/accel/kxcjk-1013.c | 51 +++++++++++++-- drivers/input/misc/adxl34x.c | 2 +- drivers/input/touchscreen/Kconfig | 1 + drivers/net/can/dev.c | 2 +- drivers/net/can/m_can/m_can.c | 4 +- drivers/net/can/ti_hecc.c | 13 ++-- .../net/can/usb/kvaser_usb/kvaser_usb_hydra.c | 2 +- drivers/net/can/usb/mcba_usb.c | 4 +- drivers/net/can/usb/peak_usb/pcan_usb_core.c | 4 +- drivers/net/dsa/mv88e6xxx/global1_vtu.c | 59 ++++++++++++++--- drivers/net/ethernet/broadcom/b44.c | 3 +- .../net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 2 +- drivers/net/ethernet/faraday/ftgmac100.c | 4 ++ drivers/net/ethernet/mellanox/mlx4/fw.c | 6 +- drivers/net/ethernet/mellanox/mlx4/fw.h | 4 +- .../net/ethernet/mellanox/mlx5/core/eswitch.c | 15 ++--- drivers/net/ethernet/mellanox/mlxsw/core.c | 3 +- drivers/net/ethernet/microchip/lan743x_main.c | 13 ++-- drivers/net/ethernet/qlogic/qed/qed_iwarp.c | 12 +++- .../ethernet/qlogic/qlcnic/qlcnic_83xx_init.c | 3 +- .../ethernet/qualcomm/rmnet/rmnet_handlers.c | 5 ++ drivers/net/geneve.c | 3 +- drivers/net/usb/qmi_wwan.c | 2 +- drivers/pinctrl/pinctrl-rockchip.c | 2 + drivers/regulator/core.c | 38 ++++++----- drivers/regulator/pfuze100-regulator.c | 13 ++-- drivers/regulator/ti-abb-regulator.c | 12 +++- drivers/s390/block/dasd.c | 6 ++ drivers/scsi/ufs/ufshcd.c | 6 +- drivers/staging/rtl8723bs/os_dep/sdio_intf.c | 1 + drivers/tty/serial/imx.c | 20 +----- fs/efivarfs/super.c | 1 + fs/ext4/ext4.h | 3 +- fs/libfs.c | 6 +- fs/super.c | 33 ++-------- fs/xfs/libxfs/xfs_rmap_btree.c | 16 ++--- fs/xfs/scrub/bmap.c | 8 +-- fs/xfs/scrub/btree.c | 45 +++++++------ include/net/ip_tunnels.h | 7 ++- kernel/fail_function.c | 5 +- kernel/ptrace.c | 16 ++--- kernel/seccomp.c | 5 +- mm/huge_memory.c | 9 ++- mm/page_alloc.c | 5 ++ net/bridge/br_device.c | 1 + net/can/af_can.c | 38 ++++++++--- net/core/devlink.c | 6 +- net/core/netpoll.c | 22 +++++-- net/ipv4/inet_diag.c | 4 +- net/ipv4/tcp_bbr.c | 2 +- net/ipv6/ah6.c | 3 +- net/mac80211/rc80211_minstrel.c | 27 ++------ net/mac80211/rc80211_minstrel.h | 1 - net/mac80211/sta_info.c | 14 ++--- net/ncsi/ncsi-manage.c | 5 -- net/ncsi/ncsi-netlink.c | 22 +------ net/ncsi/ncsi-netlink.h | 3 - net/netlabel/netlabel_unlabeled.c | 17 +++-- net/sctp/input.c | 4 +- net/sctp/sm_sideeffect.c | 4 +- net/sctp/transport.c | 2 +- net/x25/af_x25.c | 1 + sound/core/control.c | 2 +- .../fireworks/fireworks_transaction.c | 4 +- sound/pci/hda/patch_realtek.c | 50 ++++++++++++++- sound/pci/mixart/mixart_core.c | 5 +- sound/soc/qcom/lpass-platform.c | 5 +- sound/usb/quirks.c | 10 +-- tools/perf/builtin-lock.c | 2 +- tools/testing/selftests/kvm/include/x86.h | 2 +- tools/testing/selftests/kvm/lib/x86.c | 3 +- 93 files changed, 517 insertions(+), 377 deletions(-) -- 2.25.1

1 91

[PATCH 01/14] blk-cgroup: prevent rcu_sched detected stalls warnings in blkg_destroy_all()
by Yang Yingliang 28 Nov '20

28 Nov '20

From: Yu Kuai <yukuai3(a)huawei.com> hulk inclusion category: bugfix bugzilla: 46357 CVE: NA --------------------------- test procedures: a. create 20000 cgroups, and echo "8:0 10000" to blkio.throttle.write_bps_device b. echo 1 > /sys/blocd/sda/device/delete test result: rcu: INFO: rcu_sched detected stalls on CPUs/tasks: [5/1143] rcu: 0-...0: (0 ticks this GP) idle=0f2/1/0x4000000000000000 softirq=15507/15507 fq rcu: (detected by 6, t=60012 jiffies, g=119977, q=27153) NMI backtrace for cpu 0 CPU: 0 PID: 443 Comm: bash Not tainted 4.19.95-00061-g0bcc83b30eec #63 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS ?-20190727_073836-buildvm-p4 RIP: 0010:blk_throtl_update_limit_valid.isra.0+0x116/0x2a0 Code: 01 00 00 e8 7c dd 74 ff 48 83 bb 78 01 00 00 00 0f 85 54 01 00 00 48 8d bb 88 01 1 RSP: 0018:ffff8881030bf9f0 EFLAGS: 00000046 RAX: 0000000000000000 RBX: ffff8880b4f37080 RCX: ffffffff95da0afe RDX: dffffc0000000000 RSI: ffff888100373980 RDI: ffff8880b4f37208 RBP: ffff888100deca00 R08: ffffffff9528f951 R09: 0000000000000001 R10: ffffed10159dbf56 R11: ffff8880acedfab3 R12: ffff8880b9fda498 R13: ffff8880b9fda4f4 R14: 0000000000000050 R15: ffffffff98b622c0 FS: 00007feb51c51700(0000) GS:ffff888106200000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000561619547080 CR3: 0000000102bc9000 CR4: 00000000000006f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: throtl_pd_offline+0x98/0x100 blkg_destroy+0x133/0x630 ? blkcg_deactivate_policy+0x2c0/0x2c0 ? lock_timer_base+0x65/0x110 blkg_destroy_all+0x7f/0x100 blkcg_exit_queue+0x3f/0xa5 blk_exit_queue+0x69/0xa0 blk_cleanup_queue+0x226/0x360 __scsi_remove_device+0xb4/0x3c0 scsi_remove_device+0x38/0x60 sdev_store_delete+0x74/0x100 ? dev_driver_string+0xb0/0xb0 dev_attr_store+0x41/0x70 sysfs_kf_write+0x89/0xc0 kernfs_fop_write+0x1b6/0x2e0 ? sysfs_kf_bin_read+0x130/0x130 __vfs_write+0xca/0x420 ? kernel_read+0xc0/0xc0 ? __alloc_fd+0x16f/0x2d0 ? __fd_install+0x95/0x1a0 ? handle_mm_fault+0x3e0/0x560 vfs_write+0x11a/0x2f0 ksys_write+0xb9/0x1e0 ? __x64_sys_read+0x60/0x60 ? kasan_check_write+0x20/0x30 ? filp_close+0xb5/0xf0 __x64_sys_write+0x46/0x60 do_syscall_64+0xd9/0x1f0 entry_SYSCALL_64_after_hwframe+0x44/0xa9 The usage of so many blkg is very rare, however, such problem do exist in theory. In order to avoid such warnings, release 'q->queue_lock' for a while when a batch of blkg were destroyed. Signed-off-by: Yu Kuai <yukuai3(a)huawei.com> Reviewed-by: Tao Hou <houtao1(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- block/blk-cgroup.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index e592167449aa..c64f0afa27dc 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -364,16 +364,31 @@ static void blkg_destroy(struct blkcg_gq *blkg) */ static void blkg_destroy_all(struct request_queue *q) { +#define BLKG_DESTROY_BATCH 4096 struct blkcg_gq *blkg, *n; + int count; lockdep_assert_held(q->queue_lock); +again: + count = BLKG_DESTROY_BATCH; list_for_each_entry_safe(blkg, n, &q->blkg_list, q_node) { struct blkcg *blkcg = blkg->blkcg; spin_lock(&blkcg->lock); blkg_destroy(blkg); spin_unlock(&blkcg->lock); + /* + * If the list is too long, the loop can took a long time, + * thus relese the lock for a while when a batch of blkg + * were destroyed. + */ + if (!--count) { + spin_unlock_irq(q->queue_lock); + cond_resched(); + spin_lock_irq(q->queue_lock); + goto again; + } } q->root_blkg = NULL; -- 2.25.1

1 13

kernel sig meeting
by openEuler conference 26 Nov '20

26 Nov '20

2 1

[PATCH 00/26] KVM RISC-V Support
by MingWang Li 25 Nov '20

25 Nov '20

From: Mingwang Li <limingwang(a)huawei.com> euleros inclusion category: feature bugzilla: NA CVE: NA This series adds initial KVM RISC-V support. Currently, we are able to boot Linux on RV64/RV32 Guest with multiple VCPUs. This series can be found in riscv_kvm_v15 branch at: https//github.com/avpatel/linux.git Link: https://gitee.com/openeuler/kernel/issues/I26X9V Alistair Francis (1): Revert "riscv: Use latest system call ABI" Anup Patel (20): RISC-V: Add fragmented config for debug options RISC-V: Enable CPU Hotplug in defconfigs RISC-V: Add Microchip PolarFire kconfig option RISC-V: Initial DTS for Microchip ICICLE board RISC-V: Enable drivers for Microchip PolarFire ICICLE board RISC-V: Add missing jump label initialization RISC-V: Add hypervisor extension related CSR defines RISC-V: Add initial skeletal KVM support RISC-V: KVM: Implement VCPU create, init and destroy functions RISC-V: KVM: Implement VCPU interrupts and requests handling RISC-V: KVM: Implement KVM_GET_ONE_REG/KVM_SET_ONE_REG ioctls RISC-V: KVM: Implement VCPU world-switch RISC-V: KVM: Handle MMIO exits for VCPU RISC-V: KVM: Handle WFI exits for VCPU RISC-V: KVM: Implement VMID allocator RISC-V: KVM: Implement stage2 page table programming RISC-V: KVM: Implement MMU notifiers RISC-V: KVM: Document RISC-V specific parts of KVM API RISC-V: KVM: Add MAINTAINERS entry RISC-V: Enable KVM for RV64 and RV32 Atish Patra (4): RISC-V: KVM: Add timer functionality RISC-V: KVM: FP lazy save/restore RISC-V: KVM: Implement ONE REG interface for FP registers RISC-V: KVM: Add SBI v0.1 support m34782 (1): Microchip Polarfire SoC Clock Driver Documentation/virt/kvm/api.rst | 193 +++- MAINTAINERS | 11 + arch/riscv/Kconfig | 2 + arch/riscv/Kconfig.socs | 7 + arch/riscv/Makefile | 2 + arch/riscv/boot/dts/Makefile | 1 + arch/riscv/boot/dts/microchip/Makefile | 2 + .../boot/dts/microchip/icicle-kit-es.dts | 307 +++++ arch/riscv/configs/defconfig | 30 +- arch/riscv/configs/extra_debug.config | 21 + arch/riscv/configs/rv32_defconfig | 26 +- arch/riscv/include/asm/csr.h | 89 ++ arch/riscv/include/asm/kvm_host.h | 279 +++++ arch/riscv/include/asm/kvm_types.h | 7 + arch/riscv/include/asm/kvm_vcpu_timer.h | 44 + arch/riscv/include/asm/pgtable-bits.h | 1 + arch/riscv/include/uapi/asm/kvm.h | 128 +++ arch/riscv/include/uapi/asm/unistd.h | 5 +- arch/riscv/kernel/asm-offsets.c | 156 +++ arch/riscv/kernel/setup.c | 1 + arch/riscv/kvm/Kconfig | 36 + arch/riscv/kvm/Makefile | 15 + arch/riscv/kvm/main.c | 118 ++ arch/riscv/kvm/mmu.c | 860 ++++++++++++++ arch/riscv/kvm/tlb.S | 74 ++ arch/riscv/kvm/vcpu.c | 1012 +++++++++++++++++ arch/riscv/kvm/vcpu_exit.c | 701 ++++++++++++ arch/riscv/kvm/vcpu_sbi.c | 173 +++ arch/riscv/kvm/vcpu_switch.S | 400 +++++++ arch/riscv/kvm/vcpu_timer.c | 225 ++++ arch/riscv/kvm/vm.c | 81 ++ arch/riscv/kvm/vmid.c | 120 ++ drivers/clk/Kconfig | 5 + drivers/clk/Makefile | 1 + drivers/clk/microchip/Makefile | 2 + drivers/clk/microchip/clk-pfsoc.c | 508 +++++++++ drivers/clocksource/timer-riscv.c | 8 + include/clocksource/timer-riscv.h | 16 + include/uapi/linux/kvm.h | 8 + 39 files changed, 5621 insertions(+), 54 deletions(-) create mode 100644 arch/riscv/boot/dts/microchip/Makefile create mode 100644 arch/riscv/boot/dts/microchip/icicle-kit-es.dts create mode 100644 arch/riscv/configs/extra_debug.config create mode 100644 arch/riscv/include/asm/kvm_host.h create mode 100644 arch/riscv/include/asm/kvm_types.h create mode 100644 arch/riscv/include/asm/kvm_vcpu_timer.h create mode 100644 arch/riscv/include/uapi/asm/kvm.h create mode 100644 arch/riscv/kvm/Kconfig create mode 100644 arch/riscv/kvm/Makefile create mode 100644 arch/riscv/kvm/main.c create mode 100644 arch/riscv/kvm/mmu.c create mode 100644 arch/riscv/kvm/tlb.S create mode 100644 arch/riscv/kvm/vcpu.c create mode 100644 arch/riscv/kvm/vcpu_exit.c create mode 100644 arch/riscv/kvm/vcpu_sbi.c create mode 100644 arch/riscv/kvm/vcpu_switch.S create mode 100644 arch/riscv/kvm/vcpu_timer.c create mode 100644 arch/riscv/kvm/vm.c create mode 100644 arch/riscv/kvm/vmid.c create mode 100644 drivers/clk/microchip/clk-pfsoc.c create mode 100644 include/clocksource/timer-riscv.h -- 2.19.1

1 26

[PATCH 0/7] fixes CVE-2020-4788
by Yang Yingliang 24 Nov '20

24 Nov '20

Andrew Donnellan (1): powerpc: Fix __clear_user() with KUAP enabled Christophe Leroy (2): powerpc: Add a framework for user access tracking powerpc: Implement user_access_begin and friends Daniel Axtens (1): powerpc/64s: move some exception handlers out of line Nicholas Piggin (3): powerpc/64s: flush L1D on kernel entry powerpc/uaccess: Evaluate macro arguments once, before user access is allowed powerpc/64s: flush L1D after user accesses .../admin-guide/kernel-parameters.txt | 7 + .../powerpc/include/asm/book3s/64/kup-radix.h | 22 +++ arch/powerpc/include/asm/exception-64s.h | 9 +- arch/powerpc/include/asm/feature-fixups.h | 19 +++ arch/powerpc/include/asm/futex.h | 4 + arch/powerpc/include/asm/kup.h | 40 +++++ arch/powerpc/include/asm/security_features.h | 7 + arch/powerpc/include/asm/setup.h | 4 + arch/powerpc/include/asm/uaccess.h | 147 ++++++++++++++---- arch/powerpc/kernel/exceptions-64s.S | 96 +++++++----- arch/powerpc/kernel/setup_64.c | 122 ++++++++++++++- arch/powerpc/kernel/vmlinux.lds.S | 14 ++ arch/powerpc/lib/checksum_wrappers.c | 4 + arch/powerpc/lib/feature-fixups.c | 104 +++++++++++++ arch/powerpc/lib/string_32.S | 4 +- arch/powerpc/lib/string_64.S | 6 +- arch/powerpc/platforms/powernv/setup.c | 17 ++ arch/powerpc/platforms/pseries/setup.c | 8 + 18 files changed, 553 insertions(+), 81 deletions(-) create mode 100644 arch/powerpc/include/asm/book3s/64/kup-radix.h create mode 100644 arch/powerpc/include/asm/kup.h -- 2.25.1

1 7

[PATCH 00/14] backport stable-4.19.159
by Yang Yingliang 23 Nov '20

23 Nov '20

Andrew Donnellan (1): powerpc: Fix __clear_user() with KUAP enabled Christophe Leroy (3): powerpc: Add a framework for user access tracking powerpc: Implement user_access_begin and friends powerpc/8xx: Always fault when _PAGE_ACCESSED is not set Daniel Axtens (1): powerpc/64s: move some exception handlers out of line David Edmondson (1): KVM: x86: clflushopt should be treated as a no-op by emulation Greg Kroah-Hartman (1): Linux 4.19.159 Johannes Berg (1): mac80211: always wind down STA state Nicholas Piggin (3): powerpc/64s: flush L1D on kernel entry powerpc/uaccess: Evaluate macro arguments once, before user access is allowed powerpc/64s: flush L1D after user accesses Nick Desaulniers (1): ACPI: GED: fix -Wformat Salvatore Bonaccorso (1): Revert "perf cs-etm: Move definition of 'traceid_list' global variable from header file" Zhang Changzhong (1): can: proc: can_remove_proc(): silence remove_proc_entry warning .../admin-guide/kernel-parameters.txt | 7 + Makefile | 2 +- .../powerpc/include/asm/book3s/64/kup-radix.h | 22 +++ arch/powerpc/include/asm/exception-64s.h | 9 +- arch/powerpc/include/asm/feature-fixups.h | 19 +++ arch/powerpc/include/asm/futex.h | 4 + arch/powerpc/include/asm/kup.h | 40 +++++ arch/powerpc/include/asm/security_features.h | 7 + arch/powerpc/include/asm/setup.h | 4 + arch/powerpc/include/asm/uaccess.h | 147 ++++++++++++++---- arch/powerpc/kernel/exceptions-64s.S | 96 +++++++----- arch/powerpc/kernel/head_8xx.S | 8 +- arch/powerpc/kernel/setup_64.c | 122 ++++++++++++++- arch/powerpc/kernel/vmlinux.lds.S | 14 ++ arch/powerpc/lib/checksum_wrappers.c | 4 + arch/powerpc/lib/feature-fixups.c | 104 +++++++++++++ arch/powerpc/lib/string_32.S | 4 +- arch/powerpc/lib/string_64.S | 6 +- arch/powerpc/platforms/powernv/setup.c | 17 ++ arch/powerpc/platforms/pseries/setup.c | 8 + arch/x86/kvm/emulate.c | 8 +- drivers/acpi/evged.c | 2 +- net/can/proc.c | 6 +- net/mac80211/sta_info.c | 18 +++ tools/perf/util/cs-etm.c | 3 - tools/perf/util/cs-etm.h | 3 + 26 files changed, 589 insertions(+), 95 deletions(-) create mode 100644 arch/powerpc/include/asm/book3s/64/kup-radix.h create mode 100644 arch/powerpc/include/asm/kup.h -- 2.25.1

1 14

[PATCH] vt: Disable KD_FONT_OP_COPY
by Yang Yingliang 23 Nov '20

23 Nov '20

From: Daniel Vetter <daniel.vetter(a)ffwll.ch> stable inclusion from linux-4.19.156 commit 6612b754ac0c85ca8b1181b5d3ea4461a8c1bbcb CVE: CVE-2020-28974 -------------------------------- commit 3c4e0dff2095c579b142d5a0693257f1c58b4804 upstream. It's buggy: On Fri, Nov 06, 2020 at 10:30:08PM +0800, Minh Yuan wrote: > We recently discovered a slab-out-of-bounds read in fbcon in the latest > kernel ( v5.10-rc2 for now ). The root cause of this vulnerability is that > "fbcon_do_set_font" did not handle "vc->vc_font.data" and > "vc->vc_font.height" correctly, and the patch > <https://lkml.org/lkml/2020/9/27/223> for VT_RESIZEX can't handle this > issue. > > Specifically, we use KD_FONT_OP_SET to set a small font.data for tty6, and > use KD_FONT_OP_SET again to set a large font.height for tty1. After that, > we use KD_FONT_OP_COPY to assign tty6's vc_font.data to tty1's vc_font.data > in "fbcon_do_set_font", while tty1 retains the original larger > height. Obviously, this will cause an out-of-bounds read, because we can > access a smaller vc_font.data with a larger vc_font.height. Further there was only one user ever. - Android's loadfont, busybox and console-tools only ever use OP_GET and OP_SET - fbset documentation only mentions the kernel cmdline font: option, not anything else. - systemd used OP_COPY before release 232 published in Nov 2016 Now unfortunately the crucial report seems to have gone down with gmane, and the commit message doesn't say much. But the pull request hints at OP_COPY being broken https://github.com/systemd/systemd/pull/3651 So in other words, this never worked, and the only project which foolishly every tried to use it, realized that rather quickly too. Instead of trying to fix security issues here on dead code by adding missing checks, fix the entire thing by removing the functionality. Note that systemd code using the OP_COPY function ignored the return value, so it doesn't matter what we're doing here really - just in case a lone server somewhere happens to be extremely unlucky and running an affected old version of systemd. The relevant code from font_copy_to_all_vcs() in systemd was: /* copy font from active VT, where the font was uploaded to */ cfo.op = KD_FONT_OP_COPY; cfo.height = vcs.v_active-1; /* tty1 == index 0 */ (void) ioctl(vcfd, KDFONTOP, &cfo); Note this just disables the ioctl, garbage collecting the now unused callbacks is left for -next. v2: Tetsuo found the old mail, which allowed me to find it on another archive. Add the link too. Acked-by: Peilin Ye <yepeilin.cs(a)gmail.com> Reported-by: Minh Yuan <yuanmingbuaa(a)gmail.com> References: https://lists.freedesktop.org/archives/systemd-devel/2016-June/036935.html References: https://github.com/systemd/systemd/pull/3651 Cc: Greg KH <greg(a)kroah.com> Cc: Peilin Ye <yepeilin.cs(a)gmail.com> Cc: Tetsuo Handa <penguin-kernel(a)i-love.sakura.ne.jp> Signed-off-by: Daniel Vetter <daniel.vetter(a)intel.com> Link: https://lore.kernel.org/r/20201108153806.3140315-1-daniel.vetter@ffwll.ch Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/tty/vt/vt.c | 24 ++---------------------- 1 file changed, 2 insertions(+), 22 deletions(-) diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 758f522f331e..13ea0579f104 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -4574,27 +4574,6 @@ static int con_font_default(struct vc_data *vc, struct console_font_op *op) return rc; } -static int con_font_copy(struct vc_data *vc, struct console_font_op *op) -{ - int con = op->height; - int rc; - - - console_lock(); - if (vc->vc_mode != KD_TEXT) - rc = -EINVAL; - else if (!vc->vc_sw->con_font_copy) - rc = -ENOSYS; - else if (con < 0 || !vc_cons_allocated(con)) - rc = -ENOTTY; - else if (con == vc->vc_num) /* nothing to do */ - rc = 0; - else - rc = vc->vc_sw->con_font_copy(vc, con); - console_unlock(); - return rc; -} - int con_font_op(struct vc_data *vc, struct console_font_op *op) { switch (op->op) { @@ -4605,7 +4584,8 @@ int con_font_op(struct vc_data *vc, struct console_font_op *op) case KD_FONT_OP_SET_DEFAULT: return con_font_default(vc, op); case KD_FONT_OP_COPY: - return con_font_copy(vc, op); + /* was buggy and never really used */ + return -EINVAL; } return -ENOSYS; } -- 2.25.1

1 0

[PATCH 1/3] fbdev, newport_con: Move FONT_EXTRA_WORDS macros into linux/font.h
by Yang Yingliang 23 Nov '20

23 Nov '20

From: Peilin Ye <yepeilin.cs(a)gmail.com> stable inclusion from linux-4.19.151 commit 7b9eaa7241ea2cfa580b854d461be72107a4b35c CVE: CVE-2020-28915 It's prepare for fixing CVE-2020-28915. -------------------------------- commit bb0890b4cd7f8203e3aa99c6d0f062d6acdaad27 upstream. drivers/video/console/newport_con.c is borrowing FONT_EXTRA_WORDS macros from drivers/video/fbdev/core/fbcon.h. To keep things simple, move all definitions into <linux/font.h>. Since newport_con now uses four extra words, initialize the fourth word in newport_set_font() properly. Cc: stable(a)vger.kernel.org Signed-off-by: Peilin Ye <yepeilin.cs(a)gmail.com> Reviewed-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Daniel Vetter <daniel.vetter(a)ffwll.ch> Link: https://patchwork.freedesktop.org/patch/msgid/7fb8bc9b0abc676ada6b7ac0e0bd4… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/video/console/newport_con.c | 7 +------ drivers/video/fbdev/core/fbcon.h | 7 ------- drivers/video/fbdev/core/fbcon_rotate.c | 1 + drivers/video/fbdev/core/tileblit.c | 1 + include/linux/font.h | 8 ++++++++ 5 files changed, 11 insertions(+), 13 deletions(-) diff --git a/drivers/video/console/newport_con.c b/drivers/video/console/newport_con.c index 7f2526b43b33..46a6e6568994 100644 --- a/drivers/video/console/newport_con.c +++ b/drivers/video/console/newport_con.c @@ -33,12 +33,6 @@ #define FONT_DATA ((unsigned char *)font_vga_8x16.data) -/* borrowed from fbcon.c */ -#define REFCOUNT(fd) (((int *)(fd))[-1]) -#define FNTSIZE(fd) (((int *)(fd))[-2]) -#define FNTCHARCNT(fd) (((int *)(fd))[-3]) -#define FONT_EXTRA_WORDS 3 - static unsigned char *font_data[MAX_NR_CONSOLES]; static struct newport_regs *npregs; @@ -519,6 +513,7 @@ static int newport_set_font(int unit, struct console_font *op) FNTSIZE(new_data) = size; FNTCHARCNT(new_data) = op->charcount; REFCOUNT(new_data) = 0; /* usage counter */ + FNTSUM(new_data) = 0; p = new_data; for (i = 0; i < op->charcount; i++) { diff --git a/drivers/video/fbdev/core/fbcon.h b/drivers/video/fbdev/core/fbcon.h index ac11200fd2d5..609280281a47 100644 --- a/drivers/video/fbdev/core/fbcon.h +++ b/drivers/video/fbdev/core/fbcon.h @@ -152,13 +152,6 @@ static inline int attr_col_ec(int shift, struct vc_data *vc, #define attr_bgcol_ec(bgshift, vc, info) attr_col_ec(bgshift, vc, info, 0) #define attr_fgcol_ec(fgshift, vc, info) attr_col_ec(fgshift, vc, info, 1) -/* Font */ -#define REFCOUNT(fd) (((int *)(fd))[-1]) -#define FNTSIZE(fd) (((int *)(fd))[-2]) -#define FNTCHARCNT(fd) (((int *)(fd))[-3]) -#define FNTSUM(fd) (((int *)(fd))[-4]) -#define FONT_EXTRA_WORDS 4 - /* * Scroll Method */ diff --git a/drivers/video/fbdev/core/fbcon_rotate.c b/drivers/video/fbdev/core/fbcon_rotate.c index c0d445294aa7..ac72d4f85f7d 100644 --- a/drivers/video/fbdev/core/fbcon_rotate.c +++ b/drivers/video/fbdev/core/fbcon_rotate.c @@ -14,6 +14,7 @@ #include <linux/fb.h> #include <linux/vt_kern.h> #include <linux/console.h> +#include <linux/font.h> #include <asm/types.h> #include "fbcon.h" #include "fbcon_rotate.h" diff --git a/drivers/video/fbdev/core/tileblit.c b/drivers/video/fbdev/core/tileblit.c index eb664dbf96f6..adff8d6ffe6f 100644 --- a/drivers/video/fbdev/core/tileblit.c +++ b/drivers/video/fbdev/core/tileblit.c @@ -13,6 +13,7 @@ #include <linux/fb.h> #include <linux/vt_kern.h> #include <linux/console.h> +#include <linux/font.h> #include <asm/types.h> #include "fbcon.h" diff --git a/include/linux/font.h b/include/linux/font.h index d6821769dd1e..0a3639a00b3a 100644 --- a/include/linux/font.h +++ b/include/linux/font.h @@ -57,4 +57,12 @@ extern const struct font_desc *get_default_font(int xres, int yres, /* Max. length for the name of a predefined font */ #define MAX_FONT_NAME 32 +/* Extra word getters */ +#define REFCOUNT(fd) (((int *)(fd))[-1]) +#define FNTSIZE(fd) (((int *)(fd))[-2]) +#define FNTCHARCNT(fd) (((int *)(fd))[-3]) +#define FNTSUM(fd) (((int *)(fd))[-4]) + +#define FONT_EXTRA_WORDS 4 + #endif /* _VIDEO_FONT_H */ -- 2.25.1

1 2

[PATCH] speakup: Do not let the line discipline be used several times
by Yang Yingliang 23 Nov '20

23 Nov '20

From: Samuel Thibault <samuel.thibault(a)ens-lyon.org> mainline inclusion from mainline-v5.10 commit d4122754442799187d5d537a9c039a49a67e57f1 category: bugfix bugzilla: NA CVE: CVE-2020-28941 -------------------------------- Speakup has only one speakup_tty variable to store the tty it is managing. This makes sense since its codebase currently assumes that there is only one user who controls the screen reading. That however means that we have to forbid using the line discipline several times, otherwise the second closure would try to free a NULL ldisc_data, leading to general protection fault: 0000 [#1] SMP KASAN PTI RIP: 0010:spk_ttyio_ldisc_close+0x2c/0x60 Call Trace: tty_ldisc_release+0xa2/0x340 tty_release_struct+0x17/0xd0 tty_release+0x9d9/0xcc0 __fput+0x231/0x740 task_work_run+0x12c/0x1a0 do_exit+0x9b5/0x2230 ? release_task+0x1240/0x1240 ? __do_page_fault+0x562/0xa30 do_group_exit+0xd5/0x2a0 __x64_sys_exit_group+0x35/0x40 do_syscall_64+0x89/0x2b0 ? page_fault+0x8/0x30 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Cc: stable(a)vger.kernel.org Reported-by: 秦世松 <qinshisong1205(a)gmail.com> Signed-off-by: Samuel Thibault <samuel.thibault(a)ens-lyon.org> Tested-by: Shisong Qin <qinshisong1205(a)gmail.com> Link: https://lore.kernel.org/r/20201110183541.fzgnlwhjpgqzjeth@function Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: drivers/accessibility/speakup/spk_ttyio.c [yyl: spk_ttyio.c is in drivers/staging/ in kernel-4.19] Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/staging/speakup/spk_ttyio.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/staging/speakup/spk_ttyio.c b/drivers/staging/speakup/spk_ttyio.c index 93742dbdee77..6c754ddf1257 100644 --- a/drivers/staging/speakup/spk_ttyio.c +++ b/drivers/staging/speakup/spk_ttyio.c @@ -49,15 +49,25 @@ static int spk_ttyio_ldisc_open(struct tty_struct *tty) if (tty->ops->write == NULL) return -EOPNOTSUPP; + + mutex_lock(&speakup_tty_mutex); + if (speakup_tty) { + mutex_unlock(&speakup_tty_mutex); + return -EBUSY; + } speakup_tty = tty; ldisc_data = kmalloc(sizeof(struct spk_ldisc_data), GFP_KERNEL); - if (!ldisc_data) + if (!ldisc_data) { + speakup_tty = NULL; + mutex_unlock(&speakup_tty_mutex); return -ENOMEM; + } sema_init(&ldisc_data->sem, 0); ldisc_data->buf_free = true; speakup_tty->disc_data = ldisc_data; + mutex_unlock(&speakup_tty_mutex); return 0; } -- 2.25.1

1 0

[PATCH 00/98] backport stable-4.19.158
by Yang Yingliang 23 Nov '20

23 Nov '20

Al Viro (1): don't dump the threads that had been already exiting when zapped. Alexander Usyskin (1): mei: protect mei_cl_mtu from null dereference Anand Jain (1): btrfs: dev-replace: fail mount if we don't have replace item with target device Anand K Mistry (1): x86/speculation: Allow IBPB to be conditionally enabled on CPUs with always-on STIBP Andrew Jeffery (1): ARM: 9019/1: kprobes: Avoid fortify_panic() when copying optprobe template Andy Shevchenko (1): pinctrl: intel: Set default bias in case no particular value given Ard Biesheuvel (1): crypto: arm64/aes-modes - get rid of literal load of addend vector Arnaldo Carvalho de Melo (1): perf scripting python: Avoid declaring function pointers with a visibility attribute Arnaud de Turckheim (3): gpio: pcie-idio-24: Fix irq mask when masking gpio: pcie-idio-24: Fix IRQ Enable Register value gpio: pcie-idio-24: Enable PEX8311 interrupts Baolin Wang (1): mfd: sprd: Add wakeup capability for PMIC IRQ Billy Tsai (1): pinctrl: aspeed: Fix GPI only function problem. Bob Peterson (3): gfs2: Free rd_bits later in gfs2_clear_rgrpd to fix use-after-free gfs2: Add missing truncate_inode_pages_final for sd_aspace gfs2: check for live vs. read-only file system in gfs2_fitrim Boris Protopopov (1): Convert trailing spaces and periods in path components Brian Foster (1): xfs: flush new eof page on truncate to avoid post-eof corruption Chen Zhou (1): selinux: Fix error return code in sel_ib_pkey_sid_slow() Chris Brandt (1): usb: cdc-acm: Add DISABLE_ECHO for Renesas USB Download mode Christoph Hellwig (2): nbd: fix a block_device refcount leak in nbd_release xfs: fix a missing unlock on error in xfs_fs_map_blocks Chunyan Zhang (1): tick/common: Touch watchdog in tick_unfreeze() on all CPUs Coiby Xu (2): pinctrl: amd: use higher precision for 512 RtcClk pinctrl: amd: fix incorrect way to disable debounce filter Dan Carpenter (3): ALSA: hda: prevent undefined shift in snd_hdac_ext_bus_get_link() can: peak_usb: add range checking in decode operations futex: Don't enable IRQs unconditionally in put_pi_state() Darrick J. Wong (6): xfs: set xefi_discard when creating a deferred agfl free log intent item xfs: fix scrub flagging rtinherit even if there is no rt device xfs: fix flags argument to rmap lookup when converting shared file rmaps xfs: set the unwritten bit in rmap lookup flags in xchk_bmap_get_rmapextents xfs: fix rmap key and record comparison functions xfs: fix brainos in the refcount scrubber's rmap fragment processor Dinghao Liu (1): btrfs: ref-verify: fix memory leak in btrfs_ref_tree_mod Evan Nimmo (1): of/address: Fix of_node memory leak in of_dma_is_coherent Evan Quan (3): drm/amdgpu: perform srbm soft reset always on SDMA resume drm/amd/pm: perform SMC reset on suspend/hibernation drm/amd/pm: do not use ixFEATURE_STATUS for checking smc running Evgeny Novikov (1): usb: gadget: goku_udc: fix potential crashes in probe Filipe Manana (1): Btrfs: fix missing error return if writeback for extent buffer never started Gao Xiang (1): erofs: derive atime instead of leaving it empty George Spelvin (1): random32: make prandom_u32() output unpredictable Greg Kroah-Hartman (1): Linux 4.19.158 Hannes Reinecke (1): scsi: scsi_dh_alua: Avoid crash during alua_bus_detach() Heiner Kallweit (1): r8169: fix potential skb double free in an error path Jason A. Donenfeld (1): netfilter: use actual socket sk rather than skb sk when routing harder Jerry Snitselaar (1): tpm_tis: Disable interrupts on ThinkPad T490s Jing Xiangfeng (1): thunderbolt: Add the missed ida_simple_remove() in ring_request_msix() Jiri Olsa (1): perf tools: Add missing swap for ino_generation Joakim Zhang (1): can: flexcan: remove FLEXCAN_QUIRK_DISABLE_MECR quirk for LS1021A Johannes Berg (1): mac80211: fix use of skb payload instead of header Johannes Thumshirn (1): btrfs: reschedule when cloning lots of extents Josef Bacik (1): btrfs: sysfs: init devices outside of the chunk_mutex Joseph Qi (1): ext4: unlock xattr_sem properly in ext4_inline_data_truncate() Kaixu Xia (1): ext4: correctly report "not supported" for {usr,grp}jquota when !CONFIG_QUOTA Keita Suzuki (1): scsi: hpsa: Fix memory leak in hpsa_init_one() Mao Wenan (1): net: Update window_clamp if SOCK_RCVBUF is set Marc Kleine-Budde (1): can: rx-offload: don't call kfree_skb() from IRQ context Marc Zyngier (1): genirq: Let GENERIC_IRQ_IPI select IRQ_DOMAIN_HIERARCHY Martin Schiller (1): net/x25: Fix null-ptr-deref in x25_connect Martin Willi (1): vrf: Fix fast path output packet handling with async Netfilter rules Masashi Honma (1): ath9k_htc: Use appropriate rs_datalen type Matteo Croce (2): Revert "kernel/reboot.c: convert simple_strtoul to kstrtoint" reboot: fix overflow parsing reboot cpu number Matthew Wilcox (Oracle) (1): btrfs: fix potential overflow in cluster_pages_for_defrag on 32bit arch Michał Mirosław (1): regulator: defer probe when trying to get voltage from unresolved supply Mika Westerberg (1): thunderbolt: Fix memory leak if ida_simple_get() fails in enumerate_services() Olaf Hering (1): hv_balloon: disable warning when floor reached Oleksij Rempel (1): can: can_create_echo_skb(): fix echo skb generation: always use skb_clone() Oliver Hartkopp (1): can: dev: __can_get_echo_skb(): fix real payload length return value for RTR frames Oliver Herms (1): IPv6: Set SIT tunnel hard_header_len to zero Peter Zijlstra (1): perf: Fix get_recursion_context() Qian Cai (1): s390/smp: move rcu_cpu_starting() earlier Shin'ichiro Kawasaki (1): uio: Fix use-after-free in uio_unregister_device() Stefano Brivio (1): netfilter: ipset: Update byte and packet counters regardless of whether they match Stefano Stabellini (1): swiotlb: fix "x86: Don't panic if can not alloc buffer for swiotlb" Stephane Grosjean (2): can: peak_usb: peak_usb_get_ts_time(): fix timestamp wrapping can: peak_canfd: pucan_handle_can_rx(): fix echo management when loopback is on Suravee Suthikulpanit (1): iommu/amd: Increase interrupt remapping table limit to 512 entries Sven Van Asbroeck (1): lan743x: fix "BUG: invalid wait context" when setting rx mode Thinh Nguyen (2): usb: dwc3: gadget: Continue to process pending requests usb: dwc3: gadget: Reclaim extra TRBs after request completion Thomas Zimmermann (1): drm/gma500: Fix out-of-bounds access to struct drm_device.vblank[] Tommi Rantala (1): selftests: proc: fix warning: _GNU_SOURCE redefined Tyler Hicks (1): tpm: efi: Don't create binary_bios_measurements file for an empty log Ursula Braun (1): net/af_iucv: fix null pointer dereference on shutdown Vincent Mailhol (1): can: dev: can_get_echo_skb(): prevent call to kfree_skb() in hard IRQ context Wang Hai (2): cosa: Add missing kfree in error path of cosa_write tipc: fix memory leak in tipc_topsrv_start() Wengang Wang (1): ocfs2: initialize ip_next_orphan Ye Bin (1): cfg80211: regulatory: Fix inconsistent format argument Yoshihiro Shimoda (1): mmc: renesas_sdhi_core: Add missing tmio_mmc_host_free() at remove Yunsheng Lin (1): net: sch_generic: fix the missing new qdisc assignment bug Zeng Tao (1): time: Prevent undefined behaviour in timespec64_to_ns() Zhang Qilong (2): vfio: platform: fix reference leak in vfio_platform_open xhci: hisilicon: fix refercence leak in xhci_histb_probe zhuoliang zhang (1): net: xfrm: fix a race condition during allocing spi Makefile | 2 +- arch/arm/include/asm/kprobes.h | 22 +- arch/arm/probes/kprobes/opt-arm.c | 18 +- arch/arm64/crypto/aes-modes.S | 16 +- arch/s390/kernel/smp.c | 3 +- arch/x86/kernel/cpu/bugs.c | 52 +- drivers/block/nbd.c | 1 + drivers/char/random.c | 1 - drivers/char/tpm/eventlog/efi.c | 5 + drivers/char/tpm/tpm_tis.c | 29 +- drivers/gpio/gpio-pcie-idio-24.c | 62 ++- drivers/gpu/drm/amd/amdgpu/cik_sdma.c | 27 +- .../gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c | 4 + drivers/gpu/drm/amd/powerplay/inc/hwmgr.h | 1 + drivers/gpu/drm/amd/powerplay/inc/smumgr.h | 2 + .../gpu/drm/amd/powerplay/smumgr/ci_smumgr.c | 29 +- drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c | 8 + drivers/gpu/drm/gma500/psb_irq.c | 34 +- drivers/hv/hv_balloon.c | 2 +- drivers/iommu/amd_iommu_types.h | 6 +- drivers/mfd/sprd-sc27xx-spi.c | 28 +- drivers/misc/mei/client.h | 4 +- drivers/mmc/host/renesas_sdhi_core.c | 1 + drivers/net/can/dev.c | 14 +- drivers/net/can/flexcan.c | 3 +- drivers/net/can/peak_canfd/peak_canfd.c | 11 +- drivers/net/can/rx-offload.c | 4 +- drivers/net/can/usb/peak_usb/pcan_usb_core.c | 51 +- drivers/net/can/usb/peak_usb/pcan_usb_fd.c | 48 +- drivers/net/ethernet/microchip/lan743x_main.c | 12 +- drivers/net/ethernet/microchip/lan743x_main.h | 3 - drivers/net/ethernet/realtek/r8169.c | 3 +- drivers/net/vrf.c | 92 +++- drivers/net/wan/cosa.c | 1 + drivers/net/wireless/ath/ath9k/htc_drv_txrx.c | 2 +- drivers/of/address.c | 4 +- drivers/pinctrl/aspeed/pinctrl-aspeed.c | 7 +- drivers/pinctrl/intel/pinctrl-intel.c | 8 + drivers/pinctrl/pinctrl-amd.c | 6 +- drivers/regulator/core.c | 2 + drivers/scsi/device_handler/scsi_dh_alua.c | 9 +- drivers/scsi/hpsa.c | 4 +- drivers/staging/erofs/inode.c | 21 +- drivers/thunderbolt/nhi.c | 19 +- drivers/thunderbolt/xdomain.c | 1 + drivers/uio/uio.c | 10 +- drivers/usb/class/cdc-acm.c | 9 + drivers/usb/dwc3/gadget.c | 32 +- drivers/usb/gadget/udc/goku_udc.c | 2 +- drivers/usb/host/xhci-histb.c | 2 +- drivers/vfio/platform/vfio_platform_common.c | 3 +- fs/btrfs/dev-replace.c | 26 +- fs/btrfs/extent_io.c | 4 + fs/btrfs/ioctl.c | 12 +- fs/btrfs/ref-verify.c | 1 + fs/btrfs/volumes.c | 33 +- fs/cifs/cifs_unicode.c | 8 +- fs/ext4/inline.c | 1 + fs/ext4/super.c | 4 +- fs/gfs2/rgrp.c | 5 +- fs/gfs2/super.c | 1 + fs/ocfs2/super.c | 1 + fs/xfs/libxfs/xfs_alloc.c | 1 + fs/xfs/libxfs/xfs_bmap.h | 2 +- fs/xfs/libxfs/xfs_rmap.c | 2 +- fs/xfs/libxfs/xfs_rmap_btree.c | 16 +- fs/xfs/scrub/bmap.c | 2 + fs/xfs/scrub/inode.c | 3 +- fs/xfs/scrub/refcount.c | 8 +- fs/xfs/xfs_iops.c | 10 + fs/xfs/xfs_pnfs.c | 2 +- include/linux/can/skb.h | 20 +- include/linux/netfilter_ipv4.h | 2 +- include/linux/netfilter_ipv6.h | 2 +- include/linux/prandom.h | 36 +- include/linux/time64.h | 4 + kernel/dma/swiotlb.c | 6 +- kernel/events/internal.h | 2 +- kernel/exit.c | 5 +- kernel/futex.c | 5 +- kernel/irq/Kconfig | 1 + kernel/reboot.c | 28 +- kernel/time/itimer.c | 4 - kernel/time/tick-common.c | 2 + kernel/time/timer.c | 7 - lib/random32.c | 462 +++++++++++------- net/ipv4/netfilter.c | 12 +- net/ipv4/netfilter/ipt_SYNPROXY.c | 2 +- net/ipv4/netfilter/iptable_mangle.c | 2 +- net/ipv4/netfilter/nf_nat_l3proto_ipv4.c | 2 +- net/ipv4/netfilter/nf_reject_ipv4.c | 2 +- net/ipv4/netfilter/nft_chain_route_ipv4.c | 2 +- net/ipv4/syncookies.c | 9 +- net/ipv6/netfilter.c | 6 +- net/ipv6/netfilter/ip6table_mangle.c | 2 +- net/ipv6/netfilter/nf_nat_l3proto_ipv6.c | 2 +- net/ipv6/netfilter/nft_chain_route_ipv6.c | 2 +- net/ipv6/sit.c | 2 - net/ipv6/syncookies.c | 10 +- net/iucv/af_iucv.c | 3 +- net/mac80211/tx.c | 37 +- net/netfilter/ipset/ip_set_core.c | 3 +- net/netfilter/ipvs/ip_vs_core.c | 4 +- net/sched/sch_generic.c | 3 + net/tipc/topsrv.c | 10 +- net/wireless/reg.c | 2 +- net/x25/af_x25.c | 2 +- net/xfrm/xfrm_state.c | 8 +- security/selinux/ibpkey.c | 4 +- sound/hda/ext/hdac_ext_controller.c | 2 + .../scripting-engines/trace-event-python.c | 7 +- tools/perf/util/session.c | 1 + .../testing/selftests/proc/proc-loadavg-001.c | 1 - .../selftests/proc/proc-self-syscall.c | 1 - .../testing/selftests/proc/proc-uptime-002.c | 1 - 115 files changed, 1073 insertions(+), 539 deletions(-) -- 2.25.1

1 98

[PATCH 1/2] ascend: share_pool: support debug mode and refactor some functions
by Yang Yingliang 23 Nov '20

23 Nov '20

From: Ding Tianhong <dingtianhong(a)huawei.com> ascend inclusion category: feature bugzilla: NA CVE: NA ------------------------------------------------- The share pool is used widely for several accelerator, and it is difficult to debug the user problem, so add debug mode to analyse the problem, this mode is enabled by the sysctl_sp_debug_mode flag. Some functions have been refactored to protect the critical area correctly, and output message more clearly. Signed-off-by: Tang Yizhou <tangyizhou(a)huawei.com> Signed-off-by: Zhou Guanghui <zhouguanghui1(a)huawei.com> Signed-off-by: Wu Peng <wupeng58(a)huawei.com> Signed-off-by: Ding Tianhong <dingtianhong(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/linux/share_pool.h | 15 +- kernel/sysctl.c | 9 + mm/share_pool.c | 537 +++++++++++++++++++++---------------- 3 files changed, 326 insertions(+), 235 deletions(-) diff --git a/include/linux/share_pool.h b/include/linux/share_pool.h index 09afbae33d41..2557ef138122 100644 --- a/include/linux/share_pool.h +++ b/include/linux/share_pool.h @@ -5,6 +5,7 @@ #include <linux/mm_types.h> #include <linux/notifier.h> #include <linux/vmalloc.h> +#include <linux/printk.h> #define SP_HUGEPAGE (1 << 0) #define SP_HUGEPAGE_ONLY (1 << 1) @@ -35,6 +36,8 @@ extern int sysctl_share_pool_hugepage_enable; extern int sysctl_ac_mode; +extern int sysctl_sp_debug_mode; + extern int enable_ascend_share_pool; /* Processes in the same sp_group can share memory. @@ -70,7 +73,7 @@ struct sp_group { /* number of sp_area */ atomic_t spa_num; /* total size of all sp_area from sp_alloc and k2u(spg) */ - atomic_t size; + atomic64_t size; /* record the number of hugepage allocation failures */ int hugepage_failures; /* is_alive == false means it's being destroyed */ @@ -211,6 +214,12 @@ static inline bool sp_mmap_check(unsigned long flags) return false; } +static inline void sp_dump_stack(void) +{ + if (sysctl_sp_debug_mode) + dump_stack(); +} + #else static inline int sp_group_add_task(int pid, int spg_id) @@ -349,6 +358,10 @@ static inline bool sp_mmap_check(unsigned long flags) { return false; } + +static inline void sp_dump_stack(void) +{ +} #endif #endif /* LINUX_SHARE_POOL_H */ diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 61e62f1ccee4..26c215fb37dc 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1737,6 +1737,15 @@ static struct ctl_table vm_table[] = { .extra1 = &zero, .extra2 = &one, }, + { + .procname = "sharepool_debug_mode", + .data = &sysctl_sp_debug_mode, + .maxlen = sizeof(sysctl_sp_debug_mode), + .mode = 0600, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, #endif { } }; diff --git a/mm/share_pool.c b/mm/share_pool.c index fcbc831f7f8c..24c5dd680451 100644 --- a/mm/share_pool.c +++ b/mm/share_pool.c @@ -57,6 +57,8 @@ static const int mdc_default_group_id = 1; /* access control mode */ int sysctl_ac_mode = AC_NONE; +/* debug mode */ +int sysctl_sp_debug_mode; /* idr of all sp_groups */ static DEFINE_IDR(sp_group_idr); @@ -85,9 +87,11 @@ struct sp_proc_stat { /* for kthread buff_module_guard_work */ static struct sp_proc_stat kthread_stat = {0}; -/* The caller must hold sp_mutex. */ -static struct sp_proc_stat *sp_init_proc_stat(struct task_struct *tsk) -{ +/* + * The caller must hold sp_mutex and ensure no concurrency problem + * for task_struct and mm_struct. + */ +static struct sp_proc_stat *sp_init_proc_stat(struct task_struct *tsk) { struct sp_proc_stat *stat; int id = tsk->mm->sp_stat_id; int tgid = tsk->tgid; @@ -138,7 +142,7 @@ static struct sp_spa_stat spa_stat = {0}; /* statistics of all sp group born from sp_alloc and k2u(spg) */ struct sp_spg_stat { atomic_t spa_total_num; - atomic_t spa_total_size; + atomic64_t spa_total_size; }; static struct sp_spg_stat spg_stat = {0}; @@ -166,10 +170,11 @@ struct sp_area { struct list_head link; /* link to the spg->head */ struct sp_group *spg; enum spa_type type; /* where spa born from */ + struct mm_struct *mm; /* owner of k2u(task) */ }; static DEFINE_SPINLOCK(sp_area_lock); static struct rb_root sp_area_root = RB_ROOT; -bool host_svm_sp_enable = false; +static bool host_svm_sp_enable = false; int sysctl_share_pool_hugepage_enable = 1; @@ -241,7 +246,7 @@ static int spa_dec_usage(enum spa_type type, unsigned long size) return 0; } -static void *sp_mmap(struct mm_struct *mm, struct file *file, +static unsigned long sp_mmap(struct mm_struct *mm, struct file *file, struct sp_area *spa, unsigned long *populate); static void free_sp_group(struct sp_group *spg) @@ -274,7 +279,18 @@ static struct sp_group *__sp_find_spg(int pid, int spg_id) if (ret) return NULL; - spg = tsk->mm->sp_group; + /* + * Once we encounter a concurrency problem here. + * To fix it, we believe get_task_mm() and mmput() is too + * heavy because we just get the pointer of sp_group. + */ + task_lock(tsk); + if (tsk->mm == NULL) + spg = NULL; + else + spg = tsk->mm->sp_group; + task_unlock(tsk); + put_task_struct(tsk); } else { spg = idr_find(&sp_group_idr, spg_id); @@ -318,7 +334,7 @@ static struct sp_group *find_or_alloc_sp_group(int spg_id) } spg->id = spg_id; atomic_set(&spg->spa_num, 0); - atomic_set(&spg->size, 0); + atomic64_set(&spg->size, 0); spg->is_alive = true; spg->hugepage_failures = 0; spg->dvpp_multi_spaces = false; @@ -377,9 +393,6 @@ static void sp_munmap_task_areas(struct mm_struct *mm, struct list_head *stop) struct sp_area *spa, *prev = NULL; int err; - if (!mmget_not_zero(mm)) - return; - down_write(&mm->mmap_sem); spin_lock(&sp_area_lock); list_for_each_entry(spa, &mm->sp_group->spa_list, link) { @@ -406,8 +419,17 @@ static void sp_munmap_task_areas(struct mm_struct *mm, struct list_head *stop) __sp_area_drop_locked(prev); spin_unlock(&sp_area_lock); - up_write(&mm->mmap_sem); - mmput(mm); +} + +/* The caller must hold sp_mutex. */ +static void __sp_group_drop_locked(struct sp_group *spg) +{ + bool is_alive = spg->is_alive; + + if (atomic_dec_and_test(&spg->use_count)) { + BUG_ON(is_alive); + free_sp_group(spg); + } } /** @@ -446,8 +468,9 @@ int sp_group_add_task(int pid, int spg_id) spg = idr_find(&sp_group_idr, spg_id); if (!spg_valid(spg)) { mutex_unlock(&sp_mutex); - pr_err("share pool: task add group failed because group id %d hasn't been create or dead\n", - spg_id); + if (printk_ratelimit()) + pr_err("share pool: task add group failed because group id %d " + "hasn't been create or dead\n", spg_id); return -EINVAL; } mutex_unlock(&sp_mutex); @@ -457,7 +480,9 @@ int sp_group_add_task(int pid, int spg_id) spg_id = ida_alloc_range(&sp_group_id_ida, SPG_ID_AUTO_MIN, SPG_ID_AUTO_MAX, GFP_ATOMIC); if (spg_id < 0) { - pr_err("share pool: task add group failed when automatically generate group id failed\n"); + if (printk_ratelimit()) + pr_err("share pool: task add group failed when automatically " + "generate group id failed\n"); return spg_id; } } @@ -467,8 +492,9 @@ int sp_group_add_task(int pid, int spg_id) SPG_ID_DVPP_PASS_THROUGH_MIN, SPG_ID_DVPP_PASS_THROUGH_MAX, GFP_ATOMIC); if (spg_id < 0) { - pr_err("share pool: task add group failed when automatically generate group id failed" - "in DVPP pass through\n"); + if (printk_ratelimit()) + pr_err("share pool: task add group failed when automatically " + "generate group id failed in DVPP pass through\n"); return spg_id; } } @@ -494,25 +520,31 @@ int sp_group_add_task(int pid, int spg_id) ret = PTR_ERR(spg); goto out_put_task; } + atomic_inc(&spg->use_count); + /* access control permission check */ if (sysctl_ac_mode == AC_SINGLE_OWNER) { if (spg->owner != current->group_leader) { ret = -EPERM; - goto out_put_task; + goto out_drop_group; } } + mm = get_task_mm(tsk); + if (!mm) { + ret = -ESRCH; + goto out_drop_group; + } + /* per process statistics initialization */ stat = sp_init_proc_stat(tsk); if (IS_ERR(stat)) { ret = PTR_ERR(stat); pr_err("share pool: init proc stat failed, ret %lx\n", PTR_ERR(stat)); - goto out_put_task; + goto out_put_mm; } - mm = tsk->mm; mm->sp_group = spg; - atomic_inc(&spg->use_count); list_add_tail(&tsk->mm->sp_node, &spg->procs); /* * create mappings of existing shared memory segments into this @@ -523,7 +555,7 @@ int sp_group_add_task(int pid, int spg_id) list_for_each_entry(spa, &spg->spa_list, link) { unsigned long populate = 0; struct file *file = spa_file(spa); - void *p; + unsigned long addr; if (prev) __sp_area_drop_locked(prev); @@ -532,28 +564,24 @@ int sp_group_add_task(int pid, int spg_id) atomic_inc(&spa->use_count); spin_unlock(&sp_area_lock); - p = sp_mmap(mm, file, spa, &populate); - if (IS_ERR(p) && (PTR_ERR(p) != -ESPGMMEXIT)) { + down_write(&mm->mmap_sem); + addr = sp_mmap(mm, file, spa, &populate); + if (IS_ERR_VALUE(addr)) { sp_munmap_task_areas(mm, &spa->link); - ret = PTR_ERR(p); + up_write(&mm->mmap_sem); + ret = addr; pr_err("share pool: task add group sp mmap failed, ret %d\n", ret); spin_lock(&sp_area_lock); break; } - - if (PTR_ERR(p) == -ESPGMMEXIT) { - pr_err("share pool: task add group sp mmap failed, ret -ESPGMEXIT\n"); - spin_lock(&sp_area_lock); - ret = -ESPGMMEXIT; - break; - } + up_write(&mm->mmap_sem); if (populate) { ret = do_mm_populate(mm, spa->va_start, populate, 0); if (ret) { if (printk_ratelimit()) - pr_err("share pool: task add group failed when mm populate failed: %d\n", - ret); + pr_warn("share pool: task add group failed when mm populate " + "failed (potential no enough memory): %d\n", ret); sp_munmap_task_areas(mm, spa->link.next); } } @@ -567,8 +595,16 @@ int sp_group_add_task(int pid, int spg_id) if (unlikely(ret)) { idr_remove(&sp_stat_idr, mm->sp_stat_id); kfree(stat); + mm->sp_stat_id = 0; + list_del(&mm->sp_node); + mm->sp_group = NULL; } +out_put_mm: + mmput(mm); +out_drop_group: + if (unlikely(ret)) + __sp_group_drop_locked(spg); out_put_task: put_task_struct(tsk); out_unlock: @@ -609,9 +645,6 @@ void sp_group_exit(struct mm_struct *mm) bool is_alive = true; bool unlock; - if (!enable_ascend_share_pool) - return; - /* * Nothing to do if this thread group doesn't belong to any sp_group. * No need to protect this check with lock because we can add a task @@ -638,18 +671,13 @@ void sp_group_exit(struct mm_struct *mm) void sp_group_post_exit(struct mm_struct *mm) { - bool is_alive; struct sp_proc_stat *stat; bool unlock; - if (!enable_ascend_share_pool) - return; - if (!mm->sp_group) return; spg_exit_lock(&unlock); - is_alive = mm->sp_group->is_alive; /* pointer stat must be valid, we don't need to check sanity */ stat = idr_find(&sp_stat_idr, mm->sp_stat_id); @@ -673,10 +701,7 @@ void sp_group_post_exit(struct mm_struct *mm) idr_remove(&sp_stat_idr, mm->sp_stat_id); - if (atomic_dec_and_test(&mm->sp_group->use_count)) { - BUG_ON(is_alive); - free_sp_group(mm->sp_group); - } + __sp_group_drop_locked(mm->sp_group); spg_exit_unlock(unlock); kfree(stat); @@ -716,7 +741,7 @@ static void __insert_sp_area(struct sp_area *spa) static struct sp_area *sp_alloc_area(unsigned long size, unsigned long flags, struct sp_group *spg, enum spa_type type) { - struct sp_area *spa; + struct sp_area *spa, *err; struct rb_node *n; unsigned long vstart = MMAP_SHARE_POOL_START; unsigned long vend = MMAP_SHARE_POOL_16G_START; @@ -728,6 +753,11 @@ static struct sp_area *sp_alloc_area(unsigned long size, unsigned long flags, vstart = MMAP_SHARE_POOL_16G_START; vend = MMAP_SHARE_POOL_16G_START + MMAP_SHARE_POOL_16G_SIZE; } else { + if (!spg) { + if (printk_ratelimit()) + pr_err("share pool: don't allow k2u(task) in host svm multiprocess scene\n"); + return ERR_PTR(-EINVAL); + } vstart = spg->dvpp_va_start; vend = spg->dvpp_va_start + spg->dvpp_size; } @@ -735,14 +765,11 @@ static struct sp_area *sp_alloc_area(unsigned long size, unsigned long flags, addr = vstart; - if (!sysctl_share_pool_hugepage_enable) - flags &= ~(SP_HUGEPAGE_ONLY | SP_HUGEPAGE); - spa = kmalloc(sizeof(struct sp_area), GFP_KERNEL); if (unlikely(!spa)) { if (printk_ratelimit()) pr_err("share pool: alloc spa failed due to lack of memory\n"); - return NULL; + return ERR_PTR(-ENOMEM); } spin_lock(&sp_area_lock); @@ -788,6 +815,7 @@ static struct sp_area *sp_alloc_area(unsigned long size, unsigned long flags, } found: if (addr + size_align > vend) { + err = ERR_PTR(-EOVERFLOW); goto error; } @@ -799,15 +827,17 @@ static struct sp_area *sp_alloc_area(unsigned long size, unsigned long flags, atomic_set(&spa->use_count, 1); spa->type = type; - if (spa_inc_usage(type, size)) + if (spa_inc_usage(type, size)) { + err = ERR_PTR(-EINVAL); goto error; + } __insert_sp_area(spa); if (spa->spg) { atomic_inc(&spg->spa_num); - atomic_add(size, &spg->size); + atomic64_add(size, &spg->size); atomic_inc(&spg_stat.spa_total_num); - atomic_add(size, &spg_stat.spa_total_size); + atomic64_add(size, &spg_stat.spa_total_size); list_add_tail(&spa->link, &spg->spa_list); } spin_unlock(&sp_area_lock); @@ -817,7 +847,7 @@ static struct sp_area *sp_alloc_area(unsigned long size, unsigned long flags, error: spin_unlock(&sp_area_lock); kfree(spa); - return NULL; + return err; } /* the caller should hold sp_area_lock */ @@ -862,9 +892,9 @@ static void sp_free_area(struct sp_area *spa) spa_dec_usage(spa->type, spa->real_size); /* won't fail */ if (spa->spg) { atomic_dec(&spa->spg->spa_num); - atomic_sub(spa->real_size, &spa->spg->size); + atomic64_sub(spa->real_size, &spa->spg->size); atomic_dec(&spg_stat.spa_total_num); - atomic_sub(spa->real_size, &spg_stat.spa_total_size); + atomic64_sub(spa->real_size, &spg_stat.spa_total_size); list_del(&spa->link); } rb_erase(&spa->rb_node, &sp_area_root); @@ -898,7 +928,7 @@ void sp_area_drop(struct vm_area_struct *vma) { struct sp_area *spa; - if (!sp_check_vm_share_pool(vma->vm_flags)) + if (!(vma->vm_flags & VM_SHARE_POOL)) return; /* @@ -979,13 +1009,25 @@ int sp_free(unsigned long addr) } else { /* spa == NULL */ ret = -EINVAL; if (printk_ratelimit()) - pr_err("share pool: sp_free invalid input addr %pK\n", (void *)addr); + pr_err("share pool: sp free invalid input addr %pK\n", (void *)addr); goto out; } + if (spa->type != SPA_TYPE_ALLOC) { + if (printk_ratelimit()) + pr_err("share pool: sp free failed, addr %pK is not from sp_alloc\n", + (void *)addr); + } + if (!spg_valid(spa->spg)) goto drop_spa; + pr_notice("share pool: [sp free] caller %s(%d/%d); " + "group id %d addr 0x%pK, size %ld\n", + current->comm, current->tgid, current->pid, spa->spg->id, + (void *)spa->va_start, spa->real_size); + sp_dump_stack(); + __sp_free(spa->spg, spa->va_start, spa_size(spa), NULL); /* Free the memory of the backing shmem or hugetlbfs */ @@ -993,7 +1035,7 @@ int sp_free(unsigned long addr) offset = addr - MMAP_SHARE_POOL_START; ret = vfs_fallocate(spa_file(spa), mode, offset, spa_size(spa)); if (ret) - pr_err("share pool: fallocate failed: %d\n", ret); + pr_err("share pool: sp free fallocate failed: %d\n", ret); /* pointer stat may be invalid because of kthread buff_module_guard_work */ if (current->mm == NULL) { @@ -1016,7 +1058,7 @@ int sp_free(unsigned long addr) EXPORT_SYMBOL_GPL(sp_free); /* wrapper of __do_mmap() and the caller must hold down_write(&mm->mmap_sem). */ -static unsigned long __sp_mmap(struct mm_struct *mm, struct file *file, +static unsigned long sp_mmap(struct mm_struct *mm, struct file *file, struct sp_area *spa, unsigned long *populate) { unsigned long addr = spa->va_start; @@ -1033,30 +1075,13 @@ static unsigned long __sp_mmap(struct mm_struct *mm, struct file *file, if (IS_ERR_VALUE(addr)) { atomic_dec(&spa->use_count); pr_err("share pool: do_mmap fails %ld\n", addr); + } else { + BUG_ON(addr != spa->va_start); } return addr; } -static void *sp_mmap(struct mm_struct *mm, struct file *file, - struct sp_area *spa, unsigned long *populate) -{ - unsigned long addr; - - if (!mmget_not_zero(mm)) - return ERR_PTR(-ESPGMMEXIT); - down_write(&mm->mmap_sem); - addr = __sp_mmap(mm, file, spa, populate); - up_write(&mm->mmap_sem); - mmput(mm); - - if (IS_ERR_VALUE(addr)) - return ERR_PTR(addr); - - BUG_ON(addr != spa->va_start); - return (void *)addr; -} - /** * Allocate shared memory for all the processes in the same sp_group * size - the size of memory to allocate @@ -1071,12 +1096,14 @@ void *sp_alloc(unsigned long size, unsigned long sp_flags, int spg_id) struct sp_area *spa = NULL; struct sp_proc_stat *stat; unsigned long sp_addr; - void *p_mmap, *p = ERR_PTR(-ENODEV); + unsigned long mmap_addr; + void *p = ERR_PTR(-ENODEV); struct mm_struct *mm; struct file *file; unsigned long size_aligned; int ret = 0; struct mm_struct *tmp; + unsigned long mode, offset; /* mdc scene hack */ if (enable_mdc_default_group) @@ -1133,9 +1160,6 @@ void *sp_alloc(unsigned long size, unsigned long sp_flags, int spg_id) goto out; } - if (!sysctl_share_pool_hugepage_enable) - sp_flags &= ~(SP_HUGEPAGE_ONLY | SP_HUGEPAGE); - if (sp_flags & SP_HUGEPAGE) { file = spg->file_hugetlb; size_aligned = ALIGN(size, PMD_SIZE); @@ -1145,10 +1169,12 @@ void *sp_alloc(unsigned long size, unsigned long sp_flags, int spg_id) } try_again: spa = sp_alloc_area(size_aligned, sp_flags, spg, SPA_TYPE_ALLOC); - if (!spa) { + if (IS_ERR(spa)) { if (printk_ratelimit()) - pr_err("share pool: allocation failed due to alloc spa failure\n"); - p = ERR_PTR(-ENOMEM); + pr_err("share pool: allocation failed due to alloc spa failure " + "(potential no enough virtual memory when -75): %ld\n", + PTR_ERR(spa)); + p = spa; goto out; } sp_addr = spa->va_start; @@ -1158,33 +1184,34 @@ void *sp_alloc(unsigned long size, unsigned long sp_flags, int spg_id) unsigned long populate = 0; struct vm_area_struct *vma; - p_mmap = sp_mmap(mm, file, spa, &populate); - if (IS_ERR(p_mmap) && (PTR_ERR(p_mmap) != -ESPGMMEXIT)) { - p = p_mmap; + if (!mmget_not_zero(mm)) + continue; + + down_write(&mm->mmap_sem); + mmap_addr = sp_mmap(mm, file, spa, &populate); + if (IS_ERR_VALUE(mmap_addr)) { + up_write(&mm->mmap_sem); + p = (void *)mmap_addr; __sp_free(spg, sp_addr, size_aligned, mm); - pr_err("share pool: allocation sp mmap failed, ret %ld\n", PTR_ERR(p_mmap)); - break; + mmput(mm); + pr_err("share pool: allocation sp mmap failed, ret %ld\n", mmap_addr); + goto out; } - if (PTR_ERR(p_mmap) == -ESPGMMEXIT) { - pr_info("share pool: allocation sp mmap failed, ret -ESPGMMEXIT\n"); + p =(void *)mmap_addr; /* success */ + if (populate == 0) { + up_write(&mm->mmap_sem); + mmput(mm); continue; } - p = p_mmap; /* success */ - if (populate == 0) - continue; - - if (!mmget_not_zero(mm)) - continue; - down_write(&mm->mmap_sem); vma = find_vma(mm, sp_addr); if (unlikely(!vma)) { + up_write(&mm->mmap_sem); + mmput(mm); pr_err("share pool: allocation failed due to find %pK vma failure\n", (void *)sp_addr); p = ERR_PTR(-EINVAL); - up_write(&mm->mmap_sem); - mmput(mm); goto out; } /* clean PTE_RDONLY flags or trigger SMMU event */ @@ -1216,9 +1243,17 @@ void *sp_alloc(unsigned long size, unsigned long sp_flags, int spg_id) } if (printk_ratelimit()) - pr_err("share pool: allocation failed due to mm populate failed: %d\n", - ret); + pr_warn("share pool: allocation failed due to mm populate failed" + "(potential no enough memory when -12): %d\n", ret); p = ERR_PTR(ret); + __sp_area_drop(spa); + + mode = FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE; + offset = sp_addr - MMAP_SHARE_POOL_START; + ret = vfs_fallocate(spa_file(spa), mode, offset, spa_size(spa)); + if (ret) + pr_err("share pool: fallocate failed %d\n", ret); + mmput(mm); break; } @@ -1235,24 +1270,20 @@ void *sp_alloc(unsigned long size, unsigned long sp_flags, int spg_id) mutex_unlock(&sp_mutex); /* this will free spa if mmap failed */ - if (spa) + if (spa && !IS_ERR(spa)) __sp_area_drop(spa); + if (!IS_ERR(p)) { + pr_notice("share pool: [sp alloc] caller %s(%d/%d); group id %d; " + "return addr 0x%pK, size %ld\n", + current->comm, current->tgid, current->pid, spa->spg->id, + (void *)spa->va_start, spa->real_size); + sp_dump_stack(); + } return p; } EXPORT_SYMBOL_GPL(sp_alloc); -static unsigned long __sp_remap_get_pfn(unsigned long kva) -{ - unsigned long pfn; - if (is_vmalloc_addr((void *)kva)) - pfn = vmalloc_to_pfn((void *)kva); - else - pfn = virt_to_pfn(kva); - - return pfn; -} - /* * return value: >0 means this is a hugepage addr * =0 means a normal addr. <0 means an errno. @@ -1286,7 +1317,6 @@ static unsigned long sp_remap_kva_to_vma(unsigned long kva, struct sp_area *spa, struct vm_area_struct *vma; unsigned long ret_addr; unsigned long populate = 0; - unsigned long addr, buf, offset; struct file *file = NULL; int ret = 0; struct user_struct *user = NULL; @@ -1307,7 +1337,7 @@ static unsigned long sp_remap_kva_to_vma(unsigned long kva, struct sp_area *spa, } down_write(&mm->mmap_sem); - ret_addr = __sp_mmap(mm, file, spa, &populate); + ret_addr = sp_mmap(mm, file, spa, &populate); if (IS_ERR_VALUE(ret_addr)) { pr_err("share pool: k2u mmap failed %lx\n", ret_addr); goto out; @@ -1326,20 +1356,12 @@ static unsigned long sp_remap_kva_to_vma(unsigned long kva, struct sp_area *spa, goto out; } } else { - buf = ret_addr; - addr = kva; - offset = 0; - do { - ret = remap_pfn_range(vma, buf, __sp_remap_get_pfn(addr), PAGE_SIZE, - __pgprot(vma->vm_page_prot.pgprot)); - if (ret) { - ret_addr = ret; - goto out; - } - offset += PAGE_SIZE; - buf += PAGE_SIZE; - addr += PAGE_SIZE; - } while (offset < spa_size(spa)); + ret = remap_vmalloc_range(vma, (void *)kva, 0); + if (ret) { + pr_err("share pool: remap vmalloc failed, ret %d\n", ret); + ret_addr = ret; + goto out; + } } out: @@ -1380,6 +1402,13 @@ static void *sp_make_share_kva_to_task(unsigned long kva, struct sp_area *spa, } p = (void *)ret_addr; + + task_lock(tsk); + if (tsk->mm == NULL) + p = ERR_PTR(-ESRCH); + else + spa->mm = tsk->mm; + task_unlock(tsk); out: put_task_struct(tsk); return p; @@ -1438,6 +1467,7 @@ void *sp_make_share_k2u(unsigned long kva, unsigned long size, unsigned long kva_aligned; unsigned long size_aligned; unsigned int page_size = PAGE_SIZE; + enum spa_type type; int ret; if (sp_flags & ~SP_DVPP) { @@ -1453,6 +1483,7 @@ void *sp_make_share_k2u(unsigned long kva, unsigned long size, } else if (ret == 0) { /* do nothing */ } else { + pr_err("it is not vmalloc address\n"); return ERR_PTR(ret); } /* aligned down kva is convenient for caller to start with any valid kva */ @@ -1460,24 +1491,42 @@ void *sp_make_share_k2u(unsigned long kva, unsigned long size, size_aligned = ALIGN(kva + size, page_size) - kva_aligned; mutex_lock(&sp_mutex); - spg = __sp_find_spg(pid, spg_id); + spg = __sp_find_spg(pid, SPG_ID_DEFAULT); if (spg == NULL) { - spa = sp_alloc_area(size_aligned, sp_flags, NULL, SPA_TYPE_K2TASK); - if (!spa) { + type = SPA_TYPE_K2TASK; + if (spg_id != SPG_ID_NONE && spg_id != SPG_ID_DEFAULT) { mutex_unlock(&sp_mutex); if (printk_ratelimit()) - pr_err("share pool: k2u failed due to alloc spa failure\n"); - return ERR_PTR(-ENOMEM); + pr_err("share pool: k2task invalid spg id %d\n", spg_id); + return ERR_PTR(-EINVAL); + } + spa = sp_alloc_area(size_aligned, sp_flags, NULL, type); + if (IS_ERR(spa)) { + mutex_unlock(&sp_mutex); + if (printk_ratelimit()) + pr_err("share pool: k2u(task) failed due to alloc spa failure " + "(potential no enough virtual memory when -75): %ld\n", + PTR_ERR(spa)); + return spa; } uva = sp_make_share_kva_to_task(kva_aligned, spa, pid); mutex_unlock(&sp_mutex); } else if (spg_valid(spg)) { - spa = sp_alloc_area(size_aligned, sp_flags, spg, SPA_TYPE_K2SPG); - if (!spa) { + type = SPA_TYPE_K2SPG; + if (spg_id != SPG_ID_DEFAULT && spg_id != spg->id) { mutex_unlock(&sp_mutex); if (printk_ratelimit()) - pr_err("share pool: k2u failed due to alloc spa failure\n"); - return ERR_PTR(-ENOMEM); + pr_err("share pool: k2spg invalid spg id %d\n", spg_id); + return ERR_PTR(-EINVAL); + } + spa = sp_alloc_area(size_aligned, sp_flags, spg, type); + if (IS_ERR(spa)) { + mutex_unlock(&sp_mutex); + if (printk_ratelimit()) + pr_err("share pool: k2u(spg) failed due to alloc spa failure " + "(potential no enough virtual memory when -75): %ld\n", + PTR_ERR(spa)); + return spa; } uva = sp_make_share_kva_to_spg(kva_aligned, spa, spg); @@ -1492,6 +1541,17 @@ void *sp_make_share_k2u(unsigned long kva, unsigned long size, uva = uva + (kva - kva_aligned); __sp_area_drop(spa); + + if (!IS_ERR(uva)) { + if (spg_valid(spa->spg)) + spg_id = spa->spg->id; + pr_notice("share pool: [sp k2u type %d] caller %s(%d/%d); group id %d; " + "return addr 0x%pK size %ld\n", + type, current->comm, current->tgid, current->pid, spg_id, + (void *)spa->va_start, spa->real_size); + sp_dump_stack(); + } + return uva; } EXPORT_SYMBOL_GPL(sp_make_share_k2u); @@ -1531,7 +1591,8 @@ static int sp_pte_hole(unsigned long start, unsigned long end, struct mm_walk *walk) { if (printk_ratelimit()) - pr_err("share pool: hole [%pK, %pK) appeared unexpectedly\n", (void *)start, (void *)end); + pr_err("share pool: hole [%pK, %pK) appeared unexpectedly\n", + (void *)start, (void *)end); return -EFAULT; } @@ -1545,7 +1606,8 @@ static int sp_hugetlb_entry(pte_t *ptep, unsigned long hmask, if (unlikely(!pte_present(pte))) { if (printk_ratelimit()) - pr_err("share pool: the page of addr %pK unexpectedly not in RAM\n", (void *)addr); + pr_err("share pool: the page of addr %pK unexpectedly " + "not in RAM\n", (void *)addr); return -EFAULT; } @@ -1758,6 +1820,11 @@ static int sp_unshare_uva(unsigned long uva, unsigned long size, int pid, int sp } } + if (spa->type != SPA_TYPE_K2TASK && spa->type != SPA_TYPE_K2SPG) { + pr_err("share pool: this spa should not be unshare here\n"); + ret = -EINVAL; + goto out_drop_area; + } /* * 1. overflow actually won't happen due to an spa must be valid. * 2. we must unshare [spa->va_start, spa->va_start + spa->real_size) completely @@ -1771,32 +1838,57 @@ static int sp_unshare_uva(unsigned long uva, unsigned long size, int pid, int sp if (size_aligned < ALIGN(size, page_size)) { ret = -EINVAL; if (printk_ratelimit()) - pr_err("share pool: unshare uva failed due to invalid parameter size %lu\n", size); + pr_err("share pool: unshare uva failed due to invalid parameter size %lu\n", + size); goto out_drop_area; } - if (spg_id == SPG_ID_NONE) { - if (spa->spg) { - ret = -EINVAL; + if (spa->type == SPA_TYPE_K2TASK) { + if (spg_id != SPG_ID_NONE && spg_id != SPG_ID_DEFAULT) { if (printk_ratelimit()) - pr_err("share pool: unshare uva failed, SPG_ID_NONE is invalid\n"); + pr_err("share pool: unshare uva(to task) failed, " + "invalid spg id %d\n", spg_id); + ret = -EINVAL; goto out_drop_area; } rcu_read_lock(); tsk = find_task_by_vpid(pid); - if (!tsk || (tsk->flags & PF_EXITING)) - ret = -ESRCH; - else - get_task_struct(tsk); - + if (!tsk || !tsk->mm || (tsk->flags & PF_EXITING)) { + if (printk_ratelimit()) + pr_info("share pool: no need to unshare uva(to task), " + "target process not found or do_exit\n"); + ret = -EINVAL; + rcu_read_unlock(); + sp_dump_stack(); + goto out_drop_area; + } + get_task_struct(tsk); rcu_read_unlock(); - if (ret) + + if (!spa->mm || + (current->mm && (current->mm != tsk->mm || tsk->mm != spa->mm))) { + if (printk_ratelimit()) + pr_err("share pool: unshare uva(to task) failed, " + "wrong pid or invalid spa\n"); + ret = -EINVAL; goto out_drop_area; + } + + if (spa->mm != tsk->mm) { + if (printk_ratelimit()) + pr_err("share pool: unshare uva(to task) failed, " + "spa not belong to the task\n"); + ret = -EINVAL; + goto out_drop_area; + } if (!mmget_not_zero(tsk->mm)) { put_task_struct(tsk); - pr_info("share pool: no need to unshare uva, target process is exiting\n"); + if (printk_ratelimit()) + pr_info("share pool: no need to unshare uva(to task), " + "target process mm is not existing\n"); + sp_dump_stack(); goto out_drop_area; } down_write(&tsk->mm->mmap_sem); @@ -1809,32 +1901,51 @@ static int sp_unshare_uva(unsigned long uva, unsigned long size, int pid, int sp (void *)uva_aligned); } put_task_struct(tsk); - } else { - /* - * k2u to task, then unshare_uva(..., spg_id) is invalid due to potential - * spa memory leak. - */ - if (!spa->spg) { + } else if (spa->type == SPA_TYPE_K2SPG) { + if (!spa->spg || spg_id == SPG_ID_NONE) { + if (printk_ratelimit()) + pr_err("share pool: unshare uva(to group) failed, " + "invalid spg id %d\n", spg_id); ret = -EINVAL; + goto out_drop_area; + } + + spg = __sp_find_spg(pid, SPG_ID_DEFAULT); + if (!spg_valid(spg)) { if (printk_ratelimit()) - pr_err("share pool: unshare uva failed, sp group id %d is invalid\n", spg_id); + pr_err("share pool: unshare uva(to group) invalid pid, " + "process not in sp group or group is dead\n"); + ret = -EINVAL; goto out_drop_area; } - spg = __sp_find_spg(pid, spg_id); - if (spg_valid(spg)) { - __sp_free(spg, uva_aligned, size_aligned, NULL); - } else { - if (!spg) { - if (printk_ratelimit()) - pr_err("share pool: unshare uva failed, doesn't belong to group %d\n", - spg_id); - ret = -EINVAL; - goto out_drop_area; - } else { - pr_info("share pool: no need to unshare uva, target process is exiting\n"); - } + if (spa->spg != spg) { + if (printk_ratelimit()) + pr_err("share pool: unshare uva(to group) failed, " + "spa not belong to the group\n"); + ret = -EINVAL; + goto out_drop_area; } + + if (current->mm && current->mm->sp_group != spg) { + if (printk_ratelimit()) + pr_err("share pool: unshare uva(to group) failed, " + "caller process doesn't belong to target group\n"); + ret = -EINVAL; + goto out_drop_area; + } + + __sp_free(spg, uva_aligned, size_aligned, NULL); + } + + if (!ret) { + if (spg_valid(spa->spg)) + spg_id = spa->spg->id; + pr_notice("share pool: [sp unshare uva type %d] caller %s(%d/%d); " + "group id %d addr 0x%pK size %ld\n", + spa->type, current->comm, current->tgid, current->pid, + spg_id, (void *)spa->va_start, spa->real_size); + sp_dump_stack(); } out_drop_area: @@ -1864,7 +1975,8 @@ static int sp_unshare_kva(unsigned long kva, unsigned long size) step = PAGE_SIZE; is_hugepage = false; } else { - pr_err("share pool: check vmap hugepage failed, ret %d\n", ret); + if (printk_ratelimit()) + pr_err("share pool: check vmap hugepage failed, ret %d\n", ret); return -EINVAL; } @@ -1882,7 +1994,8 @@ static int sp_unshare_kva(unsigned long kva, unsigned long size) if (page) put_page(page); else - pr_err("share pool: vmalloc to hugepage failed\n"); + pr_err("share pool: vmalloc %pK to page/hugepage failed\n", + (void *)addr); } vunmap((void *)kva_aligned); @@ -1944,7 +2057,7 @@ int sp_walk_page_range(unsigned long uva, unsigned long size, get_task_struct(tsk); if (!mmget_not_zero(tsk->mm)) { put_task_struct(tsk); - return -EINVAL; + return -ESRCH; } down_write(&tsk->mm->mmap_sem); ret = __sp_walk_page_range(uva, size, tsk, sp_walk_data); @@ -1973,46 +2086,6 @@ void sp_walk_page_free(struct sp_walk_data *sp_walk_data) } EXPORT_SYMBOL_GPL(sp_walk_page_free); -/** - * Walk the mm_struct of processes in the specified sp_group - * and call CALLBACK once for each mm_struct. - * @spg_id: the ID of the specified sp_group - * @data: the param for callback function - * @func: caller specific callback function - * - * Return -errno if fail. - */ -int sp_group_walk(int spg_id, void *data, int (*func)(struct mm_struct *mm, void *)) -{ - struct sp_group *spg; - int ret = -ESRCH; - - if (!func) { - if (printk_ratelimit()) - pr_err("share pool: null func pointer\n"); - return -EINVAL; - } - - mutex_lock(&sp_mutex); - spg = idr_find(&sp_group_idr, spg_id); - if (spg_valid(spg)) { - struct mm_struct *mm; - struct mm_struct *tmp; - list_for_each_entry_safe(mm, tmp, &spg->procs, sp_node) { - if (func) { - ret = func(mm, data); - if (ret) - goto out_unlock; - } - } - } -out_unlock: - mutex_unlock(&sp_mutex); - - return ret; -} -EXPORT_SYMBOL_GPL(sp_group_walk); - int sp_register_notifier(struct notifier_block *nb) { return blocking_notifier_chain_register(&sp_notifier_chain, nb); @@ -2039,7 +2112,7 @@ bool sp_config_dvpp_range(size_t start, size_t size, int device_id, int pid) struct sp_group *spg; if (device_id < 0 || device_id >= MAX_DEVID || pid < 0 || size <= 0 || - size > MMAP_SHARE_POOL_16G_SIZE) + size> MMAP_SHARE_POOL_16G_SIZE) return false; mutex_lock(&sp_mutex); @@ -2061,11 +2134,9 @@ EXPORT_SYMBOL_GPL(sp_config_dvpp_range); /* Check whether the address belongs to the share pool. */ bool is_sharepool_addr(unsigned long addr) { - if (host_svm_sp_enable == false) - return (addr >= MMAP_SHARE_POOL_START) && - addr < (MMAP_SHARE_POOL_16G_START + MMAP_SHARE_POOL_16G_SIZE); - - return addr >= MMAP_SHARE_POOL_START && addr < MMAP_SHARE_POOL_END; + if (host_svm_sp_enable == false) + return addr >= MMAP_SHARE_POOL_START && addr < (MMAP_SHARE_POOL_16G_START + MMAP_SHARE_POOL_16G_SIZE); + return addr >= MMAP_SHARE_POOL_START && addr < MMAP_SHARE_POOL_END; } EXPORT_SYMBOL_GPL(is_sharepool_addr); @@ -2109,7 +2180,7 @@ static int idr_proc_stat_cb(int id, void *p, void *data) mutex_lock(&sp_mutex); spg = __sp_find_spg(id, SPG_ID_DEFAULT); - if (spg) { + if (spg_valid(spg)) { seq_printf(seq, "%-12d %-10d %-18ld\n", id, spg->id, byte2kb(stat->amount)); } @@ -2130,8 +2201,7 @@ static int proc_stat_show(struct seq_file *seq, void *offset) return 0; } -static void rb_spa_stat_show(struct seq_file *seq) -{ +static void rb_spa_stat_show(struct seq_file *seq) { struct rb_node *node; struct sp_area *spa; @@ -2215,8 +2285,8 @@ static int idr_spg_stat_cb(int id, void *p, void *data) struct sp_group *spg = p; struct seq_file *seq = data; - seq_printf(seq, "Group %-10d size: %13d KB, spa num: %d.\n", - id, byte2kb(atomic_read(&spg->size)), + seq_printf(seq, "Group %-10d size: %13ld KB, spa num: %d.\n", + id, byte2kb(atomic64_read(&spg->size)), atomic_read(&spg->spa_num)); return 0; @@ -2227,8 +2297,8 @@ static void spg_overview_show(struct seq_file *seq) mutex_lock(&sp_mutex); idr_for_each(&sp_group_idr, idr_spg_stat_cb, seq); mutex_unlock(&sp_mutex); - seq_printf(seq, "Share pool total size: %13d KB, spa total num: %d.\n\n", - byte2kb(atomic_read(&spg_stat.spa_total_size)), + seq_printf(seq, "Share pool total size: %13ld KB, spa total num: %d.\n\n", + byte2kb(atomic64_read(&spg_stat.spa_total_size)), atomic_read(&spg_stat.spa_total_num)); } @@ -2255,7 +2325,6 @@ void __init proc_sharepool_init(void) proc_create_single_data("sharepool/spa_stat", 0, NULL, spa_stat_show, NULL); } - struct page *sp_alloc_pages(struct vm_struct *area, gfp_t mask, unsigned int page_order, int node) { -- 2.25.1

1 1

[PATCH] powercap: restrict energy meter to root access
by Yang Yingliang 20 Nov '20

20 Nov '20

From: Len Brown <len.brown(a)intel.com> stable inclusion from linux-4.19.157 commit 900281e167f45e0c0e5df6e59fa00334b5e38133 CVE: CVE-2020-8694 -------------------------------- commit 949dd0104c496fa7c14991a23c03c62e44637e71 upstream. Remove non-privileged user access to power data contained in /sys/class/powercap/intel-rapl*/*/energy_uj Non-privileged users currently have read access to power data and can use this data to form a security attack. Some privileged drivers/applications need read access to this data, but don't expose it to non-privileged users. For example, thermald uses this data to ensure that power management works correctly. Thus removing non-privileged access is preferred over completely disabling this power reporting capability with CONFIG_INTEL_RAPL=n. Fixes: 95677a9a3847 ("PowerCap: Fix mode for energy counter") Signed-off-by: Len Brown <len.brown(a)intel.com> Cc: stable(a)vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/powercap/powercap_sys.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/powercap/powercap_sys.c b/drivers/powercap/powercap_sys.c index 9e2f274bd44f..60c8375c3c81 100644 --- a/drivers/powercap/powercap_sys.c +++ b/drivers/powercap/powercap_sys.c @@ -379,9 +379,9 @@ static void create_power_zone_common_attributes( &dev_attr_max_energy_range_uj.attr; if (power_zone->ops->get_energy_uj) { if (power_zone->ops->reset_energy_uj) - dev_attr_energy_uj.attr.mode = S_IWUSR | S_IRUGO; + dev_attr_energy_uj.attr.mode = S_IWUSR | S_IRUSR; else - dev_attr_energy_uj.attr.mode = S_IRUGO; + dev_attr_energy_uj.attr.mode = S_IRUSR; power_zone->zone_dev_attrs[count++] = &dev_attr_energy_uj.attr; } -- 2.25.1

1 0

[summit]Last Call: 不要错过在首届线下openEuler Summit 登台演讲的机会
by Liyongle (Fred) 20 Nov '20

20 Nov '20

openEuler Summit 已经得到了大家的很多支持演讲议题、支持单位、SIG、Demo 还在增加但是，申报议题和Demo今天就是 Deadline 了 11月20日23：59截至，不要错过你发展最迅速、最具活力的首届openEuler Summit 展示自己的技术实力赶快点击下方的链接报名申请链接 Call for Speaker<https://shimo.im/forms/XtCTP9jcrXKgjytD/fill> Call for Sponsor<https://shimo.im/forms/VWWtgLsVHzovmbeH/fill> Call for SIG<https://shimo.im/forms/KSMKHGPHIAsjoNHP/fill> Call for Demo<https://shimo.im/forms/lMTsArbYcy4hd2dY/fill> [cid:image001.png@01D6BF24.25DA26F0]<https://openeuler.org/zh/interaction/summit-list/> 在 openEuler Summit 人人都是主办方人人都是参与者这和开源精神的内核是一致的大会议程日益丰满并且干货满满、富有趣味 Hello openEuler 开发者体验区非常的酷因为我们想到了一个从树莓派到数据中心的全场景展示还有其他一些 “别有用心” 的 Demo 绝对的好玩 TC、SIG、Maintainer 已经准备好了 “网友见面” 现场召开 TC 工作会议现场召开 SIG 组工作会议现场接受新 SIG 申请现场的 Maintainer 共同分享学习不断增加的支持单位、友好基金会和社区媒体 [cid:image001.jpg@01D6BF21.1ADA52F0]

1 0

[PATCH] Input: sunkbd - avoid use-after-free in teardown paths
by Yang Yingliang 19 Nov '20

19 Nov '20

From: Dmitry Torokhov <dmitry.torokhov(a)gmail.com> mainline inclusion from mainline-v5.10-rc5 commit 77e70d351db7de07a46ac49b87a6c3c7a60fca7e category: bugfix bugzilla: NA CVE: CVE-2020-25669 -------------------------------- We need to make sure we cancel the reinit work before we tear down the driver structures. Reported-by: Bodong Zhao <nopitydays(a)gmail.com> Tested-by: Bodong Zhao <nopitydays(a)gmail.com> Cc: stable(a)vger.kernel.org Signed-off-by: Dmitry Torokhov <dmitry.torokhov(a)gmail.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/input/keyboard/sunkbd.c | 41 ++++++++++++++++++++++++++------- 1 file changed, 33 insertions(+), 8 deletions(-) diff --git a/drivers/input/keyboard/sunkbd.c b/drivers/input/keyboard/sunkbd.c index ad5d7f94f95a..1c7aa86c92ab 100644 --- a/drivers/input/keyboard/sunkbd.c +++ b/drivers/input/keyboard/sunkbd.c @@ -111,7 +111,8 @@ static irqreturn_t sunkbd_interrupt(struct serio *serio, switch (data) { case SUNKBD_RET_RESET: - schedule_work(&sunkbd->tq); + if (sunkbd->enabled) + schedule_work(&sunkbd->tq); sunkbd->reset = -1; break; @@ -212,16 +213,12 @@ static int sunkbd_initialize(struct sunkbd *sunkbd) } /* - * sunkbd_reinit() sets leds and beeps to a state the computer remembers they - * were in. + * sunkbd_set_leds_beeps() sets leds and beeps to a state the computer remembers + * they were in. */ -static void sunkbd_reinit(struct work_struct *work) +static void sunkbd_set_leds_beeps(struct sunkbd *sunkbd) { - struct sunkbd *sunkbd = container_of(work, struct sunkbd, tq); - - wait_event_interruptible_timeout(sunkbd->wait, sunkbd->reset >= 0, HZ); - serio_write(sunkbd->serio, SUNKBD_CMD_SETLED); serio_write(sunkbd->serio, (!!test_bit(LED_CAPSL, sunkbd->dev->led) << 3) | @@ -234,11 +231,39 @@ static void sunkbd_reinit(struct work_struct *work) SUNKBD_CMD_BELLOFF - !!test_bit(SND_BELL, sunkbd->dev->snd)); } + +/* + * sunkbd_reinit() wait for the keyboard reset to complete and restores state + * of leds and beeps. + */ + +static void sunkbd_reinit(struct work_struct *work) +{ + struct sunkbd *sunkbd = container_of(work, struct sunkbd, tq); + + /* + * It is OK that we check sunkbd->enabled without pausing serio, + * as we only want to catch true->false transition that will + * happen once and we will be woken up for it. + */ + wait_event_interruptible_timeout(sunkbd->wait, + sunkbd->reset >= 0 || !sunkbd->enabled, + HZ); + + if (sunkbd->reset >= 0 && sunkbd->enabled) + sunkbd_set_leds_beeps(sunkbd); +} + static void sunkbd_enable(struct sunkbd *sunkbd, bool enable) { serio_pause_rx(sunkbd->serio); sunkbd->enabled = enable; serio_continue_rx(sunkbd->serio); + + if (!enable) { + wake_up_interruptible(&sunkbd->wait); + cancel_work_sync(&sunkbd->tq); + } } /* -- 2.25.1

1 0

[PATCH v3 kernel-4.19] nbd: don't update block size after device is started
by lining 19 Nov '20

19 Nov '20

From: Ming Lei <ming.lei(a)redhat.com> mainline inclusion from mainline-v5.10-rc2 commit b40813ddcd6b category: bugfix bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1891363 CVE: NA backport: openEuler-20.09 Here is the testcase: 1. rbd create --size 2G rbdpool/foo 2. rbd-nbd map rbdpool/foo 3. mkfs.ext4 /dev/nbd0 4. mount /dev/nbd0 /mnt 5. rbd resize --size 4G rbdpool/foo 6. ls /mnt ls will stuck here forever. -------------------------------- [ Upstream commit b40813ddcd6bf9f01d020804e4cb8febc480b9e4 ] Mounted NBD device can be resized, one use case is rbd-nbd. Fix the issue by setting up default block size, then not touch it in nbd_size_update() any more. This kind of usage is aligned with loop which has same use case too. Cc: stable(a)vger.kernel.org Fixes: c8a83a6b54d0 ("nbd: Use set_blocksize() to set device blocksize") Reported-by: lining <lining2020x(a)163.com> Signed-off-by: Ming Lei <ming.lei(a)redhat.com> Cc: Josef Bacik <josef(a)toxicpanda.com> Cc: Jan Kara <jack(a)suse.cz> Tested-by: lining <lining2020x(a)163.com> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: lining <lining_yewu(a)cmss.chinamobile.com> --- drivers/block/nbd.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index cdf62fb94fb15..9a0fb2d52a76c 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -268,7 +268,7 @@ static void nbd_size_clear(struct nbd_device *nbd) } } -static void nbd_size_update(struct nbd_device *nbd) +static void nbd_size_update(struct nbd_device *nbd, bool start) { struct nbd_config *config = nbd->config; struct block_device *bdev = bdget_disk(nbd->disk, 0); @@ -279,7 +279,8 @@ static void nbd_size_update(struct nbd_device *nbd) if (bdev) { if (bdev->bd_disk) { bd_set_size(bdev, config->bytesize); - set_blocksize(bdev, config->blksize); + if (start) + set_blocksize(bdev, config->blksize); } else bdev->bd_invalidated = 1; bdput(bdev); @@ -294,7 +295,7 @@ static void nbd_size_set(struct nbd_device *nbd, loff_t blocksize, config->blksize = blocksize; config->bytesize = blocksize * nr_blocks; if (nbd->task_recv != NULL) - nbd_size_update(nbd); + nbd_size_update(nbd, false); } static void nbd_complete_rq(struct request *req) @@ -1231,7 +1232,7 @@ static int nbd_start_device(struct nbd_device *nbd) args->index = i; queue_work(nbd->recv_workq, &args->work); } - nbd_size_update(nbd); + nbd_size_update(nbd, true); return error; } -- 2.27.0

3 2

[PATCH 0/2] backport stable-4.19.157
by Yang Yingliang 19 Nov '20

19 Nov '20

*** BLURB HERE *** Greg Kroah-Hartman (1): Linux 4.19.157 Len Brown (1): powercap: restrict energy meter to root access Makefile | 2 +- drivers/powercap/powercap_sys.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) -- 2.25.1

1 2

[PATCH 00/66] backport stable-4.19.156
by Yang Yingliang 19 Nov '20

19 Nov '20

Alan Stern (1): USB: Add NO_LPM quirk for Kingston flash drive Alexander Aring (1): gfs2: Wake up when sd_glock_disposal becomes zero Artem Lapkin (1): ALSA: usb-audio: add usb vendor id as DSD-capable for Khadas devices Ben Hutchings (1): Revert "btrfs: flush write bio if we loop in extent_write_cache_pages" Chris Wilson (1): drm/i915: Break up error capture compression loops with cond_resched() Claire Chang (1): serial: 8250_mtk: Fix uart_get_baud_rate warning Claudiu Manoil (2): gianfar: Replace skb_realloc_headroom with skb_cow_head for PTP gianfar: Account for Tx PTP timestamp in the skb headroom Clément Péron (1): ARM: dts: sun4i-a10: fix cpu_alert temperature Daniel Vetter (1): vt: Disable KD_FONT_OP_COPY Daniele Palmas (3): net: usb: qmi_wwan: add Telit LE910Cx 0x1230 composition USB: serial: option: add LE910Cx compositions 0x1203, 0x1230, 0x1231 USB: serial: option: add Telit FN980 composition 0x1055 Eddy Wu (1): fork: fix copy_process(CLONE_PARENT) race with the exiting ->real_parent Filipe Manana (1): Btrfs: fix unwritten extent buffers and hangs on future writeback attempts Gabriel Krisman Bertazi (2): blk-cgroup: Fix memleak on error path blk-cgroup: Pre-allocate tree node on blkg_conf_prep Geoffrey D. Bennett (2): ALSA: usb-audio: Add implicit feedback quirk for Qu-16 ALSA: usb-audio: Add implicit feedback quirk for MODX Greg Kroah-Hartman (1): Linux 4.19.156 Guenter Roeck (1): tools: perf: Fix build error in v4.19.y Hoang Huu Le (1): tipc: fix use-after-free in tipc_bcast_get_mode Hoegeun Kwon (1): drm/vc4: drv: Add error handding for bind Jason Gunthorpe (1): mm: always have io_remap_pfn_range() set pgprot_decrypted() Jeff Vander Stoep (1): vsock: use ns_capable_noaudit() on socket create Johan Hovold (1): USB: serial: cyberjack: fix write-URB completion race Josef Bacik (1): btrfs: flush write bio if we loop in extent_write_cache_pages Kairui Song (1): x86/kexec: Use up-to-dated screen_info copy to fill boot params Keith Winstein (1): ALSA: usb-audio: Add implicit feedback quirk for Zoom UAC-2 Lee Jones (1): Fonts: Replace discarded const qualifier Macpaul Lin (1): usb: mtu3: fix panic in mtu3_gadget_stop() Mark Deneen (1): cadence: force nonlinear buffers to be cloned Mike Galbraith (1): futex: Handle transient "ownerless" rtmutex state correctly Ming Lei (1): scsi: core: Don't start concurrent async scan on same host Oleg Nesterov (1): ptrace: fix task_join_group_stop() for the case when current is traced Pali Rohár (1): arm64: dts: marvell: espressobin: Add ethernet switch aliases Petr Malat (1): sctp: Fix COMM_LOST/CANT_STR_ASSOC err reporting on big-endian platforms Qinglang Miao (1): serial: txx9: add missing platform_driver_unregister() on error in serial_txx9_init Qiujun Huang (1): tracing: Fix out of bounds write in get_trace_buf Qu Wenruo (11): btrfs: extent_io: Kill the forward declaration of flush_write_bio btrfs: extent_io: Move the BUG_ON() in flush_write_bio() one level up btrfs: extent_io: Handle errors better in btree_write_cache_pages() btrfs: extent_io: add proper error handling to lock_extent_buffer_for_io() btrfs: Move btrfs_check_chunk_valid() to tree-check.[ch] and export it btrfs: tree-checker: Make chunk item checker messages more readable btrfs: tree-checker: Make btrfs_check_chunk_valid() return EUCLEAN instead of EIO btrfs: tree-checker: Check chunk item at tree block read time btrfs: tree-checker: Verify dev item btrfs: tree-checker: Fix wrong check on max devid btrfs: tree-checker: fix the error message for transid error Rafael J. Wysocki (1): PM: runtime: Resume the device earlier in __device_release_driver() Shijie Luo (1): mm: mempolicy: fix potential pte_unmap_unlock pte error Steven Rostedt (VMware) (3): ring-buffer: Fix recursion protection transitions between interrupt context ftrace: Fix recursion check for NMI test ftrace: Handle tracing when switching between context Vasily Gorbik (1): lib/crc32test: remove extra local_irq_disable/enable Vinay Kumar Yadav (2): chelsio/chtls: fix memory leaks caused by a race chelsio/chtls: fix always leaking ctrl_skb Vincent Whitchurch (1): of: Fix reserved-memory overlap detection Vineet Gupta (2): ARC: stack unwinding: avoid indefinite looping Revert "ARC: entry: fix potential EFA clobber when TIF_SYSCALL_TRACE" Xiaofei Shen (1): net: dsa: read mac address from DT for slave device YueHaibing (1): sfp: Fix error handing in sfp_probe() Zhang Qilong (1): ACPI: NFIT: Fix comparison to '-ENXIO' Ziyi Cao (1): USB: serial: option: add Quectel EC200T module support Zqiang (1): kthread_worker: prevent queuing delayed work from timer_fn when it is being canceled Makefile | 2 +- arch/arc/kernel/entry.S | 16 +- arch/arc/kernel/stacktrace.c | 7 +- arch/arm/boot/dts/sun4i-a10.dtsi | 2 +- .../dts/marvell/armada-3720-espressobin.dts | 12 +- arch/x86/kernel/kexec-bzimage64.c | 3 +- block/blk-cgroup.c | 15 +- drivers/acpi/nfit/core.c | 2 +- drivers/base/dd.c | 7 +- drivers/crypto/chelsio/chtls/chtls_cm.c | 2 +- drivers/crypto/chelsio/chtls/chtls_hw.c | 3 + drivers/gpu/drm/i915/i915_gpu_error.c | 3 + drivers/gpu/drm/vc4/vc4_drv.c | 1 + drivers/net/ethernet/cadence/macb_main.c | 3 +- drivers/net/ethernet/freescale/gianfar.c | 14 +- drivers/net/phy/sfp.c | 3 +- drivers/net/usb/qmi_wwan.c | 1 + drivers/of/of_reserved_mem.c | 13 +- drivers/scsi/scsi_scan.c | 7 +- drivers/tty/serial/8250/8250_mtk.c | 2 +- drivers/tty/serial/serial_txx9.c | 3 + drivers/tty/vt/vt.c | 24 +- drivers/usb/core/quirks.c | 3 + drivers/usb/mtu3/mtu3_gadget.c | 1 + drivers/usb/serial/cyberjack.c | 7 +- drivers/usb/serial/option.c | 10 + fs/btrfs/extent_io.c | 171 +++++++++---- fs/btrfs/tree-checker.c | 236 +++++++++++++++++- fs/btrfs/tree-checker.h | 4 + fs/btrfs/volumes.c | 123 +-------- fs/btrfs/volumes.h | 9 + fs/gfs2/glock.c | 3 +- include/asm-generic/pgtable.h | 4 - include/linux/mm.h | 9 + include/net/dsa.h | 1 + kernel/fork.c | 10 +- kernel/futex.c | 16 +- kernel/kthread.c | 3 +- kernel/signal.c | 19 +- kernel/trace/ring_buffer.c | 58 ++++- kernel/trace/trace.c | 2 +- kernel/trace/trace.h | 26 +- kernel/trace/trace_selftest.c | 9 +- lib/crc32test.c | 4 - lib/fonts/font_10x18.c | 2 +- lib/fonts/font_6x10.c | 2 +- lib/fonts/font_6x11.c | 2 +- lib/fonts/font_7x14.c | 2 +- lib/fonts/font_8x16.c | 2 +- lib/fonts/font_8x8.c | 2 +- lib/fonts/font_acorn_8x8.c | 2 +- lib/fonts/font_mini_4x6.c | 2 +- lib/fonts/font_pearl_8x8.c | 2 +- lib/fonts/font_sun12x22.c | 2 +- lib/fonts/font_sun8x16.c | 2 +- mm/mempolicy.c | 6 +- net/dsa/dsa2.c | 1 + net/dsa/slave.c | 5 +- net/sctp/sm_sideeffect.c | 4 +- net/tipc/core.c | 5 + net/vmw_vsock/af_vsock.c | 2 +- sound/usb/pcm.c | 6 + sound/usb/quirks.c | 1 + tools/perf/util/util.h | 2 +- 64 files changed, 634 insertions(+), 293 deletions(-) -- 2.25.1

1 66

[PATCH] tools: perf: Fix build error in v4.19.y
by Yang Yingliang 18 Nov '20

18 Nov '20

From: Guenter Roeck <linux(a)roeck-us.net> perf may fail to build in v4.19.y with the following error. util/evsel.c: In function ‘perf_evsel__exit’: util/util.h:25:28: error: passing argument 1 of ‘free’ discards ‘const’ qualifier from pointer target type This is observed (at least) with gcc v6.5.0. The underlying problem is the following statement. zfree(&evsel->pmu_name); evsel->pmu_name is decared 'const *'. zfree in turn is defined as #define zfree(ptr) ({ free(*ptr); *ptr = NULL; }) and thus passes the const * to free(). The problem is not seen in the upstream kernel since zfree() has been rewritten there. The problem has been introduced into v4.19.y with the backport of upstream commit d4953f7ef1a2 (perf parse-events: Fix 3 use after frees found with clang ASAN). One possible fix of this problem would be to not declare pmu_name as const. This patch chooses to typecast the parameter of zfree() to void *, following the guidance from the upstream kernel which does the same since commit 7f7c536f23e6a ("tools lib: Adopt zalloc()/zfree() from tools/perf") Fixes: a0100a363098 ("perf parse-events: Fix 3 use after frees found with clang ASAN") Signed-off-by: Guenter Roeck <linux(a)roeck-us.net> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- tools/perf/util/util.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h index dc58254a2b69..8c01b2cfdb1a 100644 --- a/tools/perf/util/util.h +++ b/tools/perf/util/util.h @@ -22,7 +22,7 @@ static inline void *zalloc(size_t size) return calloc(1, size); } -#define zfree(ptr) ({ free(*ptr); *ptr = NULL; }) +#define zfree(ptr) ({ free((void *)*ptr); *ptr = NULL; }) struct dirent; struct nsinfo; -- 2.25.1

1 0

[PATCH 01/19] nvme: Fix ctrl use-after-free during sysfs deletion
by Yang Yingliang 18 Nov '20

18 Nov '20

From: Israel Rukshin <israelr(a)mellanox.com> mainline inclusion from mainline-v5.7-rc1 commit b780d7415aacec855e2f2370cbf98f918b224903 category: bugfix bugzilla: NA CVE: NA Link: https://gitee.com/openeuler/kernel/issues/I1WGZE -------------------------------- In case nvme_sysfs_delete() is called by the user before taking the ctrl reference count, the ctrl may be freed during the creation and cause the bug. Take the reference as soon as the controller is externally visible, which is done by cdev_device_add() in nvme_init_ctrl(). Also take the reference count at the core layer instead of taking it on each transport separately. Signed-off-by: Israel Rukshin <israelr(a)mellanox.com> Reviewed-by: Max Gurtovoy <maxg(a)mellanox.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Keith Busch <kbusch(a)kernel.org> Conflicts: drivers/nvme/host/tcp.c [No code about TCP in current version.] Reviewed-by: Chao Leng <lengchao(a)huawei.com> Reviewed-by: Jike Cheng <chengjike.cheng(a)huawei.com> Signed-off-by: Lijie <lijie34(a)huawei.com> Reviewed-by: Hou Tao <houtao1(a)huawei.com> Acked-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/nvme/host/core.c | 2 ++ drivers/nvme/host/fc.c | 4 +--- drivers/nvme/host/pci.c | 1 - drivers/nvme/host/rdma.c | 3 +-- drivers/nvme/target/loop.c | 3 +-- 5 files changed, 5 insertions(+), 8 deletions(-) diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index b8446637b3c2..b7a40ac4b637 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -3779,6 +3779,7 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, if (ret) goto out_release_instance; + nvme_get_ctrl(ctrl); cdev_init(&ctrl->cdev, &nvme_dev_fops); ctrl->cdev.owner = ops->module; ret = cdev_device_add(&ctrl->cdev, ctrl->device); @@ -3795,6 +3796,7 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev, return 0; out_free_name: + nvme_put_ctrl(ctrl); kfree_const(ctrl->device->kobj.name); out_release_instance: ida_simple_remove(&nvme_instance_ida, ctrl->instance); diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index ed88d5021772..d75ae1b201ad 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -3146,10 +3146,7 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, goto fail_ctrl; } - nvme_get_ctrl(&ctrl->ctrl); - if (!queue_delayed_work(nvme_wq, &ctrl->connect_work, 0)) { - nvme_put_ctrl(&ctrl->ctrl); dev_err(ctrl->ctrl.device, "NVME-FC{%d}: failed to schedule initial connect\n", ctrl->cnum); @@ -3174,6 +3171,7 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts, /* initiate nvme ctrl ref counting teardown */ nvme_uninit_ctrl(&ctrl->ctrl); + nvme_put_ctrl(&ctrl->ctrl); /* Remove core ctrl ref. */ nvme_put_ctrl(&ctrl->ctrl); diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index f194bb7ccb04..1a0eec110614 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -2558,7 +2558,6 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id) dev_info(dev->ctrl.device, "pci function %s\n", dev_name(&pdev->dev)); nvme_reset_ctrl(&dev->ctrl); - nvme_get_ctrl(&dev->ctrl); async_schedule(nvme_async_probe, dev); return 0; diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index d0cf19e593b2..0e998e85e962 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -2050,8 +2050,6 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev, dev_info(ctrl->ctrl.device, "new ctrl: NQN \"%s\", addr %pISpcs\n", ctrl->ctrl.opts->subsysnqn, &ctrl->addr); - nvme_get_ctrl(&ctrl->ctrl); - mutex_lock(&nvme_rdma_ctrl_mutex); list_add_tail(&ctrl->list, &nvme_rdma_ctrl_list); mutex_unlock(&nvme_rdma_ctrl_mutex); @@ -2061,6 +2059,7 @@ static struct nvme_ctrl *nvme_rdma_create_ctrl(struct device *dev, out_uninit_ctrl: nvme_uninit_ctrl(&ctrl->ctrl); nvme_put_ctrl(&ctrl->ctrl); + nvme_put_ctrl(&ctrl->ctrl); if (ret > 0) ret = -EIO; return ERR_PTR(ret); diff --git a/drivers/nvme/target/loop.c b/drivers/nvme/target/loop.c index 137a27fa369c..e30080c629fd 100644 --- a/drivers/nvme/target/loop.c +++ b/drivers/nvme/target/loop.c @@ -639,8 +639,6 @@ static struct nvme_ctrl *nvme_loop_create_ctrl(struct device *dev, dev_info(ctrl->ctrl.device, "new ctrl: \"%s\"\n", ctrl->ctrl.opts->subsysnqn); - nvme_get_ctrl(&ctrl->ctrl); - changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE); WARN_ON_ONCE(!changed); @@ -658,6 +656,7 @@ static struct nvme_ctrl *nvme_loop_create_ctrl(struct device *dev, kfree(ctrl->queues); out_uninit_ctrl: nvme_uninit_ctrl(&ctrl->ctrl); + nvme_put_ctrl(&ctrl->ctrl); out_put_ctrl: nvme_put_ctrl(&ctrl->ctrl); if (ret > 0) -- 2.25.1

1 18

[PATCH v2 kernel-4.19] nbd: don't update block size after device is started
by lining 17 Nov '20

17 Nov '20

mainline inclusion from mainline-v5.10-rc2 commit b40813ddcd6b category: bugfix bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1891363 CVE: NA backport: openEuler-20.09 Here is the testcase: 1. rbd create --size 2G rbdpool/foo 2. rbd-nbd map rbdpool/foo 3. mkfs.ext4 /dev/nbd0 4. mount /dev/nbd0 /mnt 5. rbd resize --size 4G rbdpool/foo 6. ls /mnt ls will stuck here forever. -------------------------------- From: Ming Lei <ming.lei(a)redhat.com> [ Upstream commit b40813ddcd6bf9f01d020804e4cb8febc480b9e4 ] Mounted NBD device can be resized, one use case is rbd-nbd. Fix the issue by setting up default block size, then not touch it in nbd_size_update() any more. This kind of usage is aligned with loop which has same use case too. Cc: stable(a)vger.kernel.org Fixes: c8a83a6b54d0 ("nbd: Use set_blocksize() to set device blocksize") Reported-by: lining <lining2020x(a)163.com> Signed-off-by: Ming Lei <ming.lei(a)redhat.com> Cc: Josef Bacik <josef(a)toxicpanda.com> Cc: Jan Kara <jack(a)suse.cz> Tested-by: lining <lining2020x(a)163.com> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: lining <lining_yewu(a)cmss.chinamobile.com> --- drivers/block/nbd.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index cdf62fb94fb15..9a0fb2d52a76c 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -268,7 +268,7 @@ static void nbd_size_clear(struct nbd_device *nbd) } } -static void nbd_size_update(struct nbd_device *nbd) +static void nbd_size_update(struct nbd_device *nbd, bool start) { struct nbd_config *config = nbd->config; struct block_device *bdev = bdget_disk(nbd->disk, 0); @@ -279,7 +279,8 @@ static void nbd_size_update(struct nbd_device *nbd) if (bdev) { if (bdev->bd_disk) { bd_set_size(bdev, config->bytesize); - set_blocksize(bdev, config->blksize); + if (start) + set_blocksize(bdev, config->blksize); } else bdev->bd_invalidated = 1; bdput(bdev); @@ -294,7 +295,7 @@ static void nbd_size_set(struct nbd_device *nbd, loff_t blocksize, config->blksize = blocksize; config->bytesize = blocksize * nr_blocks; if (nbd->task_recv != NULL) - nbd_size_update(nbd); + nbd_size_update(nbd, false); } static void nbd_complete_rq(struct request *req) @@ -1231,7 +1232,7 @@ static int nbd_start_device(struct nbd_device *nbd) args->index = i; queue_work(nbd->recv_workq, &args->work); } - nbd_size_update(nbd); + nbd_size_update(nbd, true); return error; } -- 2.27.0

2 1

[PATCH openEuler-20.09] nbd: don't update block size after device is started
by lining 17 Nov '20

17 Nov '20

mainline inclusion from mainline-v5.10-rc2 commit b40813ddcd6b category: bugfix bugzilla: 1891363 CVE: NA -------------------------------- Mounted NBD device can be resized, one use case is rbd-nbd. Fix the issue by setting up default block size, then not touch it in nbd_size_update() any more. This kind of usage is aligned with loop which has same use case too. Cc: stable(a)vger.kernel.org Fixes: c8a83a6b54d0 ("nbd: Use set_blocksize() to set device blocksize") Reported-by: lining <lining2020x(a)163.com> Signed-off-by: Ming Lei <ming.lei(a)redhat.com> Cc: Josef Bacik <josef(a)toxicpanda.com> Cc: Jan Kara <jack(a)suse.cz> Tested-by: lining <lining2020x(a)163.com> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Signed-off-by: lining <lining_yewu(a)cmss.chinamobile.com> --- drivers/block/nbd.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 816188aa841f..a913153d0b6b 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -293,7 +293,7 @@ static void nbd_size_clear(struct nbd_device *nbd) } } -static void nbd_size_update(struct nbd_device *nbd) +static void nbd_size_update(struct nbd_device *nbd, bool start) { struct nbd_config *config = nbd->config; struct block_device *bdev = bdget_disk(nbd->disk, 0); @@ -309,7 +309,8 @@ static void nbd_size_update(struct nbd_device *nbd) if (bdev) { if (bdev->bd_disk) { bd_set_size(bdev, config->bytesize); - set_blocksize(bdev, config->blksize); + if (start) + set_blocksize(bdev, config->blksize); } else bdev->bd_invalidated = 1; bdput(bdev); @@ -324,7 +325,7 @@ static void nbd_size_set(struct nbd_device *nbd, loff_t blocksize, config->blksize = blocksize; config->bytesize = blocksize * nr_blocks; if (nbd->task_recv != NULL) - nbd_size_update(nbd); + nbd_size_update(nbd, false); } static void nbd_complete_rq(struct request *req) @@ -1263,7 +1264,7 @@ static int nbd_start_device(struct nbd_device *nbd) args->index = i; queue_work(nbd->recv_workq, &args->work); } - nbd_size_update(nbd); + nbd_size_update(nbd, true); return error; } -- 2.25.1

2 2

[PATCH 00/24] bugfixes for openEuler-1.0-LTS
by Yang Yingliang 16 Nov '20

16 Nov '20

Alexander Duyck (1): e1000: Do not perform reset in reset_task if we are already down Alistair Popple (1): mm/rmap: fixup copying of soft dirty and uffd ptes Aya Levin (2): net/mlx5e: Fix VLAN cleanup flow net/mlx5e: Fix VLAN create flow Chuck Lever (1): svcrdma: Fix leak of transport addresses Cong Wang (1): tipc: fix the skb_unshare() in tipc_buf_append() Dan Aloni (1): svcrdma: fix bounce buffers for unaligned offsets and multiple pages Dumitru Ceara (1): openvswitch: handle DNAT tuple collision Eran Ben Elisha (1): net/mlx5: Don't call timecounter cyc2time directly from 1PPS flow Hanjun Guo (4): irqchip/gicv3: Call acpi_put_table() to fix memory leak tty/amba-pl011: Call acpi_put_table() to fix memory leak cpufreq : CPPC: Break out if HiSilicon CPPC workaround is matched cpufreq: CPPC: put ACPI table after using it Jonathan Lemon (1): mlx4: handle non-napi callers to napi_poll Linus Torvalds (1): tty: make FONTX ioctl use the tty pointer they were actually passed Nikolai Merinov (1): partitions/efi: Fix partition name parsing in GUID partition entry Rohit Maheshwari (1): net/tls: sendfile fails with ktls offload Sabrina Dubroca (1): xfrmi: drop ignore_df check before updating pmtu Tonghao Zhang (2): net: openvswitch: use u64 for meter bucket net: openvswitch: use div_u64() for 64-by-32 divisions Tuong Lien (1): tipc: fix memory leak in service subscripting Yang Shi (1): mm: madvise: fix vma user-after-free Yunsheng Lin (1): net: sch_generic: aviod concurrent reset and enqueue op for lockless qdisc kiyin(尹亮) (1): perf/core: Fix a memory leak in perf_event_parse_addr_filter() block/partitions/efi.c | 35 +++++++++---- block/partitions/efi.h | 2 +- drivers/cpufreq/cppc_cpufreq.c | 8 ++- drivers/irqchip/irq-gic-v3.c | 2 + drivers/net/ethernet/intel/e1000/e1000_main.c | 18 +++++-- drivers/net/ethernet/mellanox/mlx4/en_rx.c | 3 ++ drivers/net/ethernet/mellanox/mlx4/en_tx.c | 2 +- .../net/ethernet/mellanox/mlx5/core/en_fs.c | 14 ++++-- .../ethernet/mellanox/mlx5/core/lib/clock.c | 5 +- drivers/tty/serial/amba-pl011.c | 2 + drivers/tty/vt/vt_ioctl.c | 32 ++++++------ kernel/events/core.c | 12 ++--- mm/madvise.c | 2 +- mm/migrate.c | 9 +++- mm/rmap.c | 7 ++- net/openvswitch/conntrack.c | 22 +++++---- net/openvswitch/meter.c | 4 +- net/openvswitch/meter.h | 2 +- net/sched/sch_generic.c | 49 +++++++++++++------ net/sunrpc/xprtrdma/svc_rdma_backchannel.c | 1 + net/sunrpc/xprtrdma/svc_rdma_sendto.c | 3 +- net/tipc/msg.c | 3 +- net/tipc/topsrv.c | 4 +- net/tls/tls_device.c | 11 +++-- net/xfrm/xfrm_interface.c | 2 +- 25 files changed, 168 insertions(+), 86 deletions(-) -- 2.25.1

1 24

[PATCH 01/10] ascend: mm: add an owner for mm_struct
by Yang Yingliang 16 Nov '20

16 Nov '20

From: Ding Tianhong <dingtianhong(a)huawei.com> ascend inclusion category: feature bugzilla: NA CVE: NA ------------------------------------------------- The mm->owner is only used for MEMCG currently, but the ascend share pool features will use it later, so make it to a general features and select it for CONFIG_MEMCG. Signed-off-by: Tang Yizhou <tangyizhou(a)huawei.com> Signed-off-by: Ding Tianhong <dingtianhong(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/linux/mm_types.h | 2 +- init/Kconfig | 1 + kernel/exit.c | 4 ++-- kernel/fork.c | 4 ++-- mm/Kconfig | 4 ++++ mm/debug.c | 2 +- 6 files changed, 11 insertions(+), 6 deletions(-) diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 178e9dee217a..fcfa9a75c18e 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -453,7 +453,7 @@ struct mm_struct { spinlock_t ioctx_lock; struct kioctx_table __rcu *ioctx_table; #endif -#ifdef CONFIG_MEMCG +#ifdef CONFIG_MM_OWNER /* * "owner" points to a task that is regarded as the canonical * user/owner of this mm. All of the following must be true in diff --git a/init/Kconfig b/init/Kconfig index d1427ae4de9e..6880b55901bb 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -705,6 +705,7 @@ config MEMCG bool "Memory controller" select PAGE_COUNTER select EVENTFD + select MM_OWNER help Provides control over the memory footprint of tasks in a cgroup. diff --git a/kernel/exit.c b/kernel/exit.c index 891d65e3ffd5..4d6f941712b6 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -392,7 +392,7 @@ kill_orphaned_pgrp(struct task_struct *tsk, struct task_struct *parent) } } -#ifdef CONFIG_MEMCG +#ifdef CONFIG_MM_OWNER /* * A task is exiting. If it owned this mm, find a new owner for the mm. */ @@ -478,7 +478,7 @@ void mm_update_next_owner(struct mm_struct *mm) task_unlock(c); put_task_struct(c); } -#endif /* CONFIG_MEMCG */ +#endif /* CONFIG_MM_OWNER */ /* * Turn us into a lazy TLB process if we diff --git a/kernel/fork.c b/kernel/fork.c index 768fe41a7ee3..1ac49d1852cf 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -958,7 +958,7 @@ static void mm_init_aio(struct mm_struct *mm) static __always_inline void mm_clear_owner(struct mm_struct *mm, struct task_struct *p) { -#ifdef CONFIG_MEMCG +#ifdef CONFIG_MM_OWNER if (mm->owner == p) WRITE_ONCE(mm->owner, NULL); #endif @@ -966,7 +966,7 @@ static __always_inline void mm_clear_owner(struct mm_struct *mm, static void mm_init_owner(struct mm_struct *mm, struct task_struct *p) { -#ifdef CONFIG_MEMCG +#ifdef CONFIG_MM_OWNER mm->owner = p; #endif } diff --git a/mm/Kconfig b/mm/Kconfig index dddeb30d645e..76c2197a3f99 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -299,6 +299,10 @@ config VIRT_TO_BUS deprecated interface virt_to_bus(). All new architectures should probably not select this. +config MM_OWNER + bool "Enable the ownership the mm owner" + help + This option enables mm_struct's to have an owner. config MMU_NOTIFIER bool diff --git a/mm/debug.c b/mm/debug.c index 362ce581671e..2da184b16bce 100644 --- a/mm/debug.c +++ b/mm/debug.c @@ -129,7 +129,7 @@ void dump_mm(const struct mm_struct *mm) #ifdef CONFIG_AIO "ioctx_table %px\n" #endif -#ifdef CONFIG_MEMCG +#ifdef CONFIG_MM_OWNER "owner %px " #endif "exe_file %px\n" -- 2.25.1

1 9

[PATCH 1/7] scsi/hifc: add hifc driver chip resource module
by Yang Yingliang 16 Nov '20

16 Nov '20

From: Chenguangli <chenguangli2(a)huawei.com> driver inclusion category: feature bugzilla: NA ----------------------------------------------------------------------- This module includes cfg, cqm, hwdev, hwif, mgmt, sml. and are mainly used to initialize chip capabilityes and to initialize resources for communication between drivers and chip. Signed-off-by: Chenguangli <chenguangli2(a)huawei.com> Acked-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/scsi/huawei/hifc/hifc_api_cmd.c | 1155 ++++++ drivers/scsi/huawei/hifc/hifc_api_cmd.h | 268 ++ drivers/scsi/huawei/hifc/hifc_cfg.c | 823 +++++ drivers/scsi/huawei/hifc/hifc_cfg.h | 171 + drivers/scsi/huawei/hifc/hifc_cmdq.c | 1507 ++++++++ drivers/scsi/huawei/hifc/hifc_cmdq.h | 210 ++ drivers/scsi/huawei/hifc/hifc_cqm_main.c | 694 ++++ drivers/scsi/huawei/hifc/hifc_cqm_main.h | 366 ++ drivers/scsi/huawei/hifc/hifc_cqm_object.c | 3599 +++++++++++++++++++ drivers/scsi/huawei/hifc/hifc_cqm_object.h | 244 ++ drivers/scsi/huawei/hifc/hifc_eqs.c | 1347 +++++++ drivers/scsi/huawei/hifc/hifc_eqs.h | 233 ++ drivers/scsi/huawei/hifc/hifc_hw.h | 611 ++++ drivers/scsi/huawei/hifc/hifc_hwdev.c | 3675 ++++++++++++++++++++ drivers/scsi/huawei/hifc/hifc_hwdev.h | 456 +++ drivers/scsi/huawei/hifc/hifc_hwif.c | 630 ++++ drivers/scsi/huawei/hifc/hifc_hwif.h | 243 ++ drivers/scsi/huawei/hifc/hifc_mgmt.c | 1426 ++++++++ drivers/scsi/huawei/hifc/hifc_mgmt.h | 407 +++ drivers/scsi/huawei/hifc/hifc_sml.c | 361 ++ drivers/scsi/huawei/hifc/hifc_sml.h | 183 + drivers/scsi/huawei/hifc/hifc_wq.c | 624 ++++ drivers/scsi/huawei/hifc/hifc_wq.h | 165 + 23 files changed, 19398 insertions(+) create mode 100644 drivers/scsi/huawei/hifc/hifc_api_cmd.c create mode 100644 drivers/scsi/huawei/hifc/hifc_api_cmd.h create mode 100644 drivers/scsi/huawei/hifc/hifc_cfg.c create mode 100644 drivers/scsi/huawei/hifc/hifc_cfg.h create mode 100644 drivers/scsi/huawei/hifc/hifc_cmdq.c create mode 100644 drivers/scsi/huawei/hifc/hifc_cmdq.h create mode 100644 drivers/scsi/huawei/hifc/hifc_cqm_main.c create mode 100644 drivers/scsi/huawei/hifc/hifc_cqm_main.h create mode 100644 drivers/scsi/huawei/hifc/hifc_cqm_object.c create mode 100644 drivers/scsi/huawei/hifc/hifc_cqm_object.h create mode 100644 drivers/scsi/huawei/hifc/hifc_eqs.c create mode 100644 drivers/scsi/huawei/hifc/hifc_eqs.h create mode 100644 drivers/scsi/huawei/hifc/hifc_hw.h create mode 100644 drivers/scsi/huawei/hifc/hifc_hwdev.c create mode 100644 drivers/scsi/huawei/hifc/hifc_hwdev.h create mode 100644 drivers/scsi/huawei/hifc/hifc_hwif.c create mode 100644 drivers/scsi/huawei/hifc/hifc_hwif.h create mode 100644 drivers/scsi/huawei/hifc/hifc_mgmt.c create mode 100644 drivers/scsi/huawei/hifc/hifc_mgmt.h create mode 100644 drivers/scsi/huawei/hifc/hifc_sml.c create mode 100644 drivers/scsi/huawei/hifc/hifc_sml.h create mode 100644 drivers/scsi/huawei/hifc/hifc_wq.c create mode 100644 drivers/scsi/huawei/hifc/hifc_wq.h diff --git a/drivers/scsi/huawei/hifc/hifc_api_cmd.c b/drivers/scsi/huawei/hifc/hifc_api_cmd.c new file mode 100644 index 000000000000..22632f779582 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_api_cmd.c @@ -0,0 +1,1155 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": [COMM]" fmt + +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/completion.h> +#include <linux/kernel.h> +#include <linux/device.h> +#include <linux/pci.h> +#include <linux/dma-mapping.h> +#include <linux/semaphore.h> +#include <linux/jiffies.h> +#include <linux/delay.h> + +#include "hifc_knl_adp.h" +#include "hifc_hw.h" +#include "hifc_hwdev.h" +#include "hifc_hwif.h" +#include "hifc_api_cmd.h" + +#define API_CMD_CHAIN_CELL_SIZE_SHIFT 6U + +#define API_CMD_CELL_DESC_SIZE 8 +#define API_CMD_CELL_DATA_ADDR_SIZE 8 + +#define API_CHAIN_NUM_CELLS 32 +#define API_CHAIN_CELL_SIZE 128 +#define API_CHAIN_RSP_DATA_SIZE 128 + +#define API_CMD_CELL_WB_ADDR_SIZE 8 + +#define API_CHAIN_CELL_ALIGNMENT 8 + +#define API_CMD_TIMEOUT 10000 +#define API_CMD_STATUS_TIMEOUT 100000 + +#define API_CMD_BUF_SIZE 2048ULL + +#define API_CMD_NODE_ALIGN_SIZE 512ULL +#define API_PAYLOAD_ALIGN_SIZE 64ULL + +#define API_CHAIN_RESP_ALIGNMENT 64ULL + +#define COMPLETION_TIMEOUT_DEFAULT 1000UL +#define POLLING_COMPLETION_TIMEOUT_DEFAULT 1000U + +#define API_CMD_RESPONSE_DATA_PADDR(val) be64_to_cpu(*((u64 *)(val))) + +#define READ_API_CMD_PRIV_DATA(id, token) (((id) << 16) + (token)) +#define WRITE_API_CMD_PRIV_DATA(id) (((u8)id) << 16) + +#define MASKED_IDX(chain, idx) ((idx) & ((chain)->num_cells - 1)) + +#define SIZE_4BYTES(size) (ALIGN((u32)(size), 4U) >> 2) +#define SIZE_8BYTES(size) (ALIGN((u32)(size), 8U) >> 3) + +enum api_cmd_data_format { + SGL_DATA = 1, +}; + +enum api_cmd_type { + API_CMD_WRITE_TYPE = 0, + API_CMD_READ_TYPE = 1, +}; + +enum api_cmd_bypass { + NOT_BYPASS = 0, + BYPASS = 1, +}; + +enum api_cmd_resp_aeq { + NOT_TRIGGER = 0, + TRIGGER = 1, +}; + +static u8 xor_chksum_set(void *data) +{ + int idx; + u8 checksum = 0; + u8 *val = data; + + for (idx = 0; idx < 7; idx++) + checksum ^= val[idx]; + + return checksum; +} + +static void set_prod_idx(struct hifc_api_cmd_chain *chain) +{ + enum hifc_api_cmd_chain_type chain_type = chain->chain_type; + struct hifc_hwif *hwif = chain->hwdev->hwif; + u32 hw_prod_idx_addr = HIFC_CSR_API_CMD_CHAIN_PI_ADDR(chain_type); + u32 prod_idx = chain->prod_idx; + + hifc_hwif_write_reg(hwif, hw_prod_idx_addr, prod_idx); +} + +static u32 get_hw_cons_idx(struct hifc_api_cmd_chain *chain) +{ + u32 addr, val; + + addr = HIFC_CSR_API_CMD_STATUS_0_ADDR(chain->chain_type); + val = hifc_hwif_read_reg(chain->hwdev->hwif, addr); + + return HIFC_API_CMD_STATUS_GET(val, CONS_IDX); +} + +static void dump_api_chain_reg(struct hifc_api_cmd_chain *chain) +{ + void *dev = chain->hwdev->dev_hdl; + u32 addr, val; + + addr = HIFC_CSR_API_CMD_STATUS_0_ADDR(chain->chain_type); + val = hifc_hwif_read_reg(chain->hwdev->hwif, addr); + + sdk_err(dev, "Chain type: 0x%x, cpld error: 0x%x, check error: 0x%x, current fsm: 0x%x\n", + chain->chain_type, HIFC_API_CMD_STATUS_GET(val, CPLD_ERR), + HIFC_API_CMD_STATUS_GET(val, CHKSUM_ERR), + HIFC_API_CMD_STATUS_GET(val, FSM)); + + sdk_err(dev, "Chain hw current ci: 0x%x\n", + HIFC_API_CMD_STATUS_GET(val, CONS_IDX)); + + addr = HIFC_CSR_API_CMD_CHAIN_PI_ADDR(chain->chain_type); + val = hifc_hwif_read_reg(chain->hwdev->hwif, addr); + sdk_err(dev, "Chain hw current pi: 0x%x\n", val); +} + +/** + * chain_busy - check if the chain is still processing last requests + * @chain: chain to check + * Return: 0 - success, negative - failure + **/ +static int chain_busy(struct hifc_api_cmd_chain *chain) +{ + void *dev = chain->hwdev->dev_hdl; + struct hifc_api_cmd_cell_ctxt *ctxt; + u64 resp_header; + + ctxt = &chain->cell_ctxt[chain->prod_idx]; + + switch (chain->chain_type) { + case HIFC_API_CMD_MULTI_READ: + case HIFC_API_CMD_POLL_READ: + resp_header = be64_to_cpu(ctxt->resp->header); + if (ctxt->status && + !HIFC_API_CMD_RESP_HEADER_VALID(resp_header)) { + sdk_err(dev, "Context(0x%x) busy!, pi: %d, resp_header: 0x%08x%08x\n", + ctxt->status, chain->prod_idx, + upper_32_bits(resp_header), + lower_32_bits(resp_header)); + dump_api_chain_reg(chain); + return -EBUSY; + } + break; + case HIFC_API_CMD_POLL_WRITE: + case HIFC_API_CMD_WRITE_TO_MGMT_CPU: + case HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU: + chain->cons_idx = get_hw_cons_idx(chain); + + if (chain->cons_idx == MASKED_IDX(chain, chain->prod_idx + 1)) { + sdk_err(dev, "API CMD chain %d is busy, cons_idx = %d, prod_idx = %d\n", + chain->chain_type, chain->cons_idx, + chain->prod_idx); + dump_api_chain_reg(chain); + return -EBUSY; + } + break; + default: + sdk_err(dev, "Unknown Chain type %d\n", chain->chain_type); + return -EINVAL; + } + + return 0; +} + +/** + * get_cell_data_size - get the data size of specific cell type + * @type: chain type + * @cmd_size: the command size + * Return: cell_data_size + **/ +static u16 get_cell_data_size(enum hifc_api_cmd_chain_type type, u16 cmd_size) +{ + u16 cell_data_size = 0; + + switch (type) { + case HIFC_API_CMD_POLL_READ: + cell_data_size = ALIGN(API_CMD_CELL_DESC_SIZE + + API_CMD_CELL_WB_ADDR_SIZE + + API_CMD_CELL_DATA_ADDR_SIZE, + API_CHAIN_CELL_ALIGNMENT); + break; + + case HIFC_API_CMD_WRITE_TO_MGMT_CPU: + case HIFC_API_CMD_POLL_WRITE: + case HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU: + cell_data_size = ALIGN(API_CMD_CELL_DESC_SIZE + + API_CMD_CELL_DATA_ADDR_SIZE, + API_CHAIN_CELL_ALIGNMENT); + break; + default: + break; + } + + return cell_data_size; +} + +/** + * prepare_cell_ctrl - prepare the ctrl of the cell for the command + * @cell_ctrl: the control of the cell to set the control into it + * @cell_len: the size of the cell + **/ +static void prepare_cell_ctrl(u64 *cell_ctrl, u16 cell_len) +{ + u64 ctrl; + u8 chksum; + + ctrl = HIFC_API_CMD_CELL_CTRL_SET(SIZE_8BYTES(cell_len), CELL_LEN) | + HIFC_API_CMD_CELL_CTRL_SET(0ULL, RD_DMA_ATTR_OFF) | + HIFC_API_CMD_CELL_CTRL_SET(0ULL, WR_DMA_ATTR_OFF); + + chksum = xor_chksum_set(&ctrl); + + ctrl |= HIFC_API_CMD_CELL_CTRL_SET(chksum, XOR_CHKSUM); + + /* The data in the HW should be in Big Endian Format */ + *cell_ctrl = cpu_to_be64(ctrl); +} + +/** + * prepare_api_cmd - prepare API CMD command + * @chain: chain for the command + * @cell: the cell of the command + * @dest: destination node on the card that will receive the command + * @cmd: command data + * @cmd_size: the command size + **/ +static void prepare_api_cmd(struct hifc_api_cmd_chain *chain, + struct hifc_api_cmd_cell *cell, + enum hifc_node_id dest, + const void *cmd, u16 cmd_size) +{ + struct hifc_api_cmd_cell_ctxt *cell_ctxt; + u32 priv; + + cell_ctxt = &chain->cell_ctxt[chain->prod_idx]; + + switch (chain->chain_type) { + case HIFC_API_CMD_POLL_READ: + priv = READ_API_CMD_PRIV_DATA(chain->chain_type, + cell_ctxt->saved_prod_idx); + cell->desc = HIFC_API_CMD_DESC_SET(SGL_DATA, API_TYPE) | + HIFC_API_CMD_DESC_SET(API_CMD_READ_TYPE, RD_WR) | + HIFC_API_CMD_DESC_SET(BYPASS, MGMT_BYPASS) | + HIFC_API_CMD_DESC_SET(NOT_TRIGGER, RESP_AEQE_EN) | + HIFC_API_CMD_DESC_SET(priv, PRIV_DATA); + break; + case HIFC_API_CMD_POLL_WRITE: + priv = WRITE_API_CMD_PRIV_DATA(chain->chain_type); + cell->desc = HIFC_API_CMD_DESC_SET(SGL_DATA, API_TYPE) | + HIFC_API_CMD_DESC_SET(API_CMD_WRITE_TYPE, RD_WR) | + HIFC_API_CMD_DESC_SET(BYPASS, MGMT_BYPASS) | + HIFC_API_CMD_DESC_SET(NOT_TRIGGER, RESP_AEQE_EN) | + HIFC_API_CMD_DESC_SET(priv, PRIV_DATA); + break; + case HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU: + case HIFC_API_CMD_WRITE_TO_MGMT_CPU: + priv = WRITE_API_CMD_PRIV_DATA(chain->chain_type); + cell->desc = HIFC_API_CMD_DESC_SET(SGL_DATA, API_TYPE) | + HIFC_API_CMD_DESC_SET(API_CMD_WRITE_TYPE, RD_WR) | + HIFC_API_CMD_DESC_SET(NOT_BYPASS, MGMT_BYPASS) | + HIFC_API_CMD_DESC_SET(TRIGGER, RESP_AEQE_EN) | + HIFC_API_CMD_DESC_SET(priv, PRIV_DATA); + break; + default: + sdk_err(chain->hwdev->dev_hdl, "Unknown Chain type: %d\n", + chain->chain_type); + return; + } + + cell->desc |= HIFC_API_CMD_DESC_SET(dest, DEST) | + HIFC_API_CMD_DESC_SET(SIZE_4BYTES(cmd_size), SIZE); + + cell->desc |= HIFC_API_CMD_DESC_SET(xor_chksum_set(&cell->desc), + XOR_CHKSUM); + + /* The data in the HW should be in Big Endian Format */ + cell->desc = cpu_to_be64(cell->desc); + + memcpy(cell_ctxt->api_cmd_vaddr, cmd, cmd_size); +} + +/** + * prepare_cell - prepare cell ctrl and cmd in the current producer cell + * @chain: chain for the command + * @dest: destination node on the card that will receive the command + * @cmd: command data + * @cmd_size: the command size + **/ +static void prepare_cell(struct hifc_api_cmd_chain *chain, + enum hifc_node_id dest, + void *cmd, u16 cmd_size) +{ + struct hifc_api_cmd_cell *curr_node; + u16 cell_size; + + curr_node = chain->curr_node; + + cell_size = get_cell_data_size(chain->chain_type, cmd_size); + + prepare_cell_ctrl(&curr_node->ctrl, cell_size); + prepare_api_cmd(chain, curr_node, dest, cmd, cmd_size); +} + +static inline void cmd_chain_prod_idx_inc(struct hifc_api_cmd_chain *chain) +{ + chain->prod_idx = MASKED_IDX(chain, chain->prod_idx + 1); +} + +static void issue_api_cmd(struct hifc_api_cmd_chain *chain) +{ + set_prod_idx(chain); +} + +/** + * api_cmd_status_update - update the status of the chain + * @chain: chain to update + **/ +static void api_cmd_status_update(struct hifc_api_cmd_chain *chain) +{ + struct hifc_api_cmd_status *wb_status; + enum hifc_api_cmd_chain_type chain_type; + u64 status_header; + u32 buf_desc; + + wb_status = chain->wb_status; + + buf_desc = be32_to_cpu(wb_status->buf_desc); + if (HIFC_API_CMD_STATUS_GET(buf_desc, CHKSUM_ERR)) + return; + + status_header = be64_to_cpu(wb_status->header); + chain_type = HIFC_API_CMD_STATUS_HEADER_GET(status_header, CHAIN_ID); + if (chain_type >= HIFC_API_CMD_MAX) + return; + + if (chain_type != chain->chain_type) + return; + + chain->cons_idx = HIFC_API_CMD_STATUS_GET(buf_desc, CONS_IDX); +} + +/** + * wait_for_status_poll - wait for write to mgmt command to complete + * @chain: the chain of the command + * Return: 0 - success, negative - failure + **/ +static int wait_for_status_poll(struct hifc_api_cmd_chain *chain) +{ + int err = -ETIMEDOUT; + u32 cnt = 0; + + while (cnt < API_CMD_STATUS_TIMEOUT && + chain->hwdev->chip_present_flag) { + api_cmd_status_update(chain); + + /* SYNC API CMD cmd should start after prev cmd finished */ + if (chain->cons_idx == chain->prod_idx) { + err = 0; + break; + } + + usleep_range(50, 100); + cnt++; + } + + return err; +} + +static void copy_resp_data(struct hifc_api_cmd_cell_ctxt *ctxt, void *ack, + u16 ack_size) +{ + struct hifc_api_cmd_resp_fmt *resp = ctxt->resp; + + memcpy(ack, &resp->resp_data, ack_size); + ctxt->status = 0; +} + +/** + * prepare_cell - polling for respense data of the read api-command + * @chain: pointer to api cmd chain + * + * Return: 0 - success, negative - failure + **/ +static int wait_for_resp_polling(struct hifc_api_cmd_cell_ctxt *ctxt) +{ + u64 resp_header; + int ret = -ETIMEDOUT; + u32 cnt = 0; + + while (cnt < POLLING_COMPLETION_TIMEOUT_DEFAULT) { + resp_header = be64_to_cpu(ctxt->resp->header); + + rmb(); /* read the latest header */ + + if (HIFC_API_CMD_RESP_HEADER_VALID(resp_header)) { + ret = 0; + break; + } + usleep_range(100, 1000); + cnt++; + } + + if (ret) + pr_err("Wait for api chain response timeout\n"); + + return ret; +} + +/** + * wait_for_api_cmd_completion - wait for command to complete + * @chain: chain for the command + * Return: 0 - success, negative - failure + **/ +static int wait_for_api_cmd_completion(struct hifc_api_cmd_chain *chain, + struct hifc_api_cmd_cell_ctxt *ctxt, + void *ack, u16 ack_size) +{ + void *dev = chain->hwdev->dev_hdl; + int err = 0; + + switch (chain->chain_type) { + case HIFC_API_CMD_POLL_READ: + err = wait_for_resp_polling(ctxt); + if (!err) + copy_resp_data(ctxt, ack, ack_size); + break; + case HIFC_API_CMD_POLL_WRITE: + case HIFC_API_CMD_WRITE_TO_MGMT_CPU: + err = wait_for_status_poll(chain); + if (err) { + sdk_err(dev, "API CMD Poll status timeout, chain type: %d\n", + chain->chain_type); + break; + } + break; + case HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU: + /* No need to wait */ + break; + default: + sdk_err(dev, "Unknown API CMD Chain type: %d\n", + chain->chain_type); + err = -EINVAL; + break; + } + + if (err) + dump_api_chain_reg(chain); + + return err; +} + +static inline void update_api_cmd_ctxt(struct hifc_api_cmd_chain *chain, + struct hifc_api_cmd_cell_ctxt *ctxt) +{ + ctxt->status = 1; + ctxt->saved_prod_idx = chain->prod_idx; + if (ctxt->resp) { + ctxt->resp->header = 0; + + /* make sure "header" was cleared */ + wmb(); + } +} + +/** + * api_cmd - API CMD command + * @chain: chain for the command + * @dest: destination node on the card that will receive the command + * @cmd: command data + * @size: the command size + * Return: 0 - success, negative - failure + **/ +static int api_cmd(struct hifc_api_cmd_chain *chain, + enum hifc_node_id dest, + void *cmd, u16 cmd_size, void *ack, u16 ack_size) +{ + struct hifc_api_cmd_cell_ctxt *ctxt; + + if (chain->chain_type == HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU) + spin_lock(&chain->async_lock); + else + down(&chain->sem); + ctxt = &chain->cell_ctxt[chain->prod_idx]; + if (chain_busy(chain)) { + if (chain->chain_type == HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU) + spin_unlock(&chain->async_lock); + else + up(&chain->sem); + return -EBUSY; + } + update_api_cmd_ctxt(chain, ctxt); + + prepare_cell(chain, dest, cmd, cmd_size); + + cmd_chain_prod_idx_inc(chain); + + wmb(); /* issue the command */ + + issue_api_cmd(chain); + + /* incremented prod idx, update ctxt */ + + chain->curr_node = chain->cell_ctxt[chain->prod_idx].cell_vaddr; + if (chain->chain_type == HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU) + spin_unlock(&chain->async_lock); + else + up(&chain->sem); + + return wait_for_api_cmd_completion(chain, ctxt, ack, ack_size); +} + +/** + * hifc_api_cmd_write - Write API CMD command + * @chain: chain for write command + * @dest: destination node on the card that will receive the command + * @cmd: command data + * @size: the command size + * Return: 0 - success, negative - failure + **/ +int hifc_api_cmd_write(struct hifc_api_cmd_chain *chain, + enum hifc_node_id dest, void *cmd, u16 size) +{ + /* Verify the chain type */ + return api_cmd(chain, dest, cmd, size, NULL, 0); +} + +int hifc_api_cmd_read(struct hifc_api_cmd_chain *chain, + enum hifc_node_id dest, + void *cmd, u16 size, void *ack, u16 ack_size) +{ + return api_cmd(chain, dest, cmd, size, ack, ack_size); +} + +/** + * api_cmd_hw_restart - restart the chain in the HW + * @chain: the API CMD specific chain to restart + **/ +static int api_cmd_hw_restart(struct hifc_api_cmd_chain *cmd_chain) +{ + struct hifc_hwif *hwif = cmd_chain->hwdev->hwif; + u32 reg_addr, val; + int err; + u32 cnt = 0; + + /* Read Modify Write */ + reg_addr = HIFC_CSR_API_CMD_CHAIN_REQ_ADDR(cmd_chain->chain_type); + val = hifc_hwif_read_reg(hwif, reg_addr); + + val = HIFC_API_CMD_CHAIN_REQ_CLEAR(val, RESTART); + val |= HIFC_API_CMD_CHAIN_REQ_SET(1, RESTART); + + hifc_hwif_write_reg(hwif, reg_addr, val); + + err = -ETIMEDOUT; + while (cnt < API_CMD_TIMEOUT) { + val = hifc_hwif_read_reg(hwif, reg_addr); + + if (!HIFC_API_CMD_CHAIN_REQ_GET(val, RESTART)) { + err = 0; + break; + } + + usleep_range(900, 1000); + cnt++; + } + + return err; +} + +/** + * api_cmd_ctrl_init - set the control register of a chain + * @chain: the API CMD specific chain to set control register for + **/ +static void api_cmd_ctrl_init(struct hifc_api_cmd_chain *chain) +{ + struct hifc_hwif *hwif = chain->hwdev->hwif; + u32 reg_addr, ctrl; + u32 size; + + /* Read Modify Write */ + reg_addr = HIFC_CSR_API_CMD_CHAIN_CTRL_ADDR(chain->chain_type); + + size = (u32)ilog2(chain->cell_size >> API_CMD_CHAIN_CELL_SIZE_SHIFT); + + ctrl = hifc_hwif_read_reg(hwif, reg_addr); + + ctrl = HIFC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, AEQE_EN) & + HIFC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, CELL_SIZE); + + ctrl |= HIFC_API_CMD_CHAIN_CTRL_SET(0, AEQE_EN) | + HIFC_API_CMD_CHAIN_CTRL_SET(size, CELL_SIZE); + + hifc_hwif_write_reg(hwif, reg_addr, ctrl); +} + +/** + * api_cmd_set_status_addr - set the status address of a chain in the HW + * @chain: the API CMD specific chain to set status address for + **/ +static void api_cmd_set_status_addr(struct hifc_api_cmd_chain *chain) +{ + struct hifc_hwif *hwif = chain->hwdev->hwif; + u32 addr, val; + + addr = HIFC_CSR_API_CMD_STATUS_HI_ADDR(chain->chain_type); + val = upper_32_bits(chain->wb_status_paddr); + hifc_hwif_write_reg(hwif, addr, val); + + addr = HIFC_CSR_API_CMD_STATUS_LO_ADDR(chain->chain_type); + val = lower_32_bits(chain->wb_status_paddr); + hifc_hwif_write_reg(hwif, addr, val); +} + +/** + * api_cmd_set_num_cells - set the number cells of a chain in the HW + * @chain: the API CMD specific chain to set the number of cells for + **/ +static void api_cmd_set_num_cells(struct hifc_api_cmd_chain *chain) +{ + struct hifc_hwif *hwif = chain->hwdev->hwif; + u32 addr, val; + + addr = HIFC_CSR_API_CMD_CHAIN_NUM_CELLS_ADDR(chain->chain_type); + val = chain->num_cells; + hifc_hwif_write_reg(hwif, addr, val); +} + +/** + * api_cmd_head_init - set the head cell of a chain in the HW + * @chain: the API CMD specific chain to set the head for + **/ +static void api_cmd_head_init(struct hifc_api_cmd_chain *chain) +{ + struct hifc_hwif *hwif = chain->hwdev->hwif; + u32 addr, val; + + addr = HIFC_CSR_API_CMD_CHAIN_HEAD_HI_ADDR(chain->chain_type); + val = upper_32_bits(chain->head_cell_paddr); + hifc_hwif_write_reg(hwif, addr, val); + + addr = HIFC_CSR_API_CMD_CHAIN_HEAD_LO_ADDR(chain->chain_type); + val = lower_32_bits(chain->head_cell_paddr); + hifc_hwif_write_reg(hwif, addr, val); +} + +/** + * wait_for_ready_chain - wait for the chain to be ready + * @chain: the API CMD specific chain to wait for + * Return: 0 - success, negative - failure + **/ +static int wait_for_ready_chain(struct hifc_api_cmd_chain *chain) +{ + struct hifc_hwif *hwif = chain->hwdev->hwif; + u32 addr, val; + u32 hw_cons_idx; + u32 cnt = 0; + int err; + + addr = HIFC_CSR_API_CMD_STATUS_0_ADDR(chain->chain_type); + err = -ETIMEDOUT; + while (cnt < API_CMD_TIMEOUT) { + val = hifc_hwif_read_reg(hwif, addr); + hw_cons_idx = HIFC_API_CMD_STATUS_GET(val, CONS_IDX); + + /* wait for HW cons idx to be updated */ + if (hw_cons_idx == chain->cons_idx) { + err = 0; + break; + } + + usleep_range(900, 1000); + cnt++; + } + + return err; +} + +/** + * api_cmd_chain_hw_clean - clean the HW + * @chain: the API CMD specific chain + **/ +static void api_cmd_chain_hw_clean(struct hifc_api_cmd_chain *chain) +{ + struct hifc_hwif *hwif = chain->hwdev->hwif; + u32 addr, ctrl; + + addr = HIFC_CSR_API_CMD_CHAIN_CTRL_ADDR(chain->chain_type); + + ctrl = hifc_hwif_read_reg(hwif, addr); + ctrl = HIFC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, RESTART_EN) & + HIFC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, XOR_ERR) & + HIFC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, AEQE_EN) & + HIFC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, XOR_CHK_EN) & + HIFC_API_CMD_CHAIN_CTRL_CLEAR(ctrl, CELL_SIZE); + + hifc_hwif_write_reg(hwif, addr, ctrl); +} + +/** + * api_cmd_chain_hw_init - initialize the chain in the HW + * @chain: the API CMD specific chain to initialize in HW + * Return: 0 - success, negative - failure + **/ +static int api_cmd_chain_hw_init(struct hifc_api_cmd_chain *chain) +{ + api_cmd_chain_hw_clean(chain); + + api_cmd_set_status_addr(chain); + + if (api_cmd_hw_restart(chain)) { + sdk_err(chain->hwdev->dev_hdl, "Failed to restart api_cmd_hw\n"); + return -EBUSY; + } + + api_cmd_ctrl_init(chain); + api_cmd_set_num_cells(chain); + api_cmd_head_init(chain); + + return wait_for_ready_chain(chain); +} + +/** + * alloc_cmd_buf - allocate a dma buffer for API CMD command + * @chain: the API CMD specific chain for the cmd + * @cell: the cell in the HW for the cmd + * @cell_idx: the index of the cell + * Return: 0 - success, negative - failure + **/ +static int alloc_cmd_buf(struct hifc_api_cmd_chain *chain, + struct hifc_api_cmd_cell *cell, u32 cell_idx) +{ + struct hifc_api_cmd_cell_ctxt *cell_ctxt; + void *dev = chain->hwdev->dev_hdl; + void *buf_vaddr; + u64 buf_paddr; + int err = 0; + + buf_vaddr = (u8 *)((u64)chain->buf_vaddr_base + + chain->buf_size_align * cell_idx); + buf_paddr = chain->buf_paddr_base + + chain->buf_size_align * cell_idx; + + cell_ctxt = &chain->cell_ctxt[cell_idx]; + + cell_ctxt->api_cmd_vaddr = buf_vaddr; + + /* set the cmd DMA address in the cell */ + switch (chain->chain_type) { + case HIFC_API_CMD_POLL_READ: + cell->read.hw_cmd_paddr = cpu_to_be64(buf_paddr); + break; + case HIFC_API_CMD_WRITE_TO_MGMT_CPU: + case HIFC_API_CMD_POLL_WRITE: + case HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU: + /* The data in the HW should be in Big Endian Format */ + cell->write.hw_cmd_paddr = cpu_to_be64(buf_paddr); + break; + default: + sdk_err(dev, "Unknown API CMD Chain type: %d\n", + chain->chain_type); + err = -EINVAL; + break; + } + + return err; +} + +static void alloc_resp_buf(struct hifc_api_cmd_chain *chain, + struct hifc_api_cmd_cell *cell, u32 cell_idx) +{ + struct hifc_api_cmd_cell_ctxt *cell_ctxt; + void *resp_vaddr; + u64 resp_paddr; + + resp_vaddr = (u8 *)((u64)chain->rsp_vaddr_base + + chain->rsp_size_align * cell_idx); + resp_paddr = chain->rsp_paddr_base + + chain->rsp_size_align * cell_idx; + + cell_ctxt = &chain->cell_ctxt[cell_idx]; + + cell_ctxt->resp = resp_vaddr; + cell->read.hw_wb_resp_paddr = cpu_to_be64(resp_paddr); +} + +static int hifc_alloc_api_cmd_cell_buf(struct hifc_api_cmd_chain *chain, + u32 cell_idx, + struct hifc_api_cmd_cell *node) +{ + void *dev = chain->hwdev->dev_hdl; + int err; + + /* For read chain, we should allocate buffer for the response data */ + if (chain->chain_type == HIFC_API_CMD_MULTI_READ || + chain->chain_type == HIFC_API_CMD_POLL_READ) + alloc_resp_buf(chain, node, cell_idx); + + switch (chain->chain_type) { + case HIFC_API_CMD_WRITE_TO_MGMT_CPU: + case HIFC_API_CMD_POLL_WRITE: + case HIFC_API_CMD_POLL_READ: + case HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU: + err = alloc_cmd_buf(chain, node, cell_idx); + if (err) { + sdk_err(dev, "Failed to allocate cmd buffer\n"); + goto alloc_cmd_buf_err; + } + break; + /* For api command write and api command read, the data section + * is directly inserted in the cell, so no need to allocate. + */ + case HIFC_API_CMD_MULTI_READ: + chain->cell_ctxt[cell_idx].api_cmd_vaddr = + &node->read.hw_cmd_paddr; + break; + default: + sdk_err(dev, "Unsupported API CMD chain type\n"); + err = -EINVAL; + goto alloc_cmd_buf_err; + } + + return 0; + +alloc_cmd_buf_err: + + return err; +} + +/** + * api_cmd_create_cell - create API CMD cell of specific chain + * @chain: the API CMD specific chain to create its cell + * @cell_idx: the cell index to create + * @pre_node: previous cell + * @node_vaddr: the virt addr of the cell + * Return: 0 - success, negative - failure + **/ +static int api_cmd_create_cell(struct hifc_api_cmd_chain *chain, u32 cell_idx, + struct hifc_api_cmd_cell *pre_node, + struct hifc_api_cmd_cell **node_vaddr) +{ + struct hifc_api_cmd_cell_ctxt *cell_ctxt; + struct hifc_api_cmd_cell *node; + void *cell_vaddr; + u64 cell_paddr; + int err; + + cell_vaddr = (void *)((u64)chain->cell_vaddr_base + + chain->cell_size_align * cell_idx); + cell_paddr = chain->cell_paddr_base + + chain->cell_size_align * cell_idx; + + cell_ctxt = &chain->cell_ctxt[cell_idx]; + cell_ctxt->cell_vaddr = cell_vaddr; + node = cell_ctxt->cell_vaddr; + + if (!pre_node) { + chain->head_node = cell_vaddr; + chain->head_cell_paddr = cell_paddr; + } else { + /* The data in the HW should be in Big Endian Format */ + pre_node->next_cell_paddr = cpu_to_be64(cell_paddr); + } + + /* Driver software should make sure that there is an empty API + * command cell at the end the chain + */ + node->next_cell_paddr = 0; + + err = hifc_alloc_api_cmd_cell_buf(chain, cell_idx, node); + if (err) + return err; + + *node_vaddr = node; + + return 0; +} + +/** + * api_cmd_create_cells - create API CMD cells for specific chain + * @chain: the API CMD specific chain + * Return: 0 - success, negative - failure + **/ +static int api_cmd_create_cells(struct hifc_api_cmd_chain *chain) +{ + struct hifc_api_cmd_cell *node = NULL, *pre_node = NULL; + void *dev = chain->hwdev->dev_hdl; + u32 cell_idx; + int err; + + for (cell_idx = 0; cell_idx < chain->num_cells; cell_idx++) { + err = api_cmd_create_cell(chain, cell_idx, pre_node, &node); + if (err) { + sdk_err(dev, "Failed to create API CMD cell\n"); + return err; + } + + pre_node = node; + } + + if (!node) + return -EFAULT; + + /* set the Final node to point on the start */ + node->next_cell_paddr = cpu_to_be64(chain->head_cell_paddr); + + /* set the current node to be the head */ + chain->curr_node = chain->head_node; + return 0; +} + +/** + * api_chain_init - initialize API CMD specific chain + * @chain: the API CMD specific chain to initialize + * @attr: attributes to set in the chain + * Return: 0 - success, negative - failure + **/ +static int api_chain_init(struct hifc_api_cmd_chain *chain, + struct hifc_api_cmd_chain_attr *attr) +{ + void *dev = chain->hwdev->dev_hdl; + size_t cell_ctxt_size; + size_t cells_buf_size; + int err; + + chain->chain_type = attr->chain_type; + chain->num_cells = attr->num_cells; + chain->cell_size = attr->cell_size; + chain->rsp_size = attr->rsp_size; + + chain->prod_idx = 0; + chain->cons_idx = 0; + + if (chain->chain_type == HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU) + spin_lock_init(&chain->async_lock); + else + sema_init(&chain->sem, 1); + + cell_ctxt_size = chain->num_cells * sizeof(*chain->cell_ctxt); + if (!cell_ctxt_size) { + sdk_err(dev, "Api chain cell size cannot be zero\n"); + err = -EINVAL; + goto alloc_cell_ctxt_err; + } + + chain->cell_ctxt = kzalloc(cell_ctxt_size, GFP_KERNEL); + if (!chain->cell_ctxt) { + sdk_err(dev, "Failed to allocate cell contexts for a chain\n"); + err = -ENOMEM; + goto alloc_cell_ctxt_err; + } + + chain->wb_status = dma_zalloc_coherent(dev, + sizeof(*chain->wb_status), + &chain->wb_status_paddr, + GFP_KERNEL); + if (!chain->wb_status) { + sdk_err(dev, "Failed to allocate DMA wb status\n"); + err = -ENOMEM; + goto alloc_wb_status_err; + } + + chain->cell_size_align = ALIGN((u64)chain->cell_size, + API_CMD_NODE_ALIGN_SIZE); + chain->rsp_size_align = ALIGN((u64)chain->rsp_size, + API_CHAIN_RESP_ALIGNMENT); + chain->buf_size_align = ALIGN(API_CMD_BUF_SIZE, API_PAYLOAD_ALIGN_SIZE); + + cells_buf_size = (chain->cell_size_align + chain->rsp_size_align + + chain->buf_size_align) * chain->num_cells; + + err = hifc_dma_zalloc_coherent_align(dev, cells_buf_size, + API_CMD_NODE_ALIGN_SIZE, + GFP_KERNEL, + &chain->cells_addr); + if (err) { + sdk_err(dev, "Failed to allocate API CMD cells buffer\n"); + goto alloc_cells_buf_err; + } + + chain->cell_vaddr_base = chain->cells_addr.align_vaddr; + chain->cell_paddr_base = chain->cells_addr.align_paddr; + + chain->rsp_vaddr_base = (u8 *)((u64)chain->cell_vaddr_base + + chain->cell_size_align * chain->num_cells); + chain->rsp_paddr_base = chain->cell_paddr_base + + chain->cell_size_align * chain->num_cells; + + chain->buf_vaddr_base = (u8 *)((u64)chain->rsp_vaddr_base + + chain->rsp_size_align * chain->num_cells); + chain->buf_paddr_base = chain->rsp_paddr_base + + chain->rsp_size_align * chain->num_cells; + + return 0; + +alloc_cells_buf_err: + dma_free_coherent(dev, sizeof(*chain->wb_status), + chain->wb_status, chain->wb_status_paddr); + +alloc_wb_status_err: + kfree(chain->cell_ctxt); + +alloc_cell_ctxt_err: + return err; +} + +/** + * api_chain_free - free API CMD specific chain + * @chain: the API CMD specific chain to free + **/ +static void api_chain_free(struct hifc_api_cmd_chain *chain) +{ + void *dev = chain->hwdev->dev_hdl; + + hifc_dma_free_coherent_align(dev, &chain->cells_addr); + + dma_free_coherent(dev, sizeof(*chain->wb_status), + chain->wb_status, chain->wb_status_paddr); + kfree(chain->cell_ctxt); +} + +/** + * api_cmd_create_chain - create API CMD specific chain + * @chain: the API CMD specific chain to create + * @attr: attributes to set in the chain + * Return: 0 - success, negative - failure + **/ +static int api_cmd_create_chain(struct hifc_api_cmd_chain **cmd_chain, + struct hifc_api_cmd_chain_attr *attr) +{ + struct hifc_hwdev *hwdev = attr->hwdev; + struct hifc_api_cmd_chain *chain; + int err; + + if (attr->num_cells & (attr->num_cells - 1)) { + sdk_err(hwdev->dev_hdl, "Invalid number of cells, must be power of 2\n"); + return -EINVAL; + } + + chain = kzalloc(sizeof(*chain), GFP_KERNEL); + if (!chain) + return -ENOMEM; + + chain->hwdev = hwdev; + + err = api_chain_init(chain, attr); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to initialize chain\n"); + goto chain_init_err; + } + + err = api_cmd_create_cells(chain); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to create cells for API CMD chain\n"); + goto create_cells_err; + } + + err = api_cmd_chain_hw_init(chain); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to initialize chain HW\n"); + goto chain_hw_init_err; + } + + *cmd_chain = chain; + return 0; + +chain_hw_init_err: +create_cells_err: + api_chain_free(chain); + +chain_init_err: + kfree(chain); + return err; +} + +/** + * api_cmd_destroy_chain - destroy API CMD specific chain + * @chain: the API CMD specific chain to destroy + **/ +static void api_cmd_destroy_chain(struct hifc_api_cmd_chain *chain) +{ + api_chain_free(chain); + kfree(chain); +} + +/** + * hifc_api_cmd_init - Initialize all the API CMD chains + * @hwdev: the pointer to hw device + * @chain: the API CMD chains that will be initialized + * Return: 0 - success, negative - failure + **/ +int hifc_api_cmd_init(struct hifc_hwdev *hwdev, + struct hifc_api_cmd_chain **chain) +{ + void *dev = hwdev->dev_hdl; + struct hifc_api_cmd_chain_attr attr; + enum hifc_api_cmd_chain_type chain_type, i; + int err; + + attr.hwdev = hwdev; + attr.num_cells = API_CHAIN_NUM_CELLS; + attr.cell_size = API_CHAIN_CELL_SIZE; + attr.rsp_size = API_CHAIN_RSP_DATA_SIZE; + + chain_type = HIFC_API_CMD_WRITE_TO_MGMT_CPU; + for (; chain_type < HIFC_API_CMD_MAX; chain_type++) { + attr.chain_type = chain_type; + + err = api_cmd_create_chain(&chain[chain_type], &attr); + if (err) { + sdk_err(dev, "Failed to create chain %d\n", chain_type); + goto create_chain_err; + } + } + + return 0; + +create_chain_err: + i = HIFC_API_CMD_WRITE_TO_MGMT_CPU; + for (; i < chain_type; i++) + api_cmd_destroy_chain(chain[i]); + + return err; +} + +/** + * hifc_api_cmd_free - free the API CMD chains + * @chain: the API CMD chains that will be freed + **/ +void hifc_api_cmd_free(struct hifc_api_cmd_chain **chain) +{ + enum hifc_api_cmd_chain_type chain_type; + + chain_type = HIFC_API_CMD_WRITE_TO_MGMT_CPU; + + for (; chain_type < HIFC_API_CMD_MAX; chain_type++) + api_cmd_destroy_chain(chain[chain_type]); +} + diff --git a/drivers/scsi/huawei/hifc/hifc_api_cmd.h b/drivers/scsi/huawei/hifc/hifc_api_cmd.h new file mode 100644 index 000000000000..bd14db34a119 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_api_cmd.h @@ -0,0 +1,268 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#ifndef HIFC_API_CMD_H_ +#define HIFC_API_CMD_H_ + +#define HIFC_API_CMD_CELL_CTRL_CELL_LEN_SHIFT 0 +#define HIFC_API_CMD_CELL_CTRL_RD_DMA_ATTR_OFF_SHIFT 16 +#define HIFC_API_CMD_CELL_CTRL_WR_DMA_ATTR_OFF_SHIFT 24 +#define HIFC_API_CMD_CELL_CTRL_XOR_CHKSUM_SHIFT 56 + +#define HIFC_API_CMD_CELL_CTRL_CELL_LEN_MASK 0x3FU +#define HIFC_API_CMD_CELL_CTRL_RD_DMA_ATTR_OFF_MASK 0x3FU +#define HIFC_API_CMD_CELL_CTRL_WR_DMA_ATTR_OFF_MASK 0x3FU +#define HIFC_API_CMD_CELL_CTRL_XOR_CHKSUM_MASK 0xFFU + +#define HIFC_API_CMD_CELL_CTRL_SET(val, member) \ + ((((u64)val) & HIFC_API_CMD_CELL_CTRL_##member##_MASK) << \ + HIFC_API_CMD_CELL_CTRL_##member##_SHIFT) + +#define HIFC_API_CMD_DESC_API_TYPE_SHIFT 0 +#define HIFC_API_CMD_DESC_RD_WR_SHIFT 1 +#define HIFC_API_CMD_DESC_MGMT_BYPASS_SHIFT 2 +#define HIFC_API_CMD_DESC_RESP_AEQE_EN_SHIFT 3 +#define HIFC_API_CMD_DESC_PRIV_DATA_SHIFT 8 +#define HIFC_API_CMD_DESC_DEST_SHIFT 32 +#define HIFC_API_CMD_DESC_SIZE_SHIFT 40 +#define HIFC_API_CMD_DESC_XOR_CHKSUM_SHIFT 56 + +#define HIFC_API_CMD_DESC_API_TYPE_MASK 0x1U +#define HIFC_API_CMD_DESC_RD_WR_MASK 0x1U +#define HIFC_API_CMD_DESC_MGMT_BYPASS_MASK 0x1U +#define HIFC_API_CMD_DESC_RESP_AEQE_EN_MASK 0x1U +#define HIFC_API_CMD_DESC_DEST_MASK 0x1FU +#define HIFC_API_CMD_DESC_SIZE_MASK 0x7FFU +#define HIFC_API_CMD_DESC_XOR_CHKSUM_MASK 0xFFU +#define HIFC_API_CMD_DESC_PRIV_DATA_MASK 0xFFFFFFU + +#define HIFC_API_CMD_DESC_SET(val, member) \ + ((((u64)val) & HIFC_API_CMD_DESC_##member##_MASK) << \ + HIFC_API_CMD_DESC_##member##_SHIFT) +#define HIFC_API_CMD_STATUS_HEADER_VALID_SHIFT 0 +#define HIFC_API_CMD_STATUS_HEADER_CHAIN_ID_SHIFT 16 + +#define HIFC_API_CMD_STATUS_HEADER_VALID_MASK 0xFFU +#define HIFC_API_CMD_STATUS_HEADER_CHAIN_ID_MASK 0xFFU +#define HIFC_API_CMD_STATUS_HEADER_GET(val, member) \ + (((val) >> HIFC_API_CMD_STATUS_HEADER_##member##_SHIFT) & \ + HIFC_API_CMD_STATUS_HEADER_##member##_MASK) +#define HIFC_API_CMD_CHAIN_REQ_RESTART_SHIFT 1 +#define HIFC_API_CMD_CHAIN_REQ_RESTART_MASK 0x1U +#define HIFC_API_CMD_CHAIN_REQ_WB_TRIGGER_MASK 0x1U +#define HIFC_API_CMD_CHAIN_REQ_SET(val, member) \ + (((val) & HIFC_API_CMD_CHAIN_REQ_##member##_MASK) << \ + HIFC_API_CMD_CHAIN_REQ_##member##_SHIFT) + +#define HIFC_API_CMD_CHAIN_REQ_GET(val, member) \ + (((val) >> HIFC_API_CMD_CHAIN_REQ_##member##_SHIFT) & \ + HIFC_API_CMD_CHAIN_REQ_##member##_MASK) + +#define HIFC_API_CMD_CHAIN_REQ_CLEAR(val, member) \ + ((val) & (~(HIFC_API_CMD_CHAIN_REQ_##member##_MASK \ + << HIFC_API_CMD_CHAIN_REQ_##member##_SHIFT))) + +#define HIFC_API_CMD_CHAIN_CTRL_RESTART_EN_SHIFT 1 +#define HIFC_API_CMD_CHAIN_CTRL_XOR_ERR_SHIFT 2 +#define HIFC_API_CMD_CHAIN_CTRL_AEQE_EN_SHIFT 4 +#define HIFC_API_CMD_CHAIN_CTRL_AEQ_ID_SHIFT 8 +#define HIFC_API_CMD_CHAIN_CTRL_XOR_CHK_EN_SHIFT 28 +#define HIFC_API_CMD_CHAIN_CTRL_CELL_SIZE_SHIFT 30 + +#define HIFC_API_CMD_CHAIN_CTRL_RESTART_EN_MASK 0x1U +#define HIFC_API_CMD_CHAIN_CTRL_XOR_ERR_MASK 0x1U +#define HIFC_API_CMD_CHAIN_CTRL_AEQE_EN_MASK 0x1U +#define HIFC_API_CMD_CHAIN_CTRL_AEQ_ID_MASK 0x3U +#define HIFC_API_CMD_CHAIN_CTRL_XOR_CHK_EN_MASK 0x3U +#define HIFC_API_CMD_CHAIN_CTRL_CELL_SIZE_MASK 0x3U + +#define HIFC_API_CMD_CHAIN_CTRL_SET(val, member) \ + (((val) & HIFC_API_CMD_CHAIN_CTRL_##member##_MASK) << \ + HIFC_API_CMD_CHAIN_CTRL_##member##_SHIFT) + +#define HIFC_API_CMD_CHAIN_CTRL_CLEAR(val, member) \ + ((val) & (~(HIFC_API_CMD_CHAIN_CTRL_##member##_MASK \ + << HIFC_API_CMD_CHAIN_CTRL_##member##_SHIFT))) + +#define HIFC_API_CMD_RESP_HEAD_VALID_MASK 0xFF +#define HIFC_API_CMD_RESP_HEAD_VALID_CODE 0xFF + +#define HIFC_API_CMD_RESP_HEADER_VALID(val) \ + (((val) & HIFC_API_CMD_RESP_HEAD_VALID_MASK) == \ + HIFC_API_CMD_RESP_HEAD_VALID_CODE) +#define HIFC_API_CMD_STATUS_CONS_IDX_MASK 0xFFFFFFU +#define HIFC_API_CMD_STATUS_CONS_IDX_SHIFT 0 +#define HIFC_API_CMD_STATUS_FSM_MASK 0xFU +#define HIFC_API_CMD_STATUS_FSM_SHIFT 24 +#define HIFC_API_CMD_STATUS_CHKSUM_ERR_MASK 0x3U +#define HIFC_API_CMD_STATUS_CHKSUM_ERR_SHIFT 28 +#define HIFC_API_CMD_STATUS_CPLD_ERR_MASK 0x1U +#define HIFC_API_CMD_STATUS_CPLD_ERR_SHIFT 30 + +#define HIFC_API_CMD_STATUS_GET(val, member) \ + (((val) >> HIFC_API_CMD_STATUS_##member##_SHIFT) & \ + HIFC_API_CMD_STATUS_##member##_MASK) + +/* API CMD registers */ +#define HIFC_CSR_API_CMD_BASE 0xF000 + +#define HIFC_CSR_API_CMD_STRIDE 0x100 + +#define HIFC_CSR_API_CMD_CHAIN_HEAD_HI_ADDR(idx) \ + (HIFC_CSR_API_CMD_BASE + 0x0 + (idx) * HIFC_CSR_API_CMD_STRIDE) + +#define HIFC_CSR_API_CMD_CHAIN_HEAD_LO_ADDR(idx) \ + (HIFC_CSR_API_CMD_BASE + 0x4 + (idx) * HIFC_CSR_API_CMD_STRIDE) + +#define HIFC_CSR_API_CMD_STATUS_HI_ADDR(idx) \ + (HIFC_CSR_API_CMD_BASE + 0x8 + (idx) * HIFC_CSR_API_CMD_STRIDE) + +#define HIFC_CSR_API_CMD_STATUS_LO_ADDR(idx) \ + (HIFC_CSR_API_CMD_BASE + 0xC + (idx) * HIFC_CSR_API_CMD_STRIDE) + +#define HIFC_CSR_API_CMD_CHAIN_NUM_CELLS_ADDR(idx) \ + (HIFC_CSR_API_CMD_BASE + 0x10 + (idx) * HIFC_CSR_API_CMD_STRIDE) + +#define HIFC_CSR_API_CMD_CHAIN_CTRL_ADDR(idx) \ + (HIFC_CSR_API_CMD_BASE + 0x14 + (idx) * HIFC_CSR_API_CMD_STRIDE) + +#define HIFC_CSR_API_CMD_CHAIN_PI_ADDR(idx) \ + (HIFC_CSR_API_CMD_BASE + 0x1C + (idx) * HIFC_CSR_API_CMD_STRIDE) + +#define HIFC_CSR_API_CMD_CHAIN_REQ_ADDR(idx) \ + (HIFC_CSR_API_CMD_BASE + 0x20 + (idx) * HIFC_CSR_API_CMD_STRIDE) + +#define HIFC_CSR_API_CMD_STATUS_0_ADDR(idx) \ + (HIFC_CSR_API_CMD_BASE + 0x30 + (idx) * HIFC_CSR_API_CMD_STRIDE) + +enum hifc_api_cmd_chain_type { + /* write command with completion notification */ + HIFC_API_CMD_WRITE = 0, + /* read command with completion notification */ + HIFC_API_CMD_READ = 1, + /* write to mgmt cpu command with completion */ + HIFC_API_CMD_WRITE_TO_MGMT_CPU = 2, + /* multi read command with completion notification - not used */ + HIFC_API_CMD_MULTI_READ = 3, + /* write command without completion notification */ + HIFC_API_CMD_POLL_WRITE = 4, + /* read command without completion notification */ + HIFC_API_CMD_POLL_READ = 5, + /* read from mgmt cpu command with completion */ + HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU = 6, + HIFC_API_CMD_MAX, +}; + +struct hifc_api_cmd_status { + u64 header; + u32 buf_desc; + u32 cell_addr_hi; + u32 cell_addr_lo; + u32 rsvd0; + u64 rsvd1; +}; + +/* HW struct */ +struct hifc_api_cmd_cell { + u64 ctrl; + + /* address is 64 bit in HW struct */ + u64 next_cell_paddr; + + u64 desc; + + /* HW struct */ + union { + struct { + u64 hw_cmd_paddr; + } write; + + struct { + u64 hw_wb_resp_paddr; + u64 hw_cmd_paddr; + } read; + }; +}; + +struct hifc_api_cmd_resp_fmt { + u64 header; + u64 rsvd[3]; + u64 resp_data; +}; + +struct hifc_api_cmd_cell_ctxt { + struct hifc_api_cmd_cell *cell_vaddr; + + void *api_cmd_vaddr; + + struct hifc_api_cmd_resp_fmt *resp; + + struct completion done; + int status; + + u32 saved_prod_idx; +}; + +struct hifc_api_cmd_chain_attr { + struct hifc_hwdev *hwdev; + enum hifc_api_cmd_chain_type chain_type; + + u32 num_cells; + u16 rsp_size; + u16 cell_size; +}; + +struct hifc_api_cmd_chain { + struct hifc_hwdev *hwdev; + enum hifc_api_cmd_chain_type chain_type; + + u32 num_cells; + u16 cell_size; + u16 rsp_size; + + /* HW members is 24 bit format */ + u32 prod_idx; + u32 cons_idx; + + struct semaphore sem; + /* Async cmd can not be scheduling */ + spinlock_t async_lock; + + dma_addr_t wb_status_paddr; + struct hifc_api_cmd_status *wb_status; + + dma_addr_t head_cell_paddr; + struct hifc_api_cmd_cell *head_node; + + struct hifc_api_cmd_cell_ctxt *cell_ctxt; + struct hifc_api_cmd_cell *curr_node; + + struct hifc_dma_addr_align cells_addr; + + u8 *cell_vaddr_base; + u64 cell_paddr_base; + u8 *rsp_vaddr_base; + u64 rsp_paddr_base; + u8 *buf_vaddr_base; + u64 buf_paddr_base; + u64 cell_size_align; + u64 rsp_size_align; + u64 buf_size_align; +}; + +int hifc_api_cmd_write(struct hifc_api_cmd_chain *chain, + enum hifc_node_id dest, void *cmd, u16 size); + +int hifc_api_cmd_read(struct hifc_api_cmd_chain *chain, + enum hifc_node_id dest, void *cmd, u16 size, + void *ack, u16 ack_size); + +int hifc_api_cmd_init(struct hifc_hwdev *hwdev, + struct hifc_api_cmd_chain **chain); + +void hifc_api_cmd_free(struct hifc_api_cmd_chain **chain); + +#endif diff --git a/drivers/scsi/huawei/hifc/hifc_cfg.c b/drivers/scsi/huawei/hifc/hifc_cfg.c new file mode 100644 index 000000000000..5ebe5d754c41 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_cfg.c @@ -0,0 +1,823 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ +#define pr_fmt(fmt) KBUILD_MODNAME ": [COMM]" fmt + +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/mutex.h> +#include <linux/device.h> +#include <linux/pci.h> +#include <linux/module.h> +#include <linux/completion.h> +#include <linux/semaphore.h> +#include <linux/vmalloc.h> + +#include "hifc_knl_adp.h" +#include "hifc_hw.h" +#include "hifc_hwdev.h" +#include "hifc_hwif.h" +#include "hifc_cqm_main.h" +#include "hifc_api_cmd.h" +#include "hifc_hw.h" +#include "hifc_mgmt.h" +#include "hifc_cfg.h" + +uint intr_mode; + +int hifc_sync_time(void *hwdev, u64 time) +{ + struct hifc_sync_time_info time_info = {0}; + u16 out_size = sizeof(time_info); + int err; + + time_info.mstime = time; + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_SYNC_TIME, &time_info, + sizeof(time_info), &time_info, &out_size, + 0); + if (err || time_info.status || !out_size) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Failed to sync time to mgmt, err: %d, status: 0x%x, out size: 0x%x\n", + err, time_info.status, out_size); + } + + return err; +} + +static void parse_pub_res_cap(struct service_cap *cap, + struct hifc_dev_cap *dev_cap, + enum func_type type) +{ + cap->port_id = dev_cap->port_id; + cap->force_up = dev_cap->force_up; + + pr_info("Get public resource capbility, force_up: 0x%x\n", + cap->force_up); + /* FC need max queue number, but max queue number info is in + * l2nic cap, we also put max queue num info in public cap, so + * FC can get correct max queue number info. + */ + cap->max_sqs = dev_cap->nic_max_sq + 1; + cap->max_rqs = dev_cap->nic_max_rq + 1; + + cap->host_total_function = dev_cap->host_total_func; + cap->host_oq_id_mask_val = dev_cap->host_oq_id_mask_val; + cap->max_connect_num = dev_cap->max_conn_num; + cap->max_stick2cache_num = dev_cap->max_stick2cache_num; + + pr_info("Get public resource capbility, svc_cap_en: 0x%x\n", + dev_cap->svc_cap_en); + pr_info("port_id=0x%x\n", cap->port_id); + pr_info("Host_total_function=0x%x, host_oq_id_mask_val=0x%x\n", + cap->host_total_function, cap->host_oq_id_mask_val); +} + +static void parse_fc_res_cap(struct service_cap *cap, + struct hifc_dev_cap *dev_cap, + enum func_type type) +{ + struct dev_fc_svc_cap *fc_cap = &cap->fc_cap.dev_fc_cap; + + fc_cap->max_parent_qpc_num = dev_cap->fc_max_pctx; + fc_cap->scq_num = dev_cap->fc_max_scq; + fc_cap->srq_num = dev_cap->fc_max_srq; + fc_cap->max_child_qpc_num = dev_cap->fc_max_cctx; + fc_cap->vp_id_start = dev_cap->fc_vp_id_start; + fc_cap->vp_id_end = dev_cap->fc_vp_id_end; + + pr_info("Get fc resource capbility\n"); + pr_info("Max_parent_qpc_num=0x%x, scq_num=0x%x, srq_num=0x%x, max_child_qpc_num=0x%x\n", + fc_cap->max_parent_qpc_num, fc_cap->scq_num, fc_cap->srq_num, + fc_cap->max_child_qpc_num); + pr_info("Vp_id_start=0x%x, vp_id_end=0x%x\n", + fc_cap->vp_id_start, fc_cap->vp_id_end); +} + +static void parse_dev_cap(struct hifc_hwdev *dev, + struct hifc_dev_cap *dev_cap, enum func_type type) +{ + struct service_cap *cap = &dev->cfg_mgmt->svc_cap; + + /* Public resource */ + parse_pub_res_cap(cap, dev_cap, type); + + /* PPF managed dynamic resource */ + + parse_fc_res_cap(cap, dev_cap, type); +} + +static int get_cap_from_fw(struct hifc_hwdev *dev, enum func_type type) +{ + struct hifc_dev_cap dev_cap = {0}; + u16 out_len = sizeof(dev_cap); + int err; + + dev_cap.version = HIFC_CMD_VER_FUNC_ID; + err = hifc_global_func_id_get(dev, &dev_cap.func_id); + if (err) + return err; + + sdk_info(dev->dev_hdl, "Get cap from fw, func_idx: %d\n", + dev_cap.func_id); + + err = hifc_msg_to_mgmt_sync(dev, HIFC_MOD_CFGM, HIFC_CFG_NIC_CAP, + &dev_cap, sizeof(dev_cap), + &dev_cap, &out_len, 0); + if (err || dev_cap.status || !out_len) { + sdk_err(dev->dev_hdl, + "Failed to get capability from FW, err: %d, status: 0x%x, out size: 0x%x\n", + err, dev_cap.status, out_len); + return -EFAULT; + } + + parse_dev_cap(dev, &dev_cap, type); + return 0; +} + +static void fc_param_fix(struct hifc_hwdev *dev) +{ + struct service_cap *cap = &dev->cfg_mgmt->svc_cap; + struct fc_service_cap *fc_cap = &cap->fc_cap; + + fc_cap->parent_qpc_size = FC_PCTX_SZ; + fc_cap->child_qpc_size = FC_CCTX_SZ; + fc_cap->sqe_size = FC_SQE_SZ; + + fc_cap->scqc_size = FC_SCQC_SZ; + fc_cap->scqe_size = FC_SCQE_SZ; + + fc_cap->srqc_size = FC_SRQC_SZ; + fc_cap->srqe_size = FC_SRQE_SZ; +} + +static void cfg_get_eq_num(struct hifc_hwdev *dev) +{ + struct cfg_eq_info *eq_info = &dev->cfg_mgmt->eq_info; + + eq_info->num_ceq = dev->hwif->attr.num_ceqs; + eq_info->num_ceq_remain = eq_info->num_ceq; +} + +static int cfg_init_eq(struct hifc_hwdev *dev) +{ + struct cfg_mgmt_info *cfg_mgmt = dev->cfg_mgmt; + struct cfg_eq *eq; + u8 num_ceq, i = 0; + + cfg_get_eq_num(dev); + num_ceq = cfg_mgmt->eq_info.num_ceq; + + sdk_info(dev->dev_hdl, "Cfg mgmt: ceqs=0x%x, remain=0x%x\n", + cfg_mgmt->eq_info.num_ceq, cfg_mgmt->eq_info.num_ceq_remain); + + if (!num_ceq) { + sdk_err(dev->dev_hdl, "Ceq num cfg in fw is zero\n"); + return -EFAULT; + } + eq = kcalloc(num_ceq, sizeof(*eq), GFP_KERNEL); + if (!eq) + return -ENOMEM; + + for (i = 0; i < num_ceq; ++i) { + eq[i].eqn = i; + eq[i].free = CFG_FREE; + eq[i].type = SERVICE_T_MAX; + } + + cfg_mgmt->eq_info.eq = eq; + + mutex_init(&cfg_mgmt->eq_info.eq_mutex); + + return 0; +} + +static int cfg_init_interrupt(struct hifc_hwdev *dev) +{ + struct cfg_mgmt_info *cfg_mgmt = dev->cfg_mgmt; + struct cfg_irq_info *irq_info = &cfg_mgmt->irq_param_info; + u16 intr_num = dev->hwif->attr.num_irqs; + + if (!intr_num) { + sdk_err(dev->dev_hdl, "Irq num cfg in fw is zero\n"); + return -EFAULT; + } + irq_info->alloc_info = kcalloc(intr_num, sizeof(*irq_info->alloc_info), + GFP_KERNEL); + if (!irq_info->alloc_info) + return -ENOMEM; + + irq_info->num_irq_hw = intr_num; + + cfg_mgmt->svc_cap.interrupt_type = intr_mode; + + mutex_init(&irq_info->irq_mutex); + + return 0; +} + +static int cfg_enable_interrupt(struct hifc_hwdev *dev) +{ + struct cfg_mgmt_info *cfg_mgmt = dev->cfg_mgmt; + u16 nreq = cfg_mgmt->irq_param_info.num_irq_hw; + + void *pcidev = dev->pcidev_hdl; + struct irq_alloc_info_st *irq_info; + struct msix_entry *entry; + u16 i = 0; + int actual_irq; + + irq_info = cfg_mgmt->irq_param_info.alloc_info; + + sdk_info(dev->dev_hdl, "Interrupt type: %d, irq num: %d.\n", + cfg_mgmt->svc_cap.interrupt_type, nreq); + + switch (cfg_mgmt->svc_cap.interrupt_type) { + case INTR_TYPE_MSIX: + + if (!nreq) { + sdk_err(dev->dev_hdl, "Interrupt number cannot be zero\n"); + return -EINVAL; + } + entry = kcalloc(nreq, sizeof(*entry), GFP_KERNEL); + if (!entry) + return -ENOMEM; + + for (i = 0; i < nreq; i++) + entry[i].entry = i; + + actual_irq = pci_enable_msix_range(pcidev, entry, + VECTOR_THRESHOLD, nreq); + if (actual_irq < 0) { + sdk_err(dev->dev_hdl, "Alloc msix entries with threshold 2 failed.\n"); + kfree(entry); + return -ENOMEM; + } + + nreq = (u16)actual_irq; + cfg_mgmt->irq_param_info.num_total = nreq; + cfg_mgmt->irq_param_info.num_irq_remain = nreq; + sdk_info(dev->dev_hdl, "Request %d msix vector success.\n", + nreq); + + for (i = 0; i < nreq; ++i) { + /* u16 driver uses to specify entry, OS writes */ + irq_info[i].info.msix_entry_idx = entry[i].entry; + /* u32 kernel uses to write allocated vector */ + irq_info[i].info.irq_id = entry[i].vector; + irq_info[i].type = SERVICE_T_MAX; + irq_info[i].free = CFG_FREE; + } + + kfree(entry); + + break; + + default: + sdk_err(dev->dev_hdl, "Unsupport interrupt type %d\n", + cfg_mgmt->svc_cap.interrupt_type); + break; + } + + return 0; +} + +int hifc_alloc_irqs(void *hwdev, enum hifc_service_type type, u16 num, + struct irq_info *irq_info_array, u16 *act_num) +{ + struct hifc_hwdev *dev = hwdev; + struct cfg_mgmt_info *cfg_mgmt; + struct cfg_irq_info *irq_info; + struct irq_alloc_info_st *alloc_info; + int max_num_irq; + u16 free_num_irq; + int i, j; + + if (!hwdev || !irq_info_array || !act_num) + return -EINVAL; + + cfg_mgmt = dev->cfg_mgmt; + irq_info = &cfg_mgmt->irq_param_info; + alloc_info = irq_info->alloc_info; + max_num_irq = irq_info->num_total; + free_num_irq = irq_info->num_irq_remain; + + mutex_lock(&irq_info->irq_mutex); + + if (num > free_num_irq) { + if (free_num_irq == 0) { + sdk_err(dev->dev_hdl, + "no free irq resource in cfg mgmt.\n"); + mutex_unlock(&irq_info->irq_mutex); + return -ENOMEM; + } + + sdk_warn(dev->dev_hdl, "only %d irq resource in cfg mgmt.\n", + free_num_irq); + num = free_num_irq; + } + + *act_num = 0; + + for (i = 0; i < num; i++) { + for (j = 0; j < max_num_irq; j++) { + if (alloc_info[j].free == CFG_FREE) { + if (irq_info->num_irq_remain == 0) { + sdk_err(dev->dev_hdl, "No free irq resource in cfg mgmt\n"); + mutex_unlock(&irq_info->irq_mutex); + return -EINVAL; + } + alloc_info[j].type = type; + alloc_info[j].free = CFG_BUSY; + + irq_info_array[i].msix_entry_idx = + alloc_info[j].info.msix_entry_idx; + irq_info_array[i].irq_id = + alloc_info[j].info.irq_id; + (*act_num)++; + irq_info->num_irq_remain--; + + break; + } + } + } + + mutex_unlock(&irq_info->irq_mutex); + return 0; +} + +void hifc_free_irq(void *hwdev, enum hifc_service_type type, u32 irq_id) +{ + struct hifc_hwdev *dev = hwdev; + struct cfg_mgmt_info *cfg_mgmt; + struct cfg_irq_info *irq_info; + struct irq_alloc_info_st *alloc_info; + int max_num_irq; + int i; + + if (!hwdev) + return; + + cfg_mgmt = dev->cfg_mgmt; + irq_info = &cfg_mgmt->irq_param_info; + alloc_info = irq_info->alloc_info; + max_num_irq = irq_info->num_total; + + mutex_lock(&irq_info->irq_mutex); + + for (i = 0; i < max_num_irq; i++) { + if (irq_id == alloc_info[i].info.irq_id && + type == alloc_info[i].type) { + if (alloc_info[i].free == CFG_BUSY) { + alloc_info[i].free = CFG_FREE; + irq_info->num_irq_remain++; + if (irq_info->num_irq_remain > max_num_irq) { + sdk_err(dev->dev_hdl, "Find target,but over range\n"); + mutex_unlock(&irq_info->irq_mutex); + return; + } + break; + } + } + } + + if (i >= max_num_irq) + sdk_warn(dev->dev_hdl, "Irq %d don`t need to free\n", irq_id); + + mutex_unlock(&irq_info->irq_mutex); +} + +int init_cfg_mgmt(struct hifc_hwdev *dev) +{ + int err; + struct cfg_mgmt_info *cfg_mgmt; + + cfg_mgmt = kzalloc(sizeof(*cfg_mgmt), GFP_KERNEL); + if (!cfg_mgmt) + return -ENOMEM; + + dev->cfg_mgmt = cfg_mgmt; + cfg_mgmt->hwdev = dev; + + err = cfg_init_eq(dev); + if (err) { + sdk_err(dev->dev_hdl, "Failed to init cfg event queue, err: %d\n", + err); + goto free_mgmt_mem; + } + + err = cfg_init_interrupt(dev); + if (err) { + sdk_err(dev->dev_hdl, "Failed to init cfg interrupt, err: %d\n", + err); + goto free_eq_mem; + } + + err = cfg_enable_interrupt(dev); + if (err) { + sdk_err(dev->dev_hdl, "Failed to enable cfg interrupt, err: %d\n", + err); + goto free_interrupt_mem; + } + + return 0; + +free_interrupt_mem: + kfree(cfg_mgmt->irq_param_info.alloc_info); + + cfg_mgmt->irq_param_info.alloc_info = NULL; + +free_eq_mem: + kfree(cfg_mgmt->eq_info.eq); + + cfg_mgmt->eq_info.eq = NULL; + +free_mgmt_mem: + kfree(cfg_mgmt); + return err; +} + +void free_cfg_mgmt(struct hifc_hwdev *dev) +{ + struct cfg_mgmt_info *cfg_mgmt = dev->cfg_mgmt; + + /* if the allocated resource were recycled */ + if (cfg_mgmt->irq_param_info.num_irq_remain != + cfg_mgmt->irq_param_info.num_total || + cfg_mgmt->eq_info.num_ceq_remain != cfg_mgmt->eq_info.num_ceq) + sdk_err(dev->dev_hdl, "Can't reclaim all irq and event queue, please check\n"); + + switch (cfg_mgmt->svc_cap.interrupt_type) { + case INTR_TYPE_MSIX: + pci_disable_msix(dev->pcidev_hdl); + break; + + case INTR_TYPE_MSI: + pci_disable_msi(dev->pcidev_hdl); + break; + + case INTR_TYPE_INT: + default: + break; + } + + kfree(cfg_mgmt->irq_param_info.alloc_info); + cfg_mgmt->irq_param_info.alloc_info = NULL; + + kfree(cfg_mgmt->eq_info.eq); + cfg_mgmt->eq_info.eq = NULL; + + kfree(cfg_mgmt); +} + +int init_capability(struct hifc_hwdev *dev) +{ + int err; + enum func_type type = HIFC_FUNC_TYPE(dev); + struct cfg_mgmt_info *cfg_mgmt = dev->cfg_mgmt; + + cfg_mgmt->svc_cap.timer_en = 1; + cfg_mgmt->svc_cap.test_xid_alloc_mode = 1; + cfg_mgmt->svc_cap.test_gpa_check_enable = 1; + + err = get_cap_from_fw(dev, type); + if (err) { + sdk_err(dev->dev_hdl, "Failed to get PF/PPF capability\n"); + return err; + } + + fc_param_fix(dev); + + if (dev->cfg_mgmt->svc_cap.force_up) + dev->feature_cap |= HIFC_FUNC_FORCE_LINK_UP; + + sdk_info(dev->dev_hdl, "Init capability success\n"); + return 0; +} + +void free_capability(struct hifc_hwdev *dev) +{ + sdk_info(dev->dev_hdl, "Free capability success"); +} + +bool hifc_support_fc(void *hwdev, struct fc_service_cap *cap) +{ + struct hifc_hwdev *dev = hwdev; + + if (!hwdev) + return false; + + if (cap) + memcpy(cap, &dev->cfg_mgmt->svc_cap.fc_cap, sizeof(*cap)); + + return true; +} + +u8 hifc_host_oq_id_mask(void *hwdev) +{ + struct hifc_hwdev *dev = hwdev; + + if (!dev) { + pr_err("Hwdev pointer is NULL for getting host oq id mask\n"); + return 0; + } + return dev->cfg_mgmt->svc_cap.host_oq_id_mask_val; +} + +u16 hifc_func_max_qnum(void *hwdev) +{ + struct hifc_hwdev *dev = hwdev; + + if (!dev) { + pr_err("Hwdev pointer is NULL for getting function max queue number\n"); + return 0; + } + return dev->cfg_mgmt->svc_cap.max_sqs; +} + +/* Caller should ensure atomicity when calling this function */ +int hifc_stateful_init(void *hwdev) +{ + struct hifc_hwdev *dev = hwdev; + int err; + + if (!dev) + return -EINVAL; + + if (dev->statufull_ref_cnt++) + return 0; + + err = cqm_init(dev); + if (err) { + sdk_err(dev->dev_hdl, "Failed to init cqm, err: %d\n", err); + goto init_cqm_err; + } + + sdk_info(dev->dev_hdl, "Initialize statefull resource success\n"); + + return 0; + +init_cqm_err: + + dev->statufull_ref_cnt--; + + return err; +} + +/* Caller should ensure atomicity when calling this function */ +void hifc_stateful_deinit(void *hwdev) +{ + struct hifc_hwdev *dev = hwdev; + + if (!dev || !dev->statufull_ref_cnt) + return; + + if (--dev->statufull_ref_cnt) + return; + + cqm_uninit(hwdev); + + sdk_info(dev->dev_hdl, "Clear statefull resource success\n"); +} + +bool hifc_is_hwdev_mod_inited(void *hwdev, enum hifc_hwdev_init_state state) +{ + struct hifc_hwdev *dev = hwdev; + + if (!hwdev || state >= HIFC_HWDEV_MAX_INVAL_INITED) + return false; + + return !!test_bit(state, &dev->func_state); +} + +static int hifc_os_dep_init(struct hifc_hwdev *hwdev) +{ + hwdev->workq = create_singlethread_workqueue(HIFC_HW_WQ_NAME); + if (!hwdev->workq) { + sdk_err(hwdev->dev_hdl, "Failed to initialize hardware workqueue\n"); + return -EFAULT; + } + + sema_init(&hwdev->fault_list_sem, 1); + + return 0; +} + +static void hifc_os_dep_deinit(struct hifc_hwdev *hwdev) +{ + destroy_workqueue(hwdev->workq); +} + +static int __hilink_phy_init(struct hifc_hwdev *hwdev) +{ + int err; + + err = hifc_phy_init_status_judge(hwdev); + if (err) { + sdk_info(hwdev->dev_hdl, "Phy init failed\n"); + return err; + } + + return 0; +} + +static int init_hwdev_and_hwif(struct hifc_init_para *para) +{ + struct hifc_hwdev *hwdev; + int err; + + if (!(*para->hwdev)) { + hwdev = kzalloc(sizeof(*hwdev), GFP_KERNEL); + if (!hwdev) + return -ENOMEM; + + *para->hwdev = hwdev; + hwdev->adapter_hdl = para->adapter_hdl; + hwdev->pcidev_hdl = para->pcidev_hdl; + hwdev->dev_hdl = para->dev_hdl; + hwdev->chip_node = para->chip_node; + + hwdev->chip_fault_stats = vzalloc(HIFC_CHIP_FAULT_SIZE); + if (!hwdev->chip_fault_stats) + goto alloc_chip_fault_stats_err; + + err = hifc_init_hwif(hwdev, para->cfg_reg_base, + para->intr_reg_base, + para->db_base_phy, para->db_base, + para->dwqe_mapping); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init hwif\n"); + goto init_hwif_err; + } + } + + return 0; + +init_hwif_err: + vfree(hwdev->chip_fault_stats); + +alloc_chip_fault_stats_err: + + *para->hwdev = NULL; + + return -EFAULT; +} + +static void deinit_hwdev_and_hwif(struct hifc_hwdev *hwdev) +{ + hifc_free_hwif(hwdev); + + vfree(hwdev->chip_fault_stats); + + kfree(hwdev); +} + +static int init_hw_cfg(struct hifc_hwdev *hwdev) +{ + int err; + + err = init_capability(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init capability\n"); + return err; + } + + err = __hilink_phy_init(hwdev); + if (err) + goto hilink_phy_init_err; + + return 0; + +hilink_phy_init_err: + free_capability(hwdev); + + return err; +} + +/* Return: + * 0: all success + * >0: partitial success + * <0: all failed + */ +int hifc_init_hwdev(struct hifc_init_para *para) +{ + struct hifc_hwdev *hwdev; + int err; + + err = init_hwdev_and_hwif(para); + if (err) + return err; + + hwdev = *para->hwdev; + + /* detect slave host according to BAR reg */ + hwdev->feature_cap = HIFC_FUNC_MGMT | HIFC_FUNC_PORT | + HIFC_FUNC_SUPP_RATE_LIMIT | HIFC_FUNC_SUPP_DFX_REG | + HIFC_FUNC_SUPP_RX_MODE | HIFC_FUNC_SUPP_SET_VF_MAC_VLAN | + HIFC_FUNC_SUPP_CHANGE_MAC; + + err = hifc_os_dep_init(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init os dependent\n"); + goto os_dep_init_err; + } + + hifc_set_chip_present(hwdev); + hifc_init_heartbeat(hwdev); + + err = init_cfg_mgmt(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init config mgmt\n"); + goto init_cfg_mgmt_err; + } + + err = hifc_init_comm_ch(hwdev); + if (err) { + if (!(hwdev->func_state & HIFC_HWDEV_INIT_MODES_MASK)) { + sdk_err(hwdev->dev_hdl, "Failed to init communication channel\n"); + goto init_comm_ch_err; + } else { + sdk_err(hwdev->dev_hdl, "Init communication channel partitail failed\n"); + return hwdev->func_state & HIFC_HWDEV_INIT_MODES_MASK; + } + } + + err = init_hw_cfg(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init hardware config\n"); + goto init_hw_cfg_err; + } + + set_bit(HIFC_HWDEV_ALL_INITED, &hwdev->func_state); + + sdk_info(hwdev->dev_hdl, "Init hwdev success\n"); + + return 0; + +init_hw_cfg_err: + return (hwdev->func_state & HIFC_HWDEV_INIT_MODES_MASK); + +init_comm_ch_err: + free_cfg_mgmt(hwdev); + +init_cfg_mgmt_err: + hifc_destroy_heartbeat(hwdev); + hifc_os_dep_deinit(hwdev); + +os_dep_init_err: + deinit_hwdev_and_hwif(hwdev); + *para->hwdev = NULL; + + return -EFAULT; +} + +void hifc_free_hwdev(void *hwdev) +{ + struct hifc_hwdev *dev = hwdev; + enum hifc_hwdev_init_state state = HIFC_HWDEV_ALL_INITED; + int flag = 0; + + if (!hwdev) + return; + + if (test_bit(HIFC_HWDEV_ALL_INITED, &dev->func_state)) { + clear_bit(HIFC_HWDEV_ALL_INITED, &dev->func_state); + + /* BM slave function not need to exec rx_tx_flush */ + + hifc_func_rx_tx_flush(hwdev); + + free_capability(dev); + } + while (state > HIFC_HWDEV_NONE_INITED) { + if (test_bit(state, &dev->func_state)) { + flag = 1; + break; + } + state--; + } + if (flag) { + hifc_uninit_comm_ch(dev); + free_cfg_mgmt(dev); + hifc_destroy_heartbeat(dev); + hifc_os_dep_deinit(dev); + } + clear_bit(HIFC_HWDEV_NONE_INITED, &dev->func_state); + + deinit_hwdev_and_hwif(dev); +} + +u64 hifc_get_func_feature_cap(void *hwdev) +{ + struct hifc_hwdev *dev = hwdev; + + if (!dev) { + pr_err("Hwdev pointer is NULL for getting function feature capability\n"); + return 0; + } + + return dev->feature_cap; +} + diff --git a/drivers/scsi/huawei/hifc/hifc_cfg.h b/drivers/scsi/huawei/hifc/hifc_cfg.h new file mode 100644 index 000000000000..b8a9dd35b1fd --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_cfg.h @@ -0,0 +1,171 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#ifndef __CFG_MGT_H__ +#define __CFG_MGT_H__ + +enum { + CFG_FREE = 0, + CFG_BUSY = 1 +}; + +/* FC */ +#define FC_PCTX_SZ 256 +#define FC_CCTX_SZ 256 +#define FC_SQE_SZ 128 +#define FC_SCQC_SZ 64 +#define FC_SCQE_SZ 64 +#define FC_SRQC_SZ 64 +#define FC_SRQE_SZ 32 + +/* device capability */ +struct service_cap { + /* Host global resources */ + u16 host_total_function; + u8 host_oq_id_mask_val; + + /* DO NOT get interrupt_type from firmware */ + enum intr_type interrupt_type; + u8 intr_chip_en; + + u8 port_id; /* PF/VF's physical port */ + u8 force_up; + + u8 timer_en; /* 0:disable, 1:enable */ + + u16 max_sqs; + u16 max_rqs; + + /* For test */ + bool test_xid_alloc_mode; + bool test_gpa_check_enable; + + u32 max_connect_num; /* PF/VF maximum connection number(1M) */ + /* The maximum connections which can be stick to cache memory, max 1K */ + u16 max_stick2cache_num; + + struct nic_service_cap nic_cap; /* NIC capability */ + struct fc_service_cap fc_cap; /* FC capability */ +}; + +struct hifc_sync_time_info { + u8 status; + u8 version; + u8 rsvd0[6]; + + u64 mstime; +}; + +struct cfg_eq { + enum hifc_service_type type; + int eqn; + int free; /* 1 - alocated, 0- freed */ +}; + +struct cfg_eq_info { + struct cfg_eq *eq; + u8 num_ceq; + u8 num_ceq_remain; + /* mutex used for allocate EQs */ + struct mutex eq_mutex; +}; + +struct irq_alloc_info_st { + enum hifc_service_type type; + int free; /* 1 - alocated, 0- freed */ + struct irq_info info; +}; + +struct cfg_irq_info { + struct irq_alloc_info_st *alloc_info; + u16 num_total; + u16 num_irq_remain; + u16 num_irq_hw; /* device max irq number */ + + /* mutex used for allocate EQs */ + struct mutex irq_mutex; +}; + +#define VECTOR_THRESHOLD 2 + +struct cfg_mgmt_info { + struct hifc_hwdev *hwdev; + struct service_cap svc_cap; + struct cfg_eq_info eq_info; /* EQ */ + struct cfg_irq_info irq_param_info; /* IRQ */ + u32 func_seq_num; /* temporary */ +}; + +enum cfg_sub_cmd { + /* PPF(PF) <-> FW */ + HIFC_CFG_NIC_CAP = 0, + CFG_FW_VERSION, + CFG_UCODE_VERSION, + HIFC_CFG_FUNC_CAP, + HIFC_CFG_MBOX_CAP = 6, +}; + +struct hifc_dev_cap { + u8 status; + u8 version; + u8 rsvd0[6]; + + /* Public resource */ + u8 sf_svc_attr; + u8 host_id; + u8 sf_en_pf; + u8 sf_en_vf; + + u8 ep_id; + u8 intr_type; + u8 max_cos_id; + u8 er_id; + u8 port_id; + u8 max_vf; + u16 svc_cap_en; + u16 host_total_func; + u8 host_oq_id_mask_val; + u8 max_vf_cos_id; + + u32 max_conn_num; + u16 max_stick2cache_num; + u16 max_bfilter_start_addr; + u16 bfilter_len; + u16 hash_bucket_num; + u8 cfg_file_ver; + u8 net_port_mode; + u8 valid_cos_bitmap; /* every bit indicate cos is valid */ + u8 force_up; + u32 pf_num; + u32 pf_id_start; + u32 vf_num; + u32 vf_id_start; + + /* shared resource */ + u32 host_pctx_num; + u8 host_sf_en; + u8 rsvd2[3]; + u32 host_ccxt_num; + u32 host_scq_num; + u32 host_srq_num; + u32 host_mpt_num; + /* l2nic */ + u16 nic_max_sq; + u16 nic_max_rq; + u32 rsvd[46]; + /* FC */ + u32 fc_max_pctx; + u32 fc_max_scq; + u32 fc_max_srq; + + u32 fc_max_cctx; + u32 fc_cctx_id_start; + + u8 fc_vp_id_start; + u8 fc_vp_id_end; + u16 func_id; +}; +#endif diff --git a/drivers/scsi/huawei/hifc/hifc_cmdq.c b/drivers/scsi/huawei/hifc/hifc_cmdq.c new file mode 100644 index 000000000000..03531017c412 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_cmdq.c @@ -0,0 +1,1507 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": [COMM]" fmt + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/device.h> +#include <linux/pci.h> +#include <linux/errno.h> +#include <linux/completion.h> +#include <linux/interrupt.h> +#include <linux/io.h> +#include <linux/spinlock.h> +#include <linux/slab.h> +#include <linux/module.h> + +#include "hifc_knl_adp.h" +#include "hifc_hw.h" +#include "hifc_hwdev.h" +#include "hifc_hwif.h" +#include "hifc_wq.h" +#include "hifc_api_cmd.h" +#include "hifc_mgmt.h" +#include "hifc_eqs.h" +#include "hifc_cmdq.h" + +#define CMDQ_CMD_TIMEOUT 1000 /* millisecond */ +#define UPPER_8_BITS(data) (((data) >> 8) & 0xFF) +#define LOWER_8_BITS(data) ((data) & 0xFF) + +#define CMDQ_DB_INFO_HI_PROD_IDX_SHIFT 0 +#define CMDQ_DB_INFO_QUEUE_TYPE_SHIFT 23 +#define CMDQ_DB_INFO_CMDQ_TYPE_SHIFT 24 +#define CMDQ_DB_INFO_SRC_TYPE_SHIFT 27 +#define CMDQ_DB_INFO_HI_PROD_IDX_MASK 0xFFU +#define CMDQ_DB_INFO_QUEUE_TYPE_MASK 0x1U +#define CMDQ_DB_INFO_CMDQ_TYPE_MASK 0x7U +#define CMDQ_DB_INFO_SRC_TYPE_MASK 0x1FU + +#define CMDQ_DB_INFO_SET(val, member) \ + (((val) & CMDQ_DB_INFO_##member##_MASK) << \ + CMDQ_DB_INFO_##member##_SHIFT) + +#define CMDQ_CTRL_PI_SHIFT 0 +#define CMDQ_CTRL_CMD_SHIFT 16 +#define CMDQ_CTRL_MOD_SHIFT 24 +#define CMDQ_CTRL_ACK_TYPE_SHIFT 29 +#define CMDQ_CTRL_HW_BUSY_BIT_SHIFT 31 +#define CMDQ_CTRL_PI_MASK 0xFFFFU +#define CMDQ_CTRL_CMD_MASK 0xFFU +#define CMDQ_CTRL_MOD_MASK 0x1FU +#define CMDQ_CTRL_ACK_TYPE_MASK 0x3U +#define CMDQ_CTRL_HW_BUSY_BIT_MASK 0x1U + +#define CMDQ_CTRL_SET(val, member) \ + (((val) & CMDQ_CTRL_##member##_MASK) \ + << CMDQ_CTRL_##member##_SHIFT) + +#define CMDQ_CTRL_GET(val, member) \ + (((val) >> CMDQ_CTRL_##member##_SHIFT) \ + & CMDQ_CTRL_##member##_MASK) + +#define CMDQ_WQE_HEADER_BUFDESC_LEN_SHIFT 0 +#define CMDQ_WQE_HEADER_COMPLETE_FMT_SHIFT 15 +#define CMDQ_WQE_HEADER_DATA_FMT_SHIFT 22 +#define CMDQ_WQE_HEADER_COMPLETE_REQ_SHIFT 23 +#define CMDQ_WQE_HEADER_COMPLETE_SECT_LEN_SHIFT 27 +#define CMDQ_WQE_HEADER_CTRL_LEN_SHIFT 29 +#define CMDQ_WQE_HEADER_HW_BUSY_BIT_SHIFT 31 + +#define CMDQ_WQE_HEADER_BUFDESC_LEN_MASK 0xFFU +#define CMDQ_WQE_HEADER_COMPLETE_FMT_MASK 0x1U +#define CMDQ_WQE_HEADER_DATA_FMT_MASK 0x1U +#define CMDQ_WQE_HEADER_COMPLETE_REQ_MASK 0x1U +#define CMDQ_WQE_HEADER_COMPLETE_SECT_LEN_MASK 0x3U +#define CMDQ_WQE_HEADER_CTRL_LEN_MASK 0x3U +#define CMDQ_WQE_HEADER_HW_BUSY_BIT_MASK 0x1U + +#define CMDQ_WQE_HEADER_SET(val, member) \ + (((val) & CMDQ_WQE_HEADER_##member##_MASK) \ + << CMDQ_WQE_HEADER_##member##_SHIFT) + +#define CMDQ_WQE_HEADER_GET(val, member) \ + (((val) >> CMDQ_WQE_HEADER_##member##_SHIFT) \ + & CMDQ_WQE_HEADER_##member##_MASK) + +#define CMDQ_CTXT_CURR_WQE_PAGE_PFN_SHIFT 0 +#define CMDQ_CTXT_EQ_ID_SHIFT 56 +#define CMDQ_CTXT_CEQ_ARM_SHIFT 61 +#define CMDQ_CTXT_CEQ_EN_SHIFT 62 +#define CMDQ_CTXT_HW_BUSY_BIT_SHIFT 63 +#define CMDQ_CTXT_CURR_WQE_PAGE_PFN_MASK 0xFFFFFFFFFFFFF +#define CMDQ_CTXT_EQ_ID_MASK 0x1F +#define CMDQ_CTXT_CEQ_ARM_MASK 0x1 +#define CMDQ_CTXT_CEQ_EN_MASK 0x1 +#define CMDQ_CTXT_HW_BUSY_BIT_MASK 0x1 + +#define CMDQ_CTXT_PAGE_INFO_SET(val, member) \ + (((u64)(val) & CMDQ_CTXT_##member##_MASK) \ + << CMDQ_CTXT_##member##_SHIFT) + +#define CMDQ_CTXT_PAGE_INFO_GET(val, member) \ + (((u64)(val) >> CMDQ_CTXT_##member##_SHIFT) \ + & CMDQ_CTXT_##member##_MASK) + +#define CMDQ_CTXT_WQ_BLOCK_PFN_SHIFT 0 +#define CMDQ_CTXT_CI_SHIFT 52 +#define CMDQ_CTXT_WQ_BLOCK_PFN_MASK 0xFFFFFFFFFFFFF +#define CMDQ_CTXT_CI_MASK 0xFFF + +#define CMDQ_CTXT_BLOCK_INFO_SET(val, member) \ + (((u64)(val) & CMDQ_CTXT_##member##_MASK) \ + << CMDQ_CTXT_##member##_SHIFT) + +#define CMDQ_CTXT_BLOCK_INFO_GET(val, member) \ + (((u64)(val) >> CMDQ_CTXT_##member##_SHIFT) \ + & CMDQ_CTXT_##member##_MASK) + +#define SAVED_DATA_ARM_SHIFT 31 +#define SAVED_DATA_ARM_MASK 0x1U + +#define SAVED_DATA_SET(val, member) \ + (((val) & SAVED_DATA_##member##_MASK) \ + << SAVED_DATA_##member##_SHIFT) + +#define SAVED_DATA_CLEAR(val, member) \ + ((val) & (~(SAVED_DATA_##member##_MASK \ + << SAVED_DATA_##member##_SHIFT))) + +#define WQE_ERRCODE_VAL_SHIFT 20 +#define WQE_ERRCODE_VAL_MASK 0xF + +#define WQE_ERRCODE_GET(val, member) \ + (((val) >> WQE_ERRCODE_##member##_SHIFT) & \ + WQE_ERRCODE_##member##_MASK) + +#define CEQE_CMDQ_TYPE_SHIFT 0 +#define CEQE_CMDQ_TYPE_MASK 0x7 + +#define CEQE_CMDQ_GET(val, member) \ + (((val) >> CEQE_CMDQ_##member##_SHIFT) & CEQE_CMDQ_##member##_MASK) + +#define WQE_COMPLETED(ctrl_info) CMDQ_CTRL_GET(ctrl_info, HW_BUSY_BIT) + +#define WQE_HEADER(wqe) ((struct hifc_cmdq_header *)(wqe)) + +#define CMDQ_DB_PI_OFF(pi) (((u16)LOWER_8_BITS(pi)) << 3) + +#define CMDQ_DB_ADDR(db_base, pi) \ + (((u8 *)(db_base) + HIFC_DB_OFF) + CMDQ_DB_PI_OFF(pi)) + +#define CMDQ_PFN_SHIFT 12 +#define CMDQ_PFN(addr) ((addr) >> CMDQ_PFN_SHIFT) + +#define FIRST_DATA_TO_WRITE_LAST sizeof(u64) +#define WQE_LCMD_SIZE 64 +#define WQE_SCMD_SIZE 64 +#define COMPLETE_LEN 3 +#define CMDQ_WQEBB_SIZE 64 +#define CMDQ_WQE_SIZE 64 +#define CMDQ_WQ_PAGE_SIZE 4096 + +#define WQE_NUM_WQEBBS(wqe_size, wq) \ + ((u16)(ALIGN((u32)(wqe_size), (wq)->wqebb_size) / (wq)->wqebb_size)) + +#define cmdq_to_cmdqs(cmdq) container_of((cmdq) - (cmdq)->cmdq_type, \ + struct hifc_cmdqs, cmdq[0]) + +#define CMDQ_SEND_CMPT_CODE 10 +#define CMDQ_COMPLETE_CMPT_CODE 11 + +#define HIFC_GET_CMDQ_FREE_WQEBBS(cmdq_wq) \ + atomic_read(&(cmdq_wq)->delta) + +enum cmdq_scmd_type { + CMDQ_SET_ARM_CMD = 2, +}; + +enum cmdq_wqe_type { + WQE_LCMD_TYPE, + WQE_SCMD_TYPE, +}; + +enum ctrl_sect_len { + CTRL_SECT_LEN = 1, + CTRL_DIRECT_SECT_LEN = 2, +}; + +enum bufdesc_len { + BUFDESC_LCMD_LEN = 2, + BUFDESC_SCMD_LEN = 3, +}; + +enum data_format { + DATA_SGE, + DATA_DIRECT, +}; + +enum completion_format { + COMPLETE_DIRECT, + COMPLETE_SGE, +}; + +enum completion_request { + CEQ_SET = 1, +}; + +enum cmdq_cmd_type { + SYNC_CMD_DIRECT_RESP, + SYNC_CMD_SGE_RESP, + ASYNC_CMD, +}; + +bool hifc_cmdq_idle(struct hifc_cmdq *cmdq) +{ + struct hifc_wq *wq = cmdq->wq; + + return (atomic_read(&wq->delta) == wq->q_depth ? true : false); +} + +struct hifc_cmd_buf *hifc_alloc_cmd_buf(void *hwdev) +{ + struct hifc_cmdqs *cmdqs; + struct hifc_cmd_buf *cmd_buf; + void *dev; + + if (!hwdev) { + pr_err("Failed to alloc cmd buf, Invalid hwdev\n"); + return NULL; + } + + cmdqs = ((struct hifc_hwdev *)hwdev)->cmdqs; + dev = ((struct hifc_hwdev *)hwdev)->dev_hdl; + + cmd_buf = kzalloc(sizeof(*cmd_buf), GFP_ATOMIC); + if (!cmd_buf) + return NULL; + + cmd_buf->buf = pci_pool_alloc(cmdqs->cmd_buf_pool, GFP_ATOMIC, + &cmd_buf->dma_addr); + if (!cmd_buf->buf) { + sdk_err(dev, "Failed to allocate cmdq cmd buf from the pool\n"); + goto alloc_pci_buf_err; + } + + return cmd_buf; + +alloc_pci_buf_err: + kfree(cmd_buf); + return NULL; +} + +void hifc_free_cmd_buf(void *hwdev, struct hifc_cmd_buf *cmd_buf) +{ + struct hifc_cmdqs *cmdqs; + + if (!hwdev || !cmd_buf) { + pr_err("Failed to free cmd buf\n"); + return; + } + + cmdqs = ((struct hifc_hwdev *)hwdev)->cmdqs; + + pci_pool_free(cmdqs->cmd_buf_pool, cmd_buf->buf, cmd_buf->dma_addr); + kfree(cmd_buf); +} + +static int cmdq_wqe_size(enum cmdq_wqe_type wqe_type) +{ + int wqe_size = 0; + + switch (wqe_type) { + case WQE_LCMD_TYPE: + wqe_size = WQE_LCMD_SIZE; + break; + case WQE_SCMD_TYPE: + wqe_size = WQE_SCMD_SIZE; + break; + } + + return wqe_size; +} + +static int cmdq_get_wqe_size(enum bufdesc_len len) +{ + int wqe_size = 0; + + switch (len) { + case BUFDESC_LCMD_LEN: + wqe_size = WQE_LCMD_SIZE; + break; + case BUFDESC_SCMD_LEN: + wqe_size = WQE_SCMD_SIZE; + break; + } + + return wqe_size; +} + +static void cmdq_set_completion(struct hifc_cmdq_completion *complete, + struct hifc_cmd_buf *buf_out) +{ + struct hifc_sge_resp *sge_resp = &complete->sge_resp; + + hifc_set_sge(&sge_resp->sge, buf_out->dma_addr, + HIFC_CMDQ_BUF_SIZE); +} + +static void cmdq_set_lcmd_bufdesc(struct hifc_cmdq_wqe_lcmd *wqe, + struct hifc_cmd_buf *buf_in) +{ + hifc_set_sge(&wqe->buf_desc.sge, buf_in->dma_addr, buf_in->size); +} + +static void cmdq_set_inline_wqe_data(struct hifc_cmdq_inline_wqe *wqe, + const void *buf_in, u32 in_size) +{ + struct hifc_cmdq_wqe_scmd *wqe_scmd = &wqe->wqe_scmd; + + wqe_scmd->buf_desc.buf_len = in_size; + memcpy(wqe_scmd->buf_desc.data, buf_in, in_size); +} + +static void cmdq_fill_db(struct hifc_cmdq_db *db, + enum hifc_cmdq_type cmdq_type, u16 prod_idx) +{ + db->db_info = CMDQ_DB_INFO_SET(UPPER_8_BITS(prod_idx), HI_PROD_IDX) | + CMDQ_DB_INFO_SET(HIFC_DB_CMDQ_TYPE, QUEUE_TYPE) | + CMDQ_DB_INFO_SET(cmdq_type, CMDQ_TYPE) | + CMDQ_DB_INFO_SET(HIFC_DB_SRC_CMDQ_TYPE, SRC_TYPE); +} + +static void cmdq_set_db(struct hifc_cmdq *cmdq, + enum hifc_cmdq_type cmdq_type, u16 prod_idx) +{ + struct hifc_cmdq_db db; + + cmdq_fill_db(&db, cmdq_type, prod_idx); + + /* The data that is written to HW should be in Big Endian Format */ + db.db_info = cpu_to_be32(db.db_info); + + wmb(); /* write all before the doorbell */ + writel(db.db_info, CMDQ_DB_ADDR(cmdq->db_base, prod_idx)); +} + +static void cmdq_wqe_fill(void *dst, const void *src) +{ + memcpy((u8 *)dst + FIRST_DATA_TO_WRITE_LAST, + (u8 *)src + FIRST_DATA_TO_WRITE_LAST, + CMDQ_WQE_SIZE - FIRST_DATA_TO_WRITE_LAST); + + wmb(); /* The first 8 bytes should be written last */ + + *(u64 *)dst = *(u64 *)src; +} + +static void cmdq_prepare_wqe_ctrl(struct hifc_cmdq_wqe *wqe, int wrapped, + enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, u16 prod_idx, + enum completion_format complete_format, + enum data_format data_format, + enum bufdesc_len buf_len) +{ + struct hifc_ctrl *ctrl; + enum ctrl_sect_len ctrl_len; + struct hifc_cmdq_wqe_lcmd *wqe_lcmd; + struct hifc_cmdq_wqe_scmd *wqe_scmd; + u32 saved_data = WQE_HEADER(wqe)->saved_data; + + if (data_format == DATA_SGE) { + wqe_lcmd = &wqe->wqe_lcmd; + + wqe_lcmd->status.status_info = 0; + ctrl = &wqe_lcmd->ctrl; + ctrl_len = CTRL_SECT_LEN; + } else { + wqe_scmd = &wqe->inline_wqe.wqe_scmd; + + wqe_scmd->status.status_info = 0; + ctrl = &wqe_scmd->ctrl; + ctrl_len = CTRL_DIRECT_SECT_LEN; + } + + ctrl->ctrl_info = CMDQ_CTRL_SET(prod_idx, PI) | + CMDQ_CTRL_SET(cmd, CMD) | + CMDQ_CTRL_SET(mod, MOD) | + CMDQ_CTRL_SET(ack_type, ACK_TYPE); + + WQE_HEADER(wqe)->header_info = + CMDQ_WQE_HEADER_SET(buf_len, BUFDESC_LEN) | + CMDQ_WQE_HEADER_SET(complete_format, COMPLETE_FMT) | + CMDQ_WQE_HEADER_SET(data_format, DATA_FMT) | + CMDQ_WQE_HEADER_SET(CEQ_SET, COMPLETE_REQ) | + CMDQ_WQE_HEADER_SET(COMPLETE_LEN, COMPLETE_SECT_LEN) | + CMDQ_WQE_HEADER_SET(ctrl_len, CTRL_LEN) | + CMDQ_WQE_HEADER_SET((u32)wrapped, HW_BUSY_BIT); + + if (cmd == CMDQ_SET_ARM_CMD && mod == HIFC_MOD_COMM) { + saved_data &= SAVED_DATA_CLEAR(saved_data, ARM); + WQE_HEADER(wqe)->saved_data = saved_data | + SAVED_DATA_SET(1, ARM); + } else { + saved_data &= SAVED_DATA_CLEAR(saved_data, ARM); + WQE_HEADER(wqe)->saved_data = saved_data; + } +} + +static void cmdq_set_lcmd_wqe(struct hifc_cmdq_wqe *wqe, + enum cmdq_cmd_type cmd_type, + struct hifc_cmd_buf *buf_in, + struct hifc_cmd_buf *buf_out, int wrapped, + enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, u16 prod_idx) +{ + struct hifc_cmdq_wqe_lcmd *wqe_lcmd = &wqe->wqe_lcmd; + enum completion_format complete_format = COMPLETE_DIRECT; + + switch (cmd_type) { + case SYNC_CMD_SGE_RESP: + if (buf_out) { + complete_format = COMPLETE_SGE; + cmdq_set_completion(&wqe_lcmd->completion, buf_out); + } + break; + case SYNC_CMD_DIRECT_RESP: + complete_format = COMPLETE_DIRECT; + wqe_lcmd->completion.direct_resp = 0; + break; + case ASYNC_CMD: + complete_format = COMPLETE_DIRECT; + wqe_lcmd->completion.direct_resp = 0; + + wqe_lcmd->buf_desc.saved_async_buf = (u64)(buf_in); + break; + } + + cmdq_prepare_wqe_ctrl(wqe, wrapped, ack_type, mod, cmd, + prod_idx, complete_format, DATA_SGE, + BUFDESC_LCMD_LEN); + + cmdq_set_lcmd_bufdesc(wqe_lcmd, buf_in); +} + +static void cmdq_set_inline_wqe(struct hifc_cmdq_wqe *wqe, + enum cmdq_cmd_type cmd_type, + void *buf_in, u16 in_size, + struct hifc_cmd_buf *buf_out, int wrapped, + enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, u16 prod_idx) +{ + struct hifc_cmdq_wqe_scmd *wqe_scmd = &wqe->inline_wqe.wqe_scmd; + enum completion_format complete_format = COMPLETE_DIRECT; + + switch (cmd_type) { + case SYNC_CMD_SGE_RESP: + complete_format = COMPLETE_SGE; + cmdq_set_completion(&wqe_scmd->completion, buf_out); + break; + case SYNC_CMD_DIRECT_RESP: + complete_format = COMPLETE_DIRECT; + wqe_scmd->completion.direct_resp = 0; + break; + default: + break; + } + + cmdq_prepare_wqe_ctrl(wqe, wrapped, ack_type, mod, cmd, prod_idx, + complete_format, DATA_DIRECT, BUFDESC_SCMD_LEN); + + cmdq_set_inline_wqe_data(&wqe->inline_wqe, buf_in, in_size); +} + +static void cmdq_update_cmd_status(struct hifc_cmdq *cmdq, u16 prod_idx, + struct hifc_cmdq_wqe *wqe) +{ + struct hifc_cmdq_cmd_info *cmd_info; + struct hifc_cmdq_wqe_lcmd *wqe_lcmd; + u32 status_info; + + wqe_lcmd = &wqe->wqe_lcmd; + cmd_info = &cmdq->cmd_infos[prod_idx]; + + if (cmd_info->errcode) { + status_info = be32_to_cpu(wqe_lcmd->status.status_info); + *cmd_info->errcode = WQE_ERRCODE_GET(status_info, VAL); + } + + if (cmd_info->direct_resp && + cmd_info->cmd_type == HIFC_CMD_TYPE_DIRECT_RESP) + *cmd_info->direct_resp = + cpu_to_be64(wqe_lcmd->completion.direct_resp); +} + +static int hifc_cmdq_sync_timeout_check(struct hifc_cmdq *cmdq, + struct hifc_cmdq_wqe *wqe, u16 pi, + enum hifc_mod_type mod, u8 cmd) +{ + struct hifc_cmdq_wqe_lcmd *wqe_lcmd; + struct hifc_ctrl *ctrl; + u32 ctrl_info; + + wqe_lcmd = &wqe->wqe_lcmd; + ctrl = &wqe_lcmd->ctrl; + ctrl_info = be32_to_cpu((ctrl)->ctrl_info); + if (!WQE_COMPLETED(ctrl_info)) { + sdk_info(cmdq->hwdev->dev_hdl, "Cmdq sync command check busy bit not set, mod: %u, cmd: 0x%x\n", + mod, cmd); + return -EFAULT; + } + + cmdq_update_cmd_status(cmdq, pi, wqe); + + sdk_info(cmdq->hwdev->dev_hdl, "Cmdq sync command check succeed, mod: %u, cmd: 0x%x\n", + mod, cmd); + return 0; +} + +static void __clear_cmd_info(struct hifc_cmdq_cmd_info *cmd_info, + struct hifc_cmdq_cmd_info *saved_cmd_info) +{ + if (cmd_info->errcode == saved_cmd_info->errcode) + cmd_info->errcode = NULL; + + if (cmd_info->done == saved_cmd_info->done) + cmd_info->done = NULL; + + if (cmd_info->direct_resp == saved_cmd_info->direct_resp) + cmd_info->direct_resp = NULL; +} + +static int +cmdq_sync_cmd_timeout_handler(struct hifc_cmdq *cmdq, + struct hifc_cmdq_cmd_info *cmd_info, + struct hifc_cmdq_cmd_info *saved_cmd_info, + struct hifc_cmdq_wqe *curr_wqe, + enum hifc_mod_type mod, u8 cmd, + u16 curr_prod_idx, u64 curr_msg_id) +{ + int err; + + spin_lock_bh(&cmdq->cmdq_lock); + + if (cmd_info->cmpt_code == saved_cmd_info->cmpt_code) + cmd_info->cmpt_code = NULL; + + if (*saved_cmd_info->cmpt_code == CMDQ_COMPLETE_CMPT_CODE) { + sdk_info(cmdq->hwdev->dev_hdl, "Cmdq sync command (mod: %u, cmd: 0x%x)has been completed\n", + mod, cmd); + spin_unlock_bh(&cmdq->cmdq_lock); + return 0; + } + + if (curr_msg_id == cmd_info->cmdq_msg_id) { + err = hifc_cmdq_sync_timeout_check(cmdq, curr_wqe, + curr_prod_idx, + mod, cmd); + if (err) + cmd_info->cmd_type = HIFC_CMD_TYPE_TIMEOUT; + else + cmd_info->cmd_type = HIFC_CMD_TYPE_FAKE_TIMEOUT; + } else { + err = -ETIMEDOUT; + sdk_err(cmdq->hwdev->dev_hdl, + "Cmdq sync command current msg id dismatch with cmd_info msg id, mod: %u, cmd: 0x%x\n", + mod, cmd); + } + + __clear_cmd_info(cmd_info, saved_cmd_info); + + spin_unlock_bh(&cmdq->cmdq_lock); + + return err; +} + +static int cmdq_sync_cmd_direct_resp(struct hifc_cmdq *cmdq, + enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, + struct hifc_cmd_buf *buf_in, + u64 *out_param, u32 timeout) +{ + struct hifc_wq *wq = cmdq->wq; + struct hifc_cmdq_wqe *curr_wqe, wqe; + struct hifc_cmdq_cmd_info *cmd_info, saved_cmd_info; + struct completion done; + u16 curr_prod_idx, next_prod_idx, num_wqebbs; + int wrapped, errcode = 0, wqe_size = cmdq_wqe_size(WQE_LCMD_TYPE); + int cmpt_code = CMDQ_SEND_CMPT_CODE; + ulong timeo; + u64 curr_msg_id; + int err; + + num_wqebbs = WQE_NUM_WQEBBS(wqe_size, wq); + + /* Keep wrapped and doorbell index correct. bh - for tasklet(ceq) */ + spin_lock_bh(&cmdq->cmdq_lock); + + /* in order to save a wqebb for setting arm_bit when + * send cmdq commands frequently resulting in cmdq full + */ + if (HIFC_GET_CMDQ_FREE_WQEBBS(wq) < num_wqebbs + 1) { + spin_unlock_bh(&cmdq->cmdq_lock); + return -EBUSY; + } + + /* WQE_SIZE = WQEBB_SIZE, we will get the wq element and not shadow */ + curr_wqe = hifc_get_wqe(cmdq->wq, num_wqebbs, &curr_prod_idx); + if (!curr_wqe) { + spin_unlock_bh(&cmdq->cmdq_lock); + sdk_err(cmdq->hwdev->dev_hdl, "Can not get avalible wqebb, mod: %u, cmd: 0x%x\n", + mod, cmd); + return -EBUSY; + } + + memset(&wqe, 0, sizeof(wqe)); + + wrapped = cmdq->wrapped; + + next_prod_idx = curr_prod_idx + num_wqebbs; + if (next_prod_idx >= wq->q_depth) { + cmdq->wrapped = !cmdq->wrapped; + next_prod_idx -= wq->q_depth; + } + + cmd_info = &cmdq->cmd_infos[curr_prod_idx]; + + init_completion(&done); + + cmd_info->done = &done; + cmd_info->errcode = &errcode; + cmd_info->direct_resp = out_param; + cmd_info->cmpt_code = &cmpt_code; + + memcpy(&saved_cmd_info, cmd_info, sizeof(*cmd_info)); + + cmdq_set_lcmd_wqe(&wqe, SYNC_CMD_DIRECT_RESP, buf_in, NULL, + wrapped, ack_type, mod, cmd, curr_prod_idx); + + /* The data that is written to HW should be in Big Endian Format */ + hifc_cpu_to_be32(&wqe, wqe_size); + + /* CMDQ WQE is not shadow, therefore wqe will be written to wq */ + cmdq_wqe_fill(curr_wqe, &wqe); + + cmd_info->cmd_type = HIFC_CMD_TYPE_DIRECT_RESP; + + (cmd_info->cmdq_msg_id)++; + curr_msg_id = cmd_info->cmdq_msg_id; + + cmdq_set_db(cmdq, HIFC_CMDQ_SYNC, next_prod_idx); + + spin_unlock_bh(&cmdq->cmdq_lock); + + timeo = msecs_to_jiffies(timeout ? timeout : CMDQ_CMD_TIMEOUT); + if (!wait_for_completion_timeout(&done, timeo)) { + err = cmdq_sync_cmd_timeout_handler(cmdq, cmd_info, + &saved_cmd_info, + curr_wqe, mod, cmd, + curr_prod_idx, curr_msg_id); + + if (!err) + goto timeout_check_ok; + + sdk_err(cmdq->hwdev->dev_hdl, "Cmdq sync command timeout, prod idx: 0x%x\n", + curr_prod_idx); + return -ETIMEDOUT; + } + +timeout_check_ok: + smp_rmb(); /* read error code after completion */ + + if (errcode > 1) + return errcode; + + return 0; +} + +static int cmdq_sync_cmd_detail_resp(struct hifc_cmdq *cmdq, + enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, + struct hifc_cmd_buf *buf_in, + struct hifc_cmd_buf *buf_out, + u32 timeout) +{ + struct hifc_wq *wq = cmdq->wq; + struct hifc_cmdq_wqe *curr_wqe, wqe; + struct hifc_cmdq_cmd_info *cmd_info, saved_cmd_info; + struct completion done; + u16 curr_prod_idx, next_prod_idx, num_wqebbs; + int wrapped, errcode = 0, wqe_size = cmdq_wqe_size(WQE_LCMD_TYPE); + int cmpt_code = CMDQ_SEND_CMPT_CODE; + ulong timeo; + u64 curr_msg_id; + int err; + + num_wqebbs = WQE_NUM_WQEBBS(wqe_size, wq); + + /* Keep wrapped and doorbell index correct. bh - for tasklet(ceq) */ + spin_lock_bh(&cmdq->cmdq_lock); + + /* in order to save a wqebb for setting arm_bit when + * send cmdq commands frequently resulting in cmdq full + */ + if (HIFC_GET_CMDQ_FREE_WQEBBS(wq) < num_wqebbs + 1) { + spin_unlock_bh(&cmdq->cmdq_lock); + return -EBUSY; + } + + /* WQE_SIZE = WQEBB_SIZE, we will get the wq element and not shadow*/ + curr_wqe = hifc_get_wqe(cmdq->wq, num_wqebbs, &curr_prod_idx); + if (!curr_wqe) { + spin_unlock_bh(&cmdq->cmdq_lock); + sdk_err(cmdq->hwdev->dev_hdl, "Can not get avalible wqebb, mod: %u, cmd: 0x%x\n", + mod, cmd); + return -EBUSY; + } + + memset(&wqe, 0, sizeof(wqe)); + + wrapped = cmdq->wrapped; + + next_prod_idx = curr_prod_idx + num_wqebbs; + if (next_prod_idx >= wq->q_depth) { + cmdq->wrapped = !cmdq->wrapped; + next_prod_idx -= wq->q_depth; + } + + cmd_info = &cmdq->cmd_infos[curr_prod_idx]; + + init_completion(&done); + + cmd_info->done = &done; + cmd_info->errcode = &errcode; + cmd_info->cmpt_code = &cmpt_code; + + memcpy(&saved_cmd_info, cmd_info, sizeof(*cmd_info)); + + cmdq_set_lcmd_wqe(&wqe, SYNC_CMD_SGE_RESP, buf_in, buf_out, + wrapped, ack_type, mod, cmd, curr_prod_idx); + + hifc_cpu_to_be32(&wqe, wqe_size); + + cmdq_wqe_fill(curr_wqe, &wqe); + + cmd_info->cmd_type = HIFC_CMD_TYPE_SGE_RESP; + + (cmd_info->cmdq_msg_id)++; + curr_msg_id = cmd_info->cmdq_msg_id; + + cmdq_set_db(cmdq, HIFC_CMDQ_SYNC, next_prod_idx); + + spin_unlock_bh(&cmdq->cmdq_lock); + + timeo = msecs_to_jiffies(timeout ? timeout : CMDQ_CMD_TIMEOUT); + if (!wait_for_completion_timeout(&done, timeo)) { + err = cmdq_sync_cmd_timeout_handler(cmdq, cmd_info, + &saved_cmd_info, + curr_wqe, mod, cmd, + curr_prod_idx, curr_msg_id); + if (!err) + goto timeout_check_ok; + + sdk_err(cmdq->hwdev->dev_hdl, "Cmdq sync command timeout, prod idx: 0x%x\n", + curr_prod_idx); + return -ETIMEDOUT; + } + +timeout_check_ok: + + smp_rmb(); /* read error code after completion */ + + if (errcode > 1) + return errcode; + + return 0; +} + +static int cmdq_async_cmd(struct hifc_cmdq *cmdq, enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, + struct hifc_cmd_buf *buf_in) +{ + struct hifc_wq *wq = cmdq->wq; + int wqe_size = cmdq_wqe_size(WQE_LCMD_TYPE); + u16 curr_prod_idx, next_prod_idx, num_wqebbs; + struct hifc_cmdq_wqe *curr_wqe, wqe; + int wrapped; + + num_wqebbs = WQE_NUM_WQEBBS(wqe_size, wq); + + spin_lock_bh(&cmdq->cmdq_lock); + + /* WQE_SIZE = WQEBB_SIZE, we will get the wq element and not shadow*/ + curr_wqe = hifc_get_wqe(cmdq->wq, num_wqebbs, &curr_prod_idx); + if (!curr_wqe) { + spin_unlock_bh(&cmdq->cmdq_lock); + return -EBUSY; + } + + memset(&wqe, 0, sizeof(wqe)); + + wrapped = cmdq->wrapped; + next_prod_idx = curr_prod_idx + num_wqebbs; + if (next_prod_idx >= cmdq->wq->q_depth) { + cmdq->wrapped = !cmdq->wrapped; + next_prod_idx -= cmdq->wq->q_depth; + } + + cmdq_set_lcmd_wqe(&wqe, ASYNC_CMD, buf_in, NULL, wrapped, + ack_type, mod, cmd, curr_prod_idx); + + /* The data that is written to HW should be in Big Endian Format */ + hifc_cpu_to_be32(&wqe, wqe_size); + + cmdq_wqe_fill(curr_wqe, &wqe); + + cmdq->cmd_infos[curr_prod_idx].cmd_type = HIFC_CMD_TYPE_ASYNC; + + cmdq_set_db(cmdq, HIFC_CMDQ_ASYNC, next_prod_idx); + + spin_unlock_bh(&cmdq->cmdq_lock); + + return 0; +} + +static int cmdq_set_arm_bit(struct hifc_cmdq *cmdq, void *buf_in, u16 in_size) +{ + struct hifc_wq *wq = cmdq->wq; + struct hifc_cmdq_wqe *curr_wqe, wqe; + u16 curr_prod_idx, next_prod_idx, num_wqebbs; + int wrapped, wqe_size = cmdq_wqe_size(WQE_SCMD_TYPE); + + num_wqebbs = WQE_NUM_WQEBBS(wqe_size, wq); + + /* Keep wrapped and doorbell index correct. bh - for tasklet(ceq) */ + spin_lock_bh(&cmdq->cmdq_lock); + + /* WQE_SIZE = WQEBB_SIZE, we will get the wq element and not shadow*/ + curr_wqe = hifc_get_wqe(cmdq->wq, num_wqebbs, &curr_prod_idx); + if (!curr_wqe) { + spin_unlock_bh(&cmdq->cmdq_lock); + sdk_err(cmdq->hwdev->dev_hdl, "Can not get avalible wqebb setting arm\n"); + return -EBUSY; + } + + memset(&wqe, 0, sizeof(wqe)); + + wrapped = cmdq->wrapped; + + next_prod_idx = curr_prod_idx + num_wqebbs; + if (next_prod_idx >= wq->q_depth) { + cmdq->wrapped = !cmdq->wrapped; + next_prod_idx -= wq->q_depth; + } + + cmdq_set_inline_wqe(&wqe, SYNC_CMD_DIRECT_RESP, buf_in, in_size, NULL, + wrapped, HIFC_ACK_TYPE_CMDQ, HIFC_MOD_COMM, + CMDQ_SET_ARM_CMD, curr_prod_idx); + + /* The data that is written to HW should be in Big Endian Format */ + hifc_cpu_to_be32(&wqe, wqe_size); + + /* cmdq wqe is not shadow, therefore wqe will be written to wq */ + cmdq_wqe_fill(curr_wqe, &wqe); + + cmdq->cmd_infos[curr_prod_idx].cmd_type = HIFC_CMD_TYPE_SET_ARM; + + cmdq_set_db(cmdq, cmdq->cmdq_type, next_prod_idx); + + spin_unlock_bh(&cmdq->cmdq_lock); + + return 0; +} + +static int cmdq_params_valid(void *hwdev, struct hifc_cmd_buf *buf_in) +{ + if (!buf_in || !hwdev) { + pr_err("Invalid CMDQ buffer addr\n"); + return -EINVAL; + } + + if (!buf_in->size || buf_in->size > HIFC_CMDQ_MAX_DATA_SIZE) { + pr_err("Invalid CMDQ buffer size: 0x%x\n", buf_in->size); + return -EINVAL; + } + + return 0; +} + +#define WAIT_CMDQ_ENABLE_TIMEOUT 300 + +static int wait_cmdqs_enable(struct hifc_cmdqs *cmdqs) +{ + unsigned long end; + + end = jiffies + msecs_to_jiffies(WAIT_CMDQ_ENABLE_TIMEOUT); + do { + if (cmdqs->status & HIFC_CMDQ_ENABLE) + return 0; + } while (time_before(jiffies, end) && cmdqs->hwdev->chip_present_flag && + !cmdqs->disable_flag); + + cmdqs->disable_flag = 1; + + return -EBUSY; +} + +int hifc_cmdq_direct_resp(void *hwdev, enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, + struct hifc_cmd_buf *buf_in, u64 *out_param, + u32 timeout) +{ + struct hifc_cmdqs *cmdqs; + int err = cmdq_params_valid(hwdev, buf_in); + + if (err) { + pr_err("Invalid CMDQ parameters\n"); + return err; + } + + cmdqs = ((struct hifc_hwdev *)hwdev)->cmdqs; + + if (!(((struct hifc_hwdev *)hwdev)->chip_present_flag) || + !hifc_is_hwdev_mod_inited(hwdev, HIFC_HWDEV_CMDQ_INITED)) + return -EPERM; + + err = wait_cmdqs_enable(cmdqs); + if (err) { + sdk_err(cmdqs->hwdev->dev_hdl, "Cmdq is disable\n"); + return err; + } + + err = cmdq_sync_cmd_direct_resp(&cmdqs->cmdq[HIFC_CMDQ_SYNC], ack_type, + mod, cmd, buf_in, out_param, timeout); + if (!(((struct hifc_hwdev *)hwdev)->chip_present_flag)) + return -ETIMEDOUT; + else + return err; +} + +int hifc_cmdq_detail_resp(void *hwdev, + enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, + struct hifc_cmd_buf *buf_in, + struct hifc_cmd_buf *buf_out, + u32 timeout) +{ + struct hifc_cmdqs *cmdqs; + int err = cmdq_params_valid(hwdev, buf_in); + + if (err) + return err; + + cmdqs = ((struct hifc_hwdev *)hwdev)->cmdqs; + + if (!(((struct hifc_hwdev *)hwdev)->chip_present_flag) || + !hifc_is_hwdev_mod_inited(hwdev, HIFC_HWDEV_CMDQ_INITED)) + return -EPERM; + + err = wait_cmdqs_enable(cmdqs); + if (err) { + sdk_err(cmdqs->hwdev->dev_hdl, "Cmdq is disable\n"); + return err; + } + + err = cmdq_sync_cmd_detail_resp(&cmdqs->cmdq[HIFC_CMDQ_SYNC], ack_type, + mod, cmd, buf_in, buf_out, timeout); + if (!(((struct hifc_hwdev *)hwdev)->chip_present_flag)) + return -ETIMEDOUT; + else + return err; +} + +int hifc_cmdq_async(void *hwdev, enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, + struct hifc_cmd_buf *buf_in) +{ + struct hifc_cmdqs *cmdqs; + int err = cmdq_params_valid(hwdev, buf_in); + + if (err) + return err; + + cmdqs = ((struct hifc_hwdev *)hwdev)->cmdqs; + + if (!(((struct hifc_hwdev *)hwdev)->chip_present_flag) || + !hifc_is_hwdev_mod_inited(hwdev, HIFC_HWDEV_CMDQ_INITED)) + return -EPERM; + + err = wait_cmdqs_enable(cmdqs); + if (err) { + sdk_err(cmdqs->hwdev->dev_hdl, "Cmdq is disable\n"); + return err; + } + + return cmdq_async_cmd(&cmdqs->cmdq[HIFC_CMDQ_ASYNC], ack_type, mod, + cmd, buf_in); +} + +int hifc_set_arm_bit(void *hwdev, enum hifc_set_arm_type q_type, u16 q_id) +{ + struct hifc_cmdqs *cmdqs; + struct hifc_cmdq *cmdq; + struct hifc_cmdq_arm_bit arm_bit; + enum hifc_cmdq_type cmdq_type = HIFC_CMDQ_SYNC; + u16 in_size; + int err; + + if (!hwdev) + return -EINVAL; + + if (!(((struct hifc_hwdev *)hwdev)->chip_present_flag) || + !hifc_is_hwdev_mod_inited(hwdev, HIFC_HWDEV_CMDQ_INITED)) + return -EPERM; + + cmdqs = ((struct hifc_hwdev *)hwdev)->cmdqs; + + if (!(cmdqs->status & HIFC_CMDQ_ENABLE)) + return -EBUSY; + + if (q_type == HIFC_SET_ARM_CMDQ) { + if (q_id >= HIFC_MAX_CMDQ_TYPES) + return -EFAULT; + + cmdq_type = q_id; + } + /* sq is using interrupt now, so we only need to set arm bit for cmdq, + * remove comment below if need to set sq arm bit + * else + * cmdq_type = HIFC_CMDQ_SYNC; + */ + + cmdq = &cmdqs->cmdq[cmdq_type]; + + arm_bit.q_type = q_type; + arm_bit.q_id = q_id; + in_size = sizeof(arm_bit); + + err = cmdq_set_arm_bit(cmdq, &arm_bit, in_size); + if (err) { + sdk_err(cmdqs->hwdev->dev_hdl, + "Failed to set arm for q_type: %d, qid %d\n", + q_type, q_id); + return err; + } + + return 0; +} + +static void clear_wqe_complete_bit(struct hifc_cmdq *cmdq, + struct hifc_cmdq_wqe *wqe, u16 ci) +{ + struct hifc_cmdq_wqe_lcmd *wqe_lcmd; + struct hifc_cmdq_inline_wqe *inline_wqe; + struct hifc_cmdq_wqe_scmd *wqe_scmd; + struct hifc_ctrl *ctrl; + u32 header_info = be32_to_cpu(WQE_HEADER(wqe)->header_info); + int buf_len = CMDQ_WQE_HEADER_GET(header_info, BUFDESC_LEN); + int wqe_size = cmdq_get_wqe_size(buf_len); + u16 num_wqebbs; + + if (wqe_size == WQE_LCMD_SIZE) { + wqe_lcmd = &wqe->wqe_lcmd; + ctrl = &wqe_lcmd->ctrl; + } else { + inline_wqe = &wqe->inline_wqe; + wqe_scmd = &inline_wqe->wqe_scmd; + ctrl = &wqe_scmd->ctrl; + } + + /* clear HW busy bit */ + ctrl->ctrl_info = 0; + cmdq->cmd_infos[ci].cmd_type = HIFC_CMD_TYPE_NONE; + + wmb(); /* verify wqe is clear */ + + num_wqebbs = WQE_NUM_WQEBBS(wqe_size, cmdq->wq); + hifc_put_wqe(cmdq->wq, num_wqebbs); +} + +static void cmdq_sync_cmd_handler(struct hifc_cmdq *cmdq, + struct hifc_cmdq_wqe *wqe, u16 cons_idx) +{ + u16 prod_idx = cons_idx; + + spin_lock(&cmdq->cmdq_lock); + + cmdq_update_cmd_status(cmdq, prod_idx, wqe); + + if (cmdq->cmd_infos[prod_idx].cmpt_code) { + *cmdq->cmd_infos[prod_idx].cmpt_code = + CMDQ_COMPLETE_CMPT_CODE; + cmdq->cmd_infos[prod_idx].cmpt_code = NULL; + } + + /* make sure cmpt_code operation before done operation */ + smp_rmb(); + + if (cmdq->cmd_infos[prod_idx].done) { + complete(cmdq->cmd_infos[prod_idx].done); + cmdq->cmd_infos[prod_idx].done = NULL; + } + + spin_unlock(&cmdq->cmdq_lock); + + clear_wqe_complete_bit(cmdq, wqe, cons_idx); +} + +static void cmdq_async_cmd_handler(struct hifc_hwdev *hwdev, + struct hifc_cmdq *cmdq, + struct hifc_cmdq_wqe *wqe, u16 ci) +{ + u64 buf = wqe->wqe_lcmd.buf_desc.saved_async_buf; + int addr_sz = sizeof(u64); + + hifc_be32_to_cpu((void *)&buf, addr_sz); + if (buf) + hifc_free_cmd_buf(hwdev, (struct hifc_cmd_buf *)buf); + + clear_wqe_complete_bit(cmdq, wqe, ci); +} + +static int cmdq_arm_ceq_handler(struct hifc_cmdq *cmdq, + struct hifc_cmdq_wqe *wqe, u16 ci) +{ + struct hifc_cmdq_inline_wqe *inline_wqe = &wqe->inline_wqe; + struct hifc_cmdq_wqe_scmd *wqe_scmd = &inline_wqe->wqe_scmd; + struct hifc_ctrl *ctrl = &wqe_scmd->ctrl; + u32 ctrl_info = be32_to_cpu((ctrl)->ctrl_info); + + if (!WQE_COMPLETED(ctrl_info)) + return -EBUSY; + + clear_wqe_complete_bit(cmdq, wqe, ci); + + return 0; +} + +#define HIFC_CMDQ_WQE_HEAD_LEN 32 +static void hifc_dump_cmdq_wqe_head(struct hifc_hwdev *hwdev, + struct hifc_cmdq_wqe *wqe) +{ + u32 i; + u32 *data = (u32 *)wqe; + + for (i = 0; i < (HIFC_CMDQ_WQE_HEAD_LEN / sizeof(u32)); i += 4) { + sdk_info(hwdev->dev_hdl, "wqe data: 0x%08x, 0x%08x, 0x%08x, 0x%08x\n", + data[i], data[i + 1], data[i + 2], + data[i + 3]);/*lint !e679*/ + } +} + +#define CMDQ_CMD_TYPE_TIMEOUT(cmd_type) \ + ((cmd_type) == HIFC_CMD_TYPE_TIMEOUT || \ + (cmd_type) == HIFC_CMD_TYPE_FAKE_TIMEOUT) + +static inline void cmdq_response_handle(struct hifc_hwdev *hwdev, + struct hifc_cmdq *cmdq, + struct hifc_cmdq_wqe *wqe, + enum hifc_cmdq_type cmdq_type, u16 ci) +{ + if (cmdq_type == HIFC_CMDQ_ASYNC) + cmdq_async_cmd_handler(hwdev, cmdq, wqe, ci); + else + cmdq_sync_cmd_handler(cmdq, wqe, ci); +} + +static inline void set_arm_bit(struct hifc_hwdev *hwdev, int set_arm, + enum hifc_cmdq_type cmdq_type) +{ + if (set_arm) + hifc_set_arm_bit(hwdev, HIFC_SET_ARM_CMDQ, cmdq_type); +} + +void hifc_cmdq_ceq_handler(void *handle, u32 ceqe_data) +{ + struct hifc_cmdqs *cmdqs = ((struct hifc_hwdev *)handle)->cmdqs; + enum hifc_cmdq_type cmdq_type = CEQE_CMDQ_GET(ceqe_data, TYPE); + struct hifc_cmdq *cmdq = &cmdqs->cmdq[cmdq_type]; + struct hifc_hwdev *hwdev = cmdqs->hwdev; + struct hifc_cmdq_wqe *wqe; + struct hifc_cmdq_wqe_lcmd *wqe_lcmd; + struct hifc_ctrl *ctrl; + struct hifc_cmdq_cmd_info *cmd_info; + u32 ctrl_info; + u16 ci; + int set_arm = 1; + + while ((wqe = hifc_read_wqe(cmdq->wq, 1, &ci)) != NULL) { + cmd_info = &cmdq->cmd_infos[ci]; + + if (cmd_info->cmd_type == HIFC_CMD_TYPE_NONE) { + set_arm = 1; + break; + } else if (CMDQ_CMD_TYPE_TIMEOUT(cmd_info->cmd_type)) { + if (cmd_info->cmd_type == HIFC_CMD_TYPE_TIMEOUT) { + sdk_info(hwdev->dev_hdl, "Cmdq timeout, q_id: %u, ci: %u\n", + cmdq_type, ci); + hifc_dump_cmdq_wqe_head(hwdev, wqe); + } + + set_arm = 1; + clear_wqe_complete_bit(cmdq, wqe, ci); + } else if (cmd_info->cmd_type == HIFC_CMD_TYPE_SET_ARM) { + /* arm_bit was set until here */ + set_arm = 0; + + if (cmdq_arm_ceq_handler(cmdq, wqe, ci)) + break; + } else { + set_arm = 1; + + /* only arm bit is using scmd wqe, the wqe is lcmd */ + wqe_lcmd = &wqe->wqe_lcmd; + ctrl = &wqe_lcmd->ctrl; + ctrl_info = be32_to_cpu((ctrl)->ctrl_info); + + if (!WQE_COMPLETED(ctrl_info)) + break; + + /* This memory barrier is needed to keep us from reading + * any other fields out of the cmdq wqe until we have + * verified the command has been processed and + * written back. + */ + dma_rmb(); + + cmdq_response_handle(hwdev, cmdq, wqe, cmdq_type, ci); + } + } + + set_arm_bit(hwdev, set_arm, cmdq_type); +} + +static void cmdq_init_queue_ctxt(struct hifc_cmdq *cmdq, + struct hifc_cmdq_pages *cmdq_pages, + struct hifc_cmdq_ctxt *cmdq_ctxt) +{ + struct hifc_cmdqs *cmdqs = cmdq_to_cmdqs(cmdq); + struct hifc_hwdev *hwdev = cmdqs->hwdev; + struct hifc_wq *wq = cmdq->wq; + struct hifc_cmdq_ctxt_info *ctxt_info = &cmdq_ctxt->ctxt_info; + u64 wq_first_page_paddr, cmdq_first_block_paddr, pfn; + u16 start_ci = (u16)wq->cons_idx; + + /* The data in the HW is in Big Endian Format */ + wq_first_page_paddr = be64_to_cpu(*wq->block_vaddr); + + pfn = CMDQ_PFN(wq_first_page_paddr); + + ctxt_info->curr_wqe_page_pfn = + CMDQ_CTXT_PAGE_INFO_SET(1, HW_BUSY_BIT) | + CMDQ_CTXT_PAGE_INFO_SET(1, CEQ_EN) | + CMDQ_CTXT_PAGE_INFO_SET(1, CEQ_ARM) | + CMDQ_CTXT_PAGE_INFO_SET(HIFC_CEQ_ID_CMDQ, EQ_ID) | + CMDQ_CTXT_PAGE_INFO_SET(pfn, CURR_WQE_PAGE_PFN); + + /* If only use one page, use 0-level CLA */ + if (cmdq->wq->num_q_pages != 1) { + cmdq_first_block_paddr = cmdq_pages->cmdq_page_paddr; + pfn = CMDQ_PFN(cmdq_first_block_paddr); + } + + ctxt_info->wq_block_pfn = CMDQ_CTXT_BLOCK_INFO_SET(start_ci, CI) | + CMDQ_CTXT_BLOCK_INFO_SET(pfn, WQ_BLOCK_PFN); + + cmdq_ctxt->func_idx = hifc_global_func_id_hw(hwdev); + cmdq_ctxt->ppf_idx = HIFC_HWIF_PPF_IDX(hwdev->hwif); + cmdq_ctxt->cmdq_id = cmdq->cmdq_type; +} + +static int init_cmdq(struct hifc_cmdq *cmdq, struct hifc_hwdev *hwdev, + struct hifc_wq *wq, enum hifc_cmdq_type q_type) +{ + void __iomem *db_base; + int err = 0; + + cmdq->wq = wq; + cmdq->cmdq_type = q_type; + cmdq->wrapped = 1; + cmdq->hwdev = hwdev; + + spin_lock_init(&cmdq->cmdq_lock); + + cmdq->cmd_infos = kcalloc(wq->q_depth, sizeof(*cmdq->cmd_infos), + GFP_KERNEL); + if (!cmdq->cmd_infos) { + err = -ENOMEM; + goto cmd_infos_err; + } + + err = hifc_alloc_db_addr(hwdev, &db_base, NULL); + if (err) + goto alloc_db_err; + + cmdq->db_base = (u8 *)db_base; + return 0; + +alloc_db_err: + kfree(cmdq->cmd_infos); + +cmd_infos_err: + + return err; +} + +static void free_cmdq(struct hifc_hwdev *hwdev, struct hifc_cmdq *cmdq) +{ + hifc_free_db_addr(hwdev, cmdq->db_base, NULL); + kfree(cmdq->cmd_infos); +} + +int hifc_set_cmdq_ctxts(struct hifc_hwdev *hwdev) +{ + struct hifc_cmdqs *cmdqs = hwdev->cmdqs; + struct hifc_cmdq_ctxt *cmdq_ctxt, cmdq_ctxt_out = {0}; + enum hifc_cmdq_type cmdq_type; + u16 in_size; + u16 out_size = sizeof(*cmdq_ctxt); + int err; + + cmdq_type = HIFC_CMDQ_SYNC; + for (; cmdq_type < HIFC_MAX_CMDQ_TYPES; cmdq_type++) { + cmdq_ctxt = &cmdqs->cmdq[cmdq_type].cmdq_ctxt; + cmdq_ctxt->func_idx = hifc_global_func_id_hw(hwdev); + in_size = sizeof(*cmdq_ctxt); + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_CMDQ_CTXT_SET, + cmdq_ctxt, in_size, + &cmdq_ctxt_out, &out_size, 0); + if (err || !out_size || cmdq_ctxt_out.status) { + sdk_err(hwdev->dev_hdl, "Failed to set cmdq ctxt, err: %d, status: 0x%x, out_size: 0x%x\n", + err, cmdq_ctxt_out.status, out_size); + return -EFAULT; + } + } + + cmdqs->status |= HIFC_CMDQ_ENABLE; + cmdqs->disable_flag = 0; + + return 0; +} + +void hifc_cmdq_flush_cmd(struct hifc_hwdev *hwdev, + struct hifc_cmdq *cmdq) +{ + struct hifc_cmdq_wqe *wqe; + struct hifc_cmdq_cmd_info *cmdq_info; + u16 ci, wqe_left, i; + u64 buf; + + spin_lock_bh(&cmdq->cmdq_lock); + wqe_left = cmdq->wq->q_depth - (u16)atomic_read(&cmdq->wq->delta); + ci = MASKED_WQE_IDX(cmdq->wq, cmdq->wq->cons_idx); + for (i = 0; i < wqe_left; i++, ci++) { + ci = MASKED_WQE_IDX(cmdq->wq, ci); + cmdq_info = &cmdq->cmd_infos[ci]; + + if (cmdq_info->cmd_type == HIFC_CMD_TYPE_SET_ARM) + continue; + + if (cmdq->cmdq_type == HIFC_CMDQ_ASYNC) { + wqe = hifc_get_wqebb_addr(cmdq->wq, ci); + buf = wqe->wqe_lcmd.buf_desc.saved_async_buf; + wqe->wqe_lcmd.buf_desc.saved_async_buf = 0; + + hifc_be32_to_cpu((void *)&buf, sizeof(u64)); + if (buf) + hifc_free_cmd_buf(hwdev, + (struct hifc_cmd_buf *)buf); + } else { + if (cmdq_info->done) { + complete(cmdq_info->done); + cmdq_info->done = NULL; + cmdq_info->cmpt_code = NULL; + cmdq_info->direct_resp = NULL; + cmdq_info->errcode = NULL; + } + } + } + + spin_unlock_bh(&cmdq->cmdq_lock); +} + +int hifc_reinit_cmdq_ctxts(struct hifc_hwdev *hwdev) +{ + struct hifc_cmdqs *cmdqs = hwdev->cmdqs; + enum hifc_cmdq_type cmdq_type; + + cmdq_type = HIFC_CMDQ_SYNC; + for (; cmdq_type < HIFC_MAX_CMDQ_TYPES; cmdq_type++) { + hifc_cmdq_flush_cmd(hwdev, &cmdqs->cmdq[cmdq_type]); + cmdqs->cmdq[cmdq_type].wrapped = 1; + hifc_wq_wqe_pg_clear(cmdqs->cmdq[cmdq_type].wq); + } + + return hifc_set_cmdq_ctxts(hwdev); +} + +int hifc_cmdqs_init(struct hifc_hwdev *hwdev) +{ + struct hifc_cmdqs *cmdqs; + struct hifc_cmdq_ctxt *cmdq_ctxt; + enum hifc_cmdq_type type, cmdq_type; + size_t saved_wqs_size; + u32 max_wqe_size; + int err; + + cmdqs = kzalloc(sizeof(*cmdqs), GFP_KERNEL); + if (!cmdqs) + return -ENOMEM; + + hwdev->cmdqs = cmdqs; + cmdqs->hwdev = hwdev; + + saved_wqs_size = HIFC_MAX_CMDQ_TYPES * sizeof(struct hifc_wq); + cmdqs->saved_wqs = kzalloc(saved_wqs_size, GFP_KERNEL); + if (!cmdqs->saved_wqs) { + sdk_err(hwdev->dev_hdl, "Failed to allocate saved wqs\n"); + err = -ENOMEM; + goto alloc_wqs_err; + } + + cmdqs->cmd_buf_pool = dma_pool_create("hifc_cmdq", hwdev->dev_hdl, + HIFC_CMDQ_BUF_SIZE, + HIFC_CMDQ_BUF_SIZE, 0ULL); + if (!cmdqs->cmd_buf_pool) { + sdk_err(hwdev->dev_hdl, "Failed to create cmdq buffer pool\n"); + err = -ENOMEM; + goto pool_create_err; + } + + max_wqe_size = (u32)cmdq_wqe_size(WQE_LCMD_TYPE); + err = hifc_cmdq_alloc(&cmdqs->cmdq_pages, cmdqs->saved_wqs, + hwdev->dev_hdl, HIFC_MAX_CMDQ_TYPES, + hwdev->wq_page_size, CMDQ_WQEBB_SIZE, + HIFC_CMDQ_DEPTH, max_wqe_size); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to allocate cmdq\n"); + goto cmdq_alloc_err; + } + + cmdq_type = HIFC_CMDQ_SYNC; + for (; cmdq_type < HIFC_MAX_CMDQ_TYPES; cmdq_type++) { + err = init_cmdq(&cmdqs->cmdq[cmdq_type], hwdev, + &cmdqs->saved_wqs[cmdq_type], cmdq_type); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to initialize cmdq type :%d\n", + cmdq_type); + goto init_cmdq_err; + } + + cmdq_ctxt = &cmdqs->cmdq[cmdq_type].cmdq_ctxt; + cmdq_init_queue_ctxt(&cmdqs->cmdq[cmdq_type], + &cmdqs->cmdq_pages, cmdq_ctxt); + } + + err = hifc_set_cmdq_ctxts(hwdev); + if (err) + goto init_cmdq_err; + + return 0; + +init_cmdq_err: + type = HIFC_CMDQ_SYNC; + for (; type < cmdq_type; type++) + free_cmdq(hwdev, &cmdqs->cmdq[type]); + + hifc_cmdq_free(&cmdqs->cmdq_pages, cmdqs->saved_wqs, + HIFC_MAX_CMDQ_TYPES); + +cmdq_alloc_err: + dma_pool_destroy(cmdqs->cmd_buf_pool); + +pool_create_err: + kfree(cmdqs->saved_wqs); + +alloc_wqs_err: + kfree(cmdqs); + + return err; +} + +void hifc_cmdqs_free(struct hifc_hwdev *hwdev) +{ + struct hifc_cmdqs *cmdqs = hwdev->cmdqs; + enum hifc_cmdq_type cmdq_type = HIFC_CMDQ_SYNC; + + cmdqs->status &= ~HIFC_CMDQ_ENABLE; + + for (; cmdq_type < HIFC_MAX_CMDQ_TYPES; cmdq_type++) { + hifc_cmdq_flush_cmd(hwdev, &cmdqs->cmdq[cmdq_type]); + free_cmdq(cmdqs->hwdev, &cmdqs->cmdq[cmdq_type]); + } + + hifc_cmdq_free(&cmdqs->cmdq_pages, cmdqs->saved_wqs, + HIFC_MAX_CMDQ_TYPES); + + dma_pool_destroy(cmdqs->cmd_buf_pool); + + kfree(cmdqs->saved_wqs); + + kfree(cmdqs); +} diff --git a/drivers/scsi/huawei/hifc/hifc_cmdq.h b/drivers/scsi/huawei/hifc/hifc_cmdq.h new file mode 100644 index 000000000000..cb2ac81c5edc --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_cmdq.h @@ -0,0 +1,210 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#ifndef HIFC_CMDQ_H_ +#define HIFC_CMDQ_H_ + +#define HIFC_DB_OFF 0x00000800 + +#define HIFC_SCMD_DATA_LEN 16 + +#define HIFC_CMDQ_DEPTH 4096 + +#define HIFC_CMDQ_BUF_SIZE 2048U +#define HIFC_CMDQ_BUF_HW_RSVD 8 +#define HIFC_CMDQ_MAX_DATA_SIZE \ + (HIFC_CMDQ_BUF_SIZE - HIFC_CMDQ_BUF_HW_RSVD) +#define WQ_PAGE_PFN_SHIFT 12 +#define WQ_BLOCK_PFN_SHIFT 9 + +#define WQ_PAGE_PFN(page_addr) ((page_addr) >> WQ_PAGE_PFN_SHIFT) +#define WQ_BLOCK_PFN(page_addr) ((page_addr) >> WQ_BLOCK_PFN_SHIFT) + +enum hifc_cmdq_type { + HIFC_CMDQ_SYNC, + HIFC_CMDQ_ASYNC, + HIFC_MAX_CMDQ_TYPES, +}; + +enum hifc_db_src_type { + HIFC_DB_SRC_CMDQ_TYPE, + HIFC_DB_SRC_L2NIC_SQ_TYPE, +}; + +enum hifc_cmdq_db_type { + HIFC_DB_SQ_RQ_TYPE, + HIFC_DB_CMDQ_TYPE, +}; + +/* CMDQ WQE CTRLS */ +struct hifc_cmdq_header { + u32 header_info; + u32 saved_data; +}; + +struct hifc_scmd_bufdesc { + u32 buf_len; + u32 rsvd; + u8 data[HIFC_SCMD_DATA_LEN]; +}; + +struct hifc_lcmd_bufdesc { + struct hifc_sge sge; + u32 rsvd1; + u64 saved_async_buf; + u64 rsvd3; +}; + +struct hifc_cmdq_db { + u32 db_info; + u32 rsvd; +}; + +struct hifc_status { + u32 status_info; +}; + +struct hifc_ctrl { + u32 ctrl_info; +}; + +struct hifc_sge_resp { + struct hifc_sge sge; + u32 rsvd; +}; + +struct hifc_cmdq_completion { + /* HW Format */ + union { + struct hifc_sge_resp sge_resp; + u64 direct_resp; + }; +}; + +struct hifc_cmdq_wqe_scmd { + struct hifc_cmdq_header header; + struct hifc_cmdq_db db; + struct hifc_status status; + struct hifc_ctrl ctrl; + struct hifc_cmdq_completion completion; + struct hifc_scmd_bufdesc buf_desc; +}; + +struct hifc_cmdq_wqe_lcmd { + struct hifc_cmdq_header header; + struct hifc_status status; + struct hifc_ctrl ctrl; + struct hifc_cmdq_completion completion; + struct hifc_lcmd_bufdesc buf_desc; +}; + +struct hifc_cmdq_inline_wqe { + struct hifc_cmdq_wqe_scmd wqe_scmd; +}; + +struct hifc_cmdq_wqe { + /* HW Format */ + union { + struct hifc_cmdq_inline_wqe inline_wqe; + struct hifc_cmdq_wqe_lcmd wqe_lcmd; + }; +}; + +struct hifc_cmdq_arm_bit { + u32 q_type; + u32 q_id; +}; + +struct hifc_cmdq_ctxt_info { + u64 curr_wqe_page_pfn; + u64 wq_block_pfn; +}; + +struct hifc_cmdq_ctxt { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_idx; + u8 cmdq_id; + u8 ppf_idx; + + u8 rsvd1[4]; + + struct hifc_cmdq_ctxt_info ctxt_info; +}; + +enum hifc_cmdq_status { + HIFC_CMDQ_ENABLE = BIT(0), +}; + +enum hifc_cmdq_cmd_type { + HIFC_CMD_TYPE_NONE, + HIFC_CMD_TYPE_SET_ARM, + HIFC_CMD_TYPE_DIRECT_RESP, + HIFC_CMD_TYPE_SGE_RESP, + HIFC_CMD_TYPE_ASYNC, + HIFC_CMD_TYPE_TIMEOUT, + HIFC_CMD_TYPE_FAKE_TIMEOUT, +}; + +struct hifc_cmdq_cmd_info { + enum hifc_cmdq_cmd_type cmd_type; + + struct completion *done; + int *errcode; + int *cmpt_code; + u64 *direct_resp; + u64 cmdq_msg_id; +}; + +struct hifc_cmdq { + struct hifc_wq *wq; + + enum hifc_cmdq_type cmdq_type; + int wrapped; + + /* spinlock for send cmdq commands */ + spinlock_t cmdq_lock; + + /* doorbell area */ + u8 __iomem *db_base; + + struct hifc_cmdq_ctxt cmdq_ctxt; + + struct hifc_cmdq_cmd_info *cmd_infos; + + struct hifc_hwdev *hwdev; +}; + +struct hifc_cmdqs { + struct hifc_hwdev *hwdev; + + struct pci_pool *cmd_buf_pool; + + struct hifc_wq *saved_wqs; + + struct hifc_cmdq_pages cmdq_pages; + struct hifc_cmdq cmdq[HIFC_MAX_CMDQ_TYPES]; + + u32 status; + u32 disable_flag; +}; + +void hifc_cmdq_ceq_handler(void *hwdev, u32 ceqe_data); + +int hifc_reinit_cmdq_ctxts(struct hifc_hwdev *hwdev); + +bool hifc_cmdq_idle(struct hifc_cmdq *cmdq); + +int hifc_cmdqs_init(struct hifc_hwdev *hwdev); + +void hifc_cmdqs_free(struct hifc_hwdev *hwdev); + +void hifc_cmdq_flush_cmd(struct hifc_hwdev *hwdev, + struct hifc_cmdq *cmdq); + +#endif diff --git a/drivers/scsi/huawei/hifc/hifc_cqm_main.c b/drivers/scsi/huawei/hifc/hifc_cqm_main.c new file mode 100644 index 000000000000..4cd048f1e662 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_cqm_main.c @@ -0,0 +1,694 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#include <linux/sched.h> +#include <linux/pci.h> +#include <linux/module.h> +#include <linux/delay.h> +#include <linux/vmalloc.h> + +#include "hifc_knl_adp.h" +#include "hifc_hw.h" +#include "hifc_hwdev.h" +#include "hifc_hwif.h" +#include "hifc_api_cmd.h" +#include "hifc_mgmt.h" +#include "hifc_cfg.h" +#include "hifc_cqm_object.h" +#include "hifc_cqm_main.h" + +#define GET_MAX(a, b) (((a) > (b)) ? (a) : (b)) +#define GET_MIN(a, b) (((a) < (b)) ? (a) : (b)) + +static void cqm_capability_init_check_ppf(void *ex_handle, + u32 *total_function_num) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct service_cap *service_capability = &handle->cfg_mgmt->svc_cap; + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + (handle->cqm_hdl); + + if (cqm_handle->func_attribute.func_type == CQM_PPF) { + *total_function_num = service_capability->host_total_function; + cqm_handle->func_capability.timer_enable = + service_capability->timer_en; + + cqm_info(handle->dev_hdl, "Cap init: total function num 0x%x\n", + *total_function_num); + cqm_info(handle->dev_hdl, "Cap init: timer_enable %d (1: enable; 0: disable)\n", + cqm_handle->func_capability.timer_enable); + } +} + +void cqm_test_mode_init(struct cqm_handle_s *cqm_handle, + struct service_cap *service_capability) +{ + cqm_handle->func_capability.xid_alloc_mode = + service_capability->test_xid_alloc_mode; + cqm_handle->func_capability.gpa_check_enable = + service_capability->test_gpa_check_enable; +} + +static s32 cqm_service_capability_init_for_each( + struct cqm_handle_s *cqm_handle, + struct service_cap *service_capability) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)cqm_handle->ex_handle; + + cqm_info(handle->dev_hdl, "Cap init: fc is valid\n"); + cqm_handle->func_capability.hash_number += + service_capability->fc_cap.dev_fc_cap.max_parent_qpc_num; + cqm_handle->func_capability.hash_basic_size = CQM_HASH_BUCKET_SIZE_64; + cqm_handle->func_capability.qpc_number += + service_capability->fc_cap.dev_fc_cap.max_parent_qpc_num; + cqm_handle->func_capability.qpc_basic_size = + GET_MAX(service_capability->fc_cap.parent_qpc_size, + cqm_handle->func_capability.qpc_basic_size); + cqm_handle->func_capability.qpc_alloc_static = true; + cqm_handle->func_capability.scqc_number += + service_capability->fc_cap.dev_fc_cap.scq_num; + cqm_handle->func_capability.scqc_basic_size = + GET_MAX(service_capability->fc_cap.scqc_size, + cqm_handle->func_capability.scqc_basic_size); + cqm_handle->func_capability.srqc_number += + service_capability->fc_cap.dev_fc_cap.srq_num; + cqm_handle->func_capability.srqc_basic_size = + GET_MAX(service_capability->fc_cap.srqc_size, + cqm_handle->func_capability.srqc_basic_size); + cqm_handle->func_capability.lun_number = CQM_LUN_FC_NUM; + cqm_handle->func_capability.lun_basic_size = CQM_LUN_SIZE_8; + cqm_handle->func_capability.taskmap_number = CQM_TASKMAP_FC_NUM; + cqm_handle->func_capability.taskmap_basic_size = PAGE_SIZE; + cqm_handle->func_capability.childc_number += + service_capability->fc_cap.dev_fc_cap.max_child_qpc_num; + cqm_handle->func_capability.childc_basic_size = + GET_MAX(service_capability->fc_cap.child_qpc_size, + cqm_handle->func_capability.childc_basic_size); + cqm_handle->func_capability.pagesize_reorder = CQM_FC_PAGESIZE_ORDER; + + return CQM_SUCCESS; +} + +s32 cqm_service_capability_init(struct cqm_handle_s *cqm_handle, + struct service_cap *service_capability) +{ + cqm_handle->service.has_register = false; + cqm_handle->service.buf_order = 0; + + if (cqm_service_capability_init_for_each( + cqm_handle, + service_capability) == CQM_FAIL) + return CQM_FAIL; + + return CQM_SUCCESS; +} + +/** + * cqm_capability_init - Initialize capability of cqm function and service, + * need to read information from the configuration management module + * @ex_handle: handle of hwdev + */ +s32 cqm_capability_init(void *ex_handle) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct service_cap *service_capability = &handle->cfg_mgmt->svc_cap; + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + (handle->cqm_hdl); + u32 total_function_num = 0; + int err = 0; + + cqm_capability_init_check_ppf(ex_handle, &total_function_num); + + cqm_handle->func_capability.flow_table_based_conn_number = + service_capability->max_connect_num; + cqm_handle->func_capability.flow_table_based_conn_cache_number = + service_capability->max_stick2cache_num; + cqm_info(handle->dev_hdl, "Cap init: cfg max_conn_num 0x%x, max_cache_conn_num 0x%x\n", + cqm_handle->func_capability.flow_table_based_conn_number, + cqm_handle->func_capability.flow_table_based_conn_cache_number); + + cqm_handle->func_capability.qpc_reserved = 0; + cqm_handle->func_capability.mpt_reserved = 0; + cqm_handle->func_capability.qpc_alloc_static = false; + cqm_handle->func_capability.scqc_alloc_static = false; + + cqm_handle->func_capability.l3i_number = CQM_L3I_COMM_NUM; + cqm_handle->func_capability.l3i_basic_size = CQM_L3I_SIZE_8; + + cqm_handle->func_capability.timer_number = CQM_TIMER_ALIGN_SCALE_NUM * + total_function_num; + cqm_handle->func_capability.timer_basic_size = CQM_TIMER_SIZE_32; + + if (cqm_service_capability_init(cqm_handle, service_capability) == + CQM_FAIL) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_service_capability_init)); + err = CQM_FAIL; + goto out; + } + + cqm_test_mode_init(cqm_handle, service_capability); + + cqm_info(handle->dev_hdl, "Cap init: pagesize_reorder %d\n", + cqm_handle->func_capability.pagesize_reorder); + cqm_info(handle->dev_hdl, "Cap init: xid_alloc_mode %d, gpa_check_enable %d\n", + cqm_handle->func_capability.xid_alloc_mode, + cqm_handle->func_capability.gpa_check_enable); + cqm_info(handle->dev_hdl, "Cap init: qpc_alloc_mode %d, scqc_alloc_mode %d\n", + cqm_handle->func_capability.qpc_alloc_static, + cqm_handle->func_capability.scqc_alloc_static); + cqm_info(handle->dev_hdl, "Cap init: hash_number 0x%x\n", + cqm_handle->func_capability.hash_number); + cqm_info(handle->dev_hdl, "Cap init: qpc_number 0x%x, qpc_reserved 0x%x\n", + cqm_handle->func_capability.qpc_number, + cqm_handle->func_capability.qpc_reserved); + cqm_info(handle->dev_hdl, "Cap init: scqc_number 0x%x scqc_reserved 0x%x\n", + cqm_handle->func_capability.scqc_number, + cqm_handle->func_capability.scq_reserved); + cqm_info(handle->dev_hdl, "Cap init: srqc_number 0x%x\n", + cqm_handle->func_capability.srqc_number); + cqm_info(handle->dev_hdl, "Cap init: mpt_number 0x%x, mpt_reserved 0x%x\n", + cqm_handle->func_capability.mpt_number, + cqm_handle->func_capability.mpt_reserved); + cqm_info(handle->dev_hdl, "Cap init: gid_number 0x%x, lun_number 0x%x\n", + cqm_handle->func_capability.gid_number, + cqm_handle->func_capability.lun_number); + cqm_info(handle->dev_hdl, "Cap init: taskmap_number 0x%x, l3i_number 0x%x\n", + cqm_handle->func_capability.taskmap_number, + cqm_handle->func_capability.l3i_number); + cqm_info(handle->dev_hdl, "Cap init: timer_number 0x%x\n", + cqm_handle->func_capability.timer_number); + cqm_info(handle->dev_hdl, "Cap init: xid2cid_number 0x%x, reorder_number 0x%x\n", + cqm_handle->func_capability.xid2cid_number, + cqm_handle->func_capability.reorder_number); + + return CQM_SUCCESS; + +out: + if (cqm_handle->func_attribute.func_type == CQM_PPF) + cqm_handle->func_capability.timer_enable = 0; + + return err; +} + +/** + * cqm_init - Initialize cqm + * @ex_handle: handle of hwdev + */ +s32 cqm_init(void *ex_handle) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + s32 ret = CQM_FAIL; + + CQM_PTR_CHECK_RET(ex_handle, return CQM_FAIL, CQM_PTR_NULL(ex_handle)); + + cqm_handle = (struct cqm_handle_s *)kmalloc(sizeof(struct cqm_handle_s), + GFP_KERNEL | __GFP_ZERO); + CQM_PTR_CHECK_RET(cqm_handle, return CQM_FAIL, + CQM_ALLOC_FAIL(cqm_handle)); + /* Clear memory to prevent other systems' memory from being cleared */ + memset(cqm_handle, 0, sizeof(struct cqm_handle_s)); + + cqm_handle->ex_handle = handle; + cqm_handle->dev = (struct pci_dev *)(handle->pcidev_hdl); + + handle->cqm_hdl = (void *)cqm_handle; + + /* Clear statistics */ + memset(&handle->hw_stats.cqm_stats, 0, sizeof(struct hifc_cqm_stats)); + + /* Read information of vf or pf */ + cqm_handle->func_attribute = handle->hwif->attr; + cqm_info(handle->dev_hdl, "Func init: function type %d\n", + cqm_handle->func_attribute.func_type); + + /* Read ability from configuration management module */ + ret = cqm_capability_init(ex_handle); + if (ret == CQM_FAIL) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_capability_init)); + goto err1; + } + + /* Initialize entries of memory table such as BAT/CLA/bitmap */ + if (cqm_mem_init(ex_handle) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_mem_init)); + goto err1; + } + + /* Initialize event callback */ + if (cqm_event_init(ex_handle) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_event_init)); + goto err2; + } + + /* Initialize doorbell */ + if (cqm_db_init(ex_handle) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_db_init)); + goto err3; + } + + /* The timer bitmap is set directly from the beginning through CQM, + * no longer set/clear the bitmap through ifconfig up/down + */ + if (hifc_func_tmr_bitmap_set(ex_handle, 1) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, "Timer start: enable timer bitmap failed\n"); + goto err5; + } + + return CQM_SUCCESS; + +err5: + cqm_db_uninit(ex_handle); +err3: + cqm_event_uninit(ex_handle); +err2: + cqm_mem_uninit(ex_handle); +err1: + handle->cqm_hdl = NULL; + kfree(cqm_handle); + return CQM_FAIL; +} + +/** + * cqm_uninit - Deinitialize the cqm, and is called once removing a function + * @ex_handle: handle of hwdev + */ +void cqm_uninit(void *ex_handle) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + s32 ret = CQM_FAIL; + + CQM_PTR_CHECK_NO_RET(ex_handle, CQM_PTR_NULL(ex_handle), return); + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + CQM_PTR_CHECK_NO_RET(cqm_handle, CQM_PTR_NULL(cqm_handle), return); + + /* The timer bitmap is set directly from the beginning through CQM, + * no longer set/clear the bitmap through ifconfig up/down + */ + cqm_info(handle->dev_hdl, "Timer stop: disable timer\n"); + if (hifc_func_tmr_bitmap_set(ex_handle, 0) != CQM_SUCCESS) + cqm_err(handle->dev_hdl, "Timer stop: disable timer bitmap failed\n"); + + /* Stopping timer, release the resource + * after a delay of one or two milliseconds + */ + if ((cqm_handle->func_attribute.func_type == CQM_PPF) && + (cqm_handle->func_capability.timer_enable == CQM_TIMER_ENABLE)) { + cqm_info(handle->dev_hdl, "Timer stop: hifc ppf timer stop\n"); + ret = hifc_ppf_tmr_stop(handle); + + if (ret != CQM_SUCCESS) { + cqm_info(handle->dev_hdl, "Timer stop: hifc ppf timer stop, ret=%d\n", + ret); + /* The timer fails to stop + * and does not affect resource release + */ + } + usleep_range(900, 1000); + } + + /* Release hardware doorbell */ + cqm_db_uninit(ex_handle); + + /* Cancel the callback of chipif */ + cqm_event_uninit(ex_handle); + + /* Release all table items + * and require the service to release all objects + */ + cqm_mem_uninit(ex_handle); + + /* Release cqm_handle */ + handle->cqm_hdl = NULL; + kfree(cqm_handle); +} + +/** + * cqm_mem_init - Initialize related memory of cqm, + * including all levels of entries + * @ex_handle: handle of hwdev + */ +s32 cqm_mem_init(void *ex_handle) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + + if (cqm_bat_init(cqm_handle) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_bat_init)); + return CQM_FAIL; + } + + if (cqm_cla_init(cqm_handle) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_cla_init)); + goto err1; + } + + if (cqm_bitmap_init(cqm_handle) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_bitmap_init)); + goto err2; + } + + if (cqm_object_table_init(cqm_handle) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_object_table_init)); + goto err3; + } + + return CQM_SUCCESS; + +err3: + cqm_bitmap_uninit(cqm_handle); +err2: + cqm_cla_uninit(cqm_handle); +err1: + cqm_bat_uninit(cqm_handle); + return CQM_FAIL; +} + +/** + * cqm_mem_uninit - Deinitialize related memory of cqm, + * including all levels of entries + * @ex_handle: handle of hwdev + */ +void cqm_mem_uninit(void *ex_handle) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + + cqm_object_table_uninit(cqm_handle); + cqm_bitmap_uninit(cqm_handle); + cqm_cla_uninit(cqm_handle); + cqm_bat_uninit(cqm_handle); +} + +/** + * cqm_event_init - Initialize the event callback of cqm + * @ex_handle: handle of hwdev + */ +s32 cqm_event_init(void *ex_handle) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + + /* Register ceq and aeq callbacks with chipif */ + if (hifc_aeq_register_swe_cb(ex_handle, + HIFC_STATEFULL_EVENT, + cqm_aeq_callback) != CHIPIF_SUCCESS) { + cqm_err(handle->dev_hdl, "Event: fail to register aeq callback\n"); + return CQM_FAIL; + } + + return CQM_SUCCESS; +} + +/** + * cqm_event_uninit - Deinitialize the event callback of cqm + * @ex_handle: handle of hwdev + */ +void cqm_event_uninit(void *ex_handle) +{ + (void)hifc_aeq_unregister_swe_cb(ex_handle, HIFC_STATEFULL_EVENT); +} + +/** + * cqm_db_addr_alloc - Apply for a page of hardware doorbell and dwqe, + * with the same index, all obtained are physical addresses + * each function has up to 1K + * @ex_handle: handle of hwdev + * @db_addr: the address of doorbell + * @dwqe_addr: the address of dwqe + */ +s32 cqm_db_addr_alloc(void *ex_handle, void __iomem **db_addr, + void __iomem **dwqe_addr) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + + CQM_PTR_CHECK_RET(ex_handle, return CQM_FAIL, CQM_PTR_NULL(ex_handle)); + CQM_PTR_CHECK_RET(db_addr, return CQM_FAIL, CQM_PTR_NULL(db_addr)); + CQM_PTR_CHECK_RET(dwqe_addr, return CQM_FAIL, CQM_PTR_NULL(dwqe_addr)); + + atomic_inc(&handle->hw_stats.cqm_stats.cqm_db_addr_alloc_cnt); + + return hifc_alloc_db_addr(ex_handle, db_addr, dwqe_addr); +} + +/** + * cqm_db_init - Initialize doorbell of cqm + * @ex_handle: handle of hwdev + */ +s32 cqm_db_init(void *ex_handle) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + struct cqm_service_s *service = NULL; + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + + /* Assign hardware doorbell for service */ + service = &cqm_handle->service; + + if (cqm_db_addr_alloc(ex_handle, + &service->hardware_db_vaddr, + &service->dwqe_vaddr) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_db_addr_alloc)); + return CQM_FAIL; + } + + return CQM_SUCCESS; +} + +/** + * cqm_db_addr_free - Release a page of hardware doorbell and dwqe + * @ex_handle: handle of hwdev + * @db_addr: the address of doorbell + * @dwqe_addr: the address of dwqe + */ +void cqm_db_addr_free(void *ex_handle, void __iomem *db_addr, + void __iomem *dwqe_addr) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + + CQM_PTR_CHECK_NO_RET(ex_handle, CQM_PTR_NULL(ex_handle), return); + + atomic_inc(&handle->hw_stats.cqm_stats.cqm_db_addr_free_cnt); + + hifc_free_db_addr(ex_handle, db_addr, dwqe_addr); +} + +/** + * cqm_db_uninit - Deinitialize doorbell of cqm + * @ex_handle: handle of hwdev + */ +void cqm_db_uninit(void *ex_handle) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + struct cqm_service_s *service = NULL; + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + + /* Release hardware doorbell */ + service = &cqm_handle->service; + + cqm_db_addr_free(ex_handle, service->hardware_db_vaddr, + service->dwqe_vaddr); +} + +/** + * cqm_aeq_callback - cqm module callback processing of aeq + * @ex_handle: handle of hwdev + * @event: the input type of event + * @data: the input data + */ +u8 cqm_aeq_callback(void *ex_handle, u8 event, u64 data) +{ +#define CQM_AEQ_BASE_T_FC 48 +#define CQM_AEQ_BASE_T_FCOE 56 + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + struct cqm_service_s *service = NULL; + struct service_register_template_s *service_template = NULL; + u8 event_level = FAULT_LEVEL_MAX; + + CQM_PTR_CHECK_RET(ex_handle, return event_level, + CQM_PTR_NULL(ex_handle)); + + atomic_inc(&handle->hw_stats.cqm_stats.cqm_aeq_callback_cnt[event]); + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + CQM_PTR_CHECK_RET(cqm_handle, return event_level, + CQM_PTR_NULL(cqm_handle)); + + if (event >= (u8)CQM_AEQ_BASE_T_FC && + (event < (u8)CQM_AEQ_BASE_T_FCOE)) { + service = &cqm_handle->service; + service_template = &service->service_template; + + if (!service_template->aeq_callback) { + cqm_err(handle->dev_hdl, "Event: service aeq_callback unregistered\n"); + } else { + service_template->aeq_callback( + service_template->service_handle, event, data); + } + + return event_level; + } + + cqm_err(handle->dev_hdl, CQM_WRONG_VALUE(event)); + return CQM_FAIL; +} + +/** + * cqm_service_register - Service driver registers callback template with cqm + * @ex_handle: handle of hwdev + * @service_template: the template of service registration + */ +s32 cqm_service_register(void *ex_handle, + struct service_register_template_s *service_template) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + struct cqm_service_s *service = NULL; + + CQM_PTR_CHECK_RET(ex_handle, return CQM_FAIL, CQM_PTR_NULL(ex_handle)); + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + CQM_PTR_CHECK_RET(cqm_handle, return CQM_FAIL, + CQM_PTR_NULL(cqm_handle)); + CQM_PTR_CHECK_RET(service_template, return CQM_FAIL, + CQM_PTR_NULL(service_template)); + + service = &cqm_handle->service; + + if (service->has_register == true) { + cqm_err(handle->dev_hdl, "Service register: service has registered\n"); + return CQM_FAIL; + } + + service->has_register = true; + (void)memcpy((void *)(&service->service_template), + (void *)service_template, + sizeof(struct service_register_template_s)); + + return CQM_SUCCESS; +} + +/** + * cqm_service_unregister - Service-driven cancellation to CQM + * @ex_handle: handle of hwdev + * @service_type: the type of service module + */ +void cqm_service_unregister(void *ex_handle) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + struct cqm_service_s *service = NULL; + + CQM_PTR_CHECK_NO_RET(ex_handle, CQM_PTR_NULL(ex_handle), return); + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + CQM_PTR_CHECK_NO_RET(cqm_handle, CQM_PTR_NULL(cqm_handle), return); + + service = &cqm_handle->service; + + service->has_register = false; + memset(&service->service_template, 0, + sizeof(struct service_register_template_s)); +} + +/** + * cqm_cmd_alloc - Apply for a cmd buffer, the buffer size is fixed at 2K, + * the buffer content is not cleared, but the service needs to be cleared + * @ex_handle: handle of hwdev + */ +struct cqm_cmd_buf_s *cqm_cmd_alloc(void *ex_handle) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + + CQM_PTR_CHECK_RET(ex_handle, return NULL, CQM_PTR_NULL(ex_handle)); + + atomic_inc(&handle->hw_stats.cqm_stats.cqm_cmd_alloc_cnt); + + return (struct cqm_cmd_buf_s *)hifc_alloc_cmd_buf(ex_handle); +} + +/** + * cqm_cmd_free - Free a cmd buffer + * @ex_handle: handle of hwdev + * @cmd_buf: the cmd buffer which needs freeing memory for + */ +void cqm_cmd_free(void *ex_handle, struct cqm_cmd_buf_s *cmd_buf) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + + CQM_PTR_CHECK_NO_RET(ex_handle, CQM_PTR_NULL(ex_handle), return); + CQM_PTR_CHECK_NO_RET(cmd_buf, CQM_PTR_NULL(cmd_buf), return); + CQM_PTR_CHECK_NO_RET(cmd_buf->buf, CQM_PTR_NULL(buf), return); + + atomic_inc(&handle->hw_stats.cqm_stats.cqm_cmd_free_cnt); + + hifc_free_cmd_buf(ex_handle, (struct hifc_cmd_buf *)cmd_buf); +} + +/** + * cqm_send_cmd_box - Send a cmd in box mode, + * the interface will hang the completed amount, causing sleep + * @ex_handle: handle of hwdev + * @ack_type: the type of ack + * @mod: the mode of cqm send + * @cmd: the input cmd + * @buf_in: the input buffer of cqm_cmd + * @buf_out: the output buffer of cqm_cmd + * @timeout: exceeding the time limit will cause sleep + */ +s32 cqm_send_cmd_box(void *ex_handle, u8 ack_type, u8 mod, u8 cmd, + struct cqm_cmd_buf_s *buf_in, + struct cqm_cmd_buf_s *buf_out, u32 timeout) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + + CQM_PTR_CHECK_RET(ex_handle, return CQM_FAIL, CQM_PTR_NULL(ex_handle)); + CQM_PTR_CHECK_RET(buf_in, return CQM_FAIL, CQM_PTR_NULL(buf_in)); + CQM_PTR_CHECK_RET(buf_in->buf, return CQM_FAIL, CQM_PTR_NULL(buf)); + + atomic_inc(&handle->hw_stats.cqm_stats.cqm_send_cmd_box_cnt); + + return hifc_cmdq_detail_resp(ex_handle, ack_type, mod, cmd, + (struct hifc_cmd_buf *)buf_in, + (struct hifc_cmd_buf *)buf_out, timeout); +} + +/** + * cqm_ring_hardware_db - Knock hardware doorbell + * @ex_handle: handle of hwdev + * @service_type: each kernel mode will be allocated a page of hardware doorbell + * @db_count: PI exceeding 64b in doorbell[7:0] + * @db: doorbell content, organized by the business, + * if there is a small-end conversion, the business needs to be completed + */ +s32 cqm_ring_hardware_db(void *ex_handle, u32 service_type, u8 db_count, u64 db) +{ + struct hifc_hwdev *handle; + struct cqm_handle_s *cqm_handle; + struct cqm_service_s *service; + + handle = (struct hifc_hwdev *)ex_handle; + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + service = &cqm_handle->service; + + /* Write all before the doorbell */ + wmb(); + *((u64 *)service->hardware_db_vaddr + db_count) = db; + + return CQM_SUCCESS; +} diff --git a/drivers/scsi/huawei/hifc/hifc_cqm_main.h b/drivers/scsi/huawei/hifc/hifc_cqm_main.h new file mode 100644 index 000000000000..70b0c9ae0609 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_cqm_main.h @@ -0,0 +1,366 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ +#ifndef __CQM_MAIN_H__ +#define __CQM_MAIN_H__ + +#define CHIPIF_SUCCESS 0 +#define CQM_TIMER_ENABLE 1 + +enum cqm_object_type_e { + CQM_OBJECT_ROOT_CTX = 0, + CQM_OBJECT_SERVICE_CTX, + CQM_OBJECT_NONRDMA_EMBEDDED_RQ = 10, + CQM_OBJECT_NONRDMA_EMBEDDED_SQ, + CQM_OBJECT_NONRDMA_SRQ, + CQM_OBJECT_NONRDMA_EMBEDDED_CQ, + CQM_OBJECT_NONRDMA_SCQ, +}; + +struct service_register_template_s { + u32 service_type; + u32 srq_ctx_size; /* srq,scq context_size config */ + u32 scq_ctx_size; + void *service_handle; /* ceq/aeq callback fun */ + + void (*aeq_callback)(void *service_handle, u8 event_type, u64 val); +}; + +struct cqm_service_s { + bool has_register; + void __iomem *hardware_db_vaddr; + void __iomem *dwqe_vaddr; + u32 buf_order; /* size of per buf 2^buf_order page */ + struct service_register_template_s service_template; +}; + +struct cqm_func_capability_s { + bool qpc_alloc_static; /* Allocate qpc memory dynamicly/statically */ + bool scqc_alloc_static; + u8 timer_enable; /* whether timer enable */ + + u32 flow_table_based_conn_number; + u32 flow_table_based_conn_cache_number; /* Maximum number in cache */ + u32 bloomfilter_length; /* Bloomfilter table size, aligned by 64B */ + /* The starting position of the bloomfilter table in the cache */ + u32 bloomfilter_addr; + u32 qpc_reserved; /* Reserved bits in bitmap */ + u32 mpt_reserved; /* There are also reserved bits in ROCE/IWARP mpt */ + /* All basic_size must be 2^n aligned */ + u32 hash_number; + /* Number of hash buckets, BAT table fill size is + * aligned with 64 buckets, at least 64 + */ + u32 hash_basic_size; + /* Hash bucket size is 64B, including 5 valid + * entries and 1 nxt_entry + */ + u32 qpc_number; + u32 qpc_basic_size; + + /* Note: for cqm specail test */ + u32 pagesize_reorder; + bool xid_alloc_mode; + bool gpa_check_enable; + u32 scq_reserved; + + u32 mpt_number; + u32 mpt_basic_size; + u32 scqc_number; + u32 scqc_basic_size; + u32 srqc_number; + u32 srqc_basic_size; + + u32 gid_number; + u32 gid_basic_size; + u32 lun_number; + u32 lun_basic_size; + u32 taskmap_number; + u32 taskmap_basic_size; + u32 l3i_number; + u32 l3i_basic_size; + u32 childc_number; + u32 childc_basic_size; + u32 child_qpc_id_start; /* Child ctx of FC is global addressing */ + /* The maximum number of child ctx in + * chip is 8096 + */ + u32 childc_number_all_function; + + u32 timer_number; + u32 timer_basic_size; + u32 xid2cid_number; + u32 xid2cid_basic_size; + u32 reorder_number; + u32 reorder_basic_size; +}; + +#define CQM_PF TYPE_PF +#define CQM_PPF TYPE_PPF +#define CQM_BAT_ENTRY_MAX (16) +#define CQM_BAT_ENTRY_SIZE (16) + +struct cqm_buf_list_s { + void *va; + dma_addr_t pa; + u32 refcount; +}; + +struct cqm_buf_s { + struct cqm_buf_list_s *buf_list; + struct cqm_buf_list_s direct; + u32 page_number; /* page_number=2^n buf_number */ + u32 buf_number; /* buf_list node count */ + u32 buf_size; /* buf_size=2^n PAGE_SIZE */ +}; + +struct cqm_bitmap_s { + ulong *table; + u32 max_num; + u32 last; + /* The index that cannot be allocated is reserved in the front */ + u32 reserved_top; + /* Lock for bitmap allocation */ + spinlock_t lock; +}; + +struct completion; +struct cqm_object_s { + u32 service_type; + u32 object_type; /* context,queue,mpt,mtt etc */ + u32 object_size; + /* for queue, ctx, MPT Byte */ + atomic_t refcount; + struct completion free; + void *cqm_handle; +}; + +struct cqm_object_table_s { + struct cqm_object_s **table; + u32 max_num; + rwlock_t lock; +}; + +struct cqm_cla_table_s { + u32 type; + u32 max_buffer_size; + u32 obj_num; + bool alloc_static; /* Whether the buffer is statically allocated */ + u32 cla_lvl; + /* The value of x calculated by the cacheline, used for chip */ + u32 cacheline_x; + /* The value of y calculated by the cacheline, used for chip */ + u32 cacheline_y; + /* The value of z calculated by the cacheline, used for chip */ + u32 cacheline_z; + /* The value of x calculated by the obj_size, used for software */ + u32 x; + /* The value of y calculated by the obj_size, used for software */ + u32 y; + /* The value of z calculated by the obj_size, used for software */ + u32 z; + struct cqm_buf_s cla_x_buf; + struct cqm_buf_s cla_y_buf; + struct cqm_buf_s cla_z_buf; + u32 trunk_order;/* A continuous physical page contains 2^order pages */ + u32 obj_size; + /* Lock for cla buffer allocation and free */ + struct mutex lock; + struct cqm_bitmap_s bitmap; + /* The association mapping table of index and object */ + struct cqm_object_table_s obj_table; +}; + +typedef void (*init_handler)(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap); + +struct cqm_cla_entry_init_s { + u32 type; + init_handler cqm_cla_init_handler; +}; + +struct cqm_bat_table_s { + u32 bat_entry_type[CQM_BAT_ENTRY_MAX]; + u8 bat[CQM_BAT_ENTRY_MAX * CQM_BAT_ENTRY_SIZE]; + struct cqm_cla_table_s entry[CQM_BAT_ENTRY_MAX]; + u32 bat_size; +}; + +struct cqm_handle_s { + struct hifc_hwdev *ex_handle; + struct pci_dev *dev; + struct hifc_func_attr func_attribute; /* vf or pf */ + struct cqm_func_capability_s func_capability; + struct cqm_service_s service; + struct cqm_bat_table_s bat_table; + + struct list_head node; +}; + +struct cqm_cmd_buf_s { + void *buf; + dma_addr_t dma; + u16 size; +}; + +struct cqm_queue_header_s { + u64 doorbell_record; + u64 ci_record; + u64 rsv1; /* the share area bettween driver and ucode */ + u64 rsv2; /* the share area bettween driver and ucode*/ +}; + +struct cqm_queue_s { + struct cqm_object_s object; + u32 index; /* embedded queue QP has not index, SRQ and SCQ have */ + void *priv; /* service driver private info */ + u32 current_q_doorbell; + u32 current_q_room; + /* nonrdma: only select q_room_buf_1 for q_room_buf */ + struct cqm_buf_s q_room_buf_1; + struct cqm_buf_s q_room_buf_2; + struct cqm_queue_header_s *q_header_vaddr; + dma_addr_t q_header_paddr; + u8 *q_ctx_vaddr; /* SRQ and SCQ ctx space */ + dma_addr_t q_ctx_paddr; + u32 valid_wqe_num; + /*add for srq*/ + u8 *tail_container; + u8 *head_container; + u8 queue_link_mode; /*link,ring */ +}; + +struct cqm_nonrdma_qinfo_s { + struct cqm_queue_s common; + u32 wqe_size; + /* The number of wqe contained in each buf (excluding link wqe), + * For srq, it is the number of wqe contained in 1 container + */ + u32 wqe_per_buf; + u32 q_ctx_size; + /* When different services use different sizes of ctx, a large ctx will + * occupy multiple consecutive indexes of the bitmap + */ + u32 index_count; + u32 container_size; +}; + +/* service context, QPC, mpt */ +struct cqm_qpc_mpt_s { + struct cqm_object_s object; + u32 xid; + dma_addr_t paddr; + void *priv; /* service driver private info */ + u8 *vaddr; +}; + +struct cqm_qpc_mpt_info_s { + struct cqm_qpc_mpt_s common; + /* When different services use different sizes of QPC, large QPC/mpt + * will occupy multiple consecutive indexes of the bitmap + */ + u32 index_count; +}; + +#define CQM_ADDR_COMBINE(high_addr, low_addr) \ + ((((dma_addr_t)(high_addr)) << 32) + ((dma_addr_t)(low_addr))) +#define CQM_ADDR_HI(addr) ((u32)((u64)(addr) >> 32)) +#define CQM_ADDR_LW(addr) ((u32)((u64)(addr) & 0xffffffff)) +#define CQM_HASH_BUCKET_SIZE_64 (64) +#define CQM_LUN_SIZE_8 (8) +#define CQM_L3I_SIZE_8 (8) +#define CQM_TIMER_SIZE_32 (32) +#define CQM_LUN_FC_NUM (64) +#define CQM_TASKMAP_FC_NUM (4) +#define CQM_L3I_COMM_NUM (64) +#define CQM_TIMER_SCALE_NUM (2*1024) +#define CQM_TIMER_ALIGN_WHEEL_NUM (8) +#define CQM_TIMER_ALIGN_SCALE_NUM \ + (CQM_TIMER_SCALE_NUM*CQM_TIMER_ALIGN_WHEEL_NUM) +#define CQM_FC_PAGESIZE_ORDER (0) +#define CQM_QHEAD_ALIGN_ORDER (6) + +s32 cqm_mem_init(void *ex_handle); +void cqm_mem_uninit(void *ex_handle); +s32 cqm_event_init(void *ex_handle); +void cqm_event_uninit(void *ex_handle); +s32 cqm_db_init(void *ex_handle); +void cqm_db_uninit(void *ex_handle); +s32 cqm_init(void *ex_handle); +void cqm_uninit(void *ex_handle); +s32 cqm_service_register(void *ex_handle, + struct service_register_template_s *service_template); +void cqm_service_unregister(void *ex_handle); +s32 cqm_ring_hardware_db(void *ex_handle, + u32 service_type, + u8 db_count, u64 db); +s32 cqm_send_cmd_box(void *ex_handle, u8 ack_type, u8 mod, u8 cmd, + struct cqm_cmd_buf_s *buf_in, + struct cqm_cmd_buf_s *buf_out, + u32 timeout); +u8 cqm_aeq_callback(void *ex_handle, u8 event, u64 data); +void cqm_object_delete(struct cqm_object_s *object); +struct cqm_cmd_buf_s *cqm_cmd_alloc(void *ex_handle); +void cqm_cmd_free(void *ex_handle, struct cqm_cmd_buf_s *cmd_buf); +struct cqm_queue_s *cqm_object_fc_srq_create( + void *ex_handle, + enum cqm_object_type_e object_type, + u32 wqe_number, + u32 wqe_size, + void *object_priv); +struct cqm_qpc_mpt_s *cqm_object_qpc_mpt_create( + void *ex_handle, + enum cqm_object_type_e object_type, + u32 object_size, + void *object_priv, + u32 index); +struct cqm_queue_s *cqm_object_nonrdma_queue_create( + void *ex_handle, + enum cqm_object_type_e object_type, + u32 wqe_number, + u32 wqe_size, + void *object_priv); + +#define CQM_PTR_NULL(x) "%s: "#x" is null\n", __func__ +#define CQM_ALLOC_FAIL(x) "%s: "#x" alloc fail\n", __func__ +#define CQM_MAP_FAIL(x) "%s: "#x" map fail\n", __func__ +#define CQM_FUNCTION_FAIL(x) "%s: "#x" return failure\n", __func__ +#define CQM_WRONG_VALUE(x) "%s: "#x" %u is wrong\n", __func__, (u32)x + +#define cqm_err(dev, format, ...) \ + dev_err(dev, "[CQM]"format, ##__VA_ARGS__) +#define cqm_warn(dev, format, ...) \ + dev_warn(dev, "[CQM]"format, ##__VA_ARGS__) +#define cqm_notice(dev, format, ...) \ + dev_notice(dev, "[CQM]"format, ##__VA_ARGS__) +#define cqm_info(dev, format, ...) \ + dev_info(dev, "[CQM]"format, ##__VA_ARGS__) +#define cqm_dbg(format, ...) + +#define CQM_PTR_CHECK_RET(ptr, ret, desc) \ + do {\ + if (unlikely(NULL == (ptr))) {\ + pr_err("[CQM]"desc);\ + ret; \ + } \ + } while (0) + +#define CQM_PTR_CHECK_NO_RET(ptr, desc, ret) \ + do {\ + if (unlikely((ptr) == NULL)) {\ + pr_err("[CQM]"desc);\ + ret; \ + } \ + } while (0) +#define CQM_CHECK_EQUAL_RET(dev_hdl, actual, expect, ret, desc) \ + do {\ + if (unlikely((expect) != (actual))) {\ + cqm_err(dev_hdl, desc);\ + ret; \ + } \ + } while (0) + +#endif /* __CQM_MAIN_H__ */ diff --git a/drivers/scsi/huawei/hifc/hifc_cqm_object.c b/drivers/scsi/huawei/hifc/hifc_cqm_object.c new file mode 100644 index 000000000000..406b13f92e64 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_cqm_object.c @@ -0,0 +1,3599 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#include <linux/types.h> +#include <linux/sched.h> +#include <linux/pci.h> +#include <linux/module.h> +#include <linux/vmalloc.h> +#include <linux/device.h> +#include <linux/gfp.h> +#include <linux/mm.h> + +#include "hifc_knl_adp.h" +#include "hifc_hw.h" +#include "hifc_hwdev.h" +#include "hifc_hwif.h" +#include "hifc_api_cmd.h" +#include "hifc_mgmt.h" +#include "hifc_cfg.h" +#include "hifc_cqm_object.h" +#include "hifc_cqm_main.h" +#define common_section + +#define CQM_MOD_CQM 8 +#define CQM_HARDWARE_DOORBELL 1 +/** + * cqm_swab64 - Convert a memory block to another endian by 8 byte basis + * @addr: start address of the memory block + * @cnt: the number of 8 byte basis in the memory block + */ +void cqm_swab64(u8 *addr, u32 cnt) +{ + u32 i = 0; + u64 *temp = (u64 *)addr; + u64 value = 0; + + for (i = 0; i < cnt; i++) { + value = __swab64(*temp); + *temp = value; + temp++; + } +} + +/** + * cqm_swab32 - Convert a memory block to another endian by 4 byte basis + * @addr: start address of the memory block + * @cnt: the number of 4 byte basis in the memory block + */ +void cqm_swab32(u8 *addr, u32 cnt) +{ + u32 i = 0; + u32 *temp = (u32 *)addr; + u32 value = 0; + + for (i = 0; i < cnt; i++) { + value = __swab32(*temp); + *temp = value; + temp++; + } +} + +/** + * cqm_shift - Find the base logarithm of two + * @data: the input data + */ +s32 cqm_shift(u32 data) +{ + s32 shift = -1; + + do { + data >>= 1; + shift++; + } while (data); + + return shift; +} + +/** + * cqm_check_align - Check whether the data is aligned as the base of 2^n + * @data: the input data + */ +bool cqm_check_align(u32 data) +{ + if (data == 0) + return false; + + do { + /* If data can be divided exactly by 2, + * it right shifts one bit + */ + if ((data & 0x1) == 0) { + data >>= 1; + } else { + /* If data can not be divided exactly by 2 + * it is not the base of 2^n,return false + */ + return false; + } + } while (data != 1); + + return true; +} + +/** + * cqm_kmalloc_align - Alloc memory whose start address is aligned as the basis + * of 2^n + * @size: the size of memory allocated + * @flags: the type of memory allocated + * @align_order: the basis for aligning + */ +static void *cqm_kmalloc_align(size_t size, gfp_t flags, u16 align_order) +{ + void *orig_addr = NULL; + void *align_addr = NULL; + void *index_addr = NULL; + + orig_addr = kmalloc(size + ((u64)1 << align_order) + sizeof(void *), + flags); + if (!orig_addr) + return NULL; + + index_addr = (void *)((char *)orig_addr + sizeof(void *)); + align_addr = (void *)((((u64)index_addr + + ((u64)1 << align_order) - 1) >> align_order) << align_order); + + /* Record the original memory address for memory release. */ + index_addr = (void *)((char *)align_addr - sizeof(void *)); + *(void **)index_addr = orig_addr; + + cqm_dbg("allocate %lu bytes aligned address: %p, original address: %p\n", + size, align_addr, orig_addr); + + return align_addr; +} + +/** + * cqm_kfree_align - Free memory whose start address is aligned as the basis of + * 2^n + * @addr: aligned address which would be free + */ +static void cqm_kfree_align(void *addr) +{ + void *index_addr = NULL; + + /* Release original memory address */ + index_addr = (void *)((char *)addr - sizeof(void *)); + + cqm_dbg("free aligned address: %p, original address: %p\n", + addr, *(void **)index_addr); + + kfree(*(void **)index_addr); +} + +/** + * cqm_buf_alloc_page - Alloc total pages memory for buffers + * @cqm_handle: handle of cqm + * @buf: the buffer which needs allocating memory for + */ +s32 cqm_buf_alloc_page(struct cqm_handle_s *cqm_handle, struct cqm_buf_s *buf) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + u32 order = 0; + void *va = NULL; + s32 i = 0; + + order = get_order(buf->buf_size); + + /*Here to allocate for every buffer's page for non-ovs*/ + for (i = 0; i < (s32)buf->buf_number; i++) { + va = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, order); + if (!va) { + cqm_err(handle->dev_hdl, CQM_ALLOC_FAIL(buf_page)); + break; + } + /* Pages should be initialized to 0 after applied + * especially related to the hash table + */ + memset(va, 0, buf->buf_size); + buf->buf_list[i].va = va; + } + + if (i != buf->buf_number) { + i--; + for (; i >= 0; i--) { + free_pages((ulong)(buf->buf_list[i].va), order); + buf->buf_list[i].va = NULL; + } + return CQM_FAIL; + } + + return CQM_SUCCESS; +} + +/** + * cqm_buf_alloc_map - Buffer pci mapping + * @cqm_handle: handle of cqm + * @buf: the buffer which needs map + */ +s32 cqm_buf_alloc_map(struct cqm_handle_s *cqm_handle, struct cqm_buf_s *buf) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct pci_dev *dev = cqm_handle->dev; + s32 i = 0; + void *va = NULL; + + for (i = 0; i < (s32)buf->buf_number; i++) { + va = buf->buf_list[i].va; + buf->buf_list[i].pa = + pci_map_single(dev, va, buf->buf_size, + PCI_DMA_BIDIRECTIONAL); + if (pci_dma_mapping_error(dev, buf->buf_list[i].pa)) { + cqm_err(handle->dev_hdl, CQM_MAP_FAIL(buf_list)); + break; + } + } + + if (i != buf->buf_number) { + i--; + for (; i >= 0; i--) { + pci_unmap_single(dev, buf->buf_list[i].pa, + buf->buf_size, PCI_DMA_BIDIRECTIONAL); + } + return CQM_FAIL; + } + + return CQM_SUCCESS; +} + +/** + * cqm_buf_alloc_direct - Buffer pci direct remapping + * @cqm_handle: handle of cqm + * @buf: the buffer which needs remap + */ +s32 cqm_buf_alloc_direct(struct cqm_handle_s *cqm_handle, + struct cqm_buf_s *buf, bool direct) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct page **pages = NULL; + u32 order = 0; + u32 i = 0; + u32 j = 0; + + order = get_order(buf->buf_size); + + if (direct == true) { + pages = (struct page **) + vmalloc(sizeof(struct page *) * buf->page_number); + if (!pages) { + cqm_err(handle->dev_hdl, CQM_ALLOC_FAIL(pages)); + return CQM_FAIL; + } + + for (i = 0; i < buf->buf_number; i++) { + for (j = 0; j < ((u32)1 << order); j++) { + pages[(i << order) + j] = (struct page *) + (void *)virt_to_page( + (u8 *)(buf->buf_list[i].va) + + (PAGE_SIZE * j)); + } + } + + /*lint -save -e648 + *Shield alarm for kernel functions' vmapping + */ + buf->direct.va = vmap(pages, buf->page_number, + VM_MAP, PAGE_KERNEL); + /*lint -restore*/ + vfree(pages); + if (!buf->direct.va) { + cqm_err(handle->dev_hdl, CQM_MAP_FAIL(buf->direct.va)); + return CQM_FAIL; + } + } else { + buf->direct.va = NULL; + } + + return CQM_SUCCESS; +} + +/** + * cqm_buf_alloc - Allocate for buffer and dma for the struct cqm_buf_s + * @cqm_handle: handle of cqm + * @buf: the buffer which needs allocating memory for and dma + */ +s32 cqm_buf_alloc(struct cqm_handle_s *cqm_handle, + struct cqm_buf_s *buf, bool direct) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct pci_dev *dev = cqm_handle->dev; + u32 order = 0; + s32 i = 0; + + order = get_order(buf->buf_size); + + /* Allocate for the descriptor space of buffer lists */ + buf->buf_list = (struct cqm_buf_list_s *) + vmalloc(buf->buf_number * + sizeof(struct cqm_buf_list_s)); + + CQM_PTR_CHECK_RET(buf->buf_list, return CQM_FAIL, + CQM_ALLOC_FAIL(buf_list)); + memset(buf->buf_list, 0, + buf->buf_number * sizeof(struct cqm_buf_list_s)); + + /* Allocate for every buffer's page */ + if (cqm_buf_alloc_page(cqm_handle, buf) == CQM_FAIL) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_buf_alloc_page)); + goto err1; + } + + /* Buffer pci remapping */ + if (cqm_buf_alloc_map(cqm_handle, buf) == CQM_FAIL) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_buf_alloc_map)); + goto err2; + } + + /* Buffer pci mapping */ + if (cqm_buf_alloc_direct(cqm_handle, buf, direct) == CQM_FAIL) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_buf_alloc_direct)); + goto err3; + } + + return CQM_SUCCESS; + +err3: + for (i = 0; i < (s32)buf->buf_number; i++) { + pci_unmap_single(dev, buf->buf_list[i].pa, buf->buf_size, + PCI_DMA_BIDIRECTIONAL); + } +err2: + for (i = 0; i < (s32)buf->buf_number; i++) { + free_pages((ulong)(buf->buf_list[i].va), order); + buf->buf_list[i].va = NULL; + } +err1: + vfree(buf->buf_list); + buf->buf_list = NULL; + return CQM_FAIL; +} + +/** + * cqm_cla_cache_invalid - Set the chip logical address cache invalid + * @cqm_handle: handle of cqm + * @gpa: global physical address + * @cache_size: chip cache size + */ +s32 cqm_cla_cache_invalid(struct cqm_handle_s *cqm_handle, dma_addr_t gpa, + u32 cache_size) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_cmd_buf_s *buf_in = NULL; + struct cqm_cla_cache_invalid_cmd_s *cmd = NULL; + s32 ret = CQM_FAIL; + + buf_in = cqm_cmd_alloc((void *)(cqm_handle->ex_handle)); + CQM_PTR_CHECK_RET(buf_in, return CQM_FAIL, + CQM_ALLOC_FAIL(buf_in)); + buf_in->size = sizeof(struct cqm_cla_cache_invalid_cmd_s); + + /* Fill command format, and turn into big endian */ + cmd = (struct cqm_cla_cache_invalid_cmd_s *)(buf_in->buf); + cmd->cache_size = cache_size; + cmd->gpa_h = CQM_ADDR_HI(gpa); + cmd->gpa_l = CQM_ADDR_LW(gpa); + + cqm_swab32((u8 *)cmd, + (sizeof(struct cqm_cla_cache_invalid_cmd_s) >> 2)); + + /* cmdq send a cmd */ + ret = cqm_send_cmd_box((void *)(cqm_handle->ex_handle), + CQM_CMD_ACK_TYPE_CMDQ, + CQM_MOD_CQM, CQM_CMD_T_CLA_CACHE_INVALID, + buf_in, NULL, CQM_CMD_TIMEOUT); + if (ret != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_send_cmd_box)); + cqm_err(handle->dev_hdl, "Cla cache invalid: cqm_send_cmd_box_ret=%d\n", + ret); + cqm_err(handle->dev_hdl, "Cla cache invalid: cla_cache_invalid_cmd: 0x%x 0x%x 0x%x\n", + cmd->gpa_h, cmd->gpa_l, cmd->cache_size); + } + + cqm_cmd_free((void *)(cqm_handle->ex_handle), buf_in); + return ret; +} + +/** + * cqm_buf_free - Free buffer space and dma for the struct cqm_buf_s + * @buf: the buffer which needs freeing memory for + * @dev: specific pci device + */ +void cqm_buf_free(struct cqm_buf_s *buf, struct pci_dev *dev) +{ + u32 order = 0; + s32 i = 0; + + order = get_order(buf->buf_size); + + if (buf->direct.va) { + vunmap(buf->direct.va); + buf->direct.va = NULL; + } + + if (buf->buf_list) { + for (i = 0; i < (s32)(buf->buf_number); i++) { + if (buf->buf_list[i].va) { + pci_unmap_single(dev, buf->buf_list[i].pa, + buf->buf_size, + PCI_DMA_BIDIRECTIONAL); + free_pages((ulong)(buf->buf_list[i].va), order); + buf->buf_list[i].va = NULL; + } + } + + vfree(buf->buf_list); + buf->buf_list = NULL; + } +} + +/** + * __free_cache_inv - Free cache and make buffer list invalid + * @cqm_handle: handle of cqm + * @buf: the buffer which needs freeing memory for + * @inv_flag: invalid or not + * @order:the basis for aligning + * @buf_idx:buffer index + */ +static void __free_cache_inv(struct cqm_handle_s *cqm_handle, + struct cqm_buf_s *buf, s32 *inv_flag, + u32 order, s32 buf_idx) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + + if (handle->chip_present_flag) { + *inv_flag = cqm_cla_cache_invalid(cqm_handle, + buf->buf_list[buf_idx].pa, + PAGE_SIZE << order); + if (*inv_flag != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, "Buffer free: fail to invalid buf_list pa cache, inv_flag=%d\n", + *inv_flag); + } + } + + pci_unmap_single(cqm_handle->dev, buf->buf_list[buf_idx].pa, + buf->buf_size, PCI_DMA_BIDIRECTIONAL); + + free_pages((unsigned long)(buf->buf_list[buf_idx].va), order); + + buf->buf_list[buf_idx].va = NULL; +} + +/** + * cqm_buf_free_cache_inv - Free cache and make buffer list invalid + * @cqm_handle: handle of cqm + * @buf: the buffer which needs freeing memory for + * @inv_flag: invalid or not + */ +void cqm_buf_free_cache_inv(struct cqm_handle_s *cqm_handle, + struct cqm_buf_s *buf, s32 *inv_flag) +{ + u32 order = 0; + s32 i = 0; + + order = get_order(buf->buf_size); + + if (buf->direct.va) { + vunmap(buf->direct.va); + buf->direct.va = NULL; + } + + if (buf->buf_list) { + for (i = 0; i < (s32)(buf->buf_number); i++) { + if (buf->buf_list[i].va) { + __free_cache_inv(cqm_handle, buf, + inv_flag, order, i); + } + } + + vfree(buf->buf_list); + buf->buf_list = NULL; + } +} + +#define bat_cla_section + +/** + * cqm_bat_update - Send cmds to the tile to update the BAT table through cmdq + * @cqm_handle: cqm handle + * Return: 0 - success, negative - failure + */ +s32 cqm_bat_update(struct cqm_handle_s *cqm_handle) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_cmd_buf_s *buf_in = NULL; + s32 ret = CQM_FAIL; + struct cqm_bat_update_cmd_s *bat_update_cmd = NULL; + + /* Allocate a cmd and fill */ + buf_in = cqm_cmd_alloc((void *)(cqm_handle->ex_handle)); + CQM_PTR_CHECK_RET(buf_in, return CQM_FAIL, CQM_ALLOC_FAIL(buf_in)); + buf_in->size = sizeof(struct cqm_bat_update_cmd_s); + + bat_update_cmd = (struct cqm_bat_update_cmd_s *)(buf_in->buf); + bat_update_cmd->byte_len = cqm_handle->bat_table.bat_size; + bat_update_cmd->offset = 0; + memcpy(bat_update_cmd->data, cqm_handle->bat_table.bat, + bat_update_cmd->byte_len); + + /*Big endian conversion*/ + cqm_swab32((u8 *)bat_update_cmd, + sizeof(struct cqm_bat_update_cmd_s) >> 2); + + /* send a cmd */ + ret = cqm_send_cmd_box((void *)(cqm_handle->ex_handle), + CQM_CMD_ACK_TYPE_CMDQ, CQM_MOD_CQM, + CQM_CMD_T_BAT_UPDATE, buf_in, + NULL, CQM_CMD_TIMEOUT); + if (ret != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_send_cmd_box)); + cqm_err(handle->dev_hdl, "Bat update: send_cmd_box ret=%d\n", + ret); + cqm_cmd_free((void *)(cqm_handle->ex_handle), buf_in); + return CQM_FAIL; + } + + /* Free a cmd */ + cqm_cmd_free((void *)(cqm_handle->ex_handle), buf_in); + + return CQM_SUCCESS; +} + +s32 cqm_bat_init_ft(struct cqm_handle_s *cqm_handle, + struct cqm_bat_table_s *bat_table, + enum func_type function_type) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + + if (function_type == CQM_PF || function_type == CQM_PPF) { + bat_table->bat_entry_type[0] = CQM_BAT_ENTRY_T_CFG; + bat_table->bat_entry_type[1] = CQM_BAT_ENTRY_T_HASH; + bat_table->bat_entry_type[2] = CQM_BAT_ENTRY_T_QPC; + bat_table->bat_entry_type[3] = CQM_BAT_ENTRY_T_SCQC; + bat_table->bat_entry_type[4] = CQM_BAT_ENTRY_T_LUN; + bat_table->bat_entry_type[5] = CQM_BAT_ENTRY_T_TASKMAP; + bat_table->bat_entry_type[6] = CQM_BAT_ENTRY_T_L3I; + bat_table->bat_entry_type[7] = CQM_BAT_ENTRY_T_CHILDC; + bat_table->bat_entry_type[8] = CQM_BAT_ENTRY_T_TIMER; + bat_table->bat_entry_type[9] = CQM_BAT_ENTRY_T_XID2CID; + bat_table->bat_entry_type[10] = CQM_BAT_ENTRY_T_REORDER; + bat_table->bat_size = CQM_BAT_SIZE_FT_PF; + } else { + cqm_err(handle->dev_hdl, CQM_WRONG_VALUE(function_type)); + return CQM_FAIL; + } + + return CQM_SUCCESS; +} + +/** + * cqm_bat_init - Initialize the BAT table, only select the items to be + * initialized and arrange the entry order, the content of the BAT table entry + * needs to be filled after the CLA allocation + * @cqm_handle: cqm handle + * Return: 0 - success, negative - failure + */ +s32 cqm_bat_init(struct cqm_handle_s *cqm_handle) +{ + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + u32 i = 0; + + memset(bat_table, 0, sizeof(struct cqm_bat_table_s)); + + /* Initialize the type of each bat entry */ + for (i = 0; i < CQM_BAT_ENTRY_MAX; i++) + bat_table->bat_entry_type[i] = CQM_BAT_ENTRY_T_INVALID; + + if (cqm_bat_init_ft(cqm_handle, bat_table, + cqm_handle->func_attribute.func_type) == CQM_FAIL) { + return CQM_FAIL; + } + + return CQM_SUCCESS; +} + +/** + * cqm_bat_uninit - Deinitialize BAT table + * @cqm_handle: cqm handle + */ +void cqm_bat_uninit(struct cqm_handle_s *cqm_handle) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + u32 i = 0; + + for (i = 0; i < CQM_BAT_ENTRY_MAX; i++) + bat_table->bat_entry_type[i] = CQM_BAT_ENTRY_T_INVALID; + + memset(bat_table->bat, 0, CQM_BAT_ENTRY_MAX * CQM_BAT_ENTRY_SIZE); + + /* Notify the chip to refresh the BAT table */ + if (cqm_bat_update(cqm_handle) != CQM_SUCCESS) + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_bat_update)); +} + +static void cqm_bat_config_entry_size( + struct cqm_cla_table_s *cla_table, + struct cqm_bat_entry_standerd_s *bat_entry_standerd) +{ + /* Except for QPC of 256/512/1024, the others are all cacheline 256B, + * and the conversion will be done inside the chip + */ + if (cla_table->obj_size > CQM_CHIP_CACHELINE) { + if (cla_table->obj_size == 512) { + bat_entry_standerd->entry_size = CQM_BAT_ENTRY_SIZE_512; + } else { + bat_entry_standerd->entry_size = + CQM_BAT_ENTRY_SIZE_1024; + } + bat_entry_standerd->max_number = + cla_table->max_buffer_size / cla_table->obj_size; + } else { + bat_entry_standerd->entry_size = CQM_BAT_ENTRY_SIZE_256; + bat_entry_standerd->max_number = + cla_table->max_buffer_size / CQM_CHIP_CACHELINE; + } +} + +void cqm_bat_fill_cla_std_entry(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + u8 *entry_base_addr, u32 entry_type, + u8 gpa_check_enable) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_bat_entry_standerd_s *bat_entry_standerd = NULL; + dma_addr_t pa = 0; + + if (cla_table->obj_num == 0) { + cqm_info(handle->dev_hdl, "Cla alloc: cla_type %u, obj_num=0, don't init bat entry\n", + cla_table->type); + return; + } + + bat_entry_standerd = (struct cqm_bat_entry_standerd_s *)entry_base_addr; + cqm_bat_config_entry_size(cla_table, bat_entry_standerd); + bat_entry_standerd->max_number = bat_entry_standerd->max_number - 1; + + bat_entry_standerd->bypass = CQM_BAT_NO_BYPASS_CACHE; + bat_entry_standerd->z = cla_table->cacheline_z; + bat_entry_standerd->y = cla_table->cacheline_y; + bat_entry_standerd->x = cla_table->cacheline_x; + bat_entry_standerd->cla_level = cla_table->cla_lvl; + + if (cla_table->cla_lvl == CQM_CLA_LVL_0) + pa = cla_table->cla_z_buf.buf_list[0].pa; + else if (cla_table->cla_lvl == CQM_CLA_LVL_1) + pa = cla_table->cla_y_buf.buf_list[0].pa; + else + pa = cla_table->cla_x_buf.buf_list[0].pa; + + bat_entry_standerd->cla_gpa_h = CQM_ADDR_HI(pa); + if (entry_type == CQM_BAT_ENTRY_T_REORDER) { + /* Reorder does not support GPA validity check */ + bat_entry_standerd->cla_gpa_l = CQM_ADDR_LW(pa); + } else { + /* GPA is valid when gpa[0]=1 */ + bat_entry_standerd->cla_gpa_l = + CQM_ADDR_LW(pa) | gpa_check_enable; + } +} + +static void cqm_bat_fill_cla_cfg(struct cqm_handle_s *cqm_handle, + u8 *entry_base_addr) +{ + struct cqm_bat_entry_cfg_s *bat_entry_cfg = + (struct cqm_bat_entry_cfg_s *)entry_base_addr; + + bat_entry_cfg->cur_conn_cache = 0; + bat_entry_cfg->max_conn_cache = + cqm_handle->func_capability.flow_table_based_conn_cache_number; + bat_entry_cfg->cur_conn_num_h_4 = 0; + bat_entry_cfg->cur_conn_num_l_16 = 0; + bat_entry_cfg->max_conn_num = + cqm_handle->func_capability.flow_table_based_conn_number; + /* Align by 64 buckets, shift right 6 bits */ + if ((cqm_handle->func_capability.hash_number >> 6) != 0) { + /* After shift right 6 bits, the value should - 1 for the hash + * value + */ + bat_entry_cfg->bucket_num = + ((cqm_handle->func_capability.hash_number >> 6) - 1); + } + if (cqm_handle->func_capability.bloomfilter_length != 0) { + bat_entry_cfg->bloom_filter_len = + cqm_handle->func_capability.bloomfilter_length - 1; + bat_entry_cfg->bloom_filter_addr = + cqm_handle->func_capability.bloomfilter_addr; + } +} + +static void cqm_bat_fill_cla_taskmap(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + u8 *entry_base_addr) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_bat_entry_taskmap_s *bat_entry_taskmap = + (struct cqm_bat_entry_taskmap_s *)entry_base_addr; + if (cqm_handle->func_capability.taskmap_number != 0) { + bat_entry_taskmap->gpa0_h = + (u32)(cla_table->cla_z_buf.buf_list[0].pa >> 32); + bat_entry_taskmap->gpa0_l = + (u32)(cla_table->cla_z_buf.buf_list[0].pa & 0xffffffff); + + bat_entry_taskmap->gpa1_h = + (u32)(cla_table->cla_z_buf.buf_list[1].pa >> 32); + bat_entry_taskmap->gpa1_l = + (u32)(cla_table->cla_z_buf.buf_list[1].pa & 0xffffffff); + + bat_entry_taskmap->gpa2_h = + (u32)(cla_table->cla_z_buf.buf_list[2].pa >> 32); + bat_entry_taskmap->gpa2_l = + (u32)(cla_table->cla_z_buf.buf_list[2].pa & 0xffffffff); + + bat_entry_taskmap->gpa3_h = + (u32)(cla_table->cla_z_buf.buf_list[3].pa >> 32); + bat_entry_taskmap->gpa3_l = + (u32)(cla_table->cla_z_buf.buf_list[3].pa & 0xffffffff); + + cqm_info(handle->dev_hdl, "Cla alloc: taskmap bat entry: 0x%x 0x%x, 0x%x 0x%x, 0x%x 0x%x, 0x%x 0x%x\n", + bat_entry_taskmap->gpa0_h, bat_entry_taskmap->gpa0_l, + bat_entry_taskmap->gpa1_h, bat_entry_taskmap->gpa1_l, + bat_entry_taskmap->gpa2_h, bat_entry_taskmap->gpa2_l, + bat_entry_taskmap->gpa3_h, bat_entry_taskmap->gpa3_l); + } +} + +/** + * cqm_bat_fill_cla - Fill the base address of the cla table into the bat table + * @cqm_handle: cqm handle + */ +void cqm_bat_fill_cla(struct cqm_handle_s *cqm_handle) +{ + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + struct cqm_cla_table_s *cla_table = NULL; + u32 entry_type = CQM_BAT_ENTRY_T_INVALID; + u8 *entry_base_addr = NULL; + u32 i = 0; + u8 gpa_check_enable = cqm_handle->func_capability.gpa_check_enable; + + /* Fill each item according to the arranged BAT table */ + entry_base_addr = bat_table->bat; + for (i = 0; i < CQM_BAT_ENTRY_MAX; i++) { + entry_type = bat_table->bat_entry_type[i]; + if (entry_type == CQM_BAT_ENTRY_T_CFG) { + cqm_bat_fill_cla_cfg(cqm_handle, entry_base_addr); + entry_base_addr += sizeof(struct cqm_bat_entry_cfg_s); + } else if (entry_type == CQM_BAT_ENTRY_T_TASKMAP) { + cqm_bat_fill_cla_taskmap(cqm_handle, + &bat_table->entry[i], + entry_base_addr); + entry_base_addr += + sizeof(struct cqm_bat_entry_taskmap_s); + } else if ((entry_type == CQM_BAT_ENTRY_T_INVALID) || + ((entry_type == CQM_BAT_ENTRY_T_TIMER) && + (cqm_handle->func_attribute.func_type != CQM_PPF))) { + /* When entry_type is invalid, or the timer entry under + * PF does not need to apply for memory and bat filling + */ + entry_base_addr += CQM_BAT_ENTRY_SIZE; + } else { + cla_table = &bat_table->entry[i]; + cqm_bat_fill_cla_std_entry(cqm_handle, cla_table, + entry_base_addr, entry_type, + gpa_check_enable); + entry_base_addr += + sizeof(struct cqm_bat_entry_standerd_s); + } + /* Checks if entry_base_addr is out of bounds */ + if (entry_base_addr >= + (bat_table->bat + CQM_BAT_ENTRY_MAX * CQM_BAT_ENTRY_SIZE)) + break; + } +} + +static void cqm_cla_xyz_cacheline_lvl1(struct cqm_cla_table_s *cla_table, + u32 trunk_size) +{ + s32 shift = 0; + + if (cla_table->obj_size >= CQM_CHIP_CACHELINE) { + cla_table->cacheline_z = cla_table->z; + cla_table->cacheline_y = cla_table->y; + cla_table->cacheline_x = cla_table->x; + } else { + shift = cqm_shift(trunk_size / CQM_CHIP_CACHELINE); + cla_table->cacheline_z = shift ? (shift - 1) : (shift); + cla_table->cacheline_y = CQM_MAX_INDEX_BIT; + cla_table->cacheline_x = 0; + } +} + +s32 cqm_cla_xyz_lvl1(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + u32 trunk_size) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_buf_s *cla_y_buf = NULL; + struct cqm_buf_s *cla_z_buf = NULL; + dma_addr_t *base = NULL; + s32 shift = 0; + u32 i = 0; + s32 ret = CQM_FAIL; + u8 gpa_check_enable = cqm_handle->func_capability.gpa_check_enable; + + if (cla_table->type == CQM_BAT_ENTRY_T_REORDER) + gpa_check_enable = 0; + + cla_table->cla_lvl = CQM_CLA_LVL_1; + + shift = cqm_shift(trunk_size / cla_table->obj_size); + cla_table->z = shift ? (shift - 1) : (shift); + cla_table->y = CQM_MAX_INDEX_BIT; + cla_table->x = 0; + cqm_cla_xyz_cacheline_lvl1(cla_table, trunk_size); + + /* Allocate y buf space */ + cla_y_buf = &cla_table->cla_y_buf; + cla_y_buf->buf_size = trunk_size; + cla_y_buf->buf_number = 1; + cla_y_buf->page_number = cla_y_buf->buf_number << + cla_table->trunk_order; + ret = cqm_buf_alloc(cqm_handle, cla_y_buf, false); + + CQM_CHECK_EQUAL_RET(handle->dev_hdl, ret, CQM_SUCCESS, return CQM_FAIL, + CQM_ALLOC_FAIL(lvl_1_y_buf)); + + /* Allocate z buf space */ + cla_z_buf = &cla_table->cla_z_buf; + cla_z_buf->buf_size = trunk_size; + cla_z_buf->buf_number = ALIGN(cla_table->max_buffer_size, trunk_size) / + trunk_size; + cla_z_buf->page_number = cla_z_buf->buf_number << + cla_table->trunk_order; + /* Requires static allocation of all buffer space */ + if (cla_table->alloc_static == true) { + if (cqm_buf_alloc(cqm_handle, cla_z_buf, false) == CQM_FAIL) { + cqm_err(handle->dev_hdl, CQM_ALLOC_FAIL(lvl_1_z_buf)); + cqm_buf_free(cla_y_buf, cqm_handle->dev); + return CQM_FAIL; + } + + /* Fill gpa of z buf list into y buf */ + base = (dma_addr_t *)(cla_y_buf->buf_list->va); + for (i = 0; i < cla_z_buf->buf_number; i++) { + /*gpa[0]=1 means this GPA is valid*/ + *base = (cla_z_buf->buf_list[i].pa | gpa_check_enable); + base++; + } + + /* big-endian conversion */ + cqm_swab64((u8 *)(cla_y_buf->buf_list->va), + cla_z_buf->buf_number); + } else { + /* Only initialize and allocate buf list space, buffer spaces + * are dynamically allocated in service + */ + cla_z_buf->buf_list = (struct cqm_buf_list_s *) + vmalloc(cla_z_buf->buf_number * + sizeof(struct cqm_buf_list_s)); + + if (!cla_z_buf->buf_list) { + cqm_err(handle->dev_hdl, CQM_ALLOC_FAIL(lvl_1_z_buf)); + cqm_buf_free(cla_y_buf, cqm_handle->dev); + return CQM_FAIL; + } + memset(cla_z_buf->buf_list, 0, + cla_z_buf->buf_number * sizeof(struct cqm_buf_list_s)); + } + + return CQM_SUCCESS; +} + +static s32 cqm_cla_yz_lvl2_static(struct cqm_handle_s *cqm_handle, + struct cqm_buf_s *cla_y_buf, + struct cqm_buf_s *cla_z_buf, + u8 gpa_check_enable) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + dma_addr_t *base = NULL; + u32 i = 0; + + if (cqm_buf_alloc(cqm_handle, cla_z_buf, false) == CQM_FAIL) { + cqm_err(handle->dev_hdl, CQM_ALLOC_FAIL(lvl_2_z_buf)); + return CQM_FAIL; + } + + /* The virtual address of y buf is remapped for software access */ + if (cqm_buf_alloc(cqm_handle, cla_y_buf, true) == CQM_FAIL) { + cqm_err(handle->dev_hdl, CQM_ALLOC_FAIL(lvl_2_y_buf)); + cqm_buf_free(cla_z_buf, cqm_handle->dev); + return CQM_FAIL; + } + + /* Fill gpa of z buf list into y buf */ + base = (dma_addr_t *)(cla_y_buf->direct.va); + for (i = 0; i < cla_z_buf->buf_number; i++) { + /*gpa[0]=1 means this GPA is valid*/ + *base = (cla_z_buf->buf_list[i].pa | gpa_check_enable); + base++; + } + + /* big-endian conversion */ + cqm_swab64((u8 *)(cla_y_buf->direct.va), cla_z_buf->buf_number); + + return CQM_SUCCESS; +} + +static void cqm_cla_yz_lvl2_init_cacheline(struct cqm_cla_table_s *cla_table, + u32 trunk_size) +{ + s32 shift = 0; + + if (cla_table->obj_size >= CQM_CHIP_CACHELINE) { + cla_table->cacheline_z = cla_table->z; + cla_table->cacheline_y = cla_table->y; + cla_table->cacheline_x = cla_table->x; + } else { + shift = cqm_shift(trunk_size / CQM_CHIP_CACHELINE); + cla_table->cacheline_z = shift ? (shift - 1) : (shift); + shift = cqm_shift(trunk_size / sizeof(dma_addr_t)); + cla_table->cacheline_y = cla_table->cacheline_z + shift; + cla_table->cacheline_x = CQM_MAX_INDEX_BIT; + } +} + +s32 cqm_cla_xyz_lvl2(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + u32 trunk_size) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_buf_s *cla_x_buf = NULL; + struct cqm_buf_s *cla_y_buf = NULL; + struct cqm_buf_s *cla_z_buf = NULL; + dma_addr_t *base = NULL; + s32 shift = 0; + u32 i = 0; + s32 ret = CQM_FAIL; + u8 gpa_check_enable = cqm_handle->func_capability.gpa_check_enable; + + if (cla_table->type == CQM_BAT_ENTRY_T_REORDER) + gpa_check_enable = 0; + + cla_table->cla_lvl = CQM_CLA_LVL_2; + + shift = cqm_shift(trunk_size / cla_table->obj_size); + cla_table->z = shift ? (shift - 1) : (shift); + shift = cqm_shift(trunk_size / sizeof(dma_addr_t)); + cla_table->y = cla_table->z + shift; + cla_table->x = CQM_MAX_INDEX_BIT; + + cqm_cla_yz_lvl2_init_cacheline(cla_table, trunk_size); + + /* Allocate x buf space */ + cla_x_buf = &cla_table->cla_x_buf; + cla_x_buf->buf_size = trunk_size; + cla_x_buf->buf_number = 1; + cla_x_buf->page_number = cla_x_buf->buf_number << + cla_table->trunk_order; + + ret = cqm_buf_alloc(cqm_handle, cla_x_buf, false); + CQM_CHECK_EQUAL_RET(handle->dev_hdl, ret, CQM_SUCCESS, return CQM_FAIL, + CQM_ALLOC_FAIL(lvl_2_x_buf)); + + /* Allocate y buf and z buf space */ + cla_z_buf = &cla_table->cla_z_buf; + cla_z_buf->buf_size = trunk_size; + cla_z_buf->buf_number = ALIGN(cla_table->max_buffer_size, trunk_size) / + trunk_size; + cla_z_buf->page_number = cla_z_buf->buf_number << + cla_table->trunk_order; + + cla_y_buf = &cla_table->cla_y_buf; + cla_y_buf->buf_size = trunk_size; + cla_y_buf->buf_number = + (ALIGN(cla_z_buf->buf_number * sizeof(dma_addr_t), + trunk_size)) / trunk_size; + + cla_y_buf->page_number = cla_y_buf->buf_number << + cla_table->trunk_order; + + /* Requires static allocation of all buffer space */ + if (cla_table->alloc_static == true) { + if (cqm_cla_yz_lvl2_static(cqm_handle, + cla_y_buf, + cla_z_buf, + gpa_check_enable) == CQM_FAIL) { + cqm_buf_free(cla_x_buf, cqm_handle->dev); + return CQM_FAIL; + } + /* Fill gpa of y buf list into x buf */ + base = (dma_addr_t *)(cla_x_buf->buf_list->va); + for (i = 0; i < cla_y_buf->buf_number; i++) { + /* gpa[0]=1 means this GPA is valid */ + *base = (cla_y_buf->buf_list[i].pa | gpa_check_enable); + base++; + } + + /* big-endian conversion */ + cqm_swab64((u8 *)(cla_x_buf->buf_list->va), + cla_y_buf->buf_number); + } else { + /* Only initialize and allocate buf list space, buffer spaces + * are allocated in service + */ + cla_z_buf->buf_list = (struct cqm_buf_list_s *) + vmalloc(cla_z_buf->buf_number * + sizeof(struct cqm_buf_list_s)); + if (!cla_z_buf->buf_list) { + cqm_err(handle->dev_hdl, CQM_ALLOC_FAIL(lvl_2_z_buf)); + cqm_buf_free(cla_x_buf, cqm_handle->dev); + return CQM_FAIL; + } + memset(cla_z_buf->buf_list, 0, + cla_z_buf->buf_number * sizeof(struct cqm_buf_list_s)); + + cla_y_buf->buf_list = (struct cqm_buf_list_s *) + vmalloc(cla_y_buf->buf_number * + sizeof(struct cqm_buf_list_s)); + + if (!cla_y_buf->buf_list) { + cqm_err(handle->dev_hdl, CQM_ALLOC_FAIL(lvl_2_y_buf)); + cqm_buf_free(cla_z_buf, cqm_handle->dev); + cqm_buf_free(cla_x_buf, cqm_handle->dev); + return CQM_FAIL; + } + memset(cla_y_buf->buf_list, 0, + cla_y_buf->buf_number * sizeof(struct cqm_buf_list_s)); + } + + return CQM_SUCCESS; +} + +static s32 cqm_cla_xyz_check(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + + if (cla_table->obj_num == 0) { + /* If the capability is set to 0, the CLA does not need to be + * initialized and exits directly + */ + cqm_info(handle->dev_hdl, "Cla alloc: cla_type %u, obj_num=0, don't alloc buffer\n", + cla_table->type); + return CQM_SUCCESS; + } + + /* Check whether obj_size is aligned with 2^n, and error is reported in + * case of 0 and 1 + */ + if (cqm_check_align(cla_table->obj_size) == false) { + cqm_err(handle->dev_hdl, "Cla alloc: cla_type %u, obj_size 0x%x is not align on 2^n\n", + cla_table->type, cla_table->obj_size); + return CQM_FAIL; + } + + return CQM_SUCCESS; +} + +/** + * cqm_cla_xyz - Calculate how many levels of cla tables and allocate space + * for each level of cla tables + * @cqm_handle: cqm handle + * @cla_table: cla table + * Return: 0 - success, negative - failure + */ +s32 cqm_cla_xyz(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_buf_s *cla_z_buf = NULL; + u32 trunk_size = 0; + s32 ret = CQM_FAIL; + + if (cqm_cla_xyz_check(cqm_handle, cla_table) == CQM_FAIL) + return CQM_FAIL; + + trunk_size = PAGE_SIZE << cla_table->trunk_order; + + if (trunk_size < cla_table->obj_size) { + cqm_err(handle->dev_hdl, "Cla alloc: cla type %u, obj_size 0x%x is out of trunk size\n", + cla_table->type, cla_table->obj_size); + return CQM_FAIL; + } + + /* Level 0 CLA: The buffer occupies little space, and can be assigned to + * cla_z_buf during initialization + */ + if (cla_table->max_buffer_size <= trunk_size) { + cla_table->cla_lvl = CQM_CLA_LVL_0; + + cla_table->z = CQM_MAX_INDEX_BIT; + cla_table->y = 0; + cla_table->x = 0; + + cla_table->cacheline_z = cla_table->z; + cla_table->cacheline_y = cla_table->y; + cla_table->cacheline_x = cla_table->x; + + /* Allocate z buf space */ + cla_z_buf = &cla_table->cla_z_buf; + cla_z_buf->buf_size = trunk_size; + cla_z_buf->buf_number = 1; + cla_z_buf->page_number = + cla_z_buf->buf_number << cla_table->trunk_order; + ret = cqm_buf_alloc(cqm_handle, cla_z_buf, false); + CQM_CHECK_EQUAL_RET( + handle->dev_hdl, ret, CQM_SUCCESS, + return CQM_FAIL, CQM_ALLOC_FAIL(lvl_0_z_buf)); + + } else if (cla_table->max_buffer_size <= + (trunk_size * (trunk_size / sizeof(dma_addr_t)))) { + /* Level 1 CLA: Cla_y_buf is allocated during initialization, + * and cla_z_buf can be allocated dynamically + */ + if (cqm_cla_xyz_lvl1(cqm_handle, + cla_table, trunk_size) == CQM_FAIL) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_cla_xyz_lvl1)); + return CQM_FAIL; + } + } else if (cla_table->max_buffer_size <= + (trunk_size * (trunk_size / sizeof(dma_addr_t)) * + (trunk_size / sizeof(dma_addr_t)))) { + /* Level 2 CLA: Cla_x_buf is allocated during initialization, + * and cla_y_buf and cla_z_buf can be dynamically allocated + */ + if (cqm_cla_xyz_lvl2(cqm_handle, cla_table, trunk_size) == + CQM_FAIL) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_cla_xyz_lvl2)); + return CQM_FAIL; + } + } else { + cqm_err(handle->dev_hdl, "Cla alloc: cla max_buffer_size 0x%x exceeds support range\n", + cla_table->max_buffer_size); + return CQM_FAIL; + } + + return CQM_SUCCESS; +} + +static void cqm_bat_entry_hash_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = capability->pagesize_reorder; + cla_table->max_buffer_size = capability->hash_number * + capability->hash_basic_size; + cla_table->obj_size = capability->hash_basic_size; + cla_table->obj_num = capability->hash_number; + cla_table->alloc_static = true; +} + +static void cqm_bat_entry_qpc_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_handle_s *handle = (struct cqm_handle_s *)cqm_handle; + struct hifc_hwdev *hwdev_handle = handle->ex_handle; + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = capability->pagesize_reorder; + cla_table->max_buffer_size = capability->qpc_number * + capability->qpc_basic_size; + cla_table->obj_size = capability->qpc_basic_size; + cla_table->obj_num = capability->qpc_number; + cla_table->alloc_static = capability->qpc_alloc_static; + cqm_info(hwdev_handle->dev_hdl, "Cla alloc: qpc alloc_static=%d\n", + cla_table->alloc_static); +} + +static void cqm_bat_entry_mpt_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = capability->pagesize_reorder; + cla_table->max_buffer_size = capability->mpt_number * + capability->mpt_basic_size; + cla_table->obj_size = capability->mpt_basic_size; + cla_table->obj_num = capability->mpt_number; + cla_table->alloc_static = true; +} + +static void cqm_bat_entry_scqc_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_handle_s *handle = (struct cqm_handle_s *)cqm_handle; + struct hifc_hwdev *hwdev_handle = handle->ex_handle; + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = capability->pagesize_reorder; + cla_table->max_buffer_size = capability->scqc_number * + capability->scqc_basic_size; + cla_table->obj_size = capability->scqc_basic_size; + cla_table->obj_num = capability->scqc_number; + cla_table->alloc_static = capability->scqc_alloc_static; + cqm_info(hwdev_handle->dev_hdl, "Cla alloc: scqc alloc_static=%d\n", + cla_table->alloc_static); +} + +static void cqm_bat_entry_srqc_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = capability->pagesize_reorder; + cla_table->max_buffer_size = capability->srqc_number * + capability->srqc_basic_size; + cla_table->obj_size = capability->srqc_basic_size; + cla_table->obj_num = capability->srqc_number; + cla_table->alloc_static = false; +} + +static void cqm_bat_entry_gid_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->max_buffer_size = capability->gid_number * + capability->gid_basic_size; + cla_table->trunk_order = (u32)cqm_shift( + ALIGN( + cla_table->max_buffer_size, + PAGE_SIZE) / PAGE_SIZE); + cla_table->obj_size = capability->gid_basic_size; + cla_table->obj_num = capability->gid_number; + cla_table->alloc_static = true; +} + +static void cqm_bat_entry_lun_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = CLA_TABLE_PAGE_ORDER; + cla_table->max_buffer_size = capability->lun_number * + capability->lun_basic_size; + cla_table->obj_size = capability->lun_basic_size; + cla_table->obj_num = capability->lun_number; + cla_table->alloc_static = true; +} + +static void cqm_bat_entry_taskmap_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = CQM_4K_PAGE_ORDER; + cla_table->max_buffer_size = capability->taskmap_number * + capability->taskmap_basic_size; + cla_table->obj_size = capability->taskmap_basic_size; + cla_table->obj_num = capability->taskmap_number; + cla_table->alloc_static = true; +} + +static void cqm_bat_entry_l3i_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = CLA_TABLE_PAGE_ORDER; + cla_table->max_buffer_size = capability->l3i_number * + capability->l3i_basic_size; + cla_table->obj_size = capability->l3i_basic_size; + cla_table->obj_num = capability->l3i_number; + cla_table->alloc_static = true; +} + +static void cqm_bat_entry_childc_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = capability->pagesize_reorder; + cla_table->max_buffer_size = capability->childc_number * + capability->childc_basic_size; + cla_table->obj_size = capability->childc_basic_size; + cla_table->obj_num = capability->childc_number; + cla_table->alloc_static = true; +} + +static void cqm_bat_entry_timer_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = CQM_4K_PAGE_ORDER; + cla_table->max_buffer_size = capability->timer_number * + capability->timer_basic_size; + cla_table->obj_size = capability->timer_basic_size; + cla_table->obj_num = capability->timer_number; + cla_table->alloc_static = true; +} + +static void cqm_bat_entry_xid2cid_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = capability->pagesize_reorder; + cla_table->max_buffer_size = capability->xid2cid_number * + capability->xid2cid_basic_size; + cla_table->obj_size = capability->xid2cid_basic_size; + cla_table->obj_num = capability->xid2cid_number; + cla_table->alloc_static = true; +} + +static void cqm_bat_entry_reoder_init(void *cqm_handle, + struct cqm_cla_table_s *cla_table, + void *cap) +{ + struct cqm_func_capability_s *capability = + (struct cqm_func_capability_s *)cap; + + cla_table->trunk_order = capability->pagesize_reorder; + cla_table->max_buffer_size = capability->reorder_number * + capability->reorder_basic_size; + cla_table->obj_size = capability->reorder_basic_size; + cla_table->obj_num = capability->reorder_number; + cla_table->alloc_static = true; +} + +struct cqm_cla_entry_init_s cqm_cla_entry_init_tbl[] = { + {CQM_BAT_ENTRY_T_HASH, cqm_bat_entry_hash_init}, + {CQM_BAT_ENTRY_T_QPC, cqm_bat_entry_qpc_init}, + {CQM_BAT_ENTRY_T_MPT, cqm_bat_entry_mpt_init}, + {CQM_BAT_ENTRY_T_SCQC, cqm_bat_entry_scqc_init}, + {CQM_BAT_ENTRY_T_SRQC, cqm_bat_entry_srqc_init}, + {CQM_BAT_ENTRY_T_GID, cqm_bat_entry_gid_init}, + {CQM_BAT_ENTRY_T_LUN, cqm_bat_entry_lun_init}, + {CQM_BAT_ENTRY_T_TASKMAP, cqm_bat_entry_taskmap_init}, + {CQM_BAT_ENTRY_T_L3I, cqm_bat_entry_l3i_init}, + {CQM_BAT_ENTRY_T_CHILDC, cqm_bat_entry_childc_init}, + {CQM_BAT_ENTRY_T_TIMER, cqm_bat_entry_timer_init}, + {CQM_BAT_ENTRY_T_XID2CID, cqm_bat_entry_xid2cid_init}, + {CQM_BAT_ENTRY_T_REORDER, cqm_bat_entry_reoder_init}, +}; + +static struct cqm_cla_entry_init_s *cqm_get_cla_init_entry( + struct cqm_handle_s *cqm_handle, + u32 type) +{ + int i; + struct cqm_cla_entry_init_s *entry = NULL; + + for (i = 0; + i < (sizeof(cqm_cla_entry_init_tbl) / + sizeof(struct cqm_cla_entry_init_s)); i++) { + entry = &cqm_cla_entry_init_tbl[i]; + if (entry->type == type) + return entry; + } + + return NULL; +} + +void cqm_cla_init_entry(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + struct cqm_func_capability_s *capability) +{ + struct cqm_cla_entry_init_s *entry; + + entry = cqm_get_cla_init_entry(cqm_handle, cla_table->type); + if (entry && entry->cqm_cla_init_handler) + entry->cqm_cla_init_handler(cqm_handle, cla_table, capability); +} + +static s32 cqm_cla_fill_entry(struct cqm_handle_s *cqm_handle) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + s32 ret = CQM_FAIL; + + /* After the allocation of CLA entry, fill in the BAT table */ + cqm_bat_fill_cla(cqm_handle); + + /* Notify the chip to refresh the BAT table */ + ret = cqm_bat_update(cqm_handle); + if (ret != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_bat_update)); + return CQM_FAIL; + } + + cqm_info(handle->dev_hdl, "Timer start: func_type=%d, timer_enable=%u\n", + cqm_handle->func_attribute.func_type, + cqm_handle->func_capability.timer_enable); + + if ((cqm_handle->func_attribute.func_type == CQM_PPF) && + (cqm_handle->func_capability.timer_enable == CQM_TIMER_ENABLE)) { + /* After the timer resource is allocated, + * the timer needs to be enabled + */ + cqm_info(handle->dev_hdl, "Timer start: hifc ppf timer start\n"); + ret = hifc_ppf_tmr_start((void *)(cqm_handle->ex_handle)); + if (ret != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, "Timer start: hifc ppf timer start, ret=%d\n", + ret); + return CQM_FAIL; + } + } + return CQM_SUCCESS; +} + +s32 cqm_cla_init(struct cqm_handle_s *cqm_handle) +{ + struct cqm_func_capability_s *capability = &cqm_handle->func_capability; + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + struct cqm_cla_table_s *cla_table = NULL; + s32 inv_flag = 0; + u32 i = 0; + u32 j = 0; + + for (i = 0; i < CQM_BAT_ENTRY_MAX; i++) { + cla_table = &bat_table->entry[i]; + cla_table->type = bat_table->bat_entry_type[i]; + + cqm_cla_init_entry(cqm_handle, cla_table, capability); + + /* Allocate CLA entry space of all levels */ + if ((cla_table->type >= CQM_BAT_ENTRY_T_HASH) && + (cla_table->type <= CQM_BAT_ENTRY_T_REORDER)) { + /* Only needs to allocate timer resources for PPF, + * 8 wheels * 2k scales * 32B * func_num, for PF, there + * is no need to allocate resources for the timer, nor + * to fill in the structure of the timer entry in the + * BAT table. + */ + if (!((cla_table->type == CQM_BAT_ENTRY_T_TIMER) && + (cqm_handle->func_attribute.func_type + != CQM_PPF))) { + if (cqm_cla_xyz(cqm_handle, cla_table) == + CQM_FAIL) + goto err; + } + } + mutex_init(&cla_table->lock); + } + if (cqm_cla_fill_entry(cqm_handle) == CQM_FAIL) + goto err; + + return CQM_SUCCESS; + +err: + for (j = 0; j < i; j++) { + cla_table = &bat_table->entry[j]; + if (cla_table->type != CQM_BAT_ENTRY_T_INVALID) { + cqm_buf_free_cache_inv(cqm_handle, + &cla_table->cla_x_buf, + &inv_flag); + cqm_buf_free_cache_inv(cqm_handle, + &cla_table->cla_y_buf, + &inv_flag); + cqm_buf_free_cache_inv(cqm_handle, + &cla_table->cla_z_buf, + &inv_flag); + } + } + + return CQM_FAIL; +} + +void cqm_cla_uninit(struct cqm_handle_s *cqm_handle) +{ + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + struct cqm_cla_table_s *cla_table = NULL; + s32 inv_flag = 0; + u32 i = 0; + + for (i = 0; i < CQM_BAT_ENTRY_MAX; i++) { + cla_table = &bat_table->entry[i]; + if (cla_table->type != CQM_BAT_ENTRY_T_INVALID) { + cqm_buf_free_cache_inv(cqm_handle, + &cla_table->cla_x_buf, + &inv_flag); + cqm_buf_free_cache_inv(cqm_handle, + &cla_table->cla_y_buf, + &inv_flag); + cqm_buf_free_cache_inv(cqm_handle, + &cla_table->cla_z_buf, + &inv_flag); + } + } +} + +s32 cqm_cla_update(struct cqm_handle_s *cqm_handle, + struct cqm_buf_list_s *buf_node_parent, + struct cqm_buf_list_s *buf_node_child, + u32 child_index, u8 cla_update_mode) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_cmd_buf_s *buf_in = NULL; + struct cqm_cla_update_cmd_s *cmd = NULL; + dma_addr_t pa = 0; + s32 ret = CQM_FAIL; + u8 gpa_check_enable = cqm_handle->func_capability.gpa_check_enable; + + buf_in = cqm_cmd_alloc((void *)(cqm_handle->ex_handle)); + CQM_PTR_CHECK_RET(buf_in, return CQM_FAIL, CQM_ALLOC_FAIL(buf_in)); + buf_in->size = sizeof(struct cqm_cla_update_cmd_s); + + /* Fill the command format and convert to big endian */ + cmd = (struct cqm_cla_update_cmd_s *)(buf_in->buf); + + pa = buf_node_parent->pa + (child_index * sizeof(dma_addr_t)); + cmd->gpa_h = CQM_ADDR_HI(pa); + cmd->gpa_l = CQM_ADDR_LW(pa); + + pa = buf_node_child->pa; + cmd->value_h = CQM_ADDR_HI(pa); + cmd->value_l = CQM_ADDR_LW(pa); + + cqm_dbg("Cla alloc: cqm_cla_update, gpa=0x%x 0x%x, value=0x%x 0x%x, cla_update_mode=0x%x\n", + cmd->gpa_h, cmd->gpa_l, cmd->value_h, cmd->value_l, + cla_update_mode); + + /* CLA GPA check */ + if (gpa_check_enable) { + switch (cla_update_mode) { + /* gpa[0]=1 means this GPA is valid */ + case CQM_CLA_RECORD_NEW_GPA: + cmd->value_l |= 1; + break; + /* gpa[0]=0 means this GPA is valid */ + case CQM_CLA_DEL_GPA_WITHOUT_CACHE_INVALID: + case CQM_CLA_DEL_GPA_WITH_CACHE_INVALID: + cmd->value_l &= (~1); + break; + default: + cqm_err(handle->dev_hdl, + "Cla alloc: cqm_cla_update, cqm_cla_update, wrong cla_update_mode=%u\n", + cla_update_mode); + break; + } + } + + cqm_swab32((u8 *)cmd, (sizeof(struct cqm_cla_update_cmd_s) >> 2)); + + ret = cqm_send_cmd_box((void *)(cqm_handle->ex_handle), + CQM_CMD_ACK_TYPE_CMDQ, + CQM_MOD_CQM, CQM_CMD_T_CLA_UPDATE, + buf_in, NULL, CQM_CMD_TIMEOUT); + if (ret != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_send_cmd_box)); + cqm_err(handle->dev_hdl, + "Cla alloc: cqm_cla_update, cqm_send_cmd_box_ret=%d\n", + ret); + cqm_err(handle->dev_hdl, "Cla alloc: cqm_cla_update, cla_update_cmd: 0x%x 0x%x 0x%x 0x%x\n", + cmd->gpa_h, cmd->gpa_l, cmd->value_h, cmd->value_l); + cqm_cmd_free((void *)(cqm_handle->ex_handle), buf_in); + return CQM_FAIL; + } + + cqm_cmd_free((void *)(cqm_handle->ex_handle), buf_in); + return CQM_SUCCESS; +} + +/** + * cqm_cla_alloc - Allocate a CLA trunk page + * @cqm_handle: cqm handle + * @cla_table: cla handle + * @buf_node_parent: the parent node whose content is to be updated + * @buf_node_child: the child node whose content is to be allocated + * @child_index: child index + * Return: 0 - success, negative - failure + */ +s32 cqm_cla_alloc(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + struct cqm_buf_list_s *buf_node_parent, + struct cqm_buf_list_s *buf_node_child, u32 child_index) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + s32 ret = CQM_FAIL; + + /* Allocate trunk page */ + buf_node_child->va = (u8 *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, + cla_table->trunk_order); + CQM_PTR_CHECK_RET(buf_node_child->va, return CQM_FAIL, + CQM_ALLOC_FAIL(va)); + + /* pci mapping */ + buf_node_child->pa = + pci_map_single(cqm_handle->dev, buf_node_child->va, + PAGE_SIZE << cla_table->trunk_order, + PCI_DMA_BIDIRECTIONAL); + if (pci_dma_mapping_error(cqm_handle->dev, buf_node_child->pa)) { + cqm_err(handle->dev_hdl, CQM_MAP_FAIL(buf_node_child->pa)); + goto err1; + } + + /* Notify the chip of trunk_pa and + * let it fill in the cla table entry + */ + ret = cqm_cla_update(cqm_handle, buf_node_parent, + buf_node_child, child_index, + CQM_CLA_RECORD_NEW_GPA); + if (ret != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_cla_update)); + goto err2; + } + + return CQM_SUCCESS; + +err2: + pci_unmap_single(cqm_handle->dev, buf_node_child->pa, + PAGE_SIZE << cla_table->trunk_order, + PCI_DMA_BIDIRECTIONAL); +err1: + free_pages((ulong)(buf_node_child->va), cla_table->trunk_order); + buf_node_child->va = NULL; + return CQM_FAIL; +} + +void cqm_cla_free(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + struct cqm_buf_list_s *buf_node_parent, + struct cqm_buf_list_s *buf_node_child, + u32 child_index, u8 cla_update_mode) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + + cqm_dbg("Cla free: cla_update_mode=%u\n", cla_update_mode); + + if (cqm_cla_update(cqm_handle, buf_node_parent, + buf_node_child, child_index, + cla_update_mode) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_cla_update)); + return; + } + + if (cla_update_mode == CQM_CLA_DEL_GPA_WITH_CACHE_INVALID) { + if (cqm_cla_cache_invalid( + cqm_handle, buf_node_child->pa, + PAGE_SIZE << cla_table->trunk_order) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_cla_cache_invalid)); + return; + } + } + + /* Unblock the pci mapping of the trunk page */ + pci_unmap_single(cqm_handle->dev, buf_node_child->pa, + PAGE_SIZE << cla_table->trunk_order, + PCI_DMA_BIDIRECTIONAL); + + /* Free trunk page */ + free_pages((ulong)(buf_node_child->va), cla_table->trunk_order); + buf_node_child->va = NULL; +} + +/** + * cqm_static_qpc_cla_get - When QPC is a static allocation, allocate the count + * of buffer from the index position in the cla table without lock + * @cqm_handle: cqm handle + * @cla_table: cla handle + * @index: the index of table + * @count: the count of buffer + * @pa: the physical address + * Return: the virtual address + */ +u8 *cqm_static_qpc_cla_get(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + u32 index, u32 count, dma_addr_t *pa) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_buf_s *cla_y_buf = &cla_table->cla_y_buf; + struct cqm_buf_s *cla_z_buf = &cla_table->cla_z_buf; + struct cqm_buf_list_s *buf_node_z = NULL; + u32 x_index = 0; + u32 y_index = 0; + u32 z_index = 0; + u32 trunk_size = PAGE_SIZE << cla_table->trunk_order; + u8 *ret_addr = NULL; + u32 offset = 0; + + if (cla_table->cla_lvl == CQM_CLA_LVL_0) { + offset = index * cla_table->obj_size; + ret_addr = (u8 *)(cla_z_buf->buf_list->va) + offset; + *pa = cla_z_buf->buf_list->pa + offset; + } else if (cla_table->cla_lvl == CQM_CLA_LVL_1) { + z_index = index & ((1 << (cla_table->z + 1)) - 1); + y_index = index >> (cla_table->z + 1); + + if (y_index >= cla_z_buf->buf_number) { + cqm_err(handle->dev_hdl, + "Static qpc cla get: index exceeds buf_number, y_index %u, z_buf_number %u\n", + y_index, cla_z_buf->buf_number); + return NULL; + } + buf_node_z = &cla_z_buf->buf_list[y_index]; + if (!buf_node_z->va) { + cqm_err(handle->dev_hdl, "Cla get: static qpc cla_z_buf[%u].va=NULL\n", + y_index); + return NULL; + } + buf_node_z->refcount += count; + offset = z_index * cla_table->obj_size; + ret_addr = (u8 *)(buf_node_z->va) + offset; + *pa = buf_node_z->pa + offset; + } else { + z_index = index & ((1 << (cla_table->z + 1)) - 1); + y_index = (index >> (cla_table->z + 1)) & + ((1 << (cla_table->y - cla_table->z)) - 1); + x_index = index >> (cla_table->y + 1); + + if ((x_index >= cla_y_buf->buf_number) || + ((x_index * (trunk_size / sizeof(dma_addr_t)) + y_index) >= + cla_z_buf->buf_number)) { + cqm_err(handle->dev_hdl, + "Static qpc cla get: index exceeds buf_number,x_index %u, y_index %u, y_buf_number %u, z_buf_number %u\n ", + x_index, y_index, cla_y_buf->buf_number, + cla_z_buf->buf_number); + return NULL; + } + + buf_node_z = &(cla_z_buf->buf_list[x_index * + (trunk_size / sizeof(dma_addr_t)) + y_index]); + if (!buf_node_z->va) { + cqm_err(handle->dev_hdl, "Cla get: static qpc cla_z_buf.va=NULL\n"); + return NULL; + } + + buf_node_z->refcount += count; + offset = z_index * cla_table->obj_size; + ret_addr = (u8 *)(buf_node_z->va) + offset; + *pa = buf_node_z->pa + offset; + } + + return ret_addr; +} + +static s32 cqm_cla_get_level_two(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + u32 index, u32 count, + dma_addr_t *pa, u8 **ret_addr) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_buf_s *cla_x_buf = &cla_table->cla_x_buf; + struct cqm_buf_s *cla_y_buf = &cla_table->cla_y_buf; + struct cqm_buf_s *cla_z_buf = &cla_table->cla_z_buf; + struct cqm_buf_list_s *buf_node_x = NULL; + struct cqm_buf_list_s *buf_node_y = NULL; + struct cqm_buf_list_s *buf_node_z = NULL; + u32 x_index = 0; + u32 y_index = 0; + u32 z_index = 0; + u32 trunk_size = PAGE_SIZE << cla_table->trunk_order; + u32 offset = 0; + + z_index = index & ((1 << (cla_table->z + 1)) - 1); + y_index = (index >> (cla_table->z + 1)) & + ((1 << (cla_table->y - cla_table->z)) - 1); + x_index = index >> (cla_table->y + 1); + + if ((x_index >= cla_y_buf->buf_number) || + ((x_index * (trunk_size / sizeof(dma_addr_t)) + y_index) >= + cla_z_buf->buf_number)) { + cqm_err(handle->dev_hdl, + "Cla get: index exceeds buf_number, x_index %u, y_index %u, y_buf_number %u, z_buf_number %u\n", + x_index, y_index, cla_y_buf->buf_number, + cla_z_buf->buf_number); + return CQM_FAIL; + } + + buf_node_x = cla_x_buf->buf_list; + buf_node_y = &cla_y_buf->buf_list[x_index]; + buf_node_z = &(cla_z_buf->buf_list[x_index * + (trunk_size / sizeof(dma_addr_t)) + y_index]); + + /* Y buf node does not exist, allocates y node page */ + if (!buf_node_y->va) { + if (cqm_cla_alloc( + cqm_handle, cla_table, + buf_node_x, buf_node_y, x_index) == CQM_FAIL) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_cla_alloc)); + return CQM_FAIL; + } + } + + /* Z buf node does not exist, allocates z node page */ + if (!buf_node_z->va) { + if (cqm_cla_alloc(cqm_handle, + cla_table, + buf_node_y, + buf_node_z, + y_index) == CQM_FAIL) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_cla_alloc)); + if (buf_node_y->refcount == 0) { + /* Free y node, needs cache_invalid */ + cqm_cla_free( + cqm_handle, cla_table, + buf_node_x, buf_node_y, x_index, + CQM_CLA_DEL_GPA_WITH_CACHE_INVALID); + } + return CQM_FAIL; + } + + cqm_dbg("Cla get: 2L: y_refcount=0x%x\n", buf_node_y->refcount); + /* Y buf node's reference count should be +1 */ + buf_node_y->refcount++; + } + + cqm_dbg("Cla get: 2L: z_refcount=0x%x, count=0x%x\n", + buf_node_z->refcount, count); + buf_node_z->refcount += count; + offset = z_index * cla_table->obj_size; + *ret_addr = (u8 *)(buf_node_z->va) + offset; + *pa = buf_node_z->pa + offset; + + return CQM_SUCCESS; +} + +static s32 cqm_cla_get_level_one(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + u32 index, u32 count, dma_addr_t *pa, + u8 **ret_addr) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_buf_s *cla_y_buf = &cla_table->cla_y_buf; + struct cqm_buf_s *cla_z_buf = &cla_table->cla_z_buf; + struct cqm_buf_list_s *buf_node_y = NULL; + struct cqm_buf_list_s *buf_node_z = NULL; + u32 y_index = 0; + u32 z_index = 0; + u32 offset = 0; + + z_index = index & ((1 << (cla_table->z + 1)) - 1); + y_index = index >> (cla_table->z + 1); + + if (y_index >= cla_z_buf->buf_number) { + cqm_err(handle->dev_hdl, + "Cla get: index exceeds buf_number, y_index %u, z_buf_number %u\n", + y_index, cla_z_buf->buf_number); + return CQM_FAIL; + } + buf_node_z = &cla_z_buf->buf_list[y_index]; + buf_node_y = cla_y_buf->buf_list; + + /* Z buf node does not exist, first allocate the page */ + if (!buf_node_z->va) { + if (cqm_cla_alloc(cqm_handle, + cla_table, + buf_node_y, + buf_node_z, + y_index) == CQM_FAIL) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_cla_alloc)); + cqm_err(handle->dev_hdl, + "Cla get: cla_table->type=%u\n", + cla_table->type); + return CQM_FAIL; + } + } + + cqm_dbg("Cla get: 1L: z_refcount=0x%x, count=0x%x\n", + buf_node_z->refcount, count); + buf_node_z->refcount += count; + offset = z_index * cla_table->obj_size; + *ret_addr = (u8 *)(buf_node_z->va) + offset; + *pa = buf_node_z->pa + offset; + + return CQM_SUCCESS; +} + +/** + * cqm_cla_get - Allocate the count of buffer from the index position in the + * cla table + * @cqm_handle: cqm handle + * @cla_table: cla table + * @index: the index of table + * @count: the count of buffer + * @pa: the physical address + * Return: the virtual address + */ +u8 *cqm_cla_get(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, u32 index, + u32 count, dma_addr_t *pa) +{ + struct cqm_buf_s *cla_z_buf = &cla_table->cla_z_buf; + u8 *ret_addr = NULL; + u32 offset = 0; + + mutex_lock(&cla_table->lock); + if (cla_table->cla_lvl == CQM_CLA_LVL_0) { + /* Level 0 CLA pages are statically allocated */ + offset = index * cla_table->obj_size; + ret_addr = (u8 *)(cla_z_buf->buf_list->va) + offset; + *pa = cla_z_buf->buf_list->pa + offset; + } else if (cla_table->cla_lvl == CQM_CLA_LVL_1) { + if (cqm_cla_get_level_one(cqm_handle, cla_table, + index, count, + pa, &ret_addr) == CQM_FAIL) { + mutex_unlock(&cla_table->lock); + return NULL; + } + } else { + if (cqm_cla_get_level_two(cqm_handle, + cla_table, + index, + count, + pa, + &ret_addr) == CQM_FAIL) { + mutex_unlock(&cla_table->lock); + return NULL; + } + } + + mutex_unlock(&cla_table->lock); + return ret_addr; +} + +/** + * cqm_cla_put -Decrease the reference count of the trunk page, if it is reduced + * to 0, release the trunk page + * @cqm_handle: cqm handle + * @cla_table: cla table + * @index: the index of table + * @count: the count of buffer + */ +void cqm_cla_put(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + u32 index, u32 count) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_buf_s *cla_x_buf = &cla_table->cla_x_buf; + struct cqm_buf_s *cla_y_buf = &cla_table->cla_y_buf; + struct cqm_buf_s *cla_z_buf = &cla_table->cla_z_buf; + struct cqm_buf_list_s *buf_node_x = NULL; + struct cqm_buf_list_s *buf_node_y = NULL; + struct cqm_buf_list_s *buf_node_z = NULL; + u32 x_index = 0; + u32 y_index = 0; + u32 trunk_size = PAGE_SIZE << cla_table->trunk_order; + + /* Buffer is statically allocated, + * no need to control the reference count + */ + if (cla_table->alloc_static == true) + return; + + mutex_lock(&cla_table->lock); + + if (cla_table->cla_lvl == CQM_CLA_LVL_1) { + y_index = index >> (cla_table->z + 1); + + if (y_index >= cla_z_buf->buf_number) { + cqm_err(handle->dev_hdl, + "Cla put: index exceeds buf_number, y_index %u, z_buf_number %u\n", + y_index, cla_z_buf->buf_number); + cqm_err(handle->dev_hdl, + "Cla put: cla_table->type=%u\n", + cla_table->type); + mutex_unlock(&cla_table->lock); + return; + } + + buf_node_z = &cla_z_buf->buf_list[y_index]; + buf_node_y = cla_y_buf->buf_list; + + /* When the z node page reference count is 0, + * release the z node page + */ + cqm_dbg("Cla put: 1L: z_refcount=0x%x, count=0x%x\n", + buf_node_z->refcount, count); + buf_node_z->refcount -= count; + if (buf_node_z->refcount == 0) { + /* Z node does not need cache invalid */ + cqm_cla_free(cqm_handle, cla_table, buf_node_y, + buf_node_z, y_index, + CQM_CLA_DEL_GPA_WITHOUT_CACHE_INVALID); + } + } else if (cla_table->cla_lvl == CQM_CLA_LVL_2) { + y_index = (index >> (cla_table->z + 1)) & + ((1 << (cla_table->y - cla_table->z)) - 1); + x_index = index >> (cla_table->y + 1); + + if ((x_index >= cla_y_buf->buf_number) || + ((x_index * (trunk_size / sizeof(dma_addr_t)) + y_index) >= + cla_z_buf->buf_number)) { + cqm_err(handle->dev_hdl, + "Cla put: index exceeds buf_number, x_index %u, y_index %u, y_buf_number %u, z_buf_number %u\n", + x_index, y_index, cla_y_buf->buf_number, + cla_z_buf->buf_number); + mutex_unlock(&cla_table->lock); + return; + } + + buf_node_x = cla_x_buf->buf_list; + buf_node_y = &cla_y_buf->buf_list[x_index]; + buf_node_z = &(cla_z_buf->buf_list[x_index * + (trunk_size / sizeof(dma_addr_t)) + y_index]); + cqm_dbg("Cla put: 2L: z_refcount=0x%x, count=0x%x\n", + buf_node_z->refcount, count); + + /* When the z node page reference count is 0, + * release the z node page + */ + buf_node_z->refcount -= count; + if (buf_node_z->refcount == 0) { + cqm_cla_free(cqm_handle, cla_table, buf_node_y, + buf_node_z, y_index, + CQM_CLA_DEL_GPA_WITHOUT_CACHE_INVALID); + + /* When the y node page reference count is 0, + * release the y node page + */ + cqm_dbg("Cla put: 2L: y_refcount=0x%x\n", + buf_node_y->refcount); + buf_node_y->refcount--; + if (buf_node_y->refcount == 0) { + /* Y node needs cache invalid */ + cqm_cla_free( + cqm_handle, cla_table, buf_node_x, + buf_node_y, x_index, + CQM_CLA_DEL_GPA_WITH_CACHE_INVALID); + } + } + } + + mutex_unlock(&cla_table->lock); +} + +/** + * cqm_cla_table_get - Find the CLA table structure corresponding to a BAT entry + * @bat_table: bat table + * @entry_type: entry type + * @count: the count of buffer + * Return: the CLA table + */ +struct cqm_cla_table_s *cqm_cla_table_get(struct cqm_bat_table_s *bat_table, + u32 entry_type) +{ + struct cqm_cla_table_s *cla_table = NULL; + u32 i = 0; + + for (i = 0; i < CQM_BAT_ENTRY_MAX; i++) { + cla_table = &bat_table->entry[i]; + if (entry_type == cla_table->type) + return cla_table; + } + + return NULL; +} + +#define bitmap_section + +/** + * __cqm_bitmap_init - Initialize a bitmap + * @bitmap: cqm bitmap table + * Return: 0 - success, negative - failure + */ +s32 __cqm_bitmap_init(struct cqm_bitmap_s *bitmap) +{ + spin_lock_init(&bitmap->lock); + + /* The max_num of bitmap is aligned by 8, and then shifted right by + * 3bits to get how many Bytes are needed + */ + bitmap->table = + (ulong *)vmalloc((ALIGN(bitmap->max_num, 8) >> 3)); + CQM_PTR_CHECK_RET(bitmap->table, return CQM_FAIL, + CQM_ALLOC_FAIL(bitmap->table)); + memset(bitmap->table, 0, (ALIGN(bitmap->max_num, 8) >> 3)); + + return CQM_SUCCESS; +} + +static s32 cqm_bitmap_init_by_type(struct cqm_handle_s *cqm_handle, + struct cqm_cla_table_s *cla_table, + struct cqm_bitmap_s *bitmap) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_func_capability_s *capability = &cqm_handle->func_capability; + s32 ret = CQM_SUCCESS; + + switch (cla_table->type) { + case CQM_BAT_ENTRY_T_QPC: + bitmap->max_num = capability->qpc_number; + bitmap->reserved_top = capability->qpc_reserved; + bitmap->last = capability->qpc_reserved; + cqm_info(handle->dev_hdl, "Bitmap init: cla_table_type=%u, max_num=0x%x\n", + cla_table->type, bitmap->max_num); + ret = __cqm_bitmap_init(bitmap); + break; + case CQM_BAT_ENTRY_T_MPT: + bitmap->max_num = capability->mpt_number; + bitmap->reserved_top = capability->mpt_reserved; + bitmap->last = capability->mpt_reserved; + cqm_info(handle->dev_hdl, "Bitmap init: cla_table_type=%u, max_num=0x%x\n", + cla_table->type, bitmap->max_num); + ret = __cqm_bitmap_init(bitmap); + break; + case CQM_BAT_ENTRY_T_SCQC: + bitmap->max_num = capability->scqc_number; + bitmap->reserved_top = capability->scq_reserved; + bitmap->last = capability->scq_reserved; + cqm_info(handle->dev_hdl, "Bitmap init: cla_table_type=%u, max_num=0x%x\n", + cla_table->type, bitmap->max_num); + ret = __cqm_bitmap_init(bitmap); + break; + case CQM_BAT_ENTRY_T_SRQC: + bitmap->max_num = capability->srqc_number; + bitmap->reserved_top = 0; + bitmap->last = 0; + cqm_info(handle->dev_hdl, "Bitmap init: cla_table_type=%u, max_num=0x%x\n", + cla_table->type, bitmap->max_num); + ret = __cqm_bitmap_init(bitmap); + break; + default: + break; + } + + return ret; +} + +/** + * cqm_bitmap_init - Initialize a bitmap + * @cqm_handle: cqm handle + * Return: 0 - success, negative - failure + */ +s32 cqm_bitmap_init(struct cqm_handle_s *cqm_handle) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + struct cqm_cla_table_s *cla_table = NULL; + struct cqm_bitmap_s *bitmap = NULL; + u32 i = 0; + s32 ret = CQM_SUCCESS; + + for (i = 0; i < CQM_BAT_ENTRY_MAX; i++) { + cla_table = &bat_table->entry[i]; + if (cla_table->obj_num == 0) { + cqm_info(handle->dev_hdl, "Cla alloc: cla_type %u, obj_num=0, don't init bitmap\n", + cla_table->type); + continue; + } + + bitmap = &cla_table->bitmap; + ret = cqm_bitmap_init_by_type(cqm_handle, cla_table, bitmap); + if (ret != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, "Bitmap init: failed to init cla_table_type=%u, obj_num=0x%x\n", + cla_table->type, cla_table->obj_num); + goto err; + } + } + + return CQM_SUCCESS; + +err: + cqm_bitmap_uninit(cqm_handle); + return CQM_FAIL; +} + +/** + * cqm_bitmap_uninit - Uninitialize a bitmap + * @cqm_handle: cqm handle + */ +void cqm_bitmap_uninit(struct cqm_handle_s *cqm_handle) +{ + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + struct cqm_cla_table_s *cla_table = NULL; + struct cqm_bitmap_s *bitmap = NULL; + u32 i = 0; + + for (i = 0; i < CQM_BAT_ENTRY_MAX; i++) { + cla_table = &bat_table->entry[i]; + bitmap = &cla_table->bitmap; + if (cla_table->type != CQM_BAT_ENTRY_T_INVALID) { + if (bitmap->table) { + vfree(bitmap->table); + bitmap->table = NULL; + } + } + } +} + +/** + * cqm_bitmap_check_range - Starting from begin, check whether count bits are + * free in the table, required: 1. This set of bits cannot cross step, 2. This + * group of bits must be 0 + * @table: bitmap table + * @step: steps + * @max_num: max num + * @begin: begin position + * @count: the count of bit to check + * Return: If check valid return begin position + */ +u32 cqm_bitmap_check_range(const ulong *table, u32 step, + u32 max_num, u32 begin, u32 count) +{ + u32 i = 0; + u32 end = (begin + (count - 1)); + + /* Single bit is not checked */ + if (count == 1) + return begin; + + /* End is out of bounds */ + if (end >= max_num) + return max_num; + + /* Bit check, if there is a bit other than 0, return next bit */ + for (i = (begin + 1); i <= end; i++) { + if (test_bit((s32)i, table)) + return i + 1; + } + + /* Check if it is in a different step */ + if ((begin & (~(step - 1))) != (end & (~(step - 1)))) + return (end & (~(step - 1))); + + /* If check pass, return begin position */ + return begin; +} + +static void cqm_bitmap_set_bit(struct cqm_bitmap_s *bitmap, u32 index, + u32 max_num, u32 count, bool update_last, + ulong *table) +{ + u32 i; + + /* Set 1 to the found bit and reset last */ + if (index < max_num) { + for (i = index; i < (index + count); i++) + set_bit(i, table); + + if (update_last) { + bitmap->last = (index + count); + if (bitmap->last >= bitmap->max_num) + bitmap->last = bitmap->reserved_top; + } + } +} + +/** + * cqm_bitmap_alloc - Allocate a bitmap index, 0 and 1 should not be used, Scan + * back from the place where you last applied, and needs to support the + * application of a series of consecutive indexes, and should not to cross trunk + * @table: bitmap table + * @step: steps + * @count: the count of bit to check + * @update_last: update last + * Return: Success - return the index, failure - return the max + */ +u32 cqm_bitmap_alloc(struct cqm_bitmap_s *bitmap, u32 step, u32 count, + bool update_last) +{ + u32 index = 0; + u32 max_num = bitmap->max_num; + u32 last = bitmap->last; + ulong *table = bitmap->table; + + spin_lock(&bitmap->lock); + + /* Search for a free bit from the last position */ + do { + index = find_next_zero_bit(table, max_num, last); + if (index < max_num) { + last = cqm_bitmap_check_range(table, step, + max_num, index, count); + } else { + break; + } + } while (last != index); + + /* The above search failed, search for a free bit from the beginning */ + if (index >= max_num) { + last = bitmap->reserved_top; + do { + index = find_next_zero_bit(table, max_num, last); + if (index < max_num) { + last = cqm_bitmap_check_range(table, step, + max_num, + index, count); + } else { + break; + } + } while (last != index); + } + cqm_bitmap_set_bit(bitmap, index, max_num, count, update_last, table); + spin_unlock(&bitmap->lock); + return index; +} + +/** + * cqm_bitmap_alloc_reserved - Allocate the reserve bit according to index + * @bitmap: bitmap table + * @count: count + * @index: the index of bitmap + * Return: Success - return the index, failure - return the max + */ +u32 cqm_bitmap_alloc_reserved(struct cqm_bitmap_s *bitmap, u32 count, u32 index) +{ + ulong *table = bitmap->table; + u32 ret_index = CQM_INDEX_INVALID; + + if ((index >= bitmap->reserved_top) || (index >= bitmap->max_num) || + (count != 1)) { + return CQM_INDEX_INVALID; + } + + spin_lock(&bitmap->lock); + + if (test_bit(index, table)) { + ret_index = CQM_INDEX_INVALID; + } else { + set_bit(index, table); + ret_index = index; + } + + spin_unlock(&bitmap->lock); + return ret_index; +} + +/** + * cqm_bitmap_free - Release a bitmap index + * @bitmap: bitmap table + * @index: the index of bitmap + * @count: count + */ +void cqm_bitmap_free(struct cqm_bitmap_s *bitmap, u32 index, u32 count) +{ + ulong i = 0; + + spin_lock(&bitmap->lock); + + for (i = index; i < (index + count); i++) + clear_bit((s32)i, bitmap->table); + + spin_unlock(&bitmap->lock); +} + +#define obj_table_section + +/** + * _cqm_object_table_init - Initialize a table of object and index + * @cqm_handle: cqm handle + * Return: 0 - success, negative - failure + */ +s32 __cqm_object_table_init(struct cqm_object_table_s *obj_table) +{ + rwlock_init(&obj_table->lock); + + obj_table->table = (struct cqm_object_s **)vmalloc(obj_table->max_num * + sizeof(void *)); + CQM_PTR_CHECK_RET(obj_table->table, return CQM_FAIL, + CQM_ALLOC_FAIL(table)); + memset(obj_table->table, 0, obj_table->max_num * sizeof(void *)); + return CQM_SUCCESS; +} + +/** + * cqm_object_table_init - Initialize the association table of object and index + * @cqm_handle: cqm handle + * Return: 0 - success, negative - failure + */ +s32 cqm_object_table_init(struct cqm_handle_s *cqm_handle) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_func_capability_s *capability = &cqm_handle->func_capability; + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + struct cqm_cla_table_s *cla_table = NULL; + struct cqm_object_table_s *obj_table = NULL; + s32 ret = CQM_SUCCESS; + u32 i = 0; + + for (i = 0; i < CQM_BAT_ENTRY_MAX; i++) { + cla_table = &bat_table->entry[i]; + if (cla_table->obj_num == 0) { + cqm_info(handle->dev_hdl, + "Obj table init: cla_table_type %u, obj_num=0, don't init obj table\n", + cla_table->type); + continue; + } + + obj_table = &cla_table->obj_table; + + switch (cla_table->type) { + case CQM_BAT_ENTRY_T_QPC: + obj_table->max_num = capability->qpc_number; + ret = __cqm_object_table_init(obj_table); + break; + case CQM_BAT_ENTRY_T_MPT: + obj_table->max_num = capability->mpt_number; + ret = __cqm_object_table_init(obj_table); + break; + case CQM_BAT_ENTRY_T_SCQC: + obj_table->max_num = capability->scqc_number; + ret = __cqm_object_table_init(obj_table); + break; + case CQM_BAT_ENTRY_T_SRQC: + obj_table->max_num = capability->srqc_number; + ret = __cqm_object_table_init(obj_table); + break; + default: + break; + } + + if (ret != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, + "Obj table init: failed to init cla_table_type=%u, obj_num=0x%x\n", + cla_table->type, cla_table->obj_num); + goto err; + } + } + + return CQM_SUCCESS; + +err: + cqm_object_table_uninit(cqm_handle); + return CQM_FAIL; +} + +/** + * cqm_object_table_uninit - Deinitialize the association table of object and + * index + * @cqm_handle: cqm handle + */ +void cqm_object_table_uninit(struct cqm_handle_s *cqm_handle) +{ + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + struct cqm_cla_table_s *cla_table = NULL; + struct cqm_object_table_s *obj_table = NULL; + u32 i = 0; + + for (i = 0; i < CQM_BAT_ENTRY_MAX; i++) { + cla_table = &bat_table->entry[i]; + obj_table = &cla_table->obj_table; + if (cla_table->type != CQM_BAT_ENTRY_T_INVALID) { + if (obj_table->table) { + vfree(obj_table->table); + obj_table->table = NULL; + } + } + } +} + +/** + * cqm_object_table_insert - Insert an object, turn off the soft interrupt + * @cqm_handle: cqm handle + * @object_table: object table + * @index: the index of table + * @obj: the object to insert + * Return: 0 - success, negative - failure + */ +s32 cqm_object_table_insert(struct cqm_handle_s *cqm_handle, + struct cqm_object_table_s *object_table, u32 index, + struct cqm_object_s *obj) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + + if (index >= object_table->max_num) { + cqm_err(handle->dev_hdl, "Obj table insert: index 0x%x exceeds max_num 0x%x\n", + index, object_table->max_num); + return CQM_FAIL; + } + + write_lock(&object_table->lock); + + if (!object_table->table[index]) { + object_table->table[index] = obj; + write_unlock(&object_table->lock); + return CQM_SUCCESS; + } + write_unlock(&object_table->lock); + cqm_err(handle->dev_hdl, "Obj table insert: object_table->table[0x%x] has been inserted\n", + index); + return CQM_FAIL; +} + +/** + * cqm_object_table_remove - remove an object + * @cqm_handle: cqm handle + * @object_table: object table + * @index: the index of table + * @obj: the object to remove + * Return: 0 - success, negative - failure + */ +void cqm_object_table_remove(struct cqm_handle_s *cqm_handle, + struct cqm_object_table_s *object_table, + u32 index, const struct cqm_object_s *obj) +{ + struct hifc_hwdev *handle = cqm_handle->ex_handle; + + if (index >= object_table->max_num) { + cqm_err(handle->dev_hdl, "Obj table remove: index 0x%x exceeds max_num 0x%x\n", + index, object_table->max_num); + return; + } + + write_lock(&object_table->lock); + + if ((object_table->table[index]) && + (object_table->table[index] == obj)) { + object_table->table[index] = NULL; + } else { + cqm_err(handle->dev_hdl, "Obj table remove: object_table->table[0x%x] has been removed\n", + index); + } + + write_unlock(&object_table->lock); +} + +/** + * cqm_srq_used_rq_delete - Delete rq in TOE SRQ mode + * @object: cqm object + */ +void cqm_srq_used_rq_delete(struct cqm_object_s *object) +{ + struct cqm_queue_s *common = container_of(object, struct cqm_queue_s, + object); + struct cqm_nonrdma_qinfo_s *qinfo = container_of( + common, + struct cqm_nonrdma_qinfo_s, + common); + u32 link_wqe_offset = qinfo->wqe_per_buf * qinfo->wqe_size; + struct cqm_srq_linkwqe_s *srq_link_wqe = NULL; + dma_addr_t addr; + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + (common->object.cqm_handle); + struct hifc_hwdev *handle = cqm_handle->ex_handle; + + /* The current SRQ solution does not support the case where RQ + * initialization without container, which may cause error when RQ + * resources are released. So RQ initializes with only one container, + * and releases only one contaienr when resourced are released. + */ + CQM_PTR_CHECK_NO_RET( + common->head_container, "Rq del: rq has no contianer to release\n", + return); + + /* Get current container pa from link wqe, and ummap it */ + srq_link_wqe = (struct cqm_srq_linkwqe_s *)(common->head_container + + link_wqe_offset); + /* Convert only the big endian of the wqe part of the link */ + cqm_swab32((u8 *)(srq_link_wqe), sizeof(struct cqm_linkwqe_s) >> 2); + + addr = CQM_ADDR_COMBINE(srq_link_wqe->current_buffer_gpa_h, + srq_link_wqe->current_buffer_gpa_l); + if (addr == 0) { + cqm_err(handle->dev_hdl, "Rq del: buffer physical addr is null\n"); + return; + } + pci_unmap_single(cqm_handle->dev, addr, qinfo->container_size, + PCI_DMA_BIDIRECTIONAL); + + /* Get current container va from link wqe, and free it */ + addr = CQM_ADDR_COMBINE(srq_link_wqe->current_buffer_addr_h, + srq_link_wqe->current_buffer_addr_l); + if (addr == 0) { + cqm_err(handle->dev_hdl, "Rq del: buffer virtual addr is null\n"); + return; + } + kfree((void *)addr); +} + +#define obj_intern_if_section + +/** + * cqm_qpc_mpt_bitmap_alloc - Apply index from bitmap when creating qpc or mpt + * @object: cqm object + * @cla_table: cla table + * Return: 0 - success, negative - failure + */ +s32 cqm_qpc_mpt_bitmap_alloc(struct cqm_object_s *object, + struct cqm_cla_table_s *cla_table) +{ + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + object->cqm_handle; + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_qpc_mpt_s *common = container_of(object, + struct cqm_qpc_mpt_s, + object); + struct cqm_qpc_mpt_info_s *qpc_mpt_info = + container_of( + common, + struct cqm_qpc_mpt_info_s, + common); + struct cqm_bitmap_s *bitmap = &cla_table->bitmap; + u32 index = 0; + u32 count = 0; + + count = (ALIGN(object->object_size, cla_table->obj_size)) / + cla_table->obj_size; + qpc_mpt_info->index_count = count; + + if (qpc_mpt_info->common.xid == CQM_INDEX_INVALID) { + /* Allocate index normally */ + index = cqm_bitmap_alloc( + bitmap, + 1 << (cla_table->z + 1), + count, + cqm_handle->func_capability.xid_alloc_mode); + if (index < bitmap->max_num) { + qpc_mpt_info->common.xid = index; + } else { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_bitmap_alloc)); + return CQM_FAIL; + } + } else { + /* Allocate reserved index */ + index = cqm_bitmap_alloc_reserved( + bitmap, count, + qpc_mpt_info->common.xid); + if (index != qpc_mpt_info->common.xid) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_bitmap_alloc_reserved)); + return CQM_FAIL; + } + } + + return CQM_SUCCESS; +} + +static struct cqm_cla_table_s *cqm_qpc_mpt_prepare_cla_table( + struct cqm_object_s *object) +{ + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + object->cqm_handle; + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + + struct cqm_cla_table_s *cla_table = NULL; + + /* Get the corresponding cla table */ + if (object->object_type == CQM_OBJECT_SERVICE_CTX) { + cla_table = cqm_cla_table_get(bat_table, CQM_BAT_ENTRY_T_QPC); + } else { + cqm_err(handle->dev_hdl, CQM_WRONG_VALUE(object->object_type)); + return NULL; + } + + CQM_PTR_CHECK_RET(cla_table, return NULL, + CQM_FUNCTION_FAIL(cqm_cla_table_get)); + + /* Allocate index for bitmap */ + if (cqm_qpc_mpt_bitmap_alloc(object, cla_table) == CQM_FAIL) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_qpc_mpt_bitmap_alloc)); + return NULL; + } + + return cla_table; +} + +/** + * cqm_qpc_mpt_create - Create qpc or mpt + * @object: cqm object + * Return: 0 - success, negative - failure + */ +s32 cqm_qpc_mpt_create(struct cqm_object_s *object) +{ + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + object->cqm_handle; + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_qpc_mpt_s *common = + container_of(object, struct cqm_qpc_mpt_s, object); + struct cqm_qpc_mpt_info_s *qpc_mpt_info = + container_of(common, struct cqm_qpc_mpt_info_s, common); + struct cqm_cla_table_s *cla_table = NULL; + struct cqm_bitmap_s *bitmap = NULL; + struct cqm_object_table_s *object_table = NULL; + u32 index = 0; + u32 count = 0; + + cla_table = cqm_qpc_mpt_prepare_cla_table(object); + CQM_PTR_CHECK_RET(cla_table, return CQM_FAIL, + CQM_FUNCTION_FAIL(cqm_qpc_mpt_prepare_cla_table)); + + bitmap = &cla_table->bitmap; + index = qpc_mpt_info->common.xid; + count = qpc_mpt_info->index_count; + + /* Find the trunk page from BAT/CLA and allocate the buffer, the + * business needs to ensure that the released buffer has been cleared + */ + if (cla_table->alloc_static == true) { + qpc_mpt_info->common.vaddr = + cqm_static_qpc_cla_get(cqm_handle, cla_table, + index, count, &common->paddr); + } else { + qpc_mpt_info->common.vaddr = + cqm_cla_get(cqm_handle, cla_table, + index, count, &common->paddr); + } + if (!qpc_mpt_info->common.vaddr) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_cla_get)); + cqm_err(handle->dev_hdl, + "Qpc mpt init: qpc mpt vaddr is null, cla_table->alloc_static=%d\n", + cla_table->alloc_static); + goto err1; + } + + /* Associate index with object, FC executes in interrupt context */ + object_table = &cla_table->obj_table; + + if (cqm_object_table_insert(cqm_handle, object_table, index, object) != + CQM_SUCCESS) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_object_table_insert)); + goto err2; + } + + return CQM_SUCCESS; + +err2: + cqm_cla_put(cqm_handle, cla_table, index, count); +err1: + cqm_bitmap_free(bitmap, index, count); + return CQM_FAIL; +} + +/** + * cqm_qpc_mpt_delete - Delete qpc or mpt + * @object: cqm object + */ +void cqm_qpc_mpt_delete(struct cqm_object_s *object) +{ + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + object->cqm_handle; + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_qpc_mpt_s *common = container_of(object, + struct cqm_qpc_mpt_s, + object); + struct cqm_qpc_mpt_info_s *qpc_mpt_info = container_of( + common, + struct cqm_qpc_mpt_info_s, + common); + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + struct cqm_cla_table_s *cla_table = NULL; + struct cqm_bitmap_s *bitmap = NULL; + struct cqm_object_table_s *object_table = NULL; + u32 index = qpc_mpt_info->common.xid; + u32 count = qpc_mpt_info->index_count; + + /* Find the response cla table */ + atomic_inc(&cqm_handle->ex_handle->hw_stats.cqm_stats.cqm_qpc_mpt_delete_cnt); + + if (object->object_type == CQM_OBJECT_SERVICE_CTX) { + cla_table = cqm_cla_table_get(bat_table, CQM_BAT_ENTRY_T_QPC); + } else { + cqm_err(handle->dev_hdl, CQM_WRONG_VALUE(object->object_type)); + return; + } + + CQM_PTR_CHECK_NO_RET( + cla_table, CQM_FUNCTION_FAIL(cqm_cla_table_get), return); + + /* Disassociate index with object */ + object_table = &cla_table->obj_table; + + cqm_object_table_remove(cqm_handle, object_table, index, object); + + /* Wait for the completion and ensure that all references to the QPC + * are completed + */ + if (atomic_dec_and_test(&object->refcount)) { + complete(&object->free); + } else { + cqm_err(handle->dev_hdl, + "Qpc mpt del: object is referred by others, has to wait for completion\n"); + } + + /* The QPC static allocation needs to be non-blocking, and the service + * guarantees that the QPC is completed when the QPC is deleted + */ + if (cla_table->alloc_static == false) + wait_for_completion(&object->free); + /* Free qpc buffer */ + cqm_cla_put(cqm_handle, cla_table, index, count); + + /* Free index into bitmap */ + bitmap = &cla_table->bitmap; + cqm_bitmap_free(bitmap, index, count); +} + +/** + * cqm_linkwqe_fill - Fill link wqe for non RDMA queue buffer + * @buf: cqm buffer + * @wqe_per_buf: not include link wqe + * @wqe_size: wqe size + * @wqe_number: not include link wqe + * @tail: true linkwqe must be at the tail, false linkwqe may not be at the tail + * @link_mode: link wqe mode + */ +void cqm_linkwqe_fill(struct cqm_buf_s *buf, + u32 wqe_per_buf, + u32 wqe_size, + u32 wqe_number, + bool tail, + u8 link_mode) +{ + struct cqm_linkwqe_s *wqe = NULL; + struct cqm_linkwqe_128b_s *linkwqe = NULL; + u8 *va = NULL; + u32 i = 0; + dma_addr_t addr; + + /* Except for the last buffer, the linkwqe of other buffers is directly + * filled to the tail + */ + for (i = 0; i < buf->buf_number; i++) { + va = (u8 *)(buf->buf_list[i].va); + + if (i != (buf->buf_number - 1)) { + wqe = (struct cqm_linkwqe_s *)(va + (u32)(wqe_size * + wqe_per_buf)); + wqe->wf = CQM_WQE_WF_LINK; + wqe->ctrlsl = CQM_LINK_WQE_CTRLSL_VALUE; + wqe->lp = CQM_LINK_WQE_LP_INVALID; + /* The Obit of valid link wqe needs to be set to 1, and + * each service needs to confirm that o-bit=1 means + * valid, o-bit=0 means invalid + */ + wqe->o = CQM_LINK_WQE_OWNER_VALID; + addr = buf->buf_list[(u32)(i + 1)].pa; + wqe->next_page_gpa_h = CQM_ADDR_HI(addr); + wqe->next_page_gpa_l = CQM_ADDR_LW(addr); + } else { + /* The last buffer of Linkwqe must fill specially */ + if (tail == true) { + /* Must be filled at the end of the page */ + wqe = (struct cqm_linkwqe_s *)(va + + (u32)(wqe_size * wqe_per_buf)); + } else { + /* The last linkwqe is filled immediately after + * the last wqe + */ + wqe = (struct cqm_linkwqe_s *) + (va + (u32)(wqe_size * + (wqe_number - wqe_per_buf * + (buf->buf_number - 1)))); + } + wqe->wf = CQM_WQE_WF_LINK; + wqe->ctrlsl = CQM_LINK_WQE_CTRLSL_VALUE; + + /* In link mode, the last link wqe is invalid, In ring + * mode, the last link wqe is valid, pointing to the + * home page, and lp is set + */ + if (link_mode == CQM_QUEUE_LINK_MODE) { + wqe->o = CQM_LINK_WQE_OWNER_INVALID; + } else { + /* The lp field of the last link_wqe is filled + * with 1,indicating that the o-bit is flipped + * from here + */ + wqe->lp = CQM_LINK_WQE_LP_VALID; + wqe->o = CQM_LINK_WQE_OWNER_VALID; + addr = buf->buf_list[0].pa; + wqe->next_page_gpa_h = CQM_ADDR_HI(addr); + wqe->next_page_gpa_l = CQM_ADDR_LW(addr); + } + } + + if (wqe_size == 128) { + /* The 128B wqe before and after 64B have obit need to be + * assigned, For IFOE, 63th penultimate bit of last 64B is + * obit, for TOE, 157th penultimate bit of last 64B is obit + */ + linkwqe = (struct cqm_linkwqe_128b_s *)wqe; + linkwqe->second_64b.third_16B.bs.toe_o = + CQM_LINK_WQE_OWNER_VALID; + linkwqe->second_64b.forth_16B.bs.ifoe_o = + CQM_LINK_WQE_OWNER_VALID; + + /* big endian conversion */ + cqm_swab32((u8 *)wqe, + sizeof(struct cqm_linkwqe_128b_s) >> 2); + } else { + /* big endian conversion */ + cqm_swab32((u8 *)wqe, + sizeof(struct cqm_linkwqe_s) >> 2); + } + } +} + +static s32 cqm_nonrdma_queue_ctx_create_srq(struct cqm_object_s *object) +{ + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + object->cqm_handle; + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_queue_s *common = container_of(object, + struct cqm_queue_s, object); + struct cqm_nonrdma_qinfo_s *qinfo = container_of( + common, + struct cqm_nonrdma_qinfo_s, + common); + s32 shift = 0; + + shift = cqm_shift(qinfo->q_ctx_size); + common->q_ctx_vaddr = (u8 *)cqm_kmalloc_align( + qinfo->q_ctx_size, + GFP_KERNEL | __GFP_ZERO, + (u16)shift); + if (!common->q_ctx_vaddr) { + cqm_err(handle->dev_hdl, CQM_ALLOC_FAIL(q_ctx_vaddr)); + return CQM_FAIL; + } + + common->q_ctx_paddr = + pci_map_single(cqm_handle->dev, common->q_ctx_vaddr, + qinfo->q_ctx_size, PCI_DMA_BIDIRECTIONAL); + if (pci_dma_mapping_error(cqm_handle->dev, common->q_ctx_paddr)) { + cqm_err(handle->dev_hdl, CQM_MAP_FAIL(q_ctx_vaddr)); + cqm_kfree_align(common->q_ctx_vaddr); + common->q_ctx_vaddr = NULL; + return CQM_FAIL; + } + + return CQM_SUCCESS; +} + +static s32 cqm_nonrdma_queue_ctx_create_scq(struct cqm_object_s *object) +{ + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + object->cqm_handle; + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_queue_s *common = container_of(object, + struct cqm_queue_s, + object); + struct cqm_nonrdma_qinfo_s *qinfo = container_of( + common, + struct cqm_nonrdma_qinfo_s, + common); + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + struct cqm_cla_table_s *cla_table = NULL; + struct cqm_bitmap_s *bitmap = NULL; + struct cqm_object_table_s *object_table = NULL; + + /* Find the corresponding cla table */ + cla_table = cqm_cla_table_get(bat_table, CQM_BAT_ENTRY_T_SCQC); + if (!cla_table) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_cla_table_get)); + return CQM_FAIL; + } + + /* Allocate index for bitmap */ + bitmap = &cla_table->bitmap; + qinfo->index_count = (ALIGN(qinfo->q_ctx_size, cla_table->obj_size)) / + cla_table->obj_size; + qinfo->common.index = cqm_bitmap_alloc(bitmap, 1 << (cla_table->z + 1), + qinfo->index_count, cqm_handle->func_capability.xid_alloc_mode); + if (qinfo->common.index >= bitmap->max_num) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_bitmap_alloc)); + return CQM_FAIL; + } + + /* Find the trunk page from BAT/CLA and allocate buffer */ + common->q_ctx_vaddr = cqm_cla_get(cqm_handle, cla_table, + qinfo->common.index, + qinfo->index_count, + &common->q_ctx_paddr); + if (!common->q_ctx_vaddr) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_cla_get)); + cqm_bitmap_free(bitmap, qinfo->common.index, + qinfo->index_count); + return CQM_FAIL; + } + + /* Associate index with object */ + object_table = &cla_table->obj_table; + + if (cqm_object_table_insert( + cqm_handle, object_table, + qinfo->common.index, object) != CQM_SUCCESS) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_object_table_insert)); + cqm_cla_put(cqm_handle, cla_table, qinfo->common.index, + qinfo->index_count); + cqm_bitmap_free(bitmap, qinfo->common.index, + qinfo->index_count); + return CQM_FAIL; + } + + return CQM_SUCCESS; +} + +s32 cqm_nonrdma_queue_ctx_create(struct cqm_object_s *object) +{ + if (object->object_type == CQM_OBJECT_NONRDMA_SRQ) + return cqm_nonrdma_queue_ctx_create_srq(object); + else if (object->object_type == CQM_OBJECT_NONRDMA_SCQ) + return cqm_nonrdma_queue_ctx_create_scq(object); + + return CQM_SUCCESS; +} + +#define CQM_NORDMA_CHECK_WEQ_NUMBER(number) \ + (((number) < CQM_CQ_DEPTH_MIN) || ((number) > CQM_CQ_DEPTH_MAX)) + +/** + * cqm_nonrdma_queue_create - Create queue for non RDMA service + * @buf: cqm object + * Return: 0 - success, negative - failure + */ +s32 cqm_nonrdma_queue_create(struct cqm_object_s *object) +{ + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + object->cqm_handle; + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_service_s *service = &cqm_handle->service; + struct cqm_queue_s *common = container_of(object, + struct cqm_queue_s, + object); + struct cqm_nonrdma_qinfo_s *qinfo = container_of( + common, + struct cqm_nonrdma_qinfo_s, + common); + struct cqm_buf_s *q_room_buf = &common->q_room_buf_1; + u32 wqe_number = qinfo->common.object.object_size; + u32 wqe_size = qinfo->wqe_size; + u32 order = service->buf_order; + u32 buf_number = 0; + u32 buf_size = 0; + bool tail = false; /* Whether linkwqe is at the end of the page */ + + /* When creating CQ/SCQ queue, the page size is 4k, linkwqe must be at + * the end of the page + */ + if (object->object_type == CQM_OBJECT_NONRDMA_SCQ) { + /* Depth must be 2^n alignment, depth range is 256~32K */ + if (CQM_NORDMA_CHECK_WEQ_NUMBER(wqe_number)) { + cqm_err(handle->dev_hdl, CQM_WRONG_VALUE(wqe_number)); + return CQM_FAIL; + } + if (cqm_check_align(wqe_number) == false) { + cqm_err(handle->dev_hdl, "Nonrdma queue alloc: wqe_number is not align on 2^n\n"); + return CQM_FAIL; + } + + order = CQM_4K_PAGE_ORDER; /* wqe page is 4k */ + tail = true; /* linkwqe must be at the end of the page */ + buf_size = CQM_4K_PAGE_SIZE; + } else { + buf_size = PAGE_SIZE << order; + } + + /* Calculate how many buffers are required, -1 is to deduct link wqe in + * a buf + */ + qinfo->wqe_per_buf = (buf_size / wqe_size) - 1; + /* The depth from service includes the number of linkwqe */ + buf_number = ALIGN((wqe_size * wqe_number), buf_size) / buf_size; + /* Allocate cqm buffer */ + q_room_buf->buf_number = buf_number; + q_room_buf->buf_size = buf_size; + q_room_buf->page_number = (buf_number << order); + if (cqm_buf_alloc(cqm_handle, q_room_buf, false) == CQM_FAIL) { + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_buf_alloc)); + return CQM_FAIL; + } + /* Fill link wqe, (wqe_number - buf_number) is the number of wqe without + * linkwqe + */ + cqm_linkwqe_fill(q_room_buf, qinfo->wqe_per_buf, wqe_size, + wqe_number - buf_number, tail, + common->queue_link_mode); + + /* Create queue header */ + qinfo->common.q_header_vaddr = + (struct cqm_queue_header_s *)cqm_kmalloc_align( + sizeof(struct cqm_queue_header_s), + GFP_KERNEL | __GFP_ZERO, CQM_QHEAD_ALIGN_ORDER); + if (!qinfo->common.q_header_vaddr) { + cqm_err(handle->dev_hdl, CQM_ALLOC_FAIL(q_header_vaddr)); + goto err1; + } + + common->q_header_paddr = + pci_map_single(cqm_handle->dev, + qinfo->common.q_header_vaddr, + sizeof(struct cqm_queue_header_s), + PCI_DMA_BIDIRECTIONAL); + if (pci_dma_mapping_error(cqm_handle->dev, common->q_header_paddr)) { + cqm_err(handle->dev_hdl, CQM_MAP_FAIL(q_header_vaddr)); + goto err2; + } + + /* Create queue ctx */ + if (cqm_nonrdma_queue_ctx_create(object) == CQM_FAIL) { + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_nonrdma_queue_ctx_create)); + goto err3; + } + + return CQM_SUCCESS; + +err3: + pci_unmap_single(cqm_handle->dev, common->q_header_paddr, + sizeof(struct cqm_queue_header_s), + PCI_DMA_BIDIRECTIONAL); +err2: + cqm_kfree_align(qinfo->common.q_header_vaddr); + qinfo->common.q_header_vaddr = NULL; +err1: + cqm_buf_free(q_room_buf, cqm_handle->dev); + return CQM_FAIL; +} + +static void cqm_nonrdma_queue_free_scq_srq(struct cqm_object_s *object, + struct cqm_cla_table_s *cla_table) +{ + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + object->cqm_handle; + struct cqm_queue_s *common = container_of(object, + struct cqm_queue_s, + object); + struct cqm_nonrdma_qinfo_s *qinfo = + container_of(common, struct cqm_nonrdma_qinfo_s, common); + struct cqm_buf_s *q_room_buf = &common->q_room_buf_1; + u32 index = qinfo->common.index; + u32 count = qinfo->index_count; + struct cqm_bitmap_s *bitmap = NULL; + + /* If it is in TOE SRQ mode, delete the RQ */ + if (common->queue_link_mode == CQM_QUEUE_TOE_SRQ_LINK_MODE) { + cqm_dbg("Nonrdma queue del: delete srq used rq\n"); + cqm_srq_used_rq_delete(&common->object); + } else { + /* Free it if exists q room */ + cqm_buf_free(q_room_buf, cqm_handle->dev); + } + /* Free SRQ or SCQ ctx */ + if (object->object_type == CQM_OBJECT_NONRDMA_SRQ) { + /* ctx of nonrdma's SRQ is applied independently */ + if (common->q_ctx_vaddr) { + pci_unmap_single(cqm_handle->dev, common->q_ctx_paddr, + qinfo->q_ctx_size, + PCI_DMA_BIDIRECTIONAL); + + cqm_kfree_align(common->q_ctx_vaddr); + common->q_ctx_vaddr = NULL; + } + } else if (object->object_type == CQM_OBJECT_NONRDMA_SCQ) { + /* The ctx of SCQ of nonrdma is managed by BAT/CLA */ + cqm_cla_put(cqm_handle, cla_table, index, count); + + /* Release index into bitmap */ + bitmap = &cla_table->bitmap; + cqm_bitmap_free(bitmap, index, count); + } +} + +/** + * cqm_nonrdma_queue_delete - Free queue for non RDMA service + * @buf: cqm object + */ +void cqm_nonrdma_queue_delete(struct cqm_object_s *object) +{ + struct cqm_handle_s *cqm_handle = (struct cqm_handle_s *) + object->cqm_handle; + struct hifc_hwdev *handle = cqm_handle->ex_handle; + struct cqm_queue_s *common = container_of(object, + struct cqm_queue_s, object); + struct cqm_nonrdma_qinfo_s *qinfo = container_of( + common, + struct cqm_nonrdma_qinfo_s, + common); + struct cqm_bat_table_s *bat_table = &cqm_handle->bat_table; + struct cqm_cla_table_s *cla_table = NULL; + struct cqm_object_table_s *object_table = NULL; + u32 index = qinfo->common.index; + + atomic_inc(&cqm_handle->ex_handle->hw_stats.cqm_stats.cqm_nonrdma_queue_delete_cnt); + + /* SCQ has independent SCQN association */ + if (object->object_type == CQM_OBJECT_NONRDMA_SCQ) { + cla_table = cqm_cla_table_get(bat_table, CQM_BAT_ENTRY_T_SCQC); + CQM_PTR_CHECK_NO_RET( + cla_table, + CQM_FUNCTION_FAIL(cqm_cla_table_get), + return); + + /* index and object disassociate */ + object_table = &cla_table->obj_table; + + cqm_object_table_remove(cqm_handle, object_table, + index, object); + } + + /* Wait for the completion and ensure that all references to the QPC + * are completed + */ + if (atomic_dec_and_test(&object->refcount)) + complete(&object->free); + else + cqm_err(handle->dev_hdl, "Nonrdma queue del: object is referred by others, has to wait for completion\n"); + wait_for_completion(&object->free); + + /* Free it if exists q header */ + if (qinfo->common.q_header_vaddr) { + pci_unmap_single(cqm_handle->dev, common->q_header_paddr, + sizeof(struct cqm_queue_header_s), + PCI_DMA_BIDIRECTIONAL); + + cqm_kfree_align(qinfo->common.q_header_vaddr); + qinfo->common.q_header_vaddr = NULL; + } + cqm_nonrdma_queue_free_scq_srq(object, cla_table); +} + +#define obj_extern_if_section + +/** + * cqm_object_qpc_mpt_create - Create QPC and MPT + * @ex_handle: hw dev handle + * @service_type: service type + * @object_type: must be mpt and ctx + * @object_size: the unit is byte + * @object_priv: the private structure for service, can be NULL + * @index: get the reserved qpn based on this value, if wants to automatically + * allocate it, the value should be CQM_INDEX_INVALID + * Return: service ctx + */ +struct cqm_qpc_mpt_s *cqm_object_qpc_mpt_create( + void *ex_handle, + enum cqm_object_type_e object_type, + u32 object_size, void *object_priv, + u32 index) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + struct cqm_qpc_mpt_info_s *qpc_mpt_info = NULL; + s32 ret = CQM_FAIL; + + CQM_PTR_CHECK_RET(ex_handle, return NULL, CQM_PTR_NULL(ex_handle)); + + atomic_inc(&handle->hw_stats.cqm_stats.cqm_qpc_mpt_create_cnt); + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + CQM_PTR_CHECK_RET(cqm_handle, return NULL, CQM_PTR_NULL(cqm_handle)); + + /* If service does not register, returns NULL */ + if (cqm_handle->service.has_register == false) { + cqm_err(handle->dev_hdl, "service is not register"); + return NULL; + } + + if (object_type != CQM_OBJECT_SERVICE_CTX) { + cqm_err(handle->dev_hdl, CQM_WRONG_VALUE(object_type)); + return NULL; + } + + qpc_mpt_info = (struct cqm_qpc_mpt_info_s *) + kmalloc(sizeof(struct cqm_qpc_mpt_info_s), + GFP_ATOMIC | __GFP_ZERO); + CQM_PTR_CHECK_RET(qpc_mpt_info, return NULL, + CQM_ALLOC_FAIL(qpc_mpt_info)); + + qpc_mpt_info->common.object.object_type = object_type; + qpc_mpt_info->common.object.object_size = object_size; + atomic_set(&qpc_mpt_info->common.object.refcount, 1); + init_completion(&qpc_mpt_info->common.object.free); + qpc_mpt_info->common.object.cqm_handle = cqm_handle; + qpc_mpt_info->common.xid = index; + qpc_mpt_info->common.priv = object_priv; + + ret = cqm_qpc_mpt_create(&qpc_mpt_info->common.object); + if (ret == CQM_SUCCESS) + return &qpc_mpt_info->common; + + cqm_err(handle->dev_hdl, CQM_FUNCTION_FAIL(cqm_qpc_mpt_create)); + kfree(qpc_mpt_info); + return NULL; +} + +/** + * cqm_object_fc_srq_create - Create RQ for FC, the number of valid wqe in the + * queue must be meet the incoming wqe number. Because linkwqe can only be + * filled at the end of the page, the actual effective number exceeds demand, + * need to inform the number of business creation. + * @ex_handle: hw dev handle + * @service_type: service type + * @object_type: must be CQM_OBJECT_NONRDMA_SRQ + * @wqe_number: valid wqe number + * @wqe_size: wqe size + * @object_priv: the private structure for service + * Return: srq structure + */ +struct cqm_queue_s *cqm_object_fc_srq_create( + void *ex_handle, + enum cqm_object_type_e object_type, + u32 wqe_number, u32 wqe_size, + void *object_priv) +{ + struct cqm_nonrdma_qinfo_s *nonrdma_qinfo = NULL; + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + struct cqm_service_s *service = NULL; + u32 valid_wqe_per_buffer = 0; + u32 wqe_sum = 0; /* includes linkwqe, normal wqe */ + u32 buf_size = 0; + u32 buf_num = 0; + s32 ret = CQM_FAIL; + + CQM_PTR_CHECK_RET(ex_handle, return NULL, CQM_PTR_NULL(ex_handle)); + + atomic_inc(&handle->hw_stats.cqm_stats.cqm_fc_srq_create_cnt); + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + CQM_PTR_CHECK_RET(cqm_handle, return NULL, CQM_PTR_NULL(cqm_handle)); + + /* service_type must be FC */ + if (cqm_handle->service.has_register == false) { + cqm_err(handle->dev_hdl, "service is not register\n"); + return NULL; + } + + /* wqe_size can not exceed PAGE_SIZE and should not be 0, and must be + * 2^n aligned. + */ + if ((wqe_size >= PAGE_SIZE) || (cqm_check_align(wqe_size) == false)) { + cqm_err(handle->dev_hdl, CQM_WRONG_VALUE(wqe_size)); + return NULL; + } + + /* FC's RQ is SRQ (unlike TOE's SRQ, fc is that all packets received by + * the stream will be put on the same rq, and TOE's srq is similar to + * rq's resource pool) + */ + if (object_type != CQM_OBJECT_NONRDMA_SRQ) { + cqm_err(handle->dev_hdl, CQM_WRONG_VALUE(object_type)); + return NULL; + } + + service = &cqm_handle->service; + buf_size = PAGE_SIZE << (service->buf_order); + valid_wqe_per_buffer = buf_size / wqe_size - 1; /* Minus 1 link wqe */ + buf_num = wqe_number / valid_wqe_per_buffer; + if (wqe_number % valid_wqe_per_buffer != 0) + buf_num++; + + /* Calculate the total number of all wqe */ + wqe_sum = buf_num * (valid_wqe_per_buffer + 1); + nonrdma_qinfo = (struct cqm_nonrdma_qinfo_s *) + kmalloc(sizeof(struct cqm_nonrdma_qinfo_s), + GFP_KERNEL | __GFP_ZERO); + + CQM_PTR_CHECK_RET(nonrdma_qinfo, return NULL, + CQM_ALLOC_FAIL(nonrdma_qinfo)); + + /* Initialize object members */ + nonrdma_qinfo->common.object.object_type = object_type; + /* The total number of all wqe */ + nonrdma_qinfo->common.object.object_size = wqe_sum; + atomic_set(&nonrdma_qinfo->common.object.refcount, 1); + init_completion(&nonrdma_qinfo->common.object.free); + nonrdma_qinfo->common.object.cqm_handle = cqm_handle; + + /* Initialize the doorbell used by the current queue, default is the + * hardware doorbell + */ + nonrdma_qinfo->common.current_q_doorbell = CQM_HARDWARE_DOORBELL; + nonrdma_qinfo->common.queue_link_mode = CQM_QUEUE_RING_MODE; + + /* Initialize external public members */ + nonrdma_qinfo->common.priv = object_priv; + nonrdma_qinfo->common.valid_wqe_num = wqe_sum - buf_num; + + /* Initialize internal private members */ + nonrdma_qinfo->wqe_size = wqe_size; + /* The SRQ for FC, which needs to create ctx */ + nonrdma_qinfo->q_ctx_size = service->service_template.srq_ctx_size; + + ret = cqm_nonrdma_queue_create(&nonrdma_qinfo->common.object); + if (ret == CQM_SUCCESS) + return &nonrdma_qinfo->common; + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_nonrdma_queue_create)); + kfree(nonrdma_qinfo); + return NULL; +} + +static int cqm_object_nonrdma_queue_create_check( + void *ex_handle, + enum cqm_object_type_e object_type, + u32 wqe_size) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + + CQM_PTR_CHECK_RET(ex_handle, return CQM_FAIL, CQM_PTR_NULL(ex_handle)); + + atomic_inc(&handle->hw_stats.cqm_stats.cqm_nonrdma_queue_create_cnt); + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + CQM_PTR_CHECK_RET(cqm_handle, return CQM_FAIL, + CQM_PTR_NULL(cqm_handle)); + + /* If service does not register, returns NULL */ + if (cqm_handle->service.has_register == false) { + cqm_err(handle->dev_hdl, "service is not register\n"); + return CQM_FAIL; + } + /* Wqe size cannot exceed PAGE_SIZE, cannot be 0, and must be 2^n + * aligned. cqm_check_align check excludes 0, 1, non 2^n alignment + */ + if ((wqe_size >= PAGE_SIZE) || (cqm_check_align(wqe_size) == false)) { + cqm_err(handle->dev_hdl, CQM_WRONG_VALUE(wqe_size)); + return CQM_FAIL; + } + + /* Supported Nonrdma queue: RQ, SQ, SRQ, CQ, SCQ */ + if ((object_type < CQM_OBJECT_NONRDMA_EMBEDDED_RQ) || + (object_type > CQM_OBJECT_NONRDMA_SCQ)) { + cqm_err(handle->dev_hdl, CQM_WRONG_VALUE(object_type)); + return CQM_FAIL; + } + + return CQM_SUCCESS; +} + +/** + * cqm_object_nonrdma_queue_create - Create queues for non-RDMA services + * @ex_handle: hw dev handle + * @service_type: service type + * @object_type: can create embedded RQ/SQ/CQ and SRQ/SCQ + * @wqe_number: wqe number, including link wqe + * @wqe_size: wqe size, nust be 2^n + * @object_priv: the private structure for service, can be NULL + * Return: srq structure + */ +struct cqm_queue_s *cqm_object_nonrdma_queue_create( + void *ex_handle, + enum cqm_object_type_e object_type, + u32 wqe_number, u32 wqe_size, + void *object_priv) +{ + struct hifc_hwdev *handle = (struct hifc_hwdev *)ex_handle; + struct cqm_handle_s *cqm_handle = NULL; + struct cqm_nonrdma_qinfo_s *nonrdma_qinfo = NULL; + struct cqm_service_s *service = NULL; + s32 ret = CQM_FAIL; + + cqm_handle = (struct cqm_handle_s *)(handle->cqm_hdl); + if (cqm_object_nonrdma_queue_create_check(ex_handle, + object_type, + wqe_size) == CQM_FAIL) { + return NULL; + } + + nonrdma_qinfo = (struct cqm_nonrdma_qinfo_s *) + kmalloc(sizeof(struct cqm_nonrdma_qinfo_s), + GFP_KERNEL | __GFP_ZERO); + CQM_PTR_CHECK_RET(nonrdma_qinfo, return NULL, + CQM_ALLOC_FAIL(nonrdma_qinfo)); + + /* Initialize object members */ + nonrdma_qinfo->common.object.object_type = object_type; + nonrdma_qinfo->common.object.object_size = wqe_number; + atomic_set(&nonrdma_qinfo->common.object.refcount, 1); + init_completion(&nonrdma_qinfo->common.object.free); + nonrdma_qinfo->common.object.cqm_handle = cqm_handle; + + /* Initialize the doorbell used by the current queue, default is the + * hardware doorbell + */ + nonrdma_qinfo->common.current_q_doorbell = CQM_HARDWARE_DOORBELL; + nonrdma_qinfo->common.queue_link_mode = CQM_QUEUE_RING_MODE; + + /* Initialize external public members */ + nonrdma_qinfo->common.priv = object_priv; + + /* Initialize internal private members */ + nonrdma_qinfo->wqe_size = wqe_size; + service = &cqm_handle->service; + switch (object_type) { + case CQM_OBJECT_NONRDMA_SCQ: + nonrdma_qinfo->q_ctx_size = + service->service_template.scq_ctx_size; + break; + case CQM_OBJECT_NONRDMA_SRQ: + /* The creation for SRQ uses a dedicated interface */ + nonrdma_qinfo->q_ctx_size = + service->service_template.srq_ctx_size; + break; + default: + break; + } + + ret = cqm_nonrdma_queue_create(&nonrdma_qinfo->common.object); + if (ret == CQM_SUCCESS) + return &nonrdma_qinfo->common; + + cqm_err(handle->dev_hdl, + CQM_FUNCTION_FAIL(cqm_nonrdma_queue_create)); + kfree(nonrdma_qinfo); + return NULL; +} + +s32 cqm_qpc_mpt_delete_ret(struct cqm_object_s *object) +{ + u32 object_type = 0; + + object_type = object->object_type; + switch (object_type) { + case CQM_OBJECT_SERVICE_CTX: + cqm_qpc_mpt_delete(object); + return CQM_SUCCESS; + default: + return CQM_FAIL; + } +} + +s32 cqm_nonrdma_queue_delete_ret(struct cqm_object_s *object) +{ + u32 object_type = 0; + + object_type = object->object_type; + switch (object_type) { + case CQM_OBJECT_NONRDMA_SCQ: + case CQM_OBJECT_NONRDMA_SRQ: + cqm_nonrdma_queue_delete(object); + return CQM_SUCCESS; + default: + return CQM_FAIL; + } +} + +/** + * cqm_object_nonrdma_queue_create - Delete the created object, the function + * will sleep and wait for all operations on the object to complete before + * returning + * @object: cqm object + */ +void cqm_object_delete(struct cqm_object_s *object) +{ + struct cqm_handle_s *cqm_handle = NULL; + struct hifc_hwdev *handle = NULL; + + CQM_PTR_CHECK_NO_RET(object, CQM_PTR_NULL(object), return); + if (!object->cqm_handle) { + pr_err("[CQM]Obj del: cqm_handle is null, refcount %d\n", + (int)object->refcount.counter); + kfree(object); + return; + } + cqm_handle = (struct cqm_handle_s *)object->cqm_handle; + + if (!cqm_handle->ex_handle) { + pr_err("[CQM]Obj del: ex_handle is null, refcount %d\n", + (int)object->refcount.counter); + kfree(object); + return; + } + handle = cqm_handle->ex_handle; + + if (cqm_qpc_mpt_delete_ret(object) == CQM_SUCCESS) { + kfree(object); + return; + } + + if (cqm_nonrdma_queue_delete_ret(object) == CQM_SUCCESS) { + kfree(object); + return; + } + + cqm_err(handle->dev_hdl, CQM_WRONG_VALUE(object->object_type)); + kfree(object); +} diff --git a/drivers/scsi/huawei/hifc/hifc_cqm_object.h b/drivers/scsi/huawei/hifc/hifc_cqm_object.h new file mode 100644 index 000000000000..308166ddd534 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_cqm_object.h @@ -0,0 +1,244 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#ifndef __CQM_OBJECT_H__ +#define __CQM_OBJECT_H__ + +#define CLA_TABLE_PAGE_ORDER (0) +#define CQM_4K_PAGE_ORDER (0) + +#define CQM_CQ_DEPTH_MAX (32768) +#define CQM_CQ_DEPTH_MIN (256) +#define CQM_BAT_SIZE_FT_PF (192) + +#define CQM_WQE_WF_LINK 1 +#define CQM_WQE_WF_NORMAL 0 +#define CQM_QUEUE_LINK_MODE 0 +#define CQM_QUEUE_RING_MODE 1 +#define CQM_4K_PAGE_SIZE 4096 + +#define CQM_SUCCESS 0 +#define CQM_FAIL -1 +#define CQM_QUEUE_TOE_SRQ_LINK_MODE 2 +#define CQM_CMD_TIMEOUT 10000 /*ms*/ + +#define CQM_INDEX_INVALID ~(0U) +#define CQM_INDEX_RESERVED (0xfffff) /* reserved by cqm alloc */ + +enum cqm_bat_entry_type_e { + CQM_BAT_ENTRY_T_CFG = 0, + CQM_BAT_ENTRY_T_HASH, + CQM_BAT_ENTRY_T_QPC, + CQM_BAT_ENTRY_T_SCQC, + CQM_BAT_ENTRY_T_SRQC, + CQM_BAT_ENTRY_T_MPT, + CQM_BAT_ENTRY_T_GID, + CQM_BAT_ENTRY_T_LUN, + CQM_BAT_ENTRY_T_TASKMAP, + CQM_BAT_ENTRY_T_L3I, + CQM_BAT_ENTRY_T_CHILDC, + CQM_BAT_ENTRY_T_TIMER, + CQM_BAT_ENTRY_T_XID2CID, + CQM_BAT_ENTRY_T_REORDER, + + CQM_BAT_ENTRY_T_INVALID = 0xff, +}; + +enum cqm_cmd_type_e { + CQM_CMD_T_INVALID = 0, + CQM_CMD_T_BAT_UPDATE, + CQM_CMD_T_CLA_UPDATE, + CQM_CMD_T_BLOOMFILTER_SET, + CQM_CMD_T_BLOOMFILTER_CLEAR, + CQM_CMD_T_COMPACT_SRQ_UPDATE, + CQM_CMD_T_CLA_CACHE_INVALID, + CQM_CMD_T_BLOOMFILTER_INIT, + QM_CMD_T_MAX +}; + +/*linkwqe*/ +#define CQM_LINK_WQE_CTRLSL_VALUE 2 +#define CQM_LINK_WQE_LP_VALID 1 +#define CQM_LINK_WQE_LP_INVALID 0 +#define CQM_LINK_WQE_OWNER_VALID 1 +#define CQM_LINK_WQE_OWNER_INVALID 0 + +/*CLA update mode*/ +#define CQM_CLA_RECORD_NEW_GPA 0 +#define CQM_CLA_DEL_GPA_WITHOUT_CACHE_INVALID 1 +#define CQM_CLA_DEL_GPA_WITH_CACHE_INVALID 2 + +#define CQM_CLA_LVL_0 0 +#define CQM_CLA_LVL_1 1 +#define CQM_CLA_LVL_2 2 + +#define CQM_MAX_INDEX_BIT 19 +#define CQM_CHIP_CACHELINE 256 +enum cqm_cmd_ack_type_e { + CQM_CMD_ACK_TYPE_CMDQ = 0, /* ack: write back to cmdq */ + CQM_CMD_ACK_TYPE_SHARE_CQN = 1, /* ack report scq by root ctx ctx */ + CQM_CMD_ACK_TYPE_APP_CQN = 2 /* ack report scq by parent ctx */ +}; + +struct cqm_bat_entry_cfg_s { + u32 cur_conn_num_h_4 :4; + u32 rsv1 :4; + u32 max_conn_num :20; + u32 rsv2 :4; + + u32 max_conn_cache :10; + u32 rsv3 :6; + u32 cur_conn_num_l_16 :16; + + u32 bloom_filter_addr :16; + u32 cur_conn_cache :10; + u32 rsv4 :6; + + u32 bucket_num :16; + u32 bloom_filter_len :16; +}; + +#define CQM_BAT_NO_BYPASS_CACHE 0 +#define CQM_BAT_ENTRY_SIZE_256 0 +#define CQM_BAT_ENTRY_SIZE_512 1 +#define CQM_BAT_ENTRY_SIZE_1024 2 + +struct cqm_bat_entry_standerd_s { + u32 entry_size :2; + u32 rsv1 :6; + u32 max_number :20; + u32 rsv2 :4; + + u32 cla_gpa_h :32; + + u32 cla_gpa_l :32; + + u32 rsv3 :8; + u32 z :5; + u32 y :5; + u32 x :5; + u32 rsv24 :1; + u32 bypass :1; + u32 cla_level :2; + u32 rsv5 :5; +}; + +struct cqm_bat_entry_taskmap_s { + u32 gpa0_h; + u32 gpa0_l; + + u32 gpa1_h; + u32 gpa1_l; + + u32 gpa2_h; + u32 gpa2_l; + + u32 gpa3_h; + u32 gpa3_l; +}; + +struct cqm_cla_cache_invalid_cmd_s { + u32 gpa_h; + u32 gpa_l; + u32 cache_size;/* CLA cache size=4096B */ +}; + +struct cqm_cla_update_cmd_s { + /* need to update gpa addr */ + u32 gpa_h; + u32 gpa_l; + + /* update value */ + u32 value_h; + u32 value_l; +}; + +struct cqm_bat_update_cmd_s { +#define CQM_BAT_MAX_SIZE 256 + u32 offset; /* byte offset,16Byte aligned */ + u32 byte_len; /* max size: 256byte */ + u8 data[CQM_BAT_MAX_SIZE]; +}; + +struct cqm_handle_s; + +struct cqm_linkwqe_s { + u32 rsv1 :14; + u32 wf :1; + u32 rsv2 :14; + u32 ctrlsl :2; + u32 o :1; + + u32 rsv3 :31; + u32 lp :1; + + u32 next_page_gpa_h; + u32 next_page_gpa_l; + + u32 next_buffer_addr_h; + u32 next_buffer_addr_l; +}; + +struct cqm_srq_linkwqe_s { + struct cqm_linkwqe_s linkwqe; + /*add by wss for srq*/ + u32 current_buffer_gpa_h; + u32 current_buffer_gpa_l; + u32 current_buffer_addr_h; + u32 current_buffer_addr_l; + + u32 fast_link_page_addr_h; + u32 fast_link_page_addr_l; + + u32 fixed_next_buffer_addr_h; + u32 fixed_next_buffer_addr_l; +}; + +union cqm_linkwqe_first_64b_s { + struct cqm_linkwqe_s basic_linkwqe; + u32 value[16]; +}; + +struct cqm_linkwqe_second_64b_s { + u32 rsvd0[4]; + u32 rsvd1[4]; + union { + struct { + u32 rsvd0[3]; + u32 rsvd1 :29; + u32 toe_o :1; + u32 resvd2 :2; + } bs; + u32 value[4]; + } third_16B; + + union { + struct { + u32 rsvd0[2]; + u32 rsvd1 :31; + u32 ifoe_o :1; + u32 rsvd2; + } bs; + u32 value[4]; + } forth_16B; + +}; + +struct cqm_linkwqe_128b_s { + union cqm_linkwqe_first_64b_s first_64b; + struct cqm_linkwqe_second_64b_s second_64b; +}; + +s32 cqm_bat_init(struct cqm_handle_s *cqm_handle); +void cqm_bat_uninit(struct cqm_handle_s *cqm_handle); +s32 cqm_cla_init(struct cqm_handle_s *cqm_handle); +void cqm_cla_uninit(struct cqm_handle_s *cqm_handle); +s32 cqm_bitmap_init(struct cqm_handle_s *cqm_handle); +void cqm_bitmap_uninit(struct cqm_handle_s *cqm_handle); +s32 cqm_object_table_init(struct cqm_handle_s *cqm_handle); +void cqm_object_table_uninit(struct cqm_handle_s *cqm_handle); + +#endif /* __CQM_OBJECT_H__ */ diff --git a/drivers/scsi/huawei/hifc/hifc_eqs.c b/drivers/scsi/huawei/hifc/hifc_eqs.c new file mode 100644 index 000000000000..803866e1fbf9 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_eqs.c @@ -0,0 +1,1347 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": [COMM]" fmt + +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/interrupt.h> +#include <linux/workqueue.h> +#include <linux/pci.h> +#include <linux/kernel.h> +#include <linux/device.h> +#include <linux/dma-mapping.h> +#include <linux/module.h> + +#include "hifc_knl_adp.h" +#include "hifc_hw.h" +#include "hifc_hwif.h" +#include "hifc_api_cmd.h" +#include "hifc_mgmt.h" +#include "hifc_hwdev.h" +#include "hifc_eqs.h" + +#define HIFC_EQS_WQ_NAME "hifc_eqs" + +#define AEQ_CTRL_0_INTR_IDX_SHIFT 0 +#define AEQ_CTRL_0_FUNC_BUSY_SHIFT 10 +#define AEQ_CTRL_0_DMA_ATTR_SHIFT 12 +#define AEQ_CTRL_0_PCI_INTF_IDX_SHIFT 20 +#define AEQ_CTRL_0_QPS_NUM_SHIFT 22 +#define AEQ_CTRL_0_INTR_MODE_SHIFT 31 + +#define AEQ_CTRL_0_INTR_IDX_MASK 0x3FFU +#define AEQ_CTRL_0_FUNC_BUSY_MASK 0x1U +#define AEQ_CTRL_0_DMA_ATTR_MASK 0x3FU +#define AEQ_CTRL_0_PCI_INTF_IDX_MASK 0x3U +#define AEQ_CTRL_0_QPS_NUM_MASK 0xFFU +#define AEQ_CTRL_0_INTR_MODE_MASK 0x1U + +#define AEQ_CTRL_0_GET(val, member) \ + (((val) >> AEQ_CTRL_0_##member##_SHIFT) & \ + AEQ_CTRL_0_##member##_MASK) + +#define AEQ_CTRL_0_SET(val, member) \ + (((val) & AEQ_CTRL_0_##member##_MASK) << \ + AEQ_CTRL_0_##member##_SHIFT) + +#define AEQ_CTRL_0_CLEAR(val, member) \ + ((val) & (~(AEQ_CTRL_0_##member##_MASK \ + << AEQ_CTRL_0_##member##_SHIFT))) + +#define AEQ_CTRL_1_LEN_SHIFT 0 +#define AEQ_CTRL_1_FUNC_OWN_SHIFT 21 +#define AEQ_CTRL_1_ELEM_SIZE_SHIFT 24 +#define AEQ_CTRL_1_PAGE_SIZE_SHIFT 28 + +#define AEQ_CTRL_1_LEN_MASK 0x1FFFFFU +#define AEQ_CTRL_1_FUNC_OWN_MASK 0x1U +#define AEQ_CTRL_1_ELEM_SIZE_MASK 0x3U +#define AEQ_CTRL_1_PAGE_SIZE_MASK 0xFU + +#define AEQ_CTRL_1_GET(val, member) \ + (((val) >> AEQ_CTRL_1_##member##_SHIFT) & \ + AEQ_CTRL_1_##member##_MASK) + +#define AEQ_CTRL_1_SET(val, member) \ + (((val) & AEQ_CTRL_1_##member##_MASK) << \ + AEQ_CTRL_1_##member##_SHIFT) + +#define AEQ_CTRL_1_CLEAR(val, member) \ + ((val) & (~(AEQ_CTRL_1_##member##_MASK \ + << AEQ_CTRL_1_##member##_SHIFT))) + +#define HIFC_EQ_PROD_IDX_MASK 0xFFFFF +#define HIFC_TASK_PROCESS_EQE_LIMIT 1024 +#define HIFC_EQ_UPDATE_CI_STEP 64 + +static uint g_aeq_len = HIFC_DEFAULT_AEQ_LEN; +module_param(g_aeq_len, uint, 0444); +MODULE_PARM_DESC(g_aeq_len, + "aeq depth, valid range is " __stringify(HIFC_MIN_AEQ_LEN) + " - " __stringify(HIFC_MAX_AEQ_LEN)); + +static uint g_ceq_len = HIFC_DEFAULT_CEQ_LEN; +module_param(g_ceq_len, uint, 0444); +MODULE_PARM_DESC(g_ceq_len, + "ceq depth, valid range is " __stringify(HIFC_MIN_CEQ_LEN) + " - " __stringify(HIFC_MAX_CEQ_LEN)); + +static uint g_num_ceqe_in_tasklet = HIFC_TASK_PROCESS_EQE_LIMIT; +module_param(g_num_ceqe_in_tasklet, uint, 0444); +MODULE_PARM_DESC(g_num_ceqe_in_tasklet, + "The max number of ceqe can be processed in tasklet, default = 1024"); + +#define CEQ_CTRL_0_INTR_IDX_SHIFT 0 +#define CEQ_CTRL_0_DMA_ATTR_SHIFT 12 +#define CEQ_CTRL_0_LIMIT_KICK_SHIFT 20 +#define CEQ_CTRL_0_PCI_INTF_IDX_SHIFT 24 +#define CEQ_CTRL_0_INTR_MODE_SHIFT 31 + +#define CEQ_CTRL_0_INTR_IDX_MASK 0x3FFU +#define CEQ_CTRL_0_DMA_ATTR_MASK 0x3FU +#define CEQ_CTRL_0_LIMIT_KICK_MASK 0xFU +#define CEQ_CTRL_0_PCI_INTF_IDX_MASK 0x3U +#define CEQ_CTRL_0_INTR_MODE_MASK 0x1U + +#define CEQ_CTRL_0_SET(val, member) \ + (((val) & CEQ_CTRL_0_##member##_MASK) << \ + CEQ_CTRL_0_##member##_SHIFT) + +#define CEQ_CTRL_1_LEN_SHIFT 0 +#define CEQ_CTRL_1_PAGE_SIZE_SHIFT 28 +#define CEQ_CTRL_1_LEN_MASK 0x1FFFFFU +#define CEQ_CTRL_1_PAGE_SIZE_MASK 0xFU + +#define CEQ_CTRL_1_SET(val, member) \ + (((val) & CEQ_CTRL_1_##member##_MASK) << \ + CEQ_CTRL_1_##member##_SHIFT) + +#define EQ_ELEM_DESC_TYPE_SHIFT 0 +#define EQ_ELEM_DESC_SRC_SHIFT 7 +#define EQ_ELEM_DESC_SIZE_SHIFT 8 +#define EQ_ELEM_DESC_WRAPPED_SHIFT 31 +#define EQ_ELEM_DESC_TYPE_MASK 0x7FU +#define EQ_ELEM_DESC_SRC_MASK 0x1U +#define EQ_ELEM_DESC_SIZE_MASK 0xFFU +#define EQ_ELEM_DESC_WRAPPED_MASK 0x1U + +#define EQ_ELEM_DESC_GET(val, member) \ + (((val) >> EQ_ELEM_DESC_##member##_SHIFT) & \ + EQ_ELEM_DESC_##member##_MASK) + +#define EQ_CONS_IDX_CONS_IDX_SHIFT 0 +#define EQ_CONS_IDX_XOR_CHKSUM_SHIFT 24 +#define EQ_CONS_IDX_INT_ARMED_SHIFT 31 +#define EQ_CONS_IDX_CONS_IDX_MASK 0x1FFFFFU +#define EQ_CONS_IDX_XOR_CHKSUM_MASK 0xFU +#define EQ_CONS_IDX_INT_ARMED_MASK 0x1U + +#define EQ_CONS_IDX_SET(val, member) \ + (((val) & EQ_CONS_IDX_##member##_MASK) << \ + EQ_CONS_IDX_##member##_SHIFT) + +#define EQ_CONS_IDX_CLEAR(val, member) \ + ((val) & (~(EQ_CONS_IDX_##member##_MASK \ + << EQ_CONS_IDX_##member##_SHIFT))) + +#define EQ_WRAPPED(eq) ((u32)(eq)->wrapped << EQ_VALID_SHIFT) + +#define EQ_CONS_IDX(eq) ((eq)->cons_idx | \ + ((u32)(eq)->wrapped << EQ_WRAPPED_SHIFT)) + +#define EQ_CONS_IDX_REG_ADDR(eq) (((eq)->type == HIFC_AEQ) ? \ + HIFC_CSR_AEQ_CONS_IDX_ADDR((eq)->q_id) : \ + HIFC_CSR_CEQ_CONS_IDX_ADDR((eq)->q_id)) + +#define EQ_PROD_IDX_REG_ADDR(eq) (((eq)->type == HIFC_AEQ) ? \ + HIFC_CSR_AEQ_PROD_IDX_ADDR((eq)->q_id) : \ + HIFC_CSR_CEQ_PROD_IDX_ADDR((eq)->q_id)) + +#define GET_EQ_NUM_PAGES(eq, size) \ + ((u16)(ALIGN((u32)((eq)->eq_len * (eq)->elem_size), \ + (size)) / (size))) + +#define GET_EQ_NUM_ELEMS(eq, pg_size) ((pg_size) / (u32)(eq)->elem_size) + +#define GET_EQ_ELEMENT(eq, idx) \ + (((u8 *)(eq)->virt_addr[(idx) / (eq)->num_elem_in_pg]) + \ + (u32)(((idx) & ((eq)->num_elem_in_pg - 1)) * (eq)->elem_size)) + +#define GET_AEQ_ELEM(eq, idx) ((struct hifc_aeq_elem *)\ + GET_EQ_ELEMENT((eq), (idx))) + +#define GET_CEQ_ELEM(eq, idx) ((u32 *)GET_EQ_ELEMENT((eq), (idx))) + +#define GET_CURR_AEQ_ELEM(eq) GET_AEQ_ELEM((eq), (eq)->cons_idx) + +#define GET_CURR_CEQ_ELEM(eq) GET_CEQ_ELEM((eq), (eq)->cons_idx) + +#define PAGE_IN_4K(page_size) ((page_size) >> 12) +#define EQ_SET_HW_PAGE_SIZE_VAL(eq) \ + ((u32)ilog2(PAGE_IN_4K((eq)->page_size))) + +#define ELEMENT_SIZE_IN_32B(eq) (((eq)->elem_size) >> 5) +#define EQ_SET_HW_ELEM_SIZE_VAL(eq) ((u32)ilog2(ELEMENT_SIZE_IN_32B(eq))) + +#define AEQ_DMA_ATTR_DEFAULT 0 +#define CEQ_DMA_ATTR_DEFAULT 0 +#define CEQ_LMT_KICK_DEFAULT 0 +#define EQ_MSIX_RESEND_TIMER_CLEAR 1 +#define EQ_WRAPPED_SHIFT 20 +#define EQ_VALID_SHIFT 31 +#define CEQE_TYPE_SHIFT 23 +#define CEQE_TYPE_MASK 0x7 + +#define CEQE_TYPE(type) (((type) >> CEQE_TYPE_SHIFT) & \ + CEQE_TYPE_MASK) +#define CEQE_DATA_MASK 0x3FFFFFF +#define CEQE_DATA(data) ((data) & CEQE_DATA_MASK) +#define EQ_MIN_PAGE_SIZE 0x1000U +#define aeq_to_aeqs(eq) \ + container_of((eq) - (eq)->q_id, struct hifc_aeqs, aeq[0]) + +#define ceq_to_ceqs(eq) \ + container_of((eq) - (eq)->q_id, struct hifc_ceqs, ceq[0]) + +/** + * aeq_interrupt - aeq interrupt handler + * @irq: irq number + * @data: the async event queue of the event + **/ +static irqreturn_t aeq_interrupt(int irq, void *data) +{ + struct hifc_eq *aeq = (struct hifc_eq *)data; + struct hifc_hwdev *hwdev = aeq->hwdev; + + struct hifc_aeqs *aeqs = aeq_to_aeqs(aeq); + struct workqueue_struct *workq = aeqs->workq; + struct hifc_eq_work *aeq_work; + + /* clear resend timer cnt register */ + hifc_misx_intr_clear_resend_bit(hwdev, aeq->eq_irq.msix_entry_idx, + EQ_MSIX_RESEND_TIMER_CLEAR); + + aeq_work = &aeq->aeq_work; + aeq_work->data = aeq; + + queue_work(workq, &aeq_work->work); + + return IRQ_HANDLED; +} + +/** + * ceq_interrupt - ceq interrupt handler + * @irq: irq number + * @data: the completion event queue of the event + **/ +static irqreturn_t ceq_interrupt(int irq, void *data) +{ + struct hifc_eq *ceq = (struct hifc_eq *)data; + struct hifc_ceq_tasklet_data *ceq_tasklet_data; + + ceq->hard_intr_jif = jiffies; + + /* clear resend timer counters */ + hifc_misx_intr_clear_resend_bit(ceq->hwdev, ceq->eq_irq.msix_entry_idx, + EQ_MSIX_RESEND_TIMER_CLEAR); + + ceq_tasklet_data = &ceq->ceq_tasklet_data; + ceq_tasklet_data->data = data; + tasklet_schedule(&ceq->ceq_tasklet); + + return IRQ_HANDLED; +} + +static u8 eq_cons_idx_checksum_set(u32 val) +{ + u8 checksum = 0; + u8 idx; + + for (idx = 0; idx < 32; idx += 4) + checksum ^= ((val >> idx) & 0xF); + + return checksum & 0xF; +} + +/** + * hifc_aeq_register_hw_cb - register aeq callback for specific event + * @hwdev: pointer to hw device + * @event: event for the handler + * @hw_cb: callback function + * Return: 0 - success, negative - failure + **/ +int hifc_aeq_register_hw_cb(void *hwdev, enum hifc_aeq_type event, + hifc_aeq_hwe_cb hwe_cb) +{ + struct hifc_aeqs *aeqs; + + if (!hwdev || !hwe_cb || event >= HIFC_MAX_AEQ_EVENTS) + return -EINVAL; + + aeqs = ((struct hifc_hwdev *)hwdev)->aeqs; + + aeqs->aeq_hwe_cb[event] = hwe_cb; + + set_bit(HIFC_AEQ_HW_CB_REG, &aeqs->aeq_hw_cb_state[event]); + + return 0; +} + +/** + * hifc_aeq_unregister_hw_cb - unregister the aeq callback for specific event + * @hwdev: pointer to hw device + * @event: event for the handler + **/ +void hifc_aeq_unregister_hw_cb(void *hwdev, enum hifc_aeq_type event) +{ + struct hifc_aeqs *aeqs; + + if (!hwdev || event >= HIFC_MAX_AEQ_EVENTS) + return; + + aeqs = ((struct hifc_hwdev *)hwdev)->aeqs; + + clear_bit(HIFC_AEQ_HW_CB_REG, &aeqs->aeq_hw_cb_state[event]); + + while (test_bit(HIFC_AEQ_HW_CB_RUNNING, &aeqs->aeq_hw_cb_state[event])) + usleep_range(900, 1000); + + aeqs->aeq_hwe_cb[event] = NULL; +} + +/** + * hifc_aeq_register_sw_cb - register aeq callback for sw event + * @hwdev: pointer to hw device + * @event: soft event for the handler + * @sw_cb: callback function + * Return: 0 - success, negative - failure + **/ +int hifc_aeq_register_swe_cb(void *hwdev, enum hifc_aeq_sw_type event, + hifc_aeq_swe_cb aeq_swe_cb) +{ + struct hifc_aeqs *aeqs; + + if (!hwdev || !aeq_swe_cb || event >= HIFC_MAX_AEQ_SW_EVENTS) + return -EINVAL; + + aeqs = ((struct hifc_hwdev *)hwdev)->aeqs; + + aeqs->aeq_swe_cb[event] = aeq_swe_cb; + + set_bit(HIFC_AEQ_SW_CB_REG, &aeqs->aeq_sw_cb_state[event]); + + return 0; +} + +/** + * hifc_aeq_unregister_sw_cb - unregister the aeq callback for sw event + * @hwdev: pointer to hw device + * @event: soft event for the handler + **/ +void hifc_aeq_unregister_swe_cb(void *hwdev, enum hifc_aeq_sw_type event) +{ + struct hifc_aeqs *aeqs; + + if (!hwdev || event >= HIFC_MAX_AEQ_SW_EVENTS) + return; + + aeqs = ((struct hifc_hwdev *)hwdev)->aeqs; + + clear_bit(HIFC_AEQ_SW_CB_REG, &aeqs->aeq_sw_cb_state[event]); + + while (test_bit(HIFC_AEQ_SW_CB_RUNNING, &aeqs->aeq_sw_cb_state[event])) + usleep_range(900, 1000); + + aeqs->aeq_swe_cb[event] = NULL; +} + +/** + * hifc_ceq_register_sw_cb - register ceq callback for specific event + * @hwdev: pointer to hw device + * @event: event for the handler + * @callback: callback function + * Return: 0 - success, negative - failure + **/ +int hifc_ceq_register_cb(void *hwdev, enum hifc_ceq_event event, + hifc_ceq_event_cb callback) +{ + struct hifc_ceqs *ceqs; + + if (!hwdev || event >= HIFC_MAX_CEQ_EVENTS) + return -EINVAL; + + ceqs = ((struct hifc_hwdev *)hwdev)->ceqs; + + ceqs->ceq_cb[event] = callback; + + set_bit(HIFC_CEQ_CB_REG, &ceqs->ceq_cb_state[event]); + + return 0; +} + +/** + * hifc_ceq_unregister_cb - unregister ceq callback for specific event + * @hwdev: pointer to hw device + * @event: event for the handler + **/ +void hifc_ceq_unregister_cb(void *hwdev, enum hifc_ceq_event event) +{ + struct hifc_ceqs *ceqs; + + if (!hwdev || event >= HIFC_MAX_CEQ_EVENTS) + return; + + ceqs = ((struct hifc_hwdev *)hwdev)->ceqs; + + clear_bit(HIFC_CEQ_CB_REG, &ceqs->ceq_cb_state[event]); + + while (test_bit(HIFC_CEQ_CB_RUNNING, &ceqs->ceq_cb_state[event])) + usleep_range(900, 1000); + + ceqs->ceq_cb[event] = NULL; +} + +/** + * set_eq_cons_idx - write the cons idx to the hw + * @eq: The event queue to update the cons idx for + * @arm_state: arm state value + **/ +static void set_eq_cons_idx(struct hifc_eq *eq, u32 arm_state) +{ + u32 eq_wrap_ci, val; + u32 addr = EQ_CONS_IDX_REG_ADDR(eq); + + eq_wrap_ci = EQ_CONS_IDX(eq); + + /* other filed is resverd, set to 0 */ + val = EQ_CONS_IDX_SET(eq_wrap_ci, CONS_IDX) | + EQ_CONS_IDX_SET(arm_state, INT_ARMED); + + val |= EQ_CONS_IDX_SET(eq_cons_idx_checksum_set(val), XOR_CHKSUM); + + hifc_hwif_write_reg(eq->hwdev->hwif, addr, val); +} + +/** + * ceq_event_handler - handle for the ceq events + * @eqs: eqs part of the chip + * @ceqe: ceq element of the event + **/ +static void ceq_event_handler(struct hifc_ceqs *ceqs, u32 ceqe) +{ + struct hifc_hwdev *hwdev = ceqs->hwdev; + enum hifc_ceq_event event = CEQE_TYPE(ceqe); + u32 ceqe_data = CEQE_DATA(ceqe); + + if (event >= HIFC_MAX_CEQ_EVENTS) { + sdk_err(hwdev->dev_hdl, "Ceq unknown event:%d, ceqe date: 0x%x\n", + event, ceqe_data); + return; + } + + set_bit(HIFC_CEQ_CB_RUNNING, &ceqs->ceq_cb_state[event]); + + if (ceqs->ceq_cb[event] && + test_bit(HIFC_CEQ_CB_REG, &ceqs->ceq_cb_state[event])) + ceqs->ceq_cb[event](hwdev, ceqe_data); + + clear_bit(HIFC_CEQ_CB_RUNNING, &ceqs->ceq_cb_state[event]); +} + +static void aeq_swe_handler(struct hifc_aeqs *aeqs, + struct hifc_aeq_elem *aeqe_pos, + enum hifc_aeq_type event) +{ + enum hifc_ucode_event_type ucode_event; + enum hifc_aeq_sw_type sw_event; + u64 aeqe_data; + u8 lev; + + ucode_event = event; + /* SW event uses only the first 8B */ + sw_event = ucode_event >= HIFC_NIC_FATAL_ERROR_MAX ? + HIFC_STATEFULL_EVENT : + HIFC_STATELESS_EVENT; + aeqe_data = be64_to_cpu((*(u64 *)aeqe_pos->aeqe_data)); + set_bit(HIFC_AEQ_SW_CB_RUNNING, + &aeqs->aeq_sw_cb_state[sw_event]); + if (aeqs->aeq_swe_cb[sw_event] && + test_bit(HIFC_AEQ_SW_CB_REG, + &aeqs->aeq_sw_cb_state[sw_event])) { + lev = aeqs->aeq_swe_cb[sw_event](aeqs->hwdev, + ucode_event, + aeqe_data); + hifc_swe_fault_handler(aeqs->hwdev, lev, + ucode_event, aeqe_data); + } + clear_bit(HIFC_AEQ_SW_CB_RUNNING, + &aeqs->aeq_sw_cb_state[sw_event]); +} + +static void aeq_hwe_handler(struct hifc_aeqs *aeqs, + struct hifc_aeq_elem *aeqe_pos, + enum hifc_aeq_type event, u32 aeqe_desc) +{ + u8 size; + + if (event < HIFC_MAX_AEQ_EVENTS) { + size = EQ_ELEM_DESC_GET(aeqe_desc, SIZE); + set_bit(HIFC_AEQ_HW_CB_RUNNING, + &aeqs->aeq_hw_cb_state[event]); + if (aeqs->aeq_hwe_cb[event] && + test_bit(HIFC_AEQ_HW_CB_REG, + &aeqs->aeq_hw_cb_state[event])) + aeqs->aeq_hwe_cb[event](aeqs->hwdev, + aeqe_pos->aeqe_data, size); + clear_bit(HIFC_AEQ_HW_CB_RUNNING, + &aeqs->aeq_hw_cb_state[event]); + + return; + } + + sdk_warn(aeqs->hwdev->dev_hdl, "Unknown aeq hw event %d\n", event); +} + +/** + * aeq_irq_handler - handler for the aeq event + * @eq: the async event queue of the event + * Return: true - success, false - failure + **/ +static bool aeq_irq_handler(struct hifc_eq *eq) +{ + struct hifc_aeqs *aeqs = aeq_to_aeqs(eq); + struct hifc_aeq_elem *aeqe_pos; + enum hifc_aeq_type event; + u32 aeqe_desc; + u32 i, eqe_cnt = 0; + + for (i = 0; i < HIFC_TASK_PROCESS_EQE_LIMIT; i++) { + aeqe_pos = GET_CURR_AEQ_ELEM(eq); + + /* Data in HW is in Big endian Format */ + aeqe_desc = be32_to_cpu(aeqe_pos->desc); + + /* HW updates wrapped bit, when it adds eq element event */ + if (EQ_ELEM_DESC_GET(aeqe_desc, WRAPPED) == eq->wrapped) + return false; + + /* This memory barrier is needed to keep us from reading + * any other fields out of the cmdq wqe until we have + * verified the command has been processed and + * written back. + */ + dma_rmb(); + + event = EQ_ELEM_DESC_GET(aeqe_desc, TYPE); + if (EQ_ELEM_DESC_GET(aeqe_desc, SRC)) + aeq_swe_handler(aeqs, aeqe_pos, event); + else + aeq_hwe_handler(aeqs, aeqe_pos, event, aeqe_desc); + + eq->cons_idx++; + + if (eq->cons_idx == eq->eq_len) { + eq->cons_idx = 0; + eq->wrapped = !eq->wrapped; + } + + if (++eqe_cnt >= HIFC_EQ_UPDATE_CI_STEP) { + eqe_cnt = 0; + set_eq_cons_idx(eq, HIFC_EQ_NOT_ARMED); + } + } + + return true; +} + +/** + * ceq_irq_handler - handler for the ceq event + * @eq: the completion event queue of the event + * Return: true - success, false - failure + **/ +static bool ceq_irq_handler(struct hifc_eq *eq) +{ + struct hifc_ceqs *ceqs = ceq_to_ceqs(eq); + u32 ceqe, eqe_cnt = 0; + u32 i; + + for (i = 0; i < g_num_ceqe_in_tasklet; i++) { + ceqe = *(GET_CURR_CEQ_ELEM(eq)); + ceqe = be32_to_cpu(ceqe); + + /* HW updates wrapped bit, when it adds eq element event */ + if (EQ_ELEM_DESC_GET(ceqe, WRAPPED) == eq->wrapped) + return false; + + ceq_event_handler(ceqs, ceqe); + + eq->cons_idx++; + + if (eq->cons_idx == eq->eq_len) { + eq->cons_idx = 0; + eq->wrapped = !eq->wrapped; + } + + if (++eqe_cnt >= HIFC_EQ_UPDATE_CI_STEP) { + eqe_cnt = 0; + set_eq_cons_idx(eq, HIFC_EQ_NOT_ARMED); + } + } + + return true; +} + +/** + * eq_irq_handler - handler for the eq event + * @data: the event queue of the event + * Return: true - success, false - failure + **/ +static bool eq_irq_handler(void *data) +{ + struct hifc_eq *eq = (struct hifc_eq *)data; + bool uncompleted; + + if (eq->type == HIFC_AEQ) + uncompleted = aeq_irq_handler(eq); + else + uncompleted = ceq_irq_handler(eq); + + set_eq_cons_idx(eq, uncompleted ? HIFC_EQ_NOT_ARMED : HIFC_EQ_ARMED); + + return uncompleted; +} + +static void reschedule_eq_handler(struct hifc_eq *eq) +{ + if (eq->type == HIFC_AEQ) { + struct hifc_aeqs *aeqs = aeq_to_aeqs(eq); + struct workqueue_struct *workq = aeqs->workq; + struct hifc_eq_work *aeq_work = &eq->aeq_work; + + queue_work(workq, &aeq_work->work); + } else { + tasklet_schedule(&eq->ceq_tasklet); + } +} + +/** + * ceq_tasklet - ceq tasklet for the event + * @ceq_data: data that will be used by the tasklet(ceq) + **/ + +static void ceq_tasklet(ulong ceq_data) +{ + struct hifc_ceq_tasklet_data *ceq_tasklet_data = + (struct hifc_ceq_tasklet_data *)ceq_data; + struct hifc_eq *eq = (struct hifc_eq *)ceq_tasklet_data->data; + + eq->soft_intr_jif = jiffies; + + if (eq_irq_handler(ceq_tasklet_data->data)) + reschedule_eq_handler(ceq_tasklet_data->data); +} + +/** + * eq_irq_work - eq work for the event + * @work: the work that is associated with the eq + **/ +static void eq_irq_work(struct work_struct *work) +{ + struct hifc_eq_work *aeq_work = + container_of(work, struct hifc_eq_work, work); + + if (eq_irq_handler(aeq_work->data)) + reschedule_eq_handler(aeq_work->data); +} + +struct hifc_ceq_ctrl_reg { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 q_id; + u32 ctrl0; + u32 ctrl1; +}; + +static int set_ceq_ctrl_reg(struct hifc_hwdev *hwdev, u16 q_id, + u32 ctrl0, u32 ctrl1) +{ + struct hifc_ceq_ctrl_reg ceq_ctrl = {0}; + u16 in_size = sizeof(ceq_ctrl); + u16 out_size = sizeof(ceq_ctrl); + int err; + + err = hifc_global_func_id_get(hwdev, &ceq_ctrl.func_id); + if (err) + return err; + + ceq_ctrl.q_id = q_id; + ceq_ctrl.ctrl0 = ctrl0; + ceq_ctrl.ctrl1 = ctrl1; + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_CEQ_CTRL_REG_WR_BY_UP, + &ceq_ctrl, in_size, + &ceq_ctrl, &out_size, 0); + if (err || !out_size || ceq_ctrl.status) { + sdk_err(hwdev->dev_hdl, "Failed to set ceq %d ctrl reg, err: %d status: 0x%x, out_size: 0x%x\n", + q_id, err, ceq_ctrl.status, out_size); + return -EFAULT; + } + + return 0; +} + +/** + * set_eq_ctrls - setting eq's ctrls registers + * @eq: the event queue for setting + * Return: 0 - success, negative - failure + **/ +static int set_eq_ctrls(struct hifc_eq *eq) +{ + enum hifc_eq_type type = eq->type; + struct hifc_hwif *hwif = eq->hwdev->hwif; + struct irq_info *eq_irq = &eq->eq_irq; + u32 addr, val, ctrl0, ctrl1, page_size_val, elem_size; + u32 pci_intf_idx = HIFC_PCI_INTF_IDX(hwif); + int err; + + if (type == HIFC_AEQ) { + /* set ctrl0 */ + addr = HIFC_CSR_AEQ_CTRL_0_ADDR(eq->q_id); + + val = hifc_hwif_read_reg(hwif, addr); + + val = AEQ_CTRL_0_CLEAR(val, INTR_IDX) & + AEQ_CTRL_0_CLEAR(val, DMA_ATTR) & + AEQ_CTRL_0_CLEAR(val, PCI_INTF_IDX) & + AEQ_CTRL_0_CLEAR(val, INTR_MODE); + + ctrl0 = AEQ_CTRL_0_SET(eq_irq->msix_entry_idx, INTR_IDX) | + AEQ_CTRL_0_SET(AEQ_DMA_ATTR_DEFAULT, DMA_ATTR) | + AEQ_CTRL_0_SET(pci_intf_idx, PCI_INTF_IDX) | + + AEQ_CTRL_0_SET(HIFC_INTR_MODE_ARMED, INTR_MODE); + + val |= ctrl0; + + hifc_hwif_write_reg(hwif, addr, val); + + /* set ctrl1 */ + addr = HIFC_CSR_AEQ_CTRL_1_ADDR(eq->q_id); + + page_size_val = EQ_SET_HW_PAGE_SIZE_VAL(eq); + elem_size = EQ_SET_HW_ELEM_SIZE_VAL(eq); + + ctrl1 = AEQ_CTRL_1_SET(eq->eq_len, LEN) | + AEQ_CTRL_1_SET(elem_size, ELEM_SIZE) | + AEQ_CTRL_1_SET(page_size_val, PAGE_SIZE); + + hifc_hwif_write_reg(hwif, addr, ctrl1); + + } else { + ctrl0 = CEQ_CTRL_0_SET(eq_irq->msix_entry_idx, INTR_IDX) | + CEQ_CTRL_0_SET(CEQ_DMA_ATTR_DEFAULT, DMA_ATTR) | + CEQ_CTRL_0_SET(CEQ_LMT_KICK_DEFAULT, LIMIT_KICK) | + CEQ_CTRL_0_SET(pci_intf_idx, PCI_INTF_IDX) | + CEQ_CTRL_0_SET(HIFC_INTR_MODE_ARMED, INTR_MODE); + + page_size_val = EQ_SET_HW_PAGE_SIZE_VAL(eq); + + ctrl1 = CEQ_CTRL_1_SET(eq->eq_len, LEN) | + CEQ_CTRL_1_SET(page_size_val, PAGE_SIZE); + + /* set ceq ctrl reg through mgmt cpu */ + err = set_ceq_ctrl_reg(eq->hwdev, eq->q_id, ctrl0, ctrl1); + if (err) + return err; + } + + return 0; +} + +/** + * ceq_elements_init - Initialize all the elements in the ceq + * @eq: the event queue + * @init_val: value to init with it the elements + **/ +static void ceq_elements_init(struct hifc_eq *eq, u32 init_val) +{ + u32 i; + u32 *ceqe; + + for (i = 0; i < eq->eq_len; i++) { + ceqe = GET_CEQ_ELEM(eq, i); + *(ceqe) = cpu_to_be32(init_val); + } + + wmb(); /* Write the init values */ +} + +/** + * aeq_elements_init - initialize all the elements in the aeq + * @eq: the event queue + * @init_val: value to init with it the elements + **/ +static void aeq_elements_init(struct hifc_eq *eq, u32 init_val) +{ + struct hifc_aeq_elem *aeqe; + u32 i; + + for (i = 0; i < eq->eq_len; i++) { + aeqe = GET_AEQ_ELEM(eq, i); + aeqe->desc = cpu_to_be32(init_val); + } + + wmb(); /* Write the init values */ +} + +static void free_eq_pages_desc(struct hifc_eq *eq) +{ + kfree(eq->virt_addr_for_free); + kfree(eq->dma_addr_for_free); + kfree(eq->virt_addr); + kfree(eq->dma_addr); +} + +static int alloc_eq_pages_desc(struct hifc_eq *eq) +{ + u64 dma_addr_size, virt_addr_size; + int err; + + dma_addr_size = eq->num_pages * sizeof(*eq->dma_addr); + virt_addr_size = eq->num_pages * sizeof(*eq->virt_addr); + + eq->dma_addr = kzalloc(dma_addr_size, GFP_KERNEL); + if (!eq->dma_addr) + return -ENOMEM; + + eq->virt_addr = kzalloc(virt_addr_size, GFP_KERNEL); + if (!eq->virt_addr) { + err = -ENOMEM; + goto virt_addr_alloc_err; + } + + eq->dma_addr_for_free = kzalloc(dma_addr_size, GFP_KERNEL); + if (!eq->dma_addr_for_free) { + err = -ENOMEM; + goto dma_addr_free_alloc_err; + } + + eq->virt_addr_for_free = kzalloc(virt_addr_size, GFP_KERNEL); + if (!eq->virt_addr_for_free) { + err = -ENOMEM; + goto virt_addr_free_alloc_err; + } + + return 0; + +virt_addr_free_alloc_err: + kfree(eq->dma_addr_for_free); +dma_addr_free_alloc_err: + kfree(eq->virt_addr); +virt_addr_alloc_err: + kfree(eq->dma_addr); + return err; +} + +#define IS_ALIGN(x, a) (((x) & ((a) - 1)) == 0) + +static int init_eq_elements(struct hifc_eq *eq) +{ + u32 init_val; + + eq->num_elem_in_pg = GET_EQ_NUM_ELEMS(eq, eq->page_size); + if (!IS_ALIGN(eq->num_elem_in_pg, eq->num_elem_in_pg)) { + sdk_err(eq->hwdev->dev_hdl, "Number element in eq page != power of 2\n"); + return -EINVAL; + } + + init_val = EQ_WRAPPED(eq); + + if (eq->type == HIFC_AEQ) + aeq_elements_init(eq, init_val); + else + ceq_elements_init(eq, init_val); + + return 0; +} + +/** + * alloc_eq_pages - allocate the pages for the queue + * @eq: the event queue + * Return: 0 - success, negative - failure + **/ +static int alloc_eq_pages(struct hifc_eq *eq) +{ + struct hifc_hwif *hwif = eq->hwdev->hwif; + u16 pg_num, i; + u32 reg; + int err; + u8 flag = 0; + + err = alloc_eq_pages_desc(eq); + if (err) { + sdk_err(eq->hwdev->dev_hdl, "Failed to alloc eq pages description\n"); + return err; + } + + for (pg_num = 0; pg_num < eq->num_pages; pg_num++) { + eq->virt_addr_for_free[pg_num] = dma_zalloc_coherent + (eq->hwdev->dev_hdl, eq->page_size, + &eq->dma_addr_for_free[pg_num], GFP_KERNEL); + if (!eq->virt_addr_for_free[pg_num]) { + err = -ENOMEM; + goto dma_alloc_err; + } + + eq->dma_addr[pg_num] = eq->dma_addr_for_free[pg_num]; + eq->virt_addr[pg_num] = eq->virt_addr_for_free[pg_num]; + if (!IS_ALIGN(eq->dma_addr_for_free[pg_num], + eq->page_size)) { + sdk_info(eq->hwdev->dev_hdl, + "Address is not aligned to %u-bytes as hardware required\n", + eq->page_size); + sdk_info(eq->hwdev->dev_hdl, "Change eq's page size %u\n", + ((eq->page_size) >> 1)); + eq->dma_addr[pg_num] = ALIGN + (eq->dma_addr_for_free[pg_num], + (u64)((eq->page_size) >> 1)); + eq->virt_addr[pg_num] = eq->virt_addr_for_free[pg_num] + + ((u64)eq->dma_addr[pg_num] + - (u64)eq->dma_addr_for_free[pg_num]); + flag = 1; + } + reg = HIFC_EQ_HI_PHYS_ADDR_REG(eq->type, eq->q_id, pg_num); + hifc_hwif_write_reg(hwif, reg, + upper_32_bits(eq->dma_addr[pg_num])); + + reg = HIFC_EQ_LO_PHYS_ADDR_REG(eq->type, eq->q_id, pg_num); + hifc_hwif_write_reg(hwif, reg, + lower_32_bits(eq->dma_addr[pg_num])); + } + + if (flag) { + eq->page_size = eq->page_size >> 1; + eq->eq_len = eq->eq_len >> 1; + } + + err = init_eq_elements(eq); + if (err) { + sdk_err(eq->hwdev->dev_hdl, "Failed to init eq elements\n"); + goto dma_alloc_err; + } + + return 0; + +dma_alloc_err: + for (i = 0; i < pg_num; i++) + dma_free_coherent(eq->hwdev->dev_hdl, eq->page_size, + eq->virt_addr_for_free[i], + eq->dma_addr_for_free[i]); + free_eq_pages_desc(eq); + return err; +} + +/** + * free_eq_pages - free the pages of the queue + * @eq: the event queue + **/ +static void free_eq_pages(struct hifc_eq *eq) +{ + struct hifc_hwdev *hwdev = eq->hwdev; + u16 pg_num; + + for (pg_num = 0; pg_num < eq->num_pages; pg_num++) + dma_free_coherent(hwdev->dev_hdl, eq->orig_page_size, + eq->virt_addr_for_free[pg_num], + eq->dma_addr_for_free[pg_num]); + + free_eq_pages_desc(eq); +} + +static inline u32 get_page_size(struct hifc_eq *eq) +{ + u32 total_size; + u16 count, n = 0; + + total_size = ALIGN((eq->eq_len * eq->elem_size), EQ_MIN_PAGE_SIZE); + + if (total_size <= (HIFC_EQ_MAX_PAGES * EQ_MIN_PAGE_SIZE)) + return EQ_MIN_PAGE_SIZE; + + count = (u16)(ALIGN((total_size / HIFC_EQ_MAX_PAGES), + EQ_MIN_PAGE_SIZE) / EQ_MIN_PAGE_SIZE); + + if (!(count & (count - 1))) + return EQ_MIN_PAGE_SIZE * count; + + while (count) { + count >>= 1; + n++; + } + + return EQ_MIN_PAGE_SIZE << n; +} + +static int request_eq_irq(struct hifc_eq *eq, enum hifc_eq_type type, + struct irq_info *entry) +{ + int err = 0; + + if (type == HIFC_AEQ) { + struct hifc_eq_work *aeq_work = &eq->aeq_work; + + INIT_WORK(&aeq_work->work, eq_irq_work); + } else { + tasklet_init(&eq->ceq_tasklet, ceq_tasklet, + (ulong)(&eq->ceq_tasklet_data)); + } + + if (type == HIFC_AEQ) { + snprintf(eq->irq_name, sizeof(eq->irq_name), + "hifc_aeq%d@pci:%s", eq->q_id, + pci_name(eq->hwdev->pcidev_hdl)); + + err = request_irq(entry->irq_id, aeq_interrupt, 0UL, + eq->irq_name, eq); + } else { + snprintf(eq->irq_name, sizeof(eq->irq_name), + "hifc_ceq%d@pci:%s", eq->q_id, + pci_name(eq->hwdev->pcidev_hdl)); + + err = request_irq(entry->irq_id, ceq_interrupt, 0UL, + eq->irq_name, eq); + } + + return err; +} + +/** + * init_eq - initialize eq + * @eq: the event queue + * @hwdev: the pointer to hw device + * @q_id: Queue id number + * @q_len: the number of EQ elements + * @type: the type of the event queue, ceq or aeq + * @entry: msix entry associated with the event queue + * Return: 0 - Success, Negative - failure + **/ +static int init_eq(struct hifc_eq *eq, struct hifc_hwdev *hwdev, u16 q_id, + u32 q_len, enum hifc_eq_type type, struct irq_info *entry) +{ + int err = 0; + + eq->hwdev = hwdev; + eq->q_id = q_id; + eq->type = type; + eq->eq_len = q_len; + + /* clear eq_len to force eqe drop in hardware */ + if (eq->type == HIFC_AEQ) + hifc_hwif_write_reg(eq->hwdev->hwif, + HIFC_CSR_AEQ_CTRL_1_ADDR(eq->q_id), 0); + else + set_ceq_ctrl_reg(eq->hwdev, eq->q_id, 0, 0); + + eq->cons_idx = 0; + eq->wrapped = 0; + + eq->elem_size = (type == HIFC_AEQ) ? + HIFC_AEQE_SIZE : HIFC_CEQE_SIZE; + + eq->page_size = get_page_size(eq); + eq->orig_page_size = eq->page_size; + eq->num_pages = GET_EQ_NUM_PAGES(eq, eq->page_size); + if (eq->num_pages > HIFC_EQ_MAX_PAGES) { + sdk_err(hwdev->dev_hdl, "Number pages:%d too many pages for eq\n", + eq->num_pages); + return -EINVAL; + } + + err = alloc_eq_pages(eq); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to allocate pages for eq\n"); + return err; + } + + eq->eq_irq.msix_entry_idx = entry->msix_entry_idx; + eq->eq_irq.irq_id = entry->irq_id; + + err = set_eq_ctrls(eq); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to allocate pages for eq\n"); + goto init_eq_ctrls_err; + } + + hifc_hwif_write_reg(eq->hwdev->hwif, EQ_PROD_IDX_REG_ADDR(eq), 0); + set_eq_cons_idx(eq, HIFC_EQ_ARMED); + + err = request_eq_irq(eq, type, entry); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to request irq for the eq, err: %d\n", + err); + goto req_irq_err; + } + + hifc_set_msix_state(hwdev, entry->msix_entry_idx, HIFC_MSIX_ENABLE); + + return 0; + +init_eq_ctrls_err: +req_irq_err: + free_eq_pages(eq); + return err; +} + +/** + * remove_eq - remove eq + * @eq: the event queue + **/ +static void remove_eq(struct hifc_eq *eq) +{ + struct irq_info *entry = &eq->eq_irq; + + hifc_set_msix_state(eq->hwdev, entry->msix_entry_idx, + HIFC_MSIX_DISABLE); + synchronize_irq(entry->irq_id); + + free_irq(entry->irq_id, eq); + + if (eq->type == HIFC_AEQ) { + struct hifc_eq_work *aeq_work = &eq->aeq_work; + + cancel_work_sync(&aeq_work->work); + + /* clear eq_len to avoid hw access host memory */ + hifc_hwif_write_reg(eq->hwdev->hwif, + HIFC_CSR_AEQ_CTRL_1_ADDR(eq->q_id), 0); + } else { + tasklet_kill(&eq->ceq_tasklet); + + set_ceq_ctrl_reg(eq->hwdev, eq->q_id, 0, 0); + } + + /* update cons_idx to avoid invalid interrupt */ + eq->cons_idx = hifc_hwif_read_reg(eq->hwdev->hwif, + EQ_PROD_IDX_REG_ADDR(eq)); + set_eq_cons_idx(eq, HIFC_EQ_NOT_ARMED); + + free_eq_pages(eq); +} + +/** + * hifc_aeqs_init - init all the aeqs + * @hwdev: the pointer to hw device + * @num_ceqs: number of AEQs + * @msix_entries: msix entries associated with the event queues + * Return: 0 - Success, Negative - failure + **/ +int hifc_aeqs_init(struct hifc_hwdev *hwdev, u16 num_aeqs, + struct irq_info *msix_entries) +{ + struct hifc_aeqs *aeqs; + int err; + u16 i, q_id; + + aeqs = kzalloc(sizeof(*aeqs), GFP_KERNEL); + if (!aeqs) + return -ENOMEM; + + hwdev->aeqs = aeqs; + aeqs->hwdev = hwdev; + aeqs->num_aeqs = num_aeqs; + + aeqs->workq = create_singlethread_workqueue(HIFC_EQS_WQ_NAME); + if (!aeqs->workq) { + sdk_err(hwdev->dev_hdl, "Failed to initialize aeq workqueue\n"); + err = -ENOMEM; + goto create_work_err; + } + + if (g_aeq_len < HIFC_MIN_AEQ_LEN || g_aeq_len > HIFC_MAX_AEQ_LEN) { + sdk_warn(hwdev->dev_hdl, "Module Parameter g_aeq_len value %d out of range, resetting to %d\n", + g_aeq_len, HIFC_DEFAULT_AEQ_LEN); + g_aeq_len = HIFC_DEFAULT_AEQ_LEN; + } + + for (q_id = 0; q_id < num_aeqs; q_id++) { + err = init_eq(&aeqs->aeq[q_id], hwdev, q_id, g_aeq_len, + HIFC_AEQ, &msix_entries[q_id]); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init aeq %d\n", + q_id); + goto init_aeq_err; + } + } + + return 0; + +init_aeq_err: + for (i = 0; i < q_id; i++) + remove_eq(&aeqs->aeq[i]); + + destroy_workqueue(aeqs->workq); + +create_work_err: + kfree(aeqs); + + return err; +} + +/** + * hifc_aeqs_free - free all the aeqs + * @hwdev: the pointer to hw device + **/ +void hifc_aeqs_free(struct hifc_hwdev *hwdev) +{ + struct hifc_aeqs *aeqs = hwdev->aeqs; + enum hifc_aeq_type aeq_event = HIFC_HW_INTER_INT; + enum hifc_aeq_sw_type sw_aeq_event = HIFC_STATELESS_EVENT; + u16 q_id; + + for (q_id = 0; q_id < aeqs->num_aeqs; q_id++) + remove_eq(&aeqs->aeq[q_id]); + + for (; sw_aeq_event < HIFC_MAX_AEQ_SW_EVENTS; sw_aeq_event++) + hifc_aeq_unregister_swe_cb(hwdev, sw_aeq_event); + + for (; aeq_event < HIFC_MAX_AEQ_EVENTS; aeq_event++) + hifc_aeq_unregister_hw_cb(hwdev, aeq_event); + + destroy_workqueue(aeqs->workq); + + kfree(aeqs); +} + +/** + * hifc_ceqs_init - init all the ceqs + * @hwdev: the pointer to hw device + * @num_ceqs: number of CEQs + * @msix_entries: msix entries associated with the event queues + * Return: 0 - Success, Negative - failure + **/ +int hifc_ceqs_init(struct hifc_hwdev *hwdev, u16 num_ceqs, + struct irq_info *msix_entries) +{ + struct hifc_ceqs *ceqs; + int err; + u16 i, q_id; + + ceqs = kzalloc(sizeof(*ceqs), GFP_KERNEL); + if (!ceqs) + return -ENOMEM; + + hwdev->ceqs = ceqs; + + ceqs->hwdev = hwdev; + ceqs->num_ceqs = num_ceqs; + + if (g_ceq_len < HIFC_MIN_CEQ_LEN || g_ceq_len > HIFC_MAX_CEQ_LEN) { + sdk_warn(hwdev->dev_hdl, "Module Parameter g_ceq_len value %d out of range, resetting to %d\n", + g_ceq_len, HIFC_DEFAULT_CEQ_LEN); + g_ceq_len = HIFC_DEFAULT_CEQ_LEN; + } + + if (!g_num_ceqe_in_tasklet) { + sdk_warn(hwdev->dev_hdl, "Module Parameter g_num_ceqe_in_tasklet can not be zero, resetting to %d\n", + HIFC_TASK_PROCESS_EQE_LIMIT); + g_num_ceqe_in_tasklet = HIFC_TASK_PROCESS_EQE_LIMIT; + } + + for (q_id = 0; q_id < num_ceqs; q_id++) { + err = init_eq(&ceqs->ceq[q_id], hwdev, q_id, g_ceq_len, + HIFC_CEQ, &msix_entries[q_id]); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init ceq %d\n", + q_id); + goto init_ceq_err; + } + } + + return 0; + +init_ceq_err: + for (i = 0; i < q_id; i++) + remove_eq(&ceqs->ceq[i]); + + kfree(ceqs); + + return err; +} + +/** + * hifc_ceqs_free - free all the ceqs + * @hwdev: the pointer to hw device + **/ +void hifc_ceqs_free(struct hifc_hwdev *hwdev) +{ + struct hifc_ceqs *ceqs = hwdev->ceqs; + enum hifc_ceq_event ceq_event = HIFC_CMDQ; + u16 q_id; + + for (q_id = 0; q_id < ceqs->num_ceqs; q_id++) + remove_eq(&ceqs->ceq[q_id]); + + for (; ceq_event < HIFC_MAX_CEQ_EVENTS; ceq_event++) + hifc_ceq_unregister_cb(hwdev, ceq_event); + + kfree(ceqs); +} + +void hifc_get_ceq_irqs(struct hifc_hwdev *hwdev, struct irq_info *irqs, + u16 *num_irqs) +{ + struct hifc_ceqs *ceqs = hwdev->ceqs; + u16 q_id; + + for (q_id = 0; q_id < ceqs->num_ceqs; q_id++) { + irqs[q_id].irq_id = ceqs->ceq[q_id].eq_irq.irq_id; + irqs[q_id].msix_entry_idx = + ceqs->ceq[q_id].eq_irq.msix_entry_idx; + } + + *num_irqs = ceqs->num_ceqs; +} + +void hifc_get_aeq_irqs(struct hifc_hwdev *hwdev, struct irq_info *irqs, + u16 *num_irqs) +{ + struct hifc_aeqs *aeqs = hwdev->aeqs; + u16 q_id; + + for (q_id = 0; q_id < aeqs->num_aeqs; q_id++) { + irqs[q_id].irq_id = aeqs->aeq[q_id].eq_irq.irq_id; + irqs[q_id].msix_entry_idx = + aeqs->aeq[q_id].eq_irq.msix_entry_idx; + } + + *num_irqs = aeqs->num_aeqs; +} + +void hifc_dump_aeq_info(struct hifc_hwdev *hwdev) +{ + struct hifc_aeq_elem *aeqe_pos; + struct hifc_eq *eq; + u32 addr, ci, pi; + int q_id; + + for (q_id = 0; q_id < hwdev->aeqs->num_aeqs; q_id++) { + eq = &hwdev->aeqs->aeq[q_id]; + addr = EQ_CONS_IDX_REG_ADDR(eq); + ci = hifc_hwif_read_reg(hwdev->hwif, addr); + addr = EQ_PROD_IDX_REG_ADDR(eq); + pi = hifc_hwif_read_reg(hwdev->hwif, addr); + aeqe_pos = GET_CURR_AEQ_ELEM(eq); + sdk_err(hwdev->dev_hdl, "Aeq id: %d, ci: 0x%08x, pi: 0x%x, work_state: 0x%x, wrap: %d, desc: 0x%x\n", + q_id, ci, pi, work_busy(&eq->aeq_work.work), + eq->wrapped, be32_to_cpu(aeqe_pos->desc)); + } +} + diff --git a/drivers/scsi/huawei/hifc/hifc_eqs.h b/drivers/scsi/huawei/hifc/hifc_eqs.h new file mode 100644 index 000000000000..2dcfc432c8f2 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_eqs.h @@ -0,0 +1,233 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + * + */ + +#ifndef HIFC_EQS_H +#define HIFC_EQS_H + +#define HIFC_MAX_AEQS 3 +#define HIFC_MAX_CEQS 32 + +#define HIFC_EQ_MAX_PAGES 8 + +#define HIFC_AEQE_SIZE 64 +#define HIFC_CEQE_SIZE 4 + +#define HIFC_AEQE_DESC_SIZE 4 +#define HIFC_AEQE_DATA_SIZE \ + (HIFC_AEQE_SIZE - HIFC_AEQE_DESC_SIZE) + +#define HIFC_DEFAULT_AEQ_LEN 4096 +#define HIFC_DEFAULT_CEQ_LEN 8192 + +#define HIFC_MIN_AEQ_LEN 64 +#define HIFC_MAX_AEQ_LEN (512 * 1024) +#define HIFC_MIN_CEQ_LEN 64 +#define HIFC_MAX_CEQ_LEN (1024 * 1024) + +#define HIFC_CEQ_ID_CMDQ 0 +#define EQ_IRQ_NAME_LEN 64 + +/* EQ registers */ +#define HIFC_AEQ_MTT_OFF_BASE_ADDR 0x200 +#define HIFC_CEQ_MTT_OFF_BASE_ADDR 0x400 + +#define HIFC_EQ_MTT_OFF_STRIDE 0x40 + +#define HIFC_CSR_AEQ_MTT_OFF(id) \ + (HIFC_AEQ_MTT_OFF_BASE_ADDR + (id) * HIFC_EQ_MTT_OFF_STRIDE) + +#define HIFC_CSR_CEQ_MTT_OFF(id) \ + (HIFC_CEQ_MTT_OFF_BASE_ADDR + (id) * HIFC_EQ_MTT_OFF_STRIDE) + +#define HIFC_CSR_EQ_PAGE_OFF_STRIDE 8 + +#define HIFC_AEQ_HI_PHYS_ADDR_REG(q_id, pg_num) \ + (HIFC_CSR_AEQ_MTT_OFF(q_id) + \ + (pg_num) * HIFC_CSR_EQ_PAGE_OFF_STRIDE) + +#define HIFC_AEQ_LO_PHYS_ADDR_REG(q_id, pg_num) \ + (HIFC_CSR_AEQ_MTT_OFF(q_id) + \ + (pg_num) * HIFC_CSR_EQ_PAGE_OFF_STRIDE + 4) + +#define HIFC_CEQ_HI_PHYS_ADDR_REG(q_id, pg_num) \ + (HIFC_CSR_CEQ_MTT_OFF(q_id) + \ + (pg_num) * HIFC_CSR_EQ_PAGE_OFF_STRIDE) + +#define HIFC_CEQ_LO_PHYS_ADDR_REG(q_id, pg_num) \ + (HIFC_CSR_CEQ_MTT_OFF(q_id) + \ + (pg_num) * HIFC_CSR_EQ_PAGE_OFF_STRIDE + 4) + +#define HIFC_EQ_HI_PHYS_ADDR_REG(type, q_id, pg_num) \ + ((u32)((type == HIFC_AEQ) ? \ + HIFC_AEQ_HI_PHYS_ADDR_REG(q_id, pg_num) : \ + HIFC_CEQ_HI_PHYS_ADDR_REG(q_id, pg_num))) + +#define HIFC_EQ_LO_PHYS_ADDR_REG(type, q_id, pg_num) \ + ((u32)((type == HIFC_AEQ) ? \ + HIFC_AEQ_LO_PHYS_ADDR_REG(q_id, pg_num) : \ + HIFC_CEQ_LO_PHYS_ADDR_REG(q_id, pg_num))) + +#define HIFC_AEQ_CTRL_0_ADDR_BASE 0xE00 +#define HIFC_AEQ_CTRL_1_ADDR_BASE 0xE04 +#define HIFC_AEQ_CONS_IDX_0_ADDR_BASE 0xE08 +#define HIFC_AEQ_CONS_IDX_1_ADDR_BASE 0xE0C + +#define HIFC_EQ_OFF_STRIDE 0x80 + +#define HIFC_CSR_AEQ_CTRL_0_ADDR(idx) \ + (HIFC_AEQ_CTRL_0_ADDR_BASE + (idx) * HIFC_EQ_OFF_STRIDE) + +#define HIFC_CSR_AEQ_CTRL_1_ADDR(idx) \ + (HIFC_AEQ_CTRL_1_ADDR_BASE + (idx) * HIFC_EQ_OFF_STRIDE) + +#define HIFC_CSR_AEQ_CONS_IDX_ADDR(idx) \ + (HIFC_AEQ_CONS_IDX_0_ADDR_BASE + (idx) * HIFC_EQ_OFF_STRIDE) + +#define HIFC_CSR_AEQ_PROD_IDX_ADDR(idx) \ + (HIFC_AEQ_CONS_IDX_1_ADDR_BASE + (idx) * HIFC_EQ_OFF_STRIDE) + +#define HIFC_CEQ_CTRL_0_ADDR_BASE 0x1000 +#define HIFC_CEQ_CTRL_1_ADDR_BASE 0x1004 +#define HIFC_CEQ_CONS_IDX_0_ADDR_BASE 0x1008 +#define HIFC_CEQ_CONS_IDX_1_ADDR_BASE 0x100C + +#define HIFC_CSR_CEQ_CTRL_0_ADDR(idx) \ + (HIFC_CEQ_CTRL_0_ADDR_BASE + (idx) * HIFC_EQ_OFF_STRIDE) + +#define HIFC_CSR_CEQ_CTRL_1_ADDR(idx) \ + (HIFC_CEQ_CTRL_1_ADDR_BASE + (idx) * HIFC_EQ_OFF_STRIDE) + +#define HIFC_CSR_CEQ_CONS_IDX_ADDR(idx) \ + (HIFC_CEQ_CONS_IDX_0_ADDR_BASE + (idx) * HIFC_EQ_OFF_STRIDE) + +#define HIFC_CSR_CEQ_PROD_IDX_ADDR(idx) \ + (HIFC_CEQ_CONS_IDX_1_ADDR_BASE + (idx) * HIFC_EQ_OFF_STRIDE) + +enum hifc_eq_type { + HIFC_AEQ, + HIFC_CEQ +}; + +enum hifc_eq_intr_mode { + HIFC_INTR_MODE_ARMED, + HIFC_INTR_MODE_ALWAYS, +}; + +enum hifc_eq_ci_arm_state { + HIFC_EQ_NOT_ARMED, + HIFC_EQ_ARMED, +}; + +struct hifc_eq_work { + struct work_struct work; + void *data; +}; + +struct hifc_ceq_tasklet_data { + void *data; +}; + +struct hifc_eq { + struct hifc_hwdev *hwdev; + u16 q_id; + enum hifc_eq_type type; + u32 page_size; + u32 orig_page_size; + u32 eq_len; + + u32 cons_idx; + u16 wrapped; + + u16 elem_size; + u16 num_pages; + u32 num_elem_in_pg; + + struct irq_info eq_irq; + char irq_name[EQ_IRQ_NAME_LEN]; + + dma_addr_t *dma_addr; + u8 **virt_addr; + dma_addr_t *dma_addr_for_free; + u8 **virt_addr_for_free; + + struct hifc_eq_work aeq_work; + struct tasklet_struct ceq_tasklet; + struct hifc_ceq_tasklet_data ceq_tasklet_data; + + u64 hard_intr_jif; + u64 soft_intr_jif; +}; + +struct hifc_aeq_elem { + u8 aeqe_data[HIFC_AEQE_DATA_SIZE]; + u32 desc; +}; + +enum hifc_aeq_cb_state { + HIFC_AEQ_HW_CB_REG = 0, + HIFC_AEQ_HW_CB_RUNNING, + HIFC_AEQ_SW_CB_REG, + HIFC_AEQ_SW_CB_RUNNING, +}; + +struct hifc_aeqs { + struct hifc_hwdev *hwdev; + + hifc_aeq_hwe_cb aeq_hwe_cb[HIFC_MAX_AEQ_EVENTS]; + hifc_aeq_swe_cb aeq_swe_cb[HIFC_MAX_AEQ_SW_EVENTS]; + unsigned long aeq_hw_cb_state[HIFC_MAX_AEQ_EVENTS]; + unsigned long aeq_sw_cb_state[HIFC_MAX_AEQ_SW_EVENTS]; + + struct hifc_eq aeq[HIFC_MAX_AEQS]; + u16 num_aeqs; + + struct workqueue_struct *workq; +}; + +enum hifc_ceq_cb_state { + HIFC_CEQ_CB_REG = 0, + HIFC_CEQ_CB_RUNNING, +}; + +struct hifc_ceqs { + struct hifc_hwdev *hwdev; + + hifc_ceq_event_cb ceq_cb[HIFC_MAX_CEQ_EVENTS]; + void *ceq_data[HIFC_MAX_CEQ_EVENTS]; + unsigned long ceq_cb_state[HIFC_MAX_CEQ_EVENTS]; + + struct hifc_eq ceq[HIFC_MAX_CEQS]; + u16 num_ceqs; +}; + +int hifc_aeqs_init(struct hifc_hwdev *hwdev, u16 num_aeqs, + struct irq_info *msix_entries); + +void hifc_aeqs_free(struct hifc_hwdev *hwdev); + +int hifc_ceqs_init(struct hifc_hwdev *hwdev, u16 num_ceqs, + struct irq_info *msix_entries); + +void hifc_ceqs_free(struct hifc_hwdev *hwdev); + +void hifc_get_ceq_irqs(struct hifc_hwdev *hwdev, struct irq_info *irqs, + u16 *num_irqs); + +void hifc_get_aeq_irqs(struct hifc_hwdev *hwdev, struct irq_info *irqs, + u16 *num_irqs); + +void hifc_dump_aeq_info(struct hifc_hwdev *hwdev); + +#endif diff --git a/drivers/scsi/huawei/hifc/hifc_hw.h b/drivers/scsi/huawei/hifc/hifc_hw.h new file mode 100644 index 000000000000..49b2edd2bac6 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_hw.h @@ -0,0 +1,611 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#ifndef HIFC_HW_H_ +#define HIFC_HW_H_ + +#ifndef __BIG_ENDIAN__ +#define __BIG_ENDIAN__ 0x4321 +#endif + +#ifndef __LITTLE_ENDIAN__ +#define __LITTLE_ENDIAN__ 0x1234 +#endif + +enum hifc_mod_type { + HIFC_MOD_COMM = 0, /* HW communication module */ + HIFC_MOD_L2NIC = 1, /* L2NIC module*/ + HIFC_MOD_FCOE = 6, + HIFC_MOD_CFGM = 7, /* Configuration module */ + HIFC_MOD_FC = 10, + HIFC_MOD_HILINK = 14, + HIFC_MOD_HW_MAX = 16, /* hardware max module id */ + + /* Software module id, for PF/VF and multi-host */ + HIFC_MOD_MAX, +}; + +struct hifc_cmd_buf { + void *buf; + dma_addr_t dma_addr; + u16 size; +}; + +enum hifc_ack_type { + HIFC_ACK_TYPE_CMDQ, + HIFC_ACK_TYPE_SHARE_CQN, + HIFC_ACK_TYPE_APP_CQN, + HIFC_MOD_ACK_MAX = 15, +}; + +int hifc_msg_to_mgmt_sync(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size, u32 timeout); + +/* PF/VF send msg to uP by api cmd, and return immediately */ +int hifc_msg_to_mgmt_async(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size); + +int hifc_api_cmd_write_nack(void *hwdev, u8 dest, + void *cmd, u16 size); + +int hifc_api_cmd_read_ack(void *hwdev, u8 dest, + void *cmd, u16 size, void *ack, u16 ack_size); +/* PF/VF send cmd to ucode by cmdq, and return if success. + * timeout=0, use default timeout. + */ +int hifc_cmdq_direct_resp(void *hwdev, enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, + struct hifc_cmd_buf *buf_in, + u64 *out_param, u32 timeout); +/* 1. whether need the timeout parameter + * 2. out_param indicates the status of the microcode processing command + */ + +/* PF/VF send cmd to ucode by cmdq, and return detailed result. + * timeout=0, use default timeout. + */ +int hifc_cmdq_detail_resp(void *hwdev, enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, + struct hifc_cmd_buf *buf_in, + struct hifc_cmd_buf *buf_out, u32 timeout); + +/* PF/VF send cmd to ucode by cmdq, and return immediately + */ +int hifc_cmdq_async(void *hwdev, enum hifc_ack_type ack_type, + enum hifc_mod_type mod, u8 cmd, + struct hifc_cmd_buf *buf_in); + +int hifc_ppf_tmr_start(void *hwdev); +int hifc_ppf_tmr_stop(void *hwdev); + +enum hifc_ceq_event { + HIFC_CMDQ = 3, + HIFC_MAX_CEQ_EVENTS = 6, +}; + +typedef void (*hifc_ceq_event_cb)(void *handle, u32 ceqe_data); +int hifc_ceq_register_cb(void *hwdev, enum hifc_ceq_event event, + hifc_ceq_event_cb callback); +void hifc_ceq_unregister_cb(void *hwdev, enum hifc_ceq_event event); + +enum hifc_aeq_type { + HIFC_HW_INTER_INT = 0, + HIFC_MBX_FROM_FUNC = 1, + HIFC_MSG_FROM_MGMT_CPU = 2, + HIFC_API_RSP = 3, + HIFC_API_CHAIN_STS = 4, + HIFC_MBX_SEND_RSLT = 5, + HIFC_MAX_AEQ_EVENTS +}; + +enum hifc_aeq_sw_type { + HIFC_STATELESS_EVENT = 0, + HIFC_STATEFULL_EVENT = 1, + HIFC_MAX_AEQ_SW_EVENTS +}; + +typedef void (*hifc_aeq_hwe_cb)(void *handle, u8 *data, u8 size); +int hifc_aeq_register_hw_cb(void *hwdev, enum hifc_aeq_type event, + hifc_aeq_hwe_cb hwe_cb); +void hifc_aeq_unregister_hw_cb(void *hwdev, enum hifc_aeq_type event); + +typedef u8 (*hifc_aeq_swe_cb)(void *handle, u8 event, u64 data); +int hifc_aeq_register_swe_cb(void *hwdev, enum hifc_aeq_sw_type event, + hifc_aeq_swe_cb aeq_swe_cb); +void hifc_aeq_unregister_swe_cb(void *hwdev, enum hifc_aeq_sw_type event); + +typedef void (*hifc_mgmt_msg_cb)(void *hwdev, void *pri_handle, + u8 cmd, void *buf_in, u16 in_size, void *buf_out, u16 *out_size); + +int hifc_register_mgmt_msg_cb(void *hwdev, + enum hifc_mod_type mod, void *pri_handle, + hifc_mgmt_msg_cb callback); +void hifc_unregister_mgmt_msg_cb(void *hwdev, enum hifc_mod_type mod); + +struct hifc_cmd_buf *hifc_alloc_cmd_buf(void *hwdev); +void hifc_free_cmd_buf(void *hwdev, struct hifc_cmd_buf *buf); + +int hifc_alloc_db_addr(void *hwdev, void __iomem **db_base, + void __iomem **dwqe_base); +void hifc_free_db_addr(void *hwdev, void __iomem *db_base, + void __iomem *dwqe_base); + +struct nic_interrupt_info { + u32 lli_set; + u32 interrupt_coalesc_set; + u16 msix_index; + u8 lli_credit_limit; + u8 lli_timer_cfg; + u8 pending_limt; + u8 coalesc_timer_cfg; + u8 resend_timer_cfg; +}; + +int hifc_get_interrupt_cfg(void *hwdev, + struct nic_interrupt_info *interrupt_info); + +int hifc_set_interrupt_cfg(void *hwdev, + struct nic_interrupt_info interrupt_info); + +/* The driver code implementation interface*/ +void hifc_misx_intr_clear_resend_bit(void *hwdev, + u16 msix_idx, u8 clear_resend_en); + +struct hifc_sq_attr { + u8 dma_attr_off; + u8 pending_limit; + u8 coalescing_time; + u8 intr_en; + u16 intr_idx; + u32 l2nic_sqn; + u64 ci_dma_base; +}; + +int hifc_set_ci_table(void *hwdev, u16 q_id, struct hifc_sq_attr *attr); + +int hifc_set_root_ctxt(void *hwdev, u16 rq_depth, u16 sq_depth, int rx_buf_sz); +int hifc_clean_root_ctxt(void *hwdev); +void hifc_record_pcie_error(void *hwdev); + +int hifc_func_rx_tx_flush(void *hwdev); + +int hifc_func_tmr_bitmap_set(void *hwdev, bool enable); + +struct hifc_init_para { + /* Record hifc_pcidev or NDIS_Adapter pointer address*/ + void *adapter_hdl; + /* Record pcidev or Handler pointer address + * for example: ioremap interface input parameter + */ + void *pcidev_hdl; + /* Record pcidev->dev or Handler pointer address which used to + * dma address application or dev_err print the parameter + */ + void *dev_hdl; + + void *cfg_reg_base; /* Configure virtual address, bar0/1*/ + /* interrupt configuration register address, bar2/3 */ + void *intr_reg_base; + u64 db_base_phy; + void *db_base; /* the doorbell address, bar4/5 higher 4M space*/ + void *dwqe_mapping;/* direct wqe 4M, follow the doorbell address space*/ + void **hwdev; + void *chip_node; + /* In bmgw x86 host, driver can't send message to mgmt cpu directly, + * need to trasmit message ppf mbox to bmgw arm host. + */ + void *ppf_hwdev; +}; + +#ifndef IFNAMSIZ +#define IFNAMSIZ 16 +#endif +#define MAX_FUNCTION_NUM 512 +#define HIFC_MAX_PF_NUM 16 +#define HIFC_MAX_COS 8 +#define INIT_FAILED 0 +#define INIT_SUCCESS 1 +#define MAX_DRV_BUF_SIZE 4096 + +struct hifc_cmd_get_light_module_abs { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 port_id; + u8 abs_status; /* 0:present, 1:absent */ + u8 rsv[2]; +}; + +#define SFP_INFO_MAX_SIZE 512 +struct hifc_cmd_get_sfp_qsfp_info { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 port_id; + u8 wire_type; + u16 out_len; + u8 sfp_qsfp_info[SFP_INFO_MAX_SIZE]; +}; + +#define HIFC_MAX_PORT_ID 4 + +struct hifc_port_routine_cmd { + bool up_send_sfp_info; + bool up_send_sfp_abs; + + struct hifc_cmd_get_sfp_qsfp_info sfp_info; + struct hifc_cmd_get_light_module_abs abs; +}; + +struct card_node { + struct list_head node; + struct list_head func_list; + char chip_name[IFNAMSIZ]; + void *log_info; + void *dbgtool_info; + void *func_handle_array[MAX_FUNCTION_NUM]; + unsigned char dp_bus_num; + u8 func_num; + struct attribute dbgtool_attr_file; + + bool cos_up_setted; + u8 cos_up[HIFC_MAX_COS]; + bool ppf_state; + u8 pf_bus_num[HIFC_MAX_PF_NUM]; + + struct hifc_port_routine_cmd rt_cmd[HIFC_MAX_PORT_ID]; + + /* mutex used for copy sfp info */ + struct mutex sfp_mutex; +}; + +enum hifc_hwdev_init_state { + HIFC_HWDEV_NONE_INITED = 0, + HIFC_HWDEV_CLP_INITED, + HIFC_HWDEV_AEQ_INITED, + HIFC_HWDEV_MGMT_INITED, + HIFC_HWDEV_MBOX_INITED, + HIFC_HWDEV_CMDQ_INITED, + HIFC_HWDEV_COMM_CH_INITED, + HIFC_HWDEV_ALL_INITED, + HIFC_HWDEV_MAX_INVAL_INITED +}; + +enum hifc_func_cap { + /* send message to mgmt cpu directly */ + HIFC_FUNC_MGMT = 1 << 0, + /* setting port attribute, pause/speed etc. */ + HIFC_FUNC_PORT = 1 << 1, + /* Enable SR-IOV in default */ + HIFC_FUNC_SRIOV_EN_DFLT = 1 << 2, + /* Can't change VF num */ + HIFC_FUNC_SRIOV_NUM_FIX = 1 << 3, + /* Fcorce pf/vf link up */ + HIFC_FUNC_FORCE_LINK_UP = 1 << 4, + /* Support rate limit */ + HIFC_FUNC_SUPP_RATE_LIMIT = 1 << 5, + HIFC_FUNC_SUPP_DFX_REG = 1 << 6, + /* Support promisc/multicast/all-multi */ + HIFC_FUNC_SUPP_RX_MODE = 1 << 7, + /* Set vf mac and vlan by ip link */ + HIFC_FUNC_SUPP_SET_VF_MAC_VLAN = 1 << 8, + /* Support set mac by ifconfig */ + HIFC_FUNC_SUPP_CHANGE_MAC = 1 << 9, + /* OVS don't support SCTP_CRC/HW_VLAN/LRO */ + HIFC_FUNC_OFFLOAD_OVS_UNSUPP = 1 << 10, +}; + +#define FUNC_SUPPORT_MGMT(hwdev) \ + (!!(hifc_get_func_feature_cap(hwdev) & HIFC_FUNC_MGMT)) +#define FUNC_SUPPORT_PORT_SETTING(hwdev) \ + (!!(hifc_get_func_feature_cap(hwdev) & HIFC_FUNC_PORT)) +#define FUNC_SUPPORT_DCB(hwdev) \ + (FUNC_SUPPORT_PORT_SETTING(hwdev)) +#define FUNC_ENABLE_SRIOV_IN_DEFAULT(hwdev) \ + (!!(hifc_get_func_feature_cap(hwdev) & \ + HIFC_FUNC_SRIOV_EN_DFLT)) +#define FUNC_SRIOV_FIX_NUM_VF(hwdev) \ + (!!(hifc_get_func_feature_cap(hwdev) & \ + HIFC_FUNC_SRIOV_NUM_FIX)) +#define FUNC_SUPPORT_RX_MODE(hwdev) \ + (!!(hifc_get_func_feature_cap(hwdev) & \ + HIFC_FUNC_SUPP_RX_MODE)) +#define FUNC_SUPPORT_RATE_LIMIT(hwdev) \ + (!!(hifc_get_func_feature_cap(hwdev) & \ + HIFC_FUNC_SUPP_RATE_LIMIT)) +#define FUNC_SUPPORT_SET_VF_MAC_VLAN(hwdev) \ + (!!(hifc_get_func_feature_cap(hwdev) & \ + HIFC_FUNC_SUPP_SET_VF_MAC_VLAN)) +#define FUNC_SUPPORT_CHANGE_MAC(hwdev) \ + (!!(hifc_get_func_feature_cap(hwdev) & \ + HIFC_FUNC_SUPP_CHANGE_MAC)) +#define FUNC_FORCE_LINK_UP(hwdev) \ + (!!(hifc_get_func_feature_cap(hwdev) & \ + HIFC_FUNC_FORCE_LINK_UP)) +#define FUNC_SUPPORT_SCTP_CRC(hwdev) \ + (!(hifc_get_func_feature_cap(hwdev) & \ + HIFC_FUNC_OFFLOAD_OVS_UNSUPP)) +#define FUNC_SUPPORT_HW_VLAN(hwdev) \ + (!(hifc_get_func_feature_cap(hwdev) & \ + HIFC_FUNC_OFFLOAD_OVS_UNSUPP)) +#define FUNC_SUPPORT_LRO(hwdev) \ + (!(hifc_get_func_feature_cap(hwdev) & \ + HIFC_FUNC_OFFLOAD_OVS_UNSUPP)) + +int hifc_init_hwdev(struct hifc_init_para *para); +void hifc_free_hwdev(void *hwdev); +int hifc_stateful_init(void *hwdev); +void hifc_stateful_deinit(void *hwdev); +bool hifc_is_hwdev_mod_inited(void *hwdev, enum hifc_hwdev_init_state state); +u64 hifc_get_func_feature_cap(void *hwdev); +int hifc_slq_init(void *dev, int num_wqs); +void hifc_slq_uninit(void *dev); +int hifc_slq_alloc(void *dev, u16 wqebb_size, u16 q_depth, + u16 page_size, u64 *cla_addr, void **handle); +void hifc_slq_free(void *dev, void *handle); +u64 hifc_slq_get_addr(void *handle, u16 index); +u64 hifc_slq_get_first_pageaddr(void *handle); + +typedef void (*comm_up_self_msg_proc)(void *handle, void *buf_in, + u16 in_size, void *buf_out, + u16 *out_size); +void hifc_comm_recv_mgmt_self_cmd_reg(void *hwdev, u8 cmd, + comm_up_self_msg_proc proc); +void hifc_comm_recv_up_self_cmd_unreg(void *hwdev, u8 cmd); + +/* defined by chip */ +enum hifc_fault_type { + FAULT_TYPE_CHIP, + FAULT_TYPE_UCODE, + FAULT_TYPE_MEM_RD_TIMEOUT, + FAULT_TYPE_MEM_WR_TIMEOUT, + FAULT_TYPE_REG_RD_TIMEOUT, + FAULT_TYPE_REG_WR_TIMEOUT, + FAULT_TYPE_PHY_FAULT, + FAULT_TYPE_MAX, +}; + +/* defined by chip */ +enum hifc_fault_err_level { + /* default err_level=FAULT_LEVEL_FATAL if + * type==FAULT_TYPE_MEM_RD_TIMEOUT || FAULT_TYPE_MEM_WR_TIMEOUT || + * FAULT_TYPE_REG_RD_TIMEOUT || FAULT_TYPE_REG_WR_TIMEOUT || + * FAULT_TYPE_UCODE + * other: err_level in event.chip.err_level if type==FAULT_TYPE_CHIP + */ + FAULT_LEVEL_FATAL, + FAULT_LEVEL_SERIOUS_RESET, + FAULT_LEVEL_SERIOUS_FLR, + FAULT_LEVEL_GENERAL, + FAULT_LEVEL_SUGGESTION, + FAULT_LEVEL_MAX +}; + +enum hifc_fault_source_type { + /* same as FAULT_TYPE_CHIP */ + HIFC_FAULT_SRC_HW_MGMT_CHIP = 0, + /* same as FAULT_TYPE_UCODE */ + HIFC_FAULT_SRC_HW_MGMT_UCODE, + /* same as FAULT_TYPE_MEM_RD_TIMEOUT */ + HIFC_FAULT_SRC_HW_MGMT_MEM_RD_TIMEOUT, + /* same as FAULT_TYPE_MEM_WR_TIMEOUT */ + HIFC_FAULT_SRC_HW_MGMT_MEM_WR_TIMEOUT, + /* same as FAULT_TYPE_REG_RD_TIMEOUT */ + HIFC_FAULT_SRC_HW_MGMT_REG_RD_TIMEOUT, + /* same as FAULT_TYPE_REG_WR_TIMEOUT */ + HIFC_FAULT_SRC_HW_MGMT_REG_WR_TIMEOUT, + HIFC_FAULT_SRC_SW_MGMT_UCODE, + HIFC_FAULT_SRC_MGMT_WATCHDOG, + HIFC_FAULT_SRC_MGMT_RESET = 8, + HIFC_FAULT_SRC_HW_PHY_FAULT, + HIFC_FAULT_SRC_HOST_HEARTBEAT_LOST = 20, + HIFC_FAULT_SRC_TYPE_MAX, +}; + +struct hifc_fault_sw_mgmt { + u8 event_id; + u64 event_data; +}; + +union hifc_fault_hw_mgmt { + u32 val[4]; + /* valid only type==FAULT_TYPE_CHIP */ + struct { + u8 node_id; + /* enum hifc_fault_err_level */ + u8 err_level; + u16 err_type; + u32 err_csr_addr; + u32 err_csr_value; + /* func_id valid only err_level==FAULT_LEVEL_SERIOUS_FLR + */ + u16 func_id; + u16 rsvd2; + } chip; + + /* valid only type==FAULT_TYPE_UCODE */ + struct { + u8 cause_id; + u8 core_id; + u8 c_id; + u8 rsvd3; + u32 epc; + u32 rsvd4; + u32 rsvd5; + } ucode; + + /* valid only type==FAULT_TYPE_MEM_RD_TIMEOUT || + * FAULT_TYPE_MEM_WR_TIMEOUT + */ + struct { + u32 err_csr_ctrl; + u32 err_csr_data; + u32 ctrl_tab; + u32 mem_index; + } mem_timeout; + + /* valid only type==FAULT_TYPE_REG_RD_TIMEOUT || + * FAULT_TYPE_REG_WR_TIMEOUT + */ + struct { + u32 err_csr; + u32 rsvd6; + u32 rsvd7; + u32 rsvd8; + } reg_timeout; + + struct { + /* 0: read; 1: write */ + u8 op_type; + u8 port_id; + u8 dev_ad; + u8 rsvd9; + u32 csr_addr; + u32 op_data; + u32 rsvd10; + } phy_fault; +}; + +/* defined by chip */ +struct hifc_fault_event { + /* enum hifc_fault_type */ + u8 type; + u8 rsvd0[3]; + union hifc_fault_hw_mgmt event; +}; + +struct hifc_fault_recover_info { + u8 fault_src; /* enum hifc_fault_source_type */ + u8 fault_lev; /* enum hifc_fault_err_level */ + u8 rsvd0[2]; + union { + union hifc_fault_hw_mgmt hw_mgmt; + struct hifc_fault_sw_mgmt sw_mgmt; + u32 mgmt_rsvd[4]; + u32 host_rsvd[4]; + } fault_data; +}; + +struct hifc_dcb_state { + u8 dcb_on; + u8 default_cos; + u8 up_cos[8]; +}; + +enum link_err_type { + LINK_ERR_MODULE_UNRECOGENIZED, + LINK_ERR_NUM, +}; + +enum port_module_event_type { + HIFC_PORT_MODULE_CABLE_PLUGGED, + HIFC_PORT_MODULE_CABLE_UNPLUGGED, + HIFC_PORT_MODULE_LINK_ERR, + HIFC_PORT_MODULE_MAX_EVENT, +}; + +struct hifc_port_module_event { + enum port_module_event_type type; + enum link_err_type err_type; +}; + +struct hifc_event_link_info { + u8 valid; + u8 port_type; + u8 autoneg_cap; + u8 autoneg_state; + u8 duplex; + u8 speed; +}; + +struct hifc_mctp_host_info { + u8 major_cmd; + u8 sub_cmd; + u8 rsvd[2]; + + u32 data_len; + void *data; +}; + +enum hifc_event_type { + HIFC_EVENT_LINK_DOWN = 0, + HIFC_EVENT_LINK_UP = 1, + HIFC_EVENT_HEART_LOST = 2, + HIFC_EVENT_FAULT = 3, + HIFC_EVENT_NOTIFY_VF_DCB_STATE = 4, + HIFC_EVENT_DCB_STATE_CHANGE = 5, + HIFC_EVENT_FMW_ACT_NTC = 6, + HIFC_EVENT_PORT_MODULE_EVENT = 7, + HIFC_EVENT_MCTP_GET_HOST_INFO, + HIFC_EVENT_MULTI_HOST_MGMT, + HIFC_EVENT_INIT_MIGRATE_PF, +}; + +struct hifc_event_info { + enum hifc_event_type type; + union { + struct hifc_event_link_info link_info; + struct hifc_fault_event info; + struct hifc_dcb_state dcb_state; + struct hifc_port_module_event module_event; + u8 vf_default_cos; + struct hifc_mctp_host_info mctp_info; + }; +}; + +enum hifc_ucode_event_type { + HIFC_INTERNAL_TSO_FATAL_ERROR = 0x0, + HIFC_INTERNAL_LRO_FATAL_ERROR = 0x1, + HIFC_INTERNAL_TX_FATAL_ERROR = 0x2, + HIFC_INTERNAL_RX_FATAL_ERROR = 0x3, + HIFC_INTERNAL_OTHER_FATAL_ERROR = 0x4, + HIFC_NIC_FATAL_ERROR_MAX = 0x8, +}; + +typedef void (*hifc_event_handler)(void *handle, + struct hifc_event_info *event); +/* only register once */ +void hifc_event_register(void *dev, void *pri_handle, + hifc_event_handler callback); +void hifc_event_unregister(void *dev); + +void hifc_detect_hw_present(void *hwdev); + +void hifc_set_chip_absent(void *hwdev); + +int hifc_get_chip_present_flag(void *hwdev); + +void hifc_set_pcie_order_cfg(void *handle); + +int hifc_get_mgmt_channel_status(void *handle); + +struct hifc_board_info { + u32 board_type; + u32 port_num; + u32 port_speed; + u32 pcie_width; + u32 host_num; + u32 pf_num; + u32 vf_total_num; + u32 tile_num; + u32 qcm_num; + u32 core_num; + u32 work_mode; + u32 service_mode; + u32 pcie_mode; + u32 cfg_addr; + u32 boot_sel; + u32 board_id; +}; + +int hifc_get_board_info(void *hwdev, struct hifc_board_info *info); + +int hifc_get_card_present_state(void *hwdev, bool *card_present_state); + +#endif diff --git a/drivers/scsi/huawei/hifc/hifc_hwdev.c b/drivers/scsi/huawei/hifc/hifc_hwdev.c new file mode 100644 index 000000000000..760e02394b05 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_hwdev.c @@ -0,0 +1,3675 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": [COMM]" fmt + +#include <linux/kernel.h> +#include <linux/pci.h> +#include <linux/pci_regs.h> +#include <linux/types.h> +#include <linux/delay.h> +#include <linux/module.h> +#include <linux/completion.h> +#include <linux/semaphore.h> +#include <linux/interrupt.h> + +#include "hifc_knl_adp.h" +#include "hifc_hw.h" +#include "hifc_hwdev.h" +#include "hifc_hwif.h" +#include "hifc_api_cmd.h" +#include "hifc_mgmt.h" +#include "hifc_eqs.h" +#include "hifc_wq.h" +#include "hifc_cmdq.h" +#include "hifc_hwif.h" + +#define HIFC_DEAULT_EQ_MSIX_PENDING_LIMIT 0 +#define HIFC_DEAULT_EQ_MSIX_COALESC_TIMER_CFG 0xFF +#define HIFC_DEAULT_EQ_MSIX_RESEND_TIMER_CFG 7 +#define HIFC_FLR_TIMEOUT 1000 +#define HIFC_HT_GPA_PAGE_SIZE 4096UL +#define HIFC_PPF_HT_GPA_SET_RETRY_TIMES 10 +#define HIFC_GET_SFP_INFO_REAL_TIME 0x1 +#define HIFC_GLB_SO_RO_CFG_SHIFT 0x0 +#define HIFC_GLB_SO_RO_CFG_MASK 0x1 +#define HIFC_DISABLE_ORDER 0 +#define HIFC_GLB_DMA_SO_RO_GET(val, member) \ + (((val) >> HIFC_GLB_##member##_SHIFT) & HIFC_GLB_##member##_MASK) + +#define HIFC_GLB_DMA_SO_R0_CLEAR(val, member) \ + ((val) & (~(HIFC_GLB_##member##_MASK << HIFC_GLB_##member##_SHIFT))) + +#define HIFC_GLB_DMA_SO_R0_SET(val, member) \ + (((val) & HIFC_GLB_##member##_MASK) << HIFC_GLB_##member##_SHIFT) + +#define HIFC_MGMT_CHANNEL_STATUS_SHIFT 0x0 +#define HIFC_MGMT_CHANNEL_STATUS_MASK 0x1 +#define HIFC_ACTIVE_STATUS_MASK 0x80000000 +#define HIFC_ACTIVE_STATUS_CLEAR 0x7FFFFFFF + +#define HIFC_GET_MGMT_CHANNEL_STATUS(val, member) \ + (((val) >> HIFC_##member##_SHIFT) & HIFC_##member##_MASK) + +#define HIFC_CLEAR_MGMT_CHANNEL_STATUS(val, member) \ + ((val) & (~(HIFC_##member##_MASK << HIFC_##member##_SHIFT))) + +#define HIFC_SET_MGMT_CHANNEL_STATUS(val, member) \ + (((val) & HIFC_##member##_MASK) << HIFC_##member##_SHIFT) + +#define HIFC_BOARD_IS_PHY(hwdev) \ + ((hwdev)->board_info.board_type == 4 && \ + (hwdev)->board_info.board_id == 24) + +struct comm_info_ht_gpa_set { + u8 status; + u8 version; + u8 rsvd0[6]; + u32 rsvd1; + u32 rsvd2; + u64 page_pa0; + u64 page_pa1; +}; + +struct hifc_cons_idx_attr { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_idx; + u8 dma_attr_off; + u8 pending_limit; + u8 coalescing_time; + u8 intr_en; + u16 intr_idx; + u32 l2nic_sqn; + u32 sq_id; + u64 ci_addr; +}; + +struct hifc_clear_doorbell { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_idx; + u8 ppf_idx; + u8 rsvd1; +}; + +struct hifc_clear_resource { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_idx; + u8 ppf_idx; + u8 rsvd1; +}; + +struct hifc_msix_config { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 msix_index; + u8 pending_cnt; + u8 coalesct_timer_cnt; + u8 lli_tmier_cnt; + u8 lli_credit_cnt; + u8 resend_timer_cnt; + u8 rsvd1[3]; +}; + +enum func_tmr_bitmap_status { + FUNC_TMR_BITMAP_DISABLE, + FUNC_TMR_BITMAP_ENABLE, +}; + +struct hifc_func_tmr_bitmap_op { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_idx; + u8 op_id; /* 0:start; 1:stop */ + u8 ppf_idx; + u32 rsvd1; +}; + +struct hifc_ppf_tmr_op { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 ppf_idx; + u8 op_id; /* 0: stop timer; 1:start timer */ + u8 rsvd1[2]; + u32 rsvd2; +}; + +struct hifc_cmd_set_res_state { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_idx; + u8 state; + u8 rsvd1; + u32 rsvd2; +}; + +int hifc_hw_rx_buf_size[] = { + HIFC_RX_BUF_SIZE_32B, + HIFC_RX_BUF_SIZE_64B, + HIFC_RX_BUF_SIZE_96B, + HIFC_RX_BUF_SIZE_128B, + HIFC_RX_BUF_SIZE_192B, + HIFC_RX_BUF_SIZE_256B, + HIFC_RX_BUF_SIZE_384B, + HIFC_RX_BUF_SIZE_512B, + HIFC_RX_BUF_SIZE_768B, + HIFC_RX_BUF_SIZE_1K, + HIFC_RX_BUF_SIZE_1_5K, + HIFC_RX_BUF_SIZE_2K, + HIFC_RX_BUF_SIZE_3K, + HIFC_RX_BUF_SIZE_4K, + HIFC_RX_BUF_SIZE_8K, + HIFC_RX_BUF_SIZE_16K, +}; + +struct hifc_comm_board_info { + u8 status; + u8 version; + u8 rsvd0[6]; + + struct hifc_board_info info; + + u32 rsvd1[4]; +}; + +#define PHY_DOING_INIT_TIMEOUT (15 * 1000) + +struct hifc_phy_init_status { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 init_status; + u8 rsvd1[3]; +}; + +enum phy_init_status_type { + PHY_INIT_DOING = 0, + PHY_INIT_SUCCESS = 1, + PHY_INIT_FAIL = 2, + PHY_NONSUPPORT = 3, +}; + +struct hifc_update_active { + u8 status; + u8 version; + u8 rsvd0[6]; + + u32 update_flag; + u32 update_status; +}; + +struct hifc_mgmt_watchdog_info { + u8 status; + u8 version; + u8 rsvd0[6]; + + u32 curr_time_h; + u32 curr_time_l; + u32 task_id; + u32 rsv; + + u32 reg[13]; + u32 pc; + u32 lr; + u32 cpsr; + + u32 stack_top; + u32 stack_bottom; + u32 sp; + u32 curr_used; + u32 peak_used; + u32 is_overflow; + + u32 stack_actlen; + u8 data[1024]; +}; + +struct hifc_fmw_act_ntc { + u8 status; + u8 version; + u8 rsvd0[6]; + + u32 rsvd1[5]; +}; + +#define HIFC_PAGE_SIZE_HW(pg_size) ((u8)ilog2((u32)((pg_size) >> 12))) + +struct hifc_wq_page_size { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_idx; + u8 ppf_idx; + /* real_size=4KB*2^page_size, range(0~20) must be checked by driver */ + u8 page_size; + + u32 rsvd1; +}; + +#define MAX_PCIE_DFX_BUF_SIZE (1024) + +struct hifc_pcie_dfx_ntc { + u8 status; + u8 version; + u8 rsvd0[6]; + + int len; + u32 rsvd; +}; + +struct hifc_pcie_dfx_info { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 host_id; + u8 last; + u8 rsvd[2]; + u32 offset; + + u8 data[MAX_PCIE_DFX_BUF_SIZE]; +}; + +struct hifc_reg_info { + u8 status; + u8 version; + u8 rsvd0[6]; + + u32 reg_addr; + u32 val_length; + + u32 data[2]; +}; + +#define HIFC_DMA_ATTR_ENTRY_ST_SHIFT 0 +#define HIFC_DMA_ATTR_ENTRY_AT_SHIFT 8 +#define HIFC_DMA_ATTR_ENTRY_PH_SHIFT 10 +#define HIFC_DMA_ATTR_ENTRY_NO_SNOOPING_SHIFT 12 +#define HIFC_DMA_ATTR_ENTRY_TPH_EN_SHIFT 13 + +#define HIFC_DMA_ATTR_ENTRY_ST_MASK 0xFF +#define HIFC_DMA_ATTR_ENTRY_AT_MASK 0x3 +#define HIFC_DMA_ATTR_ENTRY_PH_MASK 0x3 +#define HIFC_DMA_ATTR_ENTRY_NO_SNOOPING_MASK 0x1 +#define HIFC_DMA_ATTR_ENTRY_TPH_EN_MASK 0x1 + +#define HIFC_DMA_ATTR_ENTRY_SET(val, member) \ + (((u32)(val) & HIFC_DMA_ATTR_ENTRY_##member##_MASK) << \ + HIFC_DMA_ATTR_ENTRY_##member##_SHIFT) + +#define HIFC_DMA_ATTR_ENTRY_CLEAR(val, member) \ + ((val) & (~(HIFC_DMA_ATTR_ENTRY_##member##_MASK \ + << HIFC_DMA_ATTR_ENTRY_##member##_SHIFT))) + +#define HIFC_PCIE_ST_DISABLE 0 +#define HIFC_PCIE_AT_DISABLE 0 +#define HIFC_PCIE_PH_DISABLE 0 + +#define PCIE_MSIX_ATTR_ENTRY 0 + +#define HIFC_CHIP_PRESENT 1 +#define HIFC_CHIP_ABSENT 0 + +struct hifc_cmd_fault_event { + u8 status; + u8 version; + u8 rsvd0[6]; + + struct hifc_fault_event event; +}; + +#define HEARTBEAT_DRV_MAGIC_ACK 0x5A5A5A5A + +struct hifc_heartbeat_event { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 mgmt_init_state; + u8 rsvd1[3]; + u32 heart; /* increased every event */ + u32 drv_heart; +}; + +static void hifc_set_mgmt_channel_status(void *handle, bool state) +{ + struct hifc_hwdev *hwdev = handle; + u32 val; + + if (!hwdev || hifc_func_type(hwdev) == TYPE_VF || + !(hwdev->feature_cap & HIFC_FUNC_SUPP_DFX_REG)) + return; + + val = hifc_hwif_read_reg(hwdev->hwif, HIFC_ICPL_RESERVD_ADDR); + val = HIFC_CLEAR_MGMT_CHANNEL_STATUS(val, MGMT_CHANNEL_STATUS); + val |= HIFC_SET_MGMT_CHANNEL_STATUS((u32)state, MGMT_CHANNEL_STATUS); + + hifc_hwif_write_reg(hwdev->hwif, HIFC_ICPL_RESERVD_ADDR, val); +} + +static void hifc_enable_mgmt_channel(void *hwdev, void *buf_out) +{ + struct hifc_hwdev *dev = hwdev; + struct hifc_update_active *active_info = buf_out; + + if (!active_info || hifc_func_type(hwdev) == TYPE_VF || + !(dev->feature_cap & HIFC_FUNC_SUPP_DFX_REG)) + return; + + if ((!active_info->status) && + (active_info->update_status & HIFC_ACTIVE_STATUS_MASK)) { + active_info->update_status &= HIFC_ACTIVE_STATUS_CLEAR; + return; + } + + hifc_set_mgmt_channel_status(hwdev, false); +} + +int hifc_set_wq_page_size(struct hifc_hwdev *hwdev, u16 func_idx, + u32 page_size); + +#define HIFC_QUEUE_MIN_DEPTH 6 +#define HIFC_QUEUE_MAX_DEPTH 12 +#define HIFC_MAX_RX_BUFFER_SIZE 15 + +#define ROOT_CTX_QPS_VALID(root_ctxt) \ + ((root_ctxt)->rq_depth >= HIFC_QUEUE_MIN_DEPTH && \ + (root_ctxt)->rq_depth <= HIFC_QUEUE_MAX_DEPTH && \ + (root_ctxt)->sq_depth >= HIFC_QUEUE_MIN_DEPTH && \ + (root_ctxt)->sq_depth <= HIFC_QUEUE_MAX_DEPTH && \ + (root_ctxt)->rx_buf_sz <= HIFC_MAX_RX_BUFFER_SIZE) + +struct hifc_mgmt_status_log { + u8 status; + const char *log; +}; + +struct hifc_mgmt_status_log mgmt_status_log[] = { + {HIFC_MGMT_STATUS_ERR_PARAM, "Invalid parameter"}, + {HIFC_MGMT_STATUS_ERR_FAILED, "Operation failed"}, + {HIFC_MGMT_STATUS_ERR_PORT, "Invalid port"}, + {HIFC_MGMT_STATUS_ERR_TIMEOUT, "Operation time out"}, + {HIFC_MGMT_STATUS_ERR_NOMATCH, "Version not match"}, + {HIFC_MGMT_STATUS_ERR_EXIST, "Entry exists"}, + {HIFC_MGMT_STATUS_ERR_NOMEM, "Out of memory"}, + {HIFC_MGMT_STATUS_ERR_INIT, "Feature not initialized"}, + {HIFC_MGMT_STATUS_ERR_FAULT, "Invalid address"}, + {HIFC_MGMT_STATUS_ERR_PERM, "Operation not permitted"}, + {HIFC_MGMT_STATUS_ERR_EMPTY, "Table empty"}, + {HIFC_MGMT_STATUS_ERR_FULL, "Table full"}, + {HIFC_MGMT_STATUS_ERR_NOT_FOUND, "Not found"}, + {HIFC_MGMT_STATUS_ERR_BUSY, "Device or resource busy "}, + {HIFC_MGMT_STATUS_ERR_RESOURCE, "No resources for operation "}, + {HIFC_MGMT_STATUS_ERR_CONFIG, "Invalid configuration"}, + {HIFC_MGMT_STATUS_ERR_UNAVAIL, "Feature unavailable"}, + {HIFC_MGMT_STATUS_ERR_CRC, "CRC check failed"}, + {HIFC_MGMT_STATUS_ERR_NXIO, "No such device or address"}, + {HIFC_MGMT_STATUS_ERR_ROLLBACK, "Chip rollback fail"}, + {HIFC_MGMT_STATUS_ERR_LEN, "Length too short or too long"}, + {HIFC_MGMT_STATUS_ERR_UNSUPPORT, "Feature not supported"}, +}; + +static void __print_status_info(struct hifc_hwdev *dev, + enum hifc_mod_type mod, u8 cmd, int index) +{ + if (mod == HIFC_MOD_COMM) { + sdk_err(dev->dev_hdl, "Mgmt process mod(0x%x) cmd(0x%x) fail: %s", + mod, cmd, mgmt_status_log[index].log); + } else if (mod == HIFC_MOD_L2NIC || + mod == HIFC_MOD_HILINK) { + sdk_err(dev->dev_hdl, "Mgmt process mod(0x%x) cmd(0x%x) fail: %s", + mod, cmd, mgmt_status_log[index].log); + } +} + +static bool hifc_status_need_special_handle(enum hifc_mod_type mod, + u8 cmd, u8 status) +{ + if (mod == HIFC_MOD_L2NIC) { + /* optical module isn't plugged in */ + if (((cmd == HIFC_PORT_CMD_GET_STD_SFP_INFO) || + (cmd == HIFC_PORT_CMD_GET_SFP_INFO)) && + (status == HIFC_MGMT_STATUS_ERR_NXIO)) + return true; + + if ((cmd == HIFC_PORT_CMD_SET_MAC || + cmd == HIFC_PORT_CMD_UPDATE_MAC) && + status == HIFC_MGMT_STATUS_ERR_EXIST) + return true; + } + + return false; +} + +static bool print_status_info_valid(enum hifc_mod_type mod, + const void *buf_out) +{ + if (!buf_out) + return false; + + if (mod != HIFC_MOD_COMM && mod != HIFC_MOD_L2NIC && + mod != HIFC_MOD_HILINK) + return false; + + return true; +} + +static void hifc_print_status_info(void *hwdev, enum hifc_mod_type mod, + u8 cmd, const void *buf_out) +{ + struct hifc_hwdev *dev = hwdev; + int i, size; + u8 status; + + if (!print_status_info_valid(mod, buf_out)) + return; + + status = *(u8 *)buf_out; + + if (!status) + return; + + if (hifc_status_need_special_handle(mod, cmd, status)) + return; + + size = sizeof(mgmt_status_log) / sizeof(mgmt_status_log[0]); + for (i = 0; i < size; i++) { + if (status == mgmt_status_log[i].status) { + __print_status_info(dev, mod, cmd, i); + return; + } + } + + if (mod == HIFC_MOD_COMM) { + sdk_err(dev->dev_hdl, "Mgmt process mod(0x%x) cmd(0x%x) return driver unknown status(0x%x)\n", + mod, cmd, status); + } else if (mod == HIFC_MOD_L2NIC || mod == HIFC_MOD_HILINK) { + sdk_err(dev->dev_hdl, "Mgmt process mod(0x%x) cmd(0x%x) return driver unknown status(0x%x)\n", + mod, cmd, status); + } +} + +void hifc_set_chip_present(void *hwdev) +{ + ((struct hifc_hwdev *)hwdev)->chip_present_flag = HIFC_CHIP_PRESENT; +} + +void hifc_set_chip_absent(void *hwdev) +{ + struct hifc_hwdev *dev = hwdev; + + sdk_err(dev->dev_hdl, "Card not present\n"); + dev->chip_present_flag = HIFC_CHIP_ABSENT; +} + +int hifc_get_chip_present_flag(void *hwdev) +{ + int flag; + + if (!hwdev) + return -EINVAL; + flag = ((struct hifc_hwdev *)hwdev)->chip_present_flag; + return flag; +} + +void hifc_force_complete_all(void *hwdev) +{ + struct hifc_hwdev *dev = (struct hifc_hwdev *)hwdev; + struct hifc_recv_msg *recv_resp_msg; + + set_bit(HIFC_HWDEV_STATE_BUSY, &dev->func_state); + + if (hifc_func_type(dev) != TYPE_VF && + hifc_is_hwdev_mod_inited(dev, HIFC_HWDEV_MGMT_INITED)) { + recv_resp_msg = &dev->pf_to_mgmt->recv_resp_msg_from_mgmt; + if (dev->pf_to_mgmt->event_flag == SEND_EVENT_START) { + complete(&recv_resp_msg->recv_done); + dev->pf_to_mgmt->event_flag = SEND_EVENT_TIMEOUT; + } + } + + /* only flush sync cmdq to avoid blocking remove */ + if (hifc_is_hwdev_mod_inited(dev, HIFC_HWDEV_CMDQ_INITED)) + hifc_cmdq_flush_cmd(hwdev, + &dev->cmdqs->cmdq[HIFC_CMDQ_SYNC]); + + clear_bit(HIFC_HWDEV_STATE_BUSY, &dev->func_state); +} + +void hifc_detect_hw_present(void *hwdev) +{ + u32 addr, attr1; + + addr = HIFC_CSR_FUNC_ATTR1_ADDR; + attr1 = hifc_hwif_read_reg(((struct hifc_hwdev *)hwdev)->hwif, addr); + if (attr1 == HIFC_PCIE_LINK_DOWN) { + hifc_set_chip_absent(hwdev); + hifc_force_complete_all(hwdev); + } +} + +void hifc_record_pcie_error(void *hwdev) +{ + struct hifc_hwdev *dev = (struct hifc_hwdev *)hwdev; + + if (!hwdev) + return; + + atomic_inc(&dev->hw_stats.fault_event_stats.pcie_fault_stats); +} + +static inline void __set_heartbeat_ehd_detect_delay(struct hifc_hwdev *hwdev, + u32 delay_ms) +{ + hwdev->heartbeat_ehd.start_detect_jiffies = + jiffies + msecs_to_jiffies(delay_ms); +} + +static int __pf_to_mgmt_pre_handle(struct hifc_hwdev *hwdev, + enum hifc_mod_type mod, u8 cmd) +{ + if (hifc_get_mgmt_channel_status(hwdev)) { + if (mod == HIFC_MOD_COMM || mod == HIFC_MOD_L2NIC) + return HIFC_DEV_BUSY_ACTIVE_FW; + else + return -EBUSY; + } + + /* Set channel invalid, don't allowed to send other cmd */ + if (mod == HIFC_MOD_COMM && cmd == HIFC_MGMT_CMD_ACTIVATE_FW) { + hifc_set_mgmt_channel_status(hwdev, true); + /* stop heartbeat enhanced detection temporary, and will + * restart in firmware active event when mgmt is resetted + */ + __set_heartbeat_ehd_detect_delay(hwdev, + HIFC_DEV_ACTIVE_FW_TIMEOUT); + } + + return 0; +} + +static void __pf_to_mgmt_after_handle(struct hifc_hwdev *hwdev, + enum hifc_mod_type mod, u8 cmd, + int sw_status, void *mgmt_status) +{ + /* if activate fw is failed, set channel valid */ + if (mod == HIFC_MOD_COMM && + cmd == HIFC_MGMT_CMD_ACTIVATE_FW) { + if (sw_status) + hifc_set_mgmt_channel_status(hwdev, false); + else + hifc_enable_mgmt_channel(hwdev, mgmt_status); + } +} + +int hifc_pf_msg_to_mgmt_sync(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size, u32 timeout) +{ + int err; + + if (!hwdev) + return -EINVAL; + + if (!((struct hifc_hwdev *)hwdev)->chip_present_flag) + return -EPERM; + + if (!hifc_is_hwdev_mod_inited(hwdev, HIFC_HWDEV_MGMT_INITED)) + return -EPERM; + + if (in_size > HIFC_MSG_TO_MGMT_MAX_LEN) + return -EINVAL; + + err = __pf_to_mgmt_pre_handle(hwdev, mod, cmd); + if (err) + return err; + + err = hifc_pf_to_mgmt_sync(hwdev, mod, cmd, buf_in, in_size, + buf_out, out_size, timeout); + __pf_to_mgmt_after_handle(hwdev, mod, cmd, err, buf_out); + + return err; +} + +static bool is_sfp_info_cmd_cached(struct hifc_hwdev *hwdev, + enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_cmd_get_sfp_qsfp_info *sfp_info; + struct hifc_port_routine_cmd *rt_cmd; + struct card_node *chip_node = hwdev->chip_node; + + sfp_info = buf_in; + if (!chip_node->rt_cmd || sfp_info->port_id >= HIFC_MAX_PORT_ID || + *out_size < sizeof(*sfp_info)) + return false; + + if (sfp_info->version == HIFC_GET_SFP_INFO_REAL_TIME) + return false; + + rt_cmd = &chip_node->rt_cmd[sfp_info->port_id]; + mutex_lock(&chip_node->sfp_mutex); + memcpy(buf_out, &rt_cmd->sfp_info, sizeof(*sfp_info)); + mutex_unlock(&chip_node->sfp_mutex); + + return true; +} + +static bool is_sfp_abs_cmd_cached(struct hifc_hwdev *hwdev, + enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_cmd_get_light_module_abs *abs; + struct hifc_port_routine_cmd *rt_cmd; + struct card_node *chip_node = hwdev->chip_node; + + abs = buf_in; + if (!chip_node->rt_cmd || abs->port_id >= HIFC_MAX_PORT_ID || + *out_size < sizeof(*abs)) + return false; + + if (abs->version == HIFC_GET_SFP_INFO_REAL_TIME) + return false; + + rt_cmd = &chip_node->rt_cmd[abs->port_id]; + mutex_lock(&chip_node->sfp_mutex); + memcpy(buf_out, &rt_cmd->abs, sizeof(*abs)); + mutex_unlock(&chip_node->sfp_mutex); + + return true; +} + +static bool driver_processed_cmd(struct hifc_hwdev *hwdev, + enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct card_node *chip_node = hwdev->chip_node; + + if (mod == HIFC_MOD_L2NIC) { + if (cmd == HIFC_PORT_CMD_GET_SFP_INFO && + chip_node->rt_cmd->up_send_sfp_info) { + return is_sfp_info_cmd_cached(hwdev, mod, cmd, buf_in, + in_size, buf_out, + out_size); + } else if (cmd == HIFC_PORT_CMD_GET_SFP_ABS && + chip_node->rt_cmd->up_send_sfp_abs) { + return is_sfp_abs_cmd_cached(hwdev, mod, cmd, buf_in, + in_size, buf_out, + out_size); + } + } + + return false; +} + +static int send_sync_mgmt_msg(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size, u32 timeout) +{ + unsigned long end; + + end = jiffies + msecs_to_jiffies(HIFC_DEV_ACTIVE_FW_TIMEOUT); + do { + if (!hifc_get_mgmt_channel_status(hwdev) || + !hifc_get_chip_present_flag(hwdev)) + break; + + msleep(1000); + } while (time_before(jiffies, end)); + + if (driver_processed_cmd(hwdev, mod, cmd, buf_in, in_size, buf_out, + out_size)) + return 0; + + return hifc_pf_msg_to_mgmt_sync(hwdev, mod, cmd, buf_in, in_size, + buf_out, out_size, timeout); +} + +int hifc_msg_to_mgmt_sync(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size, u32 timeout) +{ + struct hifc_hwdev *dev = hwdev; + int err; + + if (!hwdev) + return -EINVAL; + + if (!(dev->chip_present_flag)) + return -EPERM; + + err = send_sync_mgmt_msg(hwdev, mod, cmd, buf_in, in_size, + buf_out, out_size, timeout); + + hifc_print_status_info(hwdev, mod, cmd, buf_out); + + return err; +} + +/* PF/VF send msg to uP by api cmd, and return immediately */ +int hifc_msg_to_mgmt_async(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size) +{ + int err; + + if (!hwdev) + return -EINVAL; + + if (!(((struct hifc_hwdev *)hwdev)->chip_present_flag) || + !hifc_is_hwdev_mod_inited(hwdev, HIFC_HWDEV_MGMT_INITED) || + hifc_get_mgmt_channel_status(hwdev)) + return -EPERM; + + if (hifc_func_type(hwdev) == TYPE_VF) { + err = -EFAULT; + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Mailbox don't support async cmd\n"); + } else { + err = hifc_pf_to_mgmt_async(hwdev, mod, cmd, buf_in, in_size); + } + + return err; +} + +int hifc_msg_to_mgmt_no_ack(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size) +{ + struct hifc_hwdev *dev = hwdev; + int err; + + if (!hwdev) + return -EINVAL; + + if (!(dev->chip_present_flag)) + return -EPERM; + + err = hifc_pf_to_mgmt_no_ack(hwdev, mod, cmd, buf_in, in_size); + + return err; +} + +/** + * hifc_cpu_to_be32 - convert data to big endian 32 bit format + * @data: the data to convert + * @len: length of data to convert, must be Multiple of 4B + **/ +void hifc_cpu_to_be32(void *data, int len) +{ + int i, chunk_sz = sizeof(u32); + u32 *mem = data; + + if (!data) + return; + + len = len / chunk_sz; + + for (i = 0; i < len; i++) { + *mem = cpu_to_be32(*mem); + mem++; + } +} + +/** + * hifc_cpu_to_be32 - convert data from big endian 32 bit format + * @data: the data to convert + * @len: length of data to convert + **/ +void hifc_be32_to_cpu(void *data, int len) +{ + int i, chunk_sz = sizeof(u32); + u32 *mem = data; + + if (!data) + return; + + len = len / chunk_sz; + + for (i = 0; i < len; i++) { + *mem = be32_to_cpu(*mem); + mem++; + } +} + +/** + * hifc_set_sge - set dma area in scatter gather entry + * @sge: scatter gather entry + * @addr: dma address + * @len: length of relevant data in the dma address + **/ +void hifc_set_sge(struct hifc_sge *sge, dma_addr_t addr, u32 len) +{ + sge->hi_addr = upper_32_bits(addr); + sge->lo_addr = lower_32_bits(addr); + sge->len = len; +} + +int hifc_set_ci_table(void *hwdev, u16 q_id, struct hifc_sq_attr *attr) +{ + struct hifc_cons_idx_attr cons_idx_attr = {0}; + u16 out_size = sizeof(cons_idx_attr); + int err; + + if (!hwdev || !attr) + return -EINVAL; + + err = hifc_global_func_id_get(hwdev, &cons_idx_attr.func_idx); + if (err) + return err; + + cons_idx_attr.dma_attr_off = attr->dma_attr_off; + cons_idx_attr.pending_limit = attr->pending_limit; + cons_idx_attr.coalescing_time = attr->coalescing_time; + + if (attr->intr_en) { + cons_idx_attr.intr_en = attr->intr_en; + cons_idx_attr.intr_idx = attr->intr_idx; + } + + cons_idx_attr.l2nic_sqn = attr->l2nic_sqn; + cons_idx_attr.sq_id = q_id; + + cons_idx_attr.ci_addr = attr->ci_dma_base; + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_L2NIC_SQ_CI_ATTR_SET, + &cons_idx_attr, sizeof(cons_idx_attr), + &cons_idx_attr, &out_size, 0); + if (err || !out_size || cons_idx_attr.status) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Failed to set ci attribute table, err: %d, status: 0x%x, out_size: 0x%x\n", + err, cons_idx_attr.status, out_size); + return -EFAULT; + } + + return 0; +} + +static int hifc_set_cmdq_depth(struct hifc_hwdev *hwdev, u16 cmdq_depth) +{ + struct hifc_root_ctxt root_ctxt = {0}; + u16 out_size = sizeof(root_ctxt); + int err; + + err = hifc_global_func_id_get(hwdev, &root_ctxt.func_idx); + if (err) + return err; + + root_ctxt.ppf_idx = hifc_ppf_idx(hwdev); + + root_ctxt.set_cmdq_depth = 1; + root_ctxt.cmdq_depth = (u8)ilog2(cmdq_depth); + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_VAT_SET, + &root_ctxt, sizeof(root_ctxt), + &root_ctxt, &out_size, 0); + if (err || !out_size || root_ctxt.status) { + sdk_err(hwdev->dev_hdl, "Failed to set cmdq depth, err: %d, status: 0x%x, out_size: 0x%x\n", + err, root_ctxt.status, out_size); + return -EFAULT; + } + + return 0; +} + +static u16 get_hw_rx_buf_size(int rx_buf_sz) +{ +#define DEFAULT_RX_BUF_SIZE ((u16)0xB) + u16 num_hw_types = + sizeof(hifc_hw_rx_buf_size) / + sizeof(hifc_hw_rx_buf_size[0]); + u16 i; + + for (i = 0; i < num_hw_types; i++) { + if (hifc_hw_rx_buf_size[i] == rx_buf_sz) + return i; + } + + pr_err("Chip can't support rx buf size of %d\n", rx_buf_sz); + + return DEFAULT_RX_BUF_SIZE; +} + +int hifc_set_root_ctxt(void *hwdev, u16 rq_depth, u16 sq_depth, int rx_buf_sz) +{ + struct hifc_root_ctxt root_ctxt = {0}; + u16 out_size = sizeof(root_ctxt); + int err; + + if (!hwdev) + return -EINVAL; + + err = hifc_global_func_id_get(hwdev, &root_ctxt.func_idx); + if (err) + return err; + + root_ctxt.ppf_idx = hifc_ppf_idx(hwdev); + + root_ctxt.set_cmdq_depth = 0; + root_ctxt.cmdq_depth = 0; + + root_ctxt.lro_en = 1; + + root_ctxt.rq_depth = (u16)ilog2(rq_depth); + root_ctxt.rx_buf_sz = get_hw_rx_buf_size(rx_buf_sz); + root_ctxt.sq_depth = (u16)ilog2(sq_depth); + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_VAT_SET, + &root_ctxt, sizeof(root_ctxt), + &root_ctxt, &out_size, 0); + if (err || !out_size || root_ctxt.status) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Failed to set root context, err: %d, status: 0x%x, out_size: 0x%x\n", + err, root_ctxt.status, out_size); + return -EFAULT; + } + + return 0; +} + +int hifc_clean_root_ctxt(void *hwdev) +{ + struct hifc_root_ctxt root_ctxt = {0}; + u16 out_size = sizeof(root_ctxt); + int err; + + if (!hwdev) + return -EINVAL; + + err = hifc_global_func_id_get(hwdev, &root_ctxt.func_idx); + if (err) + return err; + + root_ctxt.ppf_idx = hifc_ppf_idx(hwdev); + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_VAT_SET, + &root_ctxt, sizeof(root_ctxt), + &root_ctxt, &out_size, 0); + if (err || !out_size || root_ctxt.status) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Failed to set root context, err: %d, status: 0x%x, out_size: 0x%x\n", + err, root_ctxt.status, out_size); + return -EFAULT; + } + + return 0; +} + +static int wait_for_flr_finish(struct hifc_hwif *hwif) +{ + u32 cnt = 0; + enum hifc_pf_status status; + + while (cnt < HIFC_FLR_TIMEOUT) { + status = hifc_get_pf_status(hwif); + if (status == HIFC_PF_STATUS_FLR_FINISH_FLAG) { + hifc_set_pf_status(hwif, HIFC_PF_STATUS_ACTIVE_FLAG); + return 0; + } + + usleep_range(9900, 10000); + cnt++; + } + + return -EFAULT; +} + +#define HIFC_WAIT_CMDQ_IDLE_TIMEOUT 5000 + +static int wait_cmdq_stop(struct hifc_hwdev *hwdev) +{ + enum hifc_cmdq_type cmdq_type; + struct hifc_cmdqs *cmdqs = hwdev->cmdqs; + u32 cnt = 0; + int err = 0; + + if (!(cmdqs->status & HIFC_CMDQ_ENABLE)) + return 0; + + cmdqs->status &= ~HIFC_CMDQ_ENABLE; + + while (cnt < HIFC_WAIT_CMDQ_IDLE_TIMEOUT && hwdev->chip_present_flag) { + err = 0; + cmdq_type = HIFC_CMDQ_SYNC; + for (; cmdq_type < HIFC_MAX_CMDQ_TYPES; cmdq_type++) { + if (!hifc_cmdq_idle(&cmdqs->cmdq[cmdq_type])) { + err = -EBUSY; + break; + } + } + + if (!err) + return 0; + + usleep_range(500, 1000); + cnt++; + } + + cmdq_type = HIFC_CMDQ_SYNC; + for (; cmdq_type < HIFC_MAX_CMDQ_TYPES; cmdq_type++) { + if (!hifc_cmdq_idle(&cmdqs->cmdq[cmdq_type])) + sdk_err(hwdev->dev_hdl, "Cmdq %d busy\n", cmdq_type); + } + + cmdqs->status |= HIFC_CMDQ_ENABLE; + + return err; +} + +static int hifc_pf_rx_tx_flush(struct hifc_hwdev *hwdev) +{ + struct hifc_hwif *hwif = hwdev->hwif; + struct hifc_clear_doorbell clear_db = {0}; + struct hifc_clear_resource clr_res = {0}; + u16 out_size, func_id; + int err; + int ret = 0; + + /* wait ucode stop I/O */ + msleep(100); + + err = wait_cmdq_stop(hwdev); + if (err) { + sdk_warn(hwdev->dev_hdl, "CMDQ is still working, please check CMDQ timeout value is reasonable\n"); + ret = err; + } + + hifc_disable_doorbell(hwif); + + out_size = sizeof(clear_db); + func_id = hifc_global_func_id_hw(hwdev); + clear_db.func_idx = func_id; + clear_db.ppf_idx = HIFC_HWIF_PPF_IDX(hwif); + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_FLUSH_DOORBELL, &clear_db, + sizeof(clear_db), &clear_db, &out_size, 0); + if (err || !out_size || clear_db.status) { + sdk_warn(hwdev->dev_hdl, "Failed to flush doorbell, err: %d, status: 0x%x, out_size: 0x%x\n", + err, clear_db.status, out_size); + if (err) + ret = err; + else + ret = -EFAULT; + } + + hifc_set_pf_status(hwif, HIFC_PF_STATUS_FLR_START_FLAG); + + clr_res.func_idx = func_id; + clr_res.ppf_idx = HIFC_HWIF_PPF_IDX(hwif); + + err = hifc_msg_to_mgmt_no_ack(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_START_FLR, &clr_res, + sizeof(clr_res)); + if (err) { + sdk_warn(hwdev->dev_hdl, "Failed to notice flush message\n"); + ret = err; + } + + err = wait_for_flr_finish(hwif); + if (err) { + sdk_warn(hwdev->dev_hdl, "Wait firmware FLR timeout\n"); + ret = err; + } + + hifc_enable_doorbell(hwif); + + err = hifc_reinit_cmdq_ctxts(hwdev); + if (err) { + sdk_warn(hwdev->dev_hdl, "Failed to reinit cmdq\n"); + ret = err; + } + + return ret; +} + +int hifc_func_rx_tx_flush(void *hwdev) +{ + struct hifc_hwdev *dev = hwdev; + + if (!hwdev) + return -EINVAL; + + if (!dev->chip_present_flag) + return 0; + + return hifc_pf_rx_tx_flush(dev); +} + +int hifc_get_interrupt_cfg(void *hwdev, + struct nic_interrupt_info *interrupt_info) +{ + struct hifc_hwdev *nic_hwdev = hwdev; + struct hifc_msix_config msix_cfg = {0}; + u16 out_size = sizeof(msix_cfg); + int err; + + if (!hwdev || !interrupt_info) + return -EINVAL; + + err = hifc_global_func_id_get(hwdev, &msix_cfg.func_id); + if (err) + return err; + + msix_cfg.msix_index = interrupt_info->msix_index; + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_MSI_CTRL_REG_RD_BY_UP, + &msix_cfg, sizeof(msix_cfg), + &msix_cfg, &out_size, 0); + if (err || !out_size || msix_cfg.status) { + sdk_err(nic_hwdev->dev_hdl, "Failed to get interrupt config, err: %d, status: 0x%x, out size: 0x%x\n", + err, msix_cfg.status, out_size); + return -EINVAL; + } + + interrupt_info->lli_credit_limit = msix_cfg.lli_credit_cnt; + interrupt_info->lli_timer_cfg = msix_cfg.lli_tmier_cnt; + interrupt_info->pending_limt = msix_cfg.pending_cnt; + interrupt_info->coalesc_timer_cfg = msix_cfg.coalesct_timer_cnt; + interrupt_info->resend_timer_cfg = msix_cfg.resend_timer_cnt; + + return 0; +} + +int hifc_set_interrupt_cfg(void *hwdev, + struct nic_interrupt_info interrupt_info) +{ + struct hifc_hwdev *nic_hwdev = hwdev; + struct hifc_msix_config msix_cfg = {0}; + struct nic_interrupt_info temp_info; + u16 out_size = sizeof(msix_cfg); + int err; + + if (!hwdev) + return -EINVAL; + + temp_info.msix_index = interrupt_info.msix_index; + + err = hifc_get_interrupt_cfg(hwdev, &temp_info); + if (err) + return -EINVAL; + + err = hifc_global_func_id_get(hwdev, &msix_cfg.func_id); + if (err) + return err; + + msix_cfg.msix_index = (u16)interrupt_info.msix_index; + msix_cfg.lli_credit_cnt = temp_info.lli_credit_limit; + msix_cfg.lli_tmier_cnt = temp_info.lli_timer_cfg; + msix_cfg.pending_cnt = temp_info.pending_limt; + msix_cfg.coalesct_timer_cnt = temp_info.coalesc_timer_cfg; + msix_cfg.resend_timer_cnt = temp_info.resend_timer_cfg; + + if (interrupt_info.lli_set) { + msix_cfg.lli_credit_cnt = interrupt_info.lli_credit_limit; + msix_cfg.lli_tmier_cnt = interrupt_info.lli_timer_cfg; + } + + if (interrupt_info.interrupt_coalesc_set) { + msix_cfg.pending_cnt = interrupt_info.pending_limt; + msix_cfg.coalesct_timer_cnt = interrupt_info.coalesc_timer_cfg; + msix_cfg.resend_timer_cnt = interrupt_info.resend_timer_cfg; + } + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_MSI_CTRL_REG_WR_BY_UP, + &msix_cfg, sizeof(msix_cfg), + &msix_cfg, &out_size, 0); + if (err || !out_size || msix_cfg.status) { + sdk_err(nic_hwdev->dev_hdl, "Failed to set interrupt config, err: %d, status: 0x%x, out size: 0x%x\n", + err, msix_cfg.status, out_size); + return -EINVAL; + } + + return 0; +} + +#define HIFC_MSIX_CNT_RESEND_TIMER_SHIFT 29 +#define HIFC_MSIX_CNT_RESEND_TIMER_MASK 0x7U + +#define HIFC_MSIX_CNT_SET(val, member) \ + (((val) & HIFC_MSIX_CNT_##member##_MASK) << \ + HIFC_MSIX_CNT_##member##_SHIFT) + +void hifc_misx_intr_clear_resend_bit(void *hwdev, u16 msix_idx, + u8 clear_resend_en) +{ + struct hifc_hwif *hwif; + u32 msix_ctrl = 0, addr; + + if (!hwdev) + return; + + hwif = ((struct hifc_hwdev *)hwdev)->hwif; + + msix_ctrl = HIFC_MSIX_CNT_SET(clear_resend_en, RESEND_TIMER); + + addr = HIFC_CSR_MSIX_CNT_ADDR(msix_idx); + + hifc_hwif_write_reg(hwif, addr, msix_ctrl); +} + +static int init_aeqs_msix_attr(struct hifc_hwdev *hwdev) +{ + struct hifc_aeqs *aeqs = hwdev->aeqs; + struct nic_interrupt_info info = {0}; + struct hifc_eq *eq; + u16 q_id; + int err; + + info.lli_set = 0; + info.interrupt_coalesc_set = 1; + info.pending_limt = HIFC_DEAULT_EQ_MSIX_PENDING_LIMIT; + info.coalesc_timer_cfg = HIFC_DEAULT_EQ_MSIX_COALESC_TIMER_CFG; + info.resend_timer_cfg = HIFC_DEAULT_EQ_MSIX_RESEND_TIMER_CFG; + + for (q_id = 0; q_id < aeqs->num_aeqs; q_id++) { + eq = &aeqs->aeq[q_id]; + info.msix_index = eq->eq_irq.msix_entry_idx; + err = hifc_set_interrupt_cfg(hwdev, info); + if (err) { + sdk_err(hwdev->dev_hdl, "Set msix attr for aeq %d failed\n", + q_id); + return -EFAULT; + } + } + + return 0; +} + +static int init_ceqs_msix_attr(struct hifc_hwdev *hwdev) +{ + struct hifc_ceqs *ceqs = hwdev->ceqs; + struct nic_interrupt_info info = {0}; + struct hifc_eq *eq; + u16 q_id; + int err; + + info.lli_set = 0; + info.interrupt_coalesc_set = 1; + info.pending_limt = HIFC_DEAULT_EQ_MSIX_PENDING_LIMIT; + info.coalesc_timer_cfg = HIFC_DEAULT_EQ_MSIX_COALESC_TIMER_CFG; + info.resend_timer_cfg = HIFC_DEAULT_EQ_MSIX_RESEND_TIMER_CFG; + + for (q_id = 0; q_id < ceqs->num_ceqs; q_id++) { + eq = &ceqs->ceq[q_id]; + info.msix_index = eq->eq_irq.msix_entry_idx; + err = hifc_set_interrupt_cfg(hwdev, info); + if (err) { + sdk_err(hwdev->dev_hdl, "Set msix attr for ceq %d failed\n", + q_id); + return -EFAULT; + } + } + + return 0; +} + +/** + * set_pf_dma_attr_entry - set the dma attributes for entry + * @hwdev: the pointer to hw device + * @entry_idx: the entry index in the dma table + * @st: PCIE TLP steering tag + * @at: PCIE TLP AT field + * @ph: PCIE TLP Processing Hint field + * @no_snooping: PCIE TLP No snooping + * @tph_en: PCIE TLP Processing Hint Enable + **/ +static void set_pf_dma_attr_entry(struct hifc_hwdev *hwdev, u32 entry_idx, + u8 st, u8 at, u8 ph, + enum hifc_pcie_nosnoop no_snooping, + enum hifc_pcie_tph tph_en) +{ + u32 addr, val, dma_attr_entry; + + /* Read Modify Write */ + addr = HIFC_CSR_DMA_ATTR_TBL_ADDR(entry_idx); + + val = hifc_hwif_read_reg(hwdev->hwif, addr); + val = HIFC_DMA_ATTR_ENTRY_CLEAR(val, ST) & + HIFC_DMA_ATTR_ENTRY_CLEAR(val, AT) & + HIFC_DMA_ATTR_ENTRY_CLEAR(val, PH) & + HIFC_DMA_ATTR_ENTRY_CLEAR(val, NO_SNOOPING) & + HIFC_DMA_ATTR_ENTRY_CLEAR(val, TPH_EN); + + dma_attr_entry = HIFC_DMA_ATTR_ENTRY_SET(st, ST) | + HIFC_DMA_ATTR_ENTRY_SET(at, AT) | + HIFC_DMA_ATTR_ENTRY_SET(ph, PH) | + HIFC_DMA_ATTR_ENTRY_SET(no_snooping, NO_SNOOPING) | + HIFC_DMA_ATTR_ENTRY_SET(tph_en, TPH_EN); + + val |= dma_attr_entry; + hifc_hwif_write_reg(hwdev->hwif, addr, val); +} + +/** + * dma_attr_table_init - initialize the the default dma attributes + * @hwdev: the pointer to hw device + * Return: 0 - success, negative - failure + **/ +static int dma_attr_table_init(struct hifc_hwdev *hwdev) +{ + int err = 0; + + set_pf_dma_attr_entry(hwdev, PCIE_MSIX_ATTR_ENTRY, + HIFC_PCIE_ST_DISABLE, + HIFC_PCIE_AT_DISABLE, + HIFC_PCIE_PH_DISABLE, + HIFC_PCIE_SNOOP, + HIFC_PCIE_TPH_DISABLE); + + return err; +} + +static int resources_state_set(struct hifc_hwdev *hwdev, + enum hifc_res_state state) +{ + struct hifc_cmd_set_res_state res_state = {0}; + u16 out_size = sizeof(res_state); + int err; + + err = hifc_global_func_id_get(hwdev, &res_state.func_idx); + if (err) + return err; + + res_state.state = state; + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_RES_STATE_SET, + &res_state, sizeof(res_state), + &res_state, &out_size, 0); + if (err || !out_size || res_state.status) { + sdk_err(hwdev->dev_hdl, "Failed to set resources state, err: %d, status: 0x%x, out_size: 0x%x\n", + err, res_state.status, out_size); + return -EFAULT; + } + + return 0; +} + +static void comm_mgmt_msg_handler(void *hwdev, void *pri_handle, u8 cmd, + void *buf_in, u16 in_size, void *buf_out, + u16 *out_size) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt = pri_handle; + u8 cmd_idx; + u32 *mem; + u16 i; + + for (cmd_idx = 0; cmd_idx < pf_to_mgmt->proc.cmd_num; cmd_idx++) { + if (cmd == pf_to_mgmt->proc.info[cmd_idx].cmd) { + if (!pf_to_mgmt->proc.info[cmd_idx].proc) { + sdk_warn(pf_to_mgmt->hwdev->dev_hdl, + "PF recv up comm msg handle null, cmd(0x%x)\n", + cmd); + } else { + pf_to_mgmt->proc.info[cmd_idx].proc(hwdev, + buf_in, in_size, buf_out, out_size); + } + + return; + } + } + + sdk_warn(pf_to_mgmt->hwdev->dev_hdl, "Received mgmt cpu event: 0x%x\n", + cmd); + + mem = buf_in; + for (i = 0; i < (in_size / sizeof(u32)); i++) { + pr_info("0x%x\n", *mem); + mem++; + } + + *out_size = 0; +} + +static int hifc_comm_aeqs_init(struct hifc_hwdev *hwdev) +{ + struct irq_info aeq_irqs[HIFC_MAX_AEQS] = {{0} }; + u16 num_aeqs, resp_num_irq = 0, i; + int err; + + num_aeqs = HIFC_HWIF_NUM_AEQS(hwdev->hwif); + if (num_aeqs > HIFC_MAX_AEQS) { + sdk_warn(hwdev->dev_hdl, "Adjust aeq num to %d\n", + HIFC_MAX_AEQS); + num_aeqs = HIFC_MAX_AEQS; + } + err = hifc_alloc_irqs(hwdev, SERVICE_T_INTF, num_aeqs, aeq_irqs, + &resp_num_irq); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to alloc aeq irqs, num_aeqs: %d\n", + num_aeqs); + return err; + } + + if (resp_num_irq < num_aeqs) { + sdk_warn(hwdev->dev_hdl, "Adjust aeq num to %d\n", + resp_num_irq); + num_aeqs = resp_num_irq; + } + + err = hifc_aeqs_init(hwdev, num_aeqs, aeq_irqs); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init aeqs\n"); + goto aeqs_init_err; + } + + set_bit(HIFC_HWDEV_AEQ_INITED, &hwdev->func_state); + + return 0; + +aeqs_init_err: + for (i = 0; i < num_aeqs; i++) + hifc_free_irq(hwdev, SERVICE_T_INTF, aeq_irqs[i].irq_id); + + return err; +} + +static void hifc_comm_aeqs_free(struct hifc_hwdev *hwdev) +{ + struct irq_info aeq_irqs[HIFC_MAX_AEQS] = {{0} }; + u16 num_irqs, i; + + clear_bit(HIFC_HWDEV_AEQ_INITED, &hwdev->func_state); + + hifc_get_aeq_irqs(hwdev, aeq_irqs, &num_irqs); + hifc_aeqs_free(hwdev); + for (i = 0; i < num_irqs; i++) + hifc_free_irq(hwdev, SERVICE_T_INTF, aeq_irqs[i].irq_id); +} + +static int hifc_comm_ceqs_init(struct hifc_hwdev *hwdev) +{ + struct irq_info ceq_irqs[HIFC_MAX_CEQS] = {{0} }; + u16 num_ceqs, resp_num_irq = 0, i; + int err; + + num_ceqs = HIFC_HWIF_NUM_CEQS(hwdev->hwif); + if (num_ceqs > HIFC_MAX_CEQS) { + sdk_warn(hwdev->dev_hdl, "Adjust ceq num to %d\n", + HIFC_MAX_CEQS); + num_ceqs = HIFC_MAX_CEQS; + } + + err = hifc_alloc_irqs(hwdev, SERVICE_T_INTF, num_ceqs, ceq_irqs, + &resp_num_irq); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to alloc ceq irqs, num_ceqs: %d\n", + num_ceqs); + return err; + } + + if (resp_num_irq < num_ceqs) { + sdk_warn(hwdev->dev_hdl, "Adjust ceq num to %d\n", + resp_num_irq); + num_ceqs = resp_num_irq; + } + + err = hifc_ceqs_init(hwdev, num_ceqs, ceq_irqs); + if (err) { + sdk_err(hwdev->dev_hdl, + "Failed to init ceqs, err:%d\n", err); + goto ceqs_init_err; + } + + return 0; + +ceqs_init_err: + for (i = 0; i < num_ceqs; i++) + hifc_free_irq(hwdev, SERVICE_T_INTF, ceq_irqs[i].irq_id); + + return err; +} + +static void hifc_comm_ceqs_free(struct hifc_hwdev *hwdev) +{ + struct irq_info ceq_irqs[HIFC_MAX_CEQS] = {{0} }; + u16 num_irqs; + int i; + + hifc_get_ceq_irqs(hwdev, ceq_irqs, &num_irqs); + hifc_ceqs_free(hwdev); + for (i = 0; i < num_irqs; i++) + hifc_free_irq(hwdev, SERVICE_T_INTF, ceq_irqs[i].irq_id); +} + +static int hifc_comm_pf_to_mgmt_init(struct hifc_hwdev *hwdev) +{ + int err; + + if (hifc_func_type(hwdev) == TYPE_VF || + !FUNC_SUPPORT_MGMT(hwdev)) + return 0; /* VF do not support send msg to mgmt directly */ + + err = hifc_pf_to_mgmt_init(hwdev); + if (err) + return err; + + hifc_aeq_register_hw_cb(hwdev, HIFC_MSG_FROM_MGMT_CPU, + hifc_mgmt_msg_aeqe_handler); + + hifc_register_mgmt_msg_cb(hwdev, HIFC_MOD_COMM, + hwdev->pf_to_mgmt, comm_mgmt_msg_handler); + + set_bit(HIFC_HWDEV_MGMT_INITED, &hwdev->func_state); + + return 0; +} + +static void hifc_comm_pf_to_mgmt_free(struct hifc_hwdev *hwdev) +{ + if (hifc_func_type(hwdev) == TYPE_VF || + !FUNC_SUPPORT_MGMT(hwdev)) + return; /* VF do not support send msg to mgmt directly */ + + hifc_unregister_mgmt_msg_cb(hwdev, HIFC_MOD_COMM); + + hifc_aeq_unregister_hw_cb(hwdev, HIFC_MSG_FROM_MGMT_CPU); + + hifc_pf_to_mgmt_free(hwdev); +} + +static int hifc_comm_clp_to_mgmt_init(struct hifc_hwdev *hwdev) +{ + int err; + + if (hifc_func_type(hwdev) == TYPE_VF || + !FUNC_SUPPORT_MGMT(hwdev)) + return 0; + + err = hifc_clp_pf_to_mgmt_init(hwdev); + if (err) + return err; + + set_bit(HIFC_HWDEV_CLP_INITED, &hwdev->func_state); + + return 0; +} + +static void hifc_comm_clp_to_mgmt_free(struct hifc_hwdev *hwdev) +{ + if (hifc_func_type(hwdev) == TYPE_VF || + !FUNC_SUPPORT_MGMT(hwdev)) + return; + + clear_bit(HIFC_HWDEV_CLP_INITED, &hwdev->func_state); + hifc_clp_pf_to_mgmt_free(hwdev); +} + +static int hifc_comm_cmdqs_init(struct hifc_hwdev *hwdev) +{ + int err; + + err = hifc_cmdqs_init(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init cmd queues\n"); + return err; + } + + hifc_ceq_register_cb(hwdev, HIFC_CMDQ, hifc_cmdq_ceq_handler); + + err = hifc_set_cmdq_depth(hwdev, HIFC_CMDQ_DEPTH); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to set cmdq depth\n"); + goto set_cmdq_depth_err; + } + + return 0; + +set_cmdq_depth_err: + hifc_cmdqs_free(hwdev); + + return err; +} + +static void hifc_comm_cmdqs_free(struct hifc_hwdev *hwdev) +{ + hifc_ceq_unregister_cb(hwdev, HIFC_CMDQ); + hifc_cmdqs_free(hwdev); +} + +static int hifc_sync_mgmt_func_state(struct hifc_hwdev *hwdev) +{ + int err; + + hifc_set_pf_status(hwdev->hwif, HIFC_PF_STATUS_ACTIVE_FLAG); + + err = resources_state_set(hwdev, HIFC_RES_ACTIVE); + if (err) { + sdk_err(hwdev->dev_hdl, + "Failed to set function resources state\n"); + goto resources_state_set_err; + } + + hwdev->heartbeat_ehd.en = false; + if (HIFC_FUNC_TYPE(hwdev) == TYPE_PPF) { + /* heartbeat synchronize must be after set pf active status */ + hifc_comm_recv_mgmt_self_cmd_reg( + hwdev, HIFC_MGMT_CMD_HEARTBEAT_EVENT, + mgmt_heartbeat_event_handler); + } + + return 0; + +resources_state_set_err: + hifc_set_pf_status(hwdev->hwif, HIFC_PF_STATUS_INIT); + + return err; +} + +static void hifc_unsync_mgmt_func_state(struct hifc_hwdev *hwdev) +{ + hifc_set_pf_status(hwdev->hwif, HIFC_PF_STATUS_INIT); + + hwdev->heartbeat_ehd.en = false; + if (HIFC_FUNC_TYPE(hwdev) == TYPE_PPF) { + hifc_comm_recv_up_self_cmd_unreg( + hwdev, HIFC_MGMT_CMD_HEARTBEAT_EVENT); + } + + resources_state_set(hwdev, HIFC_RES_CLEAN); +} + +int hifc_set_vport_enable(void *hwdev, bool enable) +{ + struct hifc_hwdev *nic_hwdev = (struct hifc_hwdev *)hwdev; + struct hifc_vport_state en_state = {0}; + u16 out_size = sizeof(en_state); + int err; + + if (!hwdev) + return -EINVAL; + + err = hifc_global_func_id_get(hwdev, &en_state.func_id); + if (err) + return err; + + en_state.state = enable ? 1 : 0; + + err = l2nic_msg_to_mgmt_sync(hwdev, HIFC_PORT_CMD_SET_VPORT_ENABLE, + &en_state, sizeof(en_state), + &en_state, &out_size); + if (err || !out_size || en_state.status) { + sdk_err(nic_hwdev->dev_hdl, "Failed to set vport state, err: %d, status: 0x%x, out size: 0x%x\n", + err, en_state.status, out_size); + return -EINVAL; + } + + return 0; +} + +int hifc_l2nic_reset_base(struct hifc_hwdev *hwdev, u16 reset_flag) +{ + struct hifc_l2nic_reset l2nic_reset = {0}; + u16 out_size = sizeof(l2nic_reset); + int err = 0; + + err = hifc_set_vport_enable(hwdev, false); + if (err) + return err; + + msleep(100); + + sdk_info(hwdev->dev_hdl, "L2nic reset flag 0x%x\n", reset_flag); + + err = hifc_global_func_id_get(hwdev, &l2nic_reset.func_id); + if (err) + return err; + + l2nic_reset.reset_flag = reset_flag; + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_L2NIC_RESET, &l2nic_reset, + sizeof(l2nic_reset), &l2nic_reset, + &out_size, 0); + if (err || !out_size || l2nic_reset.status) { + sdk_err(hwdev->dev_hdl, "Failed to reset L2NIC resources, err: %d, status: 0x%x, out_size: 0x%x\n", + err, l2nic_reset.status, out_size); + return -EIO; + } + + return 0; +} + +static int hifc_l2nic_reset(struct hifc_hwdev *hwdev) +{ + return hifc_l2nic_reset_base(hwdev, 0); +} + +static int __get_func_misc_info(struct hifc_hwdev *hwdev) +{ + int err; + + err = hifc_get_board_info(hwdev, &hwdev->board_info); + if (err) { + sdk_err(hwdev->dev_hdl, "Get board info failed\n"); + return err; + } + + return 0; +} + +static int init_func_mode(struct hifc_hwdev *hwdev) +{ + int err; + + err = __get_func_misc_info(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to get function msic information\n"); + return err; + } + + err = hifc_l2nic_reset(hwdev); + if (err) + return err; + + return 0; +} + +static int __init_eqs_msix_attr(struct hifc_hwdev *hwdev) +{ + int err; + + err = init_aeqs_msix_attr(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init aeqs msix attr\n"); + return err; + } + + err = init_ceqs_msix_attr(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init ceqs msix attr\n"); + return err; + } + + return 0; +} + +static int init_cmdqs_channel(struct hifc_hwdev *hwdev) +{ + u16 func_id; + int err; + + dma_attr_table_init(hwdev); + + err = hifc_comm_ceqs_init(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init completion event queues\n"); + return err; + } + + err = __init_eqs_msix_attr(hwdev); + if (err) + goto init_eqs_msix_err; + + /* set default wq page_size */ + hwdev->wq_page_size = HIFC_DEFAULT_WQ_PAGE_SIZE; + + err = hifc_global_func_id_get(hwdev, &func_id); + if (err) + goto get_func_id_err; + + err = hifc_set_wq_page_size(hwdev, func_id, hwdev->wq_page_size); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to set wq page size\n"); + goto init_wq_pg_size_err; + } + + err = hifc_comm_cmdqs_init(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init cmd queues\n"); + goto cmdq_init_err; + } + + set_bit(HIFC_HWDEV_CMDQ_INITED, &hwdev->func_state); + + return 0; + +cmdq_init_err: + if (HIFC_FUNC_TYPE(hwdev) != TYPE_VF) + hifc_set_wq_page_size(hwdev, func_id, HIFC_HW_WQ_PAGE_SIZE); +init_wq_pg_size_err: +get_func_id_err: +init_eqs_msix_err: + hifc_comm_ceqs_free(hwdev); + + return err; +} + +static int init_mgmt_channel(struct hifc_hwdev *hwdev) +{ + int err; + + err = hifc_comm_clp_to_mgmt_init(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init clp\n"); + return err; + } + + err = hifc_comm_aeqs_init(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init async event queues\n"); + goto aeqs_init_err; + } + + err = hifc_comm_pf_to_mgmt_init(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init msg\n"); + goto msg_init_err; + } + + return err; + +msg_init_err: + hifc_comm_aeqs_free(hwdev); + +aeqs_init_err: + hifc_comm_clp_to_mgmt_free(hwdev); + + return err; +} + +/* initialize communication channel */ +int hifc_init_comm_ch(struct hifc_hwdev *hwdev) +{ + int err; + + err = init_mgmt_channel(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init mgmt channel\n"); + return err; + } + + err = init_func_mode(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init function mode\n"); + goto func_mode_err; + } + + err = init_cmdqs_channel(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to init cmdq channel\n"); + goto init_cmdqs_channel_err; + } + + err = hifc_sync_mgmt_func_state(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to synchronize mgmt function state\n"); + goto sync_mgmt_func_err; + } + + err = hifc_aeq_register_swe_cb(hwdev, HIFC_STATELESS_EVENT, + hifc_nic_sw_aeqe_handler); + if (err) { + sdk_err(hwdev->dev_hdl, + "Failed to register ucode aeqe handler\n"); + goto register_ucode_aeqe_err; + } + + set_bit(HIFC_HWDEV_COMM_CH_INITED, &hwdev->func_state); + + return 0; + +register_ucode_aeqe_err: + hifc_unsync_mgmt_func_state(hwdev); +sync_mgmt_func_err: + return err; + +init_cmdqs_channel_err: + +func_mode_err: + return err; +} + +static void __uninit_comm_module(struct hifc_hwdev *hwdev, + enum hifc_hwdev_init_state init_state) +{ + u16 func_id; + + switch (init_state) { + case HIFC_HWDEV_COMM_CH_INITED: + hifc_aeq_unregister_swe_cb(hwdev, + HIFC_STATELESS_EVENT); + hifc_unsync_mgmt_func_state(hwdev); + break; + case HIFC_HWDEV_CMDQ_INITED: + hifc_comm_cmdqs_free(hwdev); + /* VF can set page size of 256K only, any other value + * will return error in pf, pf will set all vf's page + * size to 4K when disable sriov + */ + if (HIFC_FUNC_TYPE(hwdev) != TYPE_VF) { + func_id = hifc_global_func_id_hw(hwdev); + hifc_set_wq_page_size(hwdev, func_id, + HIFC_HW_WQ_PAGE_SIZE); + } + + hifc_comm_ceqs_free(hwdev); + + break; + case HIFC_HWDEV_MBOX_INITED: + break; + case HIFC_HWDEV_MGMT_INITED: + hifc_comm_pf_to_mgmt_free(hwdev); + break; + case HIFC_HWDEV_AEQ_INITED: + hifc_comm_aeqs_free(hwdev); + break; + case HIFC_HWDEV_CLP_INITED: + hifc_comm_clp_to_mgmt_free(hwdev); + break; + default: + break; + } +} + +#define HIFC_FUNC_STATE_BUSY_TIMEOUT 300 +void hifc_uninit_comm_ch(struct hifc_hwdev *hwdev) +{ + enum hifc_hwdev_init_state init_state = HIFC_HWDEV_COMM_CH_INITED; + int cnt; + + while (init_state > HIFC_HWDEV_NONE_INITED) { + if (!test_bit(init_state, &hwdev->func_state)) { + init_state--; + continue; + } + clear_bit(init_state, &hwdev->func_state); + + cnt = 0; + while (test_bit(HIFC_HWDEV_STATE_BUSY, &hwdev->func_state) && + cnt++ <= HIFC_FUNC_STATE_BUSY_TIMEOUT) + usleep_range(900, 1000); + + __uninit_comm_module(hwdev, init_state); + + init_state--; + } +} + +int hifc_slq_init(void *dev, int num_wqs) +{ + struct hifc_hwdev *hwdev = dev; + int err; + + if (!dev) + return -EINVAL; + + hwdev->wqs = kzalloc(sizeof(*hwdev->wqs), GFP_KERNEL); + if (!hwdev->wqs) + return -ENOMEM; + + err = hifc_wqs_alloc(hwdev->wqs, num_wqs, hwdev->dev_hdl); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to alloc wqs\n"); + kfree(hwdev->wqs); + hwdev->wqs = NULL; + } + + return err; +} + +void hifc_slq_uninit(void *dev) +{ + struct hifc_hwdev *hwdev = dev; + + if (!hwdev) + return; + + hifc_wqs_free(hwdev->wqs); + + kfree(hwdev->wqs); +} + +int hifc_slq_alloc(void *dev, u16 wqebb_size, u16 q_depth, u16 page_size, + u64 *cla_addr, void **handle) +{ + struct hifc_hwdev *hwdev = dev; + struct hifc_wq *wq; + int err; + + if (!dev || !cla_addr || !handle) + return -EINVAL; + + wq = kzalloc(sizeof(*wq), GFP_KERNEL); + if (!wq) + return -ENOMEM; + + err = hifc_wq_allocate(hwdev->wqs, wq, wqebb_size, hwdev->wq_page_size, + q_depth, 0); + if (err) { + sdk_err(hwdev->dev_hdl, "Failed to alloc wq\n"); + kfree(wq); + return -EFAULT; + } + + *cla_addr = wq->block_paddr; + *handle = wq; + + return 0; +} + +void hifc_slq_free(void *dev, void *handle) +{ + struct hifc_hwdev *hwdev = dev; + + if (!hwdev || !handle) + return; + + hifc_wq_free(hwdev->wqs, handle); + kfree(handle); +} + +u64 hifc_slq_get_addr(void *handle, u16 index) +{ + if (!handle) + return 0; /* NULL of wqe addr */ + + return (u64)hifc_get_wqebb_addr(handle, index); +} + +u64 hifc_slq_get_first_pageaddr(void *handle) +{ + struct hifc_wq *wq = handle; + + if (!handle) + return 0; /* NULL of wqe addr */ + + return hifc_get_first_wqe_page_addr(wq); +} + +int hifc_func_tmr_bitmap_set(void *hwdev, bool en) +{ + struct hifc_func_tmr_bitmap_op bitmap_op = {0}; + u16 out_size = sizeof(bitmap_op); + int err; + + if (!hwdev) + return -EINVAL; + + err = hifc_global_func_id_get(hwdev, &bitmap_op.func_idx); + if (err) + return err; + + bitmap_op.ppf_idx = hifc_ppf_idx(hwdev); + if (en) + bitmap_op.op_id = FUNC_TMR_BITMAP_ENABLE; + else + bitmap_op.op_id = FUNC_TMR_BITMAP_DISABLE; + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_FUNC_TMR_BITMAT_SET, + &bitmap_op, sizeof(bitmap_op), + &bitmap_op, &out_size, 0); + if (err || !out_size || bitmap_op.status) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Failed to set timer bitmap, err: %d, status: 0x%x, out_size: 0x%x\n", + err, bitmap_op.status, out_size); + return -EFAULT; + } + + return 0; +} + +int ppf_ht_gpa_set(struct hifc_hwdev *hwdev, struct hifc_page_addr *pg0, + struct hifc_page_addr *pg1) +{ + struct comm_info_ht_gpa_set ht_gpa_set = {0}; + u16 out_size = sizeof(ht_gpa_set); + int ret; + + pg0->virt_addr = dma_zalloc_coherent(hwdev->dev_hdl, + HIFC_HT_GPA_PAGE_SIZE, + &pg0->phys_addr, GFP_KERNEL); + if (!pg0->virt_addr) { + sdk_err(hwdev->dev_hdl, "Alloc pg0 page addr failed\n"); + return -EFAULT; + } + + pg1->virt_addr = dma_zalloc_coherent(hwdev->dev_hdl, + HIFC_HT_GPA_PAGE_SIZE, + &pg1->phys_addr, GFP_KERNEL); + if (!pg1->virt_addr) { + sdk_err(hwdev->dev_hdl, "Alloc pg1 page addr failed\n"); + return -EFAULT; + } + + ht_gpa_set.page_pa0 = pg0->phys_addr; + ht_gpa_set.page_pa1 = pg1->phys_addr; + sdk_info(hwdev->dev_hdl, "PPF ht gpa set: page_addr0.pa=0x%llx, page_addr1.pa=0x%llx\n", + pg0->phys_addr, pg1->phys_addr); + ret = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_PPF_HT_GPA_SET, + &ht_gpa_set, sizeof(ht_gpa_set), + &ht_gpa_set, &out_size, 0); + if (ret || !out_size || ht_gpa_set.status) { + sdk_warn(hwdev->dev_hdl, "PPF ht gpa set failed, ret: %d, status: 0x%x, out_size: 0x%x\n", + ret, ht_gpa_set.status, out_size); + return -EFAULT; + } + + hwdev->page_pa0.phys_addr = pg0->phys_addr; + hwdev->page_pa0.virt_addr = pg0->virt_addr; + + hwdev->page_pa1.phys_addr = pg1->phys_addr; + hwdev->page_pa1.virt_addr = pg1->virt_addr; + + return 0; +} + +int hifc_ppf_ht_gpa_init(struct hifc_hwdev *hwdev) +{ + int ret; + int i; + int j; + int size; + + struct hifc_page_addr page_addr0[HIFC_PPF_HT_GPA_SET_RETRY_TIMES]; + struct hifc_page_addr page_addr1[HIFC_PPF_HT_GPA_SET_RETRY_TIMES]; + + size = HIFC_PPF_HT_GPA_SET_RETRY_TIMES * sizeof(page_addr0[0]); + memset(page_addr0, 0, size); + memset(page_addr1, 0, size); + + for (i = 0; i < HIFC_PPF_HT_GPA_SET_RETRY_TIMES; i++) { + ret = ppf_ht_gpa_set(hwdev, &page_addr0[i], &page_addr1[i]); + if (!ret) + break; + } + + for (j = 0; j < i; j++) { + if (page_addr0[j].virt_addr) { + dma_free_coherent(hwdev->dev_hdl, + HIFC_HT_GPA_PAGE_SIZE, + page_addr0[j].virt_addr, + page_addr0[j].phys_addr); + page_addr0[j].virt_addr = NULL; + } + if (page_addr1[j].virt_addr) { + dma_free_coherent(hwdev->dev_hdl, + HIFC_HT_GPA_PAGE_SIZE, + page_addr1[j].virt_addr, + page_addr1[j].phys_addr); + page_addr1[j].virt_addr = NULL; + } + } + + if (i >= HIFC_PPF_HT_GPA_SET_RETRY_TIMES) { + sdk_err(hwdev->dev_hdl, "PPF ht gpa init failed, retry times: %d\n", + i); + return -EFAULT; + } + + return 0; +} + +void hifc_ppf_ht_gpa_deinit(struct hifc_hwdev *hwdev) +{ + if (hwdev->page_pa0.virt_addr) { + dma_free_coherent(hwdev->dev_hdl, HIFC_HT_GPA_PAGE_SIZE, + hwdev->page_pa0.virt_addr, + hwdev->page_pa0.phys_addr); + hwdev->page_pa0.virt_addr = NULL; + } + + if (hwdev->page_pa1.virt_addr) { + dma_free_coherent(hwdev->dev_hdl, HIFC_HT_GPA_PAGE_SIZE, + hwdev->page_pa1.virt_addr, + hwdev->page_pa1.phys_addr); + hwdev->page_pa1.virt_addr = NULL; + } +} + +static int set_ppf_tmr_status(struct hifc_hwdev *hwdev, + enum ppf_tmr_status status) +{ + struct hifc_ppf_tmr_op op = {0}; + u16 out_size = sizeof(op); + int err = 0; + + if (!hwdev) + return -EINVAL; + + if (hifc_func_type(hwdev) != TYPE_PPF) + return -EFAULT; + + if (status == HIFC_PPF_TMR_FLAG_START) { + err = hifc_ppf_ht_gpa_init(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "PPF ht gpa init fail!\n"); + return -EFAULT; + } + } else { + hifc_ppf_ht_gpa_deinit(hwdev); + } + + op.op_id = status; + op.ppf_idx = hifc_ppf_idx(hwdev); + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_PPF_TMR_SET, &op, + sizeof(op), &op, &out_size, 0); + if (err || !out_size || op.status) { + sdk_err(hwdev->dev_hdl, "Failed to set ppf timer, err: %d, status: 0x%x, out_size: 0x%x\n", + err, op.status, out_size); + return -EFAULT; + } + + return 0; +} + +int hifc_ppf_tmr_start(void *hwdev) +{ + if (!hwdev) { + pr_err("Hwdev pointer is NULL for starting ppf timer\n"); + return -EINVAL; + } + + return set_ppf_tmr_status(hwdev, HIFC_PPF_TMR_FLAG_START); +} + +int hifc_ppf_tmr_stop(void *hwdev) +{ + if (!hwdev) { + pr_err("Hwdev pointer is NULL for stop ppf timer\n"); + return -EINVAL; + } + + return set_ppf_tmr_status(hwdev, HIFC_PPF_TMR_FLAG_STOP); +} + +int hifc_set_wq_page_size(struct hifc_hwdev *hwdev, u16 func_idx, + u32 page_size) +{ + struct hifc_wq_page_size page_size_info = {0}; + u16 out_size = sizeof(page_size_info); + int err; + + page_size_info.func_idx = func_idx; + page_size_info.ppf_idx = hifc_ppf_idx(hwdev); + page_size_info.page_size = HIFC_PAGE_SIZE_HW(page_size); + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_PAGESIZE_SET, + &page_size_info, sizeof(page_size_info), + &page_size_info, &out_size, 0); + if (err || !out_size || page_size_info.status) { + sdk_err(hwdev->dev_hdl, "Failed to set wq page size, err: %d, status: 0x%x, out_size: 0x%0x\n", + err, page_size_info.status, out_size); + return -EFAULT; + } + + return 0; +} + +bool hifc_mgmt_event_ack_first(u8 mod, u8 cmd) +{ + if ((mod == HIFC_MOD_COMM && cmd == HIFC_MGMT_CMD_GET_HOST_INFO) || + (mod == HIFC_MOD_COMM && cmd == HIFC_MGMT_CMD_HEARTBEAT_EVENT)) + return false; + + if (mod == HIFC_MOD_COMM || mod == HIFC_MOD_L2NIC || + mod == HIFC_MOD_HILINK) + return true; + + return false; +} + +#define FAULT_SHOW_STR_LEN 16 + +static void chip_fault_show(struct hifc_hwdev *hwdev, + struct hifc_fault_event *event) +{ + char fault_level[FAULT_LEVEL_MAX][FAULT_SHOW_STR_LEN + 1] = { + "fatal", "reset", "flr", "general", "suggestion"}; + char level_str[FAULT_SHOW_STR_LEN + 1]; + struct hifc_fault_event_stats *fault; + u8 node_id, level; + u32 pos, base; + + fault = &hwdev->hw_stats.fault_event_stats; + + memset(level_str, 0, FAULT_SHOW_STR_LEN + 1); + level = event->event.chip.err_level; + if (level < FAULT_LEVEL_MAX) + strncpy(level_str, fault_level[level], + FAULT_SHOW_STR_LEN); + else + strncpy(level_str, "Unknown", FAULT_SHOW_STR_LEN); + + if (level == FAULT_LEVEL_SERIOUS_FLR) { + sdk_err(hwdev->dev_hdl, "err_level: %d [%s], flr func_id: %d\n", + level, level_str, event->event.chip.func_id); + atomic_inc(&fault->fault_type_stat[event->type]); + } + sdk_err(hwdev->dev_hdl, "module_id: 0x%x, err_type: 0x%x, err_level: %d[%s], err_csr_addr: 0x%08x, err_csr_value: 0x%08x\n", + event->event.chip.node_id, + event->event.chip.err_type, level, level_str, + event->event.chip.err_csr_addr, + event->event.chip.err_csr_value); + + node_id = event->event.chip.node_id; + atomic_inc(&fault->chip_fault_stats[node_id][level]); + + base = event->event.chip.node_id * FAULT_LEVEL_MAX * + HIFC_CHIP_ERROR_TYPE_MAX; + pos = base + HIFC_CHIP_ERROR_TYPE_MAX * level + + event->event.chip.err_type; + if (pos < HIFC_CHIP_FAULT_SIZE) + hwdev->chip_fault_stats[pos]++; +} + +static void fault_report_show(struct hifc_hwdev *hwdev, + struct hifc_fault_event *event) +{ + char fault_type[FAULT_TYPE_MAX][FAULT_SHOW_STR_LEN + 1] = { + "chip", "ucode", "mem rd timeout", "mem wr timeout", + "reg rd timeout", "reg wr timeout", "phy fault"}; + char type_str[FAULT_SHOW_STR_LEN + 1]; + struct hifc_fault_event_stats *fault; + + sdk_err(hwdev->dev_hdl, "Fault event report received, func_id: %d.\n", + hifc_global_func_id(hwdev)); + + memset(type_str, 0, FAULT_SHOW_STR_LEN + 1); + if (event->type < FAULT_TYPE_MAX) + strncpy(type_str, fault_type[event->type], FAULT_SHOW_STR_LEN); + else + strncpy(type_str, "Unknown", FAULT_SHOW_STR_LEN); + + sdk_err(hwdev->dev_hdl, "Fault type: %d [%s]\n", event->type, type_str); + sdk_err(hwdev->dev_hdl, "Fault val[0]: 0x%08x, val[1]: 0x%08x, val[2]: 0x%08x, val[3]: 0x%08x\n", + event->event.val[0], event->event.val[1], event->event.val[2], + event->event.val[3]); + + fault = &hwdev->hw_stats.fault_event_stats; + + switch (event->type) { + case FAULT_TYPE_CHIP: + chip_fault_show(hwdev, event); + break; + case FAULT_TYPE_UCODE: + atomic_inc(&fault->fault_type_stat[event->type]); + + sdk_err(hwdev->dev_hdl, "cause_id: %d, core_id: %d, c_id: %d, epc: 0x%08x\n", + event->event.ucode.cause_id, event->event.ucode.core_id, + event->event.ucode.c_id, event->event.ucode.epc); + break; + case FAULT_TYPE_MEM_RD_TIMEOUT: + case FAULT_TYPE_MEM_WR_TIMEOUT: + atomic_inc(&fault->fault_type_stat[event->type]); + + sdk_err(hwdev->dev_hdl, "err_csr_ctrl: 0x%08x, err_csr_data: 0x%08x, ctrl_tab: 0x%08x, mem_index: 0x%08x\n", + event->event.mem_timeout.err_csr_ctrl, + event->event.mem_timeout.err_csr_data, + event->event.mem_timeout.ctrl_tab, + event->event.mem_timeout.mem_index); + break; + case FAULT_TYPE_REG_RD_TIMEOUT: + case FAULT_TYPE_REG_WR_TIMEOUT: + atomic_inc(&fault->fault_type_stat[event->type]); + sdk_err(hwdev->dev_hdl, "err_csr: 0x%08x\n", + event->event.reg_timeout.err_csr); + break; + case FAULT_TYPE_PHY_FAULT: + atomic_inc(&fault->fault_type_stat[event->type]); + sdk_err(hwdev->dev_hdl, "op_type: %u, port_id: %u, dev_ad: %u, csr_addr: 0x%08x, op_data: 0x%08x\n", + event->event.phy_fault.op_type, + event->event.phy_fault.port_id, + event->event.phy_fault.dev_ad, + event->event.phy_fault.csr_addr, + event->event.phy_fault.op_data); + break; + default: + break; + } +} + +static void hifc_refresh_history_fault(struct hifc_hwdev *hwdev, + struct hifc_fault_recover_info *info) +{ + if (!hwdev->history_fault_flag) { + hwdev->history_fault_flag = true; + memcpy(&hwdev->history_fault, info, + sizeof(struct hifc_fault_recover_info)); + } else { + if (hwdev->history_fault.fault_lev >= info->fault_lev) + memcpy(&hwdev->history_fault, info, + sizeof(struct hifc_fault_recover_info)); + } +} + +static void fault_event_handler(struct hifc_hwdev *hwdev, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size) +{ + struct hifc_cmd_fault_event *fault_event; + struct hifc_event_info event_info; + struct hifc_fault_info_node *fault_node; + + if (in_size != sizeof(*fault_event)) { + sdk_err(hwdev->dev_hdl, "Invalid fault event report, length: %d, should be %ld.\n", + in_size, sizeof(*fault_event)); + return; + } + + fault_event = buf_in; + fault_report_show(hwdev, &fault_event->event); + + if (hwdev->event_callback) { + event_info.type = HIFC_EVENT_FAULT; + memcpy(&event_info.info, &fault_event->event, + sizeof(event_info.info)); + + hwdev->event_callback(hwdev->event_pri_handle, &event_info); + } + + /* refresh history fault info */ + fault_node = kzalloc(sizeof(*fault_node), GFP_KERNEL); + if (!fault_node) { + sdk_err(hwdev->dev_hdl, "Malloc fault node memory failed\n"); + return; + } + + if (fault_event->event.type <= FAULT_TYPE_REG_WR_TIMEOUT) + fault_node->info.fault_src = fault_event->event.type; + else if (fault_event->event.type == FAULT_TYPE_PHY_FAULT) + fault_node->info.fault_src = HIFC_FAULT_SRC_HW_PHY_FAULT; + + if (fault_node->info.fault_src == HIFC_FAULT_SRC_HW_MGMT_CHIP) + fault_node->info.fault_lev = + fault_event->event.event.chip.err_level; + else + fault_node->info.fault_lev = FAULT_LEVEL_FATAL; + + memcpy(&fault_node->info.fault_data.hw_mgmt, &fault_event->event.event, + sizeof(union hifc_fault_hw_mgmt)); + hifc_refresh_history_fault(hwdev, &fault_node->info); + + down(&hwdev->fault_list_sem); + kfree(fault_node); + up(&hwdev->fault_list_sem); +} + +static void heartbeat_lost_event_handler(struct hifc_hwdev *hwdev, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_fault_info_node *fault_node; + struct hifc_event_info event_info = {0}; + + atomic_inc(&hwdev->hw_stats.heart_lost_stats); + sdk_err(hwdev->dev_hdl, "Heart lost report received, func_id: %d\n", + hifc_global_func_id(hwdev)); + + if (hwdev->event_callback) { + event_info.type = HIFC_EVENT_HEART_LOST; + hwdev->event_callback(hwdev->event_pri_handle, &event_info); + } + + /* refresh history fault info */ + fault_node = kzalloc(sizeof(*fault_node), GFP_KERNEL); + if (!fault_node) { + sdk_err(hwdev->dev_hdl, "Malloc fault node memory failed\n"); + return; + } + + fault_node->info.fault_src = HIFC_FAULT_SRC_HOST_HEARTBEAT_LOST; + fault_node->info.fault_lev = FAULT_LEVEL_FATAL; + hifc_refresh_history_fault(hwdev, &fault_node->info); + + down(&hwdev->fault_list_sem); + kfree(fault_node); + up(&hwdev->fault_list_sem); +} + +static void sw_watchdog_timeout_info_show(struct hifc_hwdev *hwdev, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_mgmt_watchdog_info *watchdog_info; + u32 *dump_addr, *reg, stack_len, i, j; + + if (in_size != sizeof(*watchdog_info)) { + sdk_err(hwdev->dev_hdl, "Invalid mgmt watchdog report, length: %d, should be %ld.\n", + in_size, sizeof(*watchdog_info)); + return; + } + + watchdog_info = buf_in; + + sdk_err(hwdev->dev_hdl, "Mgmt deadloop time: 0x%x 0x%x, task id: 0x%x, sp: 0x%x\n", + watchdog_info->curr_time_h, watchdog_info->curr_time_l, + watchdog_info->task_id, watchdog_info->sp); + sdk_err(hwdev->dev_hdl, "Stack current used: 0x%x, peak used: 0x%x, overflow flag: 0x%x, top: 0x%x, bottom: 0x%x\n", + watchdog_info->curr_used, watchdog_info->peak_used, + watchdog_info->is_overflow, watchdog_info->stack_top, + watchdog_info->stack_bottom); + + sdk_err(hwdev->dev_hdl, "Mgmt pc: 0x%08x, lr: 0x%08x, cpsr:0x%08x\n", + watchdog_info->pc, watchdog_info->lr, watchdog_info->cpsr); + + sdk_err(hwdev->dev_hdl, "Mgmt register info\n"); + + for (i = 0; i < 3; i++) { + reg = watchdog_info->reg + (u64)(u32)(4 * i); + sdk_err(hwdev->dev_hdl, "0x%08x 0x%08x 0x%08x 0x%08x\n", + *(reg), *(reg + 1), *(reg + 2), *(reg + 3)); + } + + sdk_err(hwdev->dev_hdl, "0x%08x\n", watchdog_info->reg[12]); + + if (watchdog_info->stack_actlen <= 1024) { + stack_len = watchdog_info->stack_actlen; + } else { + sdk_err(hwdev->dev_hdl, "Oops stack length: 0x%x is wrong\n", + watchdog_info->stack_actlen); + stack_len = 1024; + } + + sdk_err(hwdev->dev_hdl, "Mgmt dump stack, 16Bytes per line(start from sp)\n"); + for (i = 0; i < (stack_len / 16); i++) { + dump_addr = (u32 *)(watchdog_info->data + ((u64)(u32)(i * 16))); + sdk_err(hwdev->dev_hdl, "0x%08x 0x%08x 0x%08x 0x%08x\n", + *dump_addr, *(dump_addr + 1), *(dump_addr + 2), + *(dump_addr + 3)); + } + + for (j = 0; j < ((stack_len % 16) / 4); j++) { + dump_addr = (u32 *)(watchdog_info->data + + ((u64)(u32)(i * 16 + j * 4))); + sdk_err(hwdev->dev_hdl, "0x%08x ", *dump_addr); + } + + *out_size = sizeof(*watchdog_info); + watchdog_info = buf_out; + watchdog_info->status = 0; +} + +static void mgmt_watchdog_timeout_event_handler(struct hifc_hwdev *hwdev, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_fault_info_node *fault_node; + + sw_watchdog_timeout_info_show(hwdev, buf_in, in_size, + buf_out, out_size); + + /* refresh history fault info */ + fault_node = kzalloc(sizeof(*fault_node), GFP_KERNEL); + if (!fault_node) { + sdk_err(hwdev->dev_hdl, "Malloc fault node memory failed\n"); + return; + } + + fault_node->info.fault_src = HIFC_FAULT_SRC_MGMT_WATCHDOG; + fault_node->info.fault_lev = FAULT_LEVEL_FATAL; + hifc_refresh_history_fault(hwdev, &fault_node->info); + + down(&hwdev->fault_list_sem); + kfree(fault_node); + up(&hwdev->fault_list_sem); +} + +static void mgmt_reset_event_handler(struct hifc_hwdev *hwdev, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + sdk_info(hwdev->dev_hdl, "Mgmt is reset\n"); + + /* mgmt reset only occurred when hot update or Mgmt deadloop, + * if Mgmt deadloop, mgmt will report an event with + * mod=0, cmd=0x56, and will reported fault to os, + * so mgmt reset event don't need to report fault + */ +} + +static void hifc_fmw_act_ntc_handler(struct hifc_hwdev *hwdev, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_event_info event_info = {0}; + struct hifc_fmw_act_ntc *notice_info; + + if (in_size != sizeof(*notice_info)) { + sdk_err(hwdev->dev_hdl, "Invalid mgmt firmware active notice, length: %d, should be %ld.\n", + in_size, sizeof(*notice_info)); + return; + } + + /* mgmt is active now, restart heartbeat enhanced detection */ + __set_heartbeat_ehd_detect_delay(hwdev, 0); + + if (!hwdev->event_callback) + return; + + event_info.type = HIFC_EVENT_FMW_ACT_NTC; + hwdev->event_callback(hwdev->event_pri_handle, &event_info); + + *out_size = sizeof(*notice_info); + notice_info = buf_out; + notice_info->status = 0; +} + +static void hifc_pcie_dfx_event_handler(struct hifc_hwdev *hwdev, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_pcie_dfx_ntc *notice_info = buf_in; + struct hifc_pcie_dfx_info *dfx_info; + u16 size = 0; + u16 cnt = 0; + u32 num = 0; + u32 i, j; + int err; + u32 *reg; + + if (in_size != sizeof(*notice_info)) { + sdk_err(hwdev->dev_hdl, "Invalid mgmt firmware active notice, length: %d, should be %ld.\n", + in_size, sizeof(*notice_info)); + return; + } + + dfx_info = kzalloc(sizeof(*dfx_info), GFP_KERNEL); + if (!dfx_info) { + sdk_err(hwdev->dev_hdl, "Malloc dfx_info memory failed\n"); + return; + } + + ((struct hifc_pcie_dfx_ntc *)buf_out)->status = 0; + *out_size = sizeof(*notice_info); + num = (u32)(notice_info->len / 1024); + sdk_info(hwdev->dev_hdl, "INFO LEN: %d\n", notice_info->len); + sdk_info(hwdev->dev_hdl, "PCIE DFX:\n"); + dfx_info->host_id = 0; + for (i = 0; i < num; i++) { + dfx_info->offset = i * MAX_PCIE_DFX_BUF_SIZE; + if (i == (num - 1)) + dfx_info->last = 1; + size = sizeof(*dfx_info); + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_PCIE_DFX_GET, + dfx_info, sizeof(*dfx_info), + dfx_info, &size, 0); + if (err || dfx_info->status || !size) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Failed to get pcie dfx info, err: %d, status: 0x%x, out size: 0x%x\n", + err, dfx_info->status, size); + kfree(dfx_info); + return; + } + + reg = (u32 *)dfx_info->data; + for (j = 0; j < 256; j = j + 8) { + /*lint -save -e661 -e662*/ + sdk_info(hwdev->dev_hdl, "0x%04x: 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x\n", + cnt, reg[j], reg[(u32)(j + 1)], + reg[(u32)(j + 2)], reg[(u32)(j + 3)], + reg[(u32)(j + 4)], reg[(u32)(j + 5)], + reg[(u32)(j + 6)], reg[(u32)(j + 7)]); + /*lint -restore*/ + cnt = cnt + 32; + } + memset(dfx_info->data, 0, MAX_PCIE_DFX_BUF_SIZE); + } + kfree(dfx_info); +} + +struct hifc_mctp_get_host_info { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 huawei_cmd; + u8 sub_cmd; + u8 rsvd[2]; + + u32 actual_len; + + u8 data[1024]; +}; + +static void hifc_mctp_get_host_info_event_handler(struct hifc_hwdev *hwdev, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_event_info event_info = {0}; + struct hifc_mctp_get_host_info *mctp_out, *mctp_in; + struct hifc_mctp_host_info *host_info; + + if (in_size != sizeof(*mctp_in)) { + sdk_err(hwdev->dev_hdl, "Invalid mgmt mctp info, length: %d, should be %ld\n", + in_size, sizeof(*mctp_in)); + return; + } + + *out_size = sizeof(*mctp_out); + mctp_out = buf_out; + mctp_out->status = 0; + + if (!hwdev->event_callback) { + mctp_out->status = HIFC_MGMT_STATUS_ERR_INIT; + return; + } + + mctp_in = buf_in; + host_info = &event_info.mctp_info; + host_info->major_cmd = mctp_in->huawei_cmd; + host_info->sub_cmd = mctp_in->sub_cmd; + host_info->data = mctp_out->data; + + event_info.type = HIFC_EVENT_MCTP_GET_HOST_INFO; + hwdev->event_callback(hwdev->event_pri_handle, &event_info); + + mctp_out->actual_len = host_info->data_len; +} + +char *__hw_to_char_fec[HILINK_FEC_MAX_TYPE] = {"RS-FEC", "BASE-FEC", "NO-FEC"}; + +char *__hw_to_char_port_type[LINK_PORT_MAX_TYPE] = { + "Unknown", "Fibre", "Electric", "Direct Attach Copper", "AOC", + "Back plane", "BaseT" +}; + +static void __get_port_type(struct hifc_hwdev *hwdev, + struct hifc_link_info *info, char **port_type) +{ + if (info->cable_absent) { + sdk_info(hwdev->dev_hdl, "Cable unpresent\n"); + return; + } + + if (info->port_type < LINK_PORT_MAX_TYPE) + *port_type = __hw_to_char_port_type[info->port_type]; + else + sdk_info(hwdev->dev_hdl, "Unknown port type: %u\n", + info->port_type); + if (info->port_type == LINK_PORT_FIBRE) { + if (info->port_sub_type == FIBRE_SUBTYPE_SR) + *port_type = "Fibre-SR"; + else if (info->port_sub_type == FIBRE_SUBTYPE_LR) + *port_type = "Fibre-LR"; + } +} + +static void __print_cable_info(struct hifc_hwdev *hwdev, + struct hifc_link_info *info) +{ + char tmp_str[512] = {0}; + char tmp_vendor[17] = {0}; + char *port_type = "Unknown port type"; + int i; + + __get_port_type(hwdev, info, &port_type); + + for (i = sizeof(info->vendor_name) - 1; i >= 0; i--) { + if (info->vendor_name[i] == ' ') + info->vendor_name[i] = '\0'; + else + break; + } + + memcpy(tmp_vendor, info->vendor_name, + sizeof(info->vendor_name)); + snprintf(tmp_str, sizeof(tmp_str) - 1, + "Vendor: %s, %s, length: %um, max_speed: %uGbps", + tmp_vendor, port_type, info->cable_length, + info->cable_max_speed); + if (info->port_type == LINK_PORT_FIBRE || + info->port_type == LINK_PORT_AOC) { + snprintf(tmp_str, sizeof(tmp_str) - 1, + "%s, %s, Temperature: %u", tmp_str, + info->sfp_type ? "SFP" : "QSFP", info->cable_temp); + if (info->sfp_type) { + snprintf(tmp_str, sizeof(tmp_str) - 1, + "%s, rx power: %uuW, tx power: %uuW", + tmp_str, info->power[0], info->power[1]); + } else { + snprintf(tmp_str, sizeof(tmp_str) - 1, + "%s, rx power: %uuw %uuW %uuW %uuW", + tmp_str, info->power[0], info->power[1], + info->power[2], info->power[3]); + } + } + + sdk_info(hwdev->dev_hdl, "Cable information: %s\n", + tmp_str); +} + +static void __hi30_lane_info(struct hifc_hwdev *hwdev, + struct hilink_lane *lane) +{ + struct hi30_ffe_data *ffe_data; + struct hi30_ctle_data *ctle_data; + + ffe_data = (struct hi30_ffe_data *)lane->hi30_ffe; + ctle_data = (struct hi30_ctle_data *)lane->hi30_ctle; + + sdk_info(hwdev->dev_hdl, "TX_FFE: PRE1=%s%d; PRE2=%s%d; MAIN=%d; POST1=%s%d; POST1X=%s%d\n", + (ffe_data->PRE1 & 0x10) ? "-" : "", + (int)(ffe_data->PRE1 & 0xf), + (ffe_data->PRE2 & 0x10) ? "-" : "", + (int)(ffe_data->PRE2 & 0xf), + (int)ffe_data->MAIN, + (ffe_data->POST1 & 0x10) ? "-" : "", + (int)(ffe_data->POST1 & 0xf), + (ffe_data->POST2 & 0x10) ? "-" : "", + (int)(ffe_data->POST2 & 0xf)); + sdk_info(hwdev->dev_hdl, "RX_CTLE: Gain1~3=%u %u %u; Boost1~3=%u %u %u; Zero1~3=%u %u %u; Squelch1~3=%u %u %u\n", + ctle_data->ctlebst[0], ctle_data->ctlebst[1], + ctle_data->ctlebst[2], ctle_data->ctlecmband[0], + ctle_data->ctlecmband[1], ctle_data->ctlecmband[2], + ctle_data->ctlermband[0], ctle_data->ctlermband[1], + ctle_data->ctlermband[2], ctle_data->ctleza[0], + ctle_data->ctleza[1], ctle_data->ctleza[2]); +} + +static void __print_hi30_status(struct hifc_hwdev *hwdev, + struct hifc_link_info *info) +{ + struct hilink_lane *lane; + int lane_used_num = 0, i; + + for (i = 0; i < HILINK_MAX_LANE; i++) { + lane = (struct hilink_lane *)(info->lane2 + i * sizeof(*lane)); + if (!lane->lane_used) + continue; + + __hi30_lane_info(hwdev, lane); + lane_used_num++; + } + + /* in new firmware, all lane info setted in lane2 */ + if (lane_used_num) + return; + + /* compatible old firmware */ + __hi30_lane_info(hwdev, (struct hilink_lane *)info->lane1); +} + +static void __print_link_info(struct hifc_hwdev *hwdev, + struct hifc_link_info *info, + enum hilink_info_print_event type) +{ + char *fec = "None"; + + if (info->fec < HILINK_FEC_MAX_TYPE) + fec = __hw_to_char_fec[info->fec]; + else + sdk_info(hwdev->dev_hdl, "Unknown fec type: %u\n", + info->fec); + + if (type == HILINK_EVENT_LINK_UP || !info->an_state) { + sdk_info(hwdev->dev_hdl, "Link information: speed %dGbps, %s, autoneg %s\n", + info->speed, fec, info->an_state ? "on" : "off"); + } else { + sdk_info(hwdev->dev_hdl, "Link information: antoneg: %s\n", + info->an_state ? "on" : "off"); + } +} + +static char *hilink_info_report_type[HILINK_EVENT_MAX_TYPE] = { + "", "link up", "link down", "cable plugged" +}; + +void print_hilink_info(struct hifc_hwdev *hwdev, + enum hilink_info_print_event type, + struct hifc_link_info *info) +{ + __print_cable_info(hwdev, info); + + __print_link_info(hwdev, info, type); + + __print_hi30_status(hwdev, info); + + if (type == HILINK_EVENT_LINK_UP) + return; + + if (type == HILINK_EVENT_CABLE_PLUGGED) { + sdk_info(hwdev->dev_hdl, "alos: %u, rx_los: %u\n", + info->alos, info->rx_los); + return; + } + + sdk_info(hwdev->dev_hdl, "PMA ctrl: %s, MAC tx %s, MAC rx %s, PMA debug info reg: 0x%x, PMA signal ok reg: 0x%x, RF/LF status reg: 0x%x\n", + info->pma_status == 1 ? "off" : "on", + info->mac_tx_en ? "enable" : "disable", + info->mac_rx_en ? "enable" : "disable", info->pma_dbg_info_reg, + info->pma_signal_ok_reg, info->rf_lf_status_reg); + sdk_info(hwdev->dev_hdl, "alos: %u, rx_los: %u, PCS block counter reg: 0x%x, PCS link: 0x%x, MAC link: 0x%x PCS_err_cnt: 0x%x\n", + info->alos, info->rx_los, info->pcs_err_blk_cnt_reg, + info->pcs_link_reg, info->mac_link_reg, info->pcs_err_cnt); +} + +static void hifc_print_hilink_info(struct hifc_hwdev *hwdev, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size) +{ + struct hifc_hilink_link_info *hilink_info = buf_in; + struct hifc_link_info *info; + enum hilink_info_print_event type; + + if (in_size != sizeof(*hilink_info)) { + sdk_err(hwdev->dev_hdl, "Invalid hilink info message size %d, should be %ld\n", + in_size, sizeof(*hilink_info)); + return; + } + + ((struct hifc_hilink_link_info *)buf_out)->status = 0; + *out_size = sizeof(*hilink_info); + + info = &hilink_info->info; + type = hilink_info->info_type; + + if (type < HILINK_EVENT_LINK_UP || type >= HILINK_EVENT_MAX_TYPE) { + sdk_info(hwdev->dev_hdl, "Invalid hilink info report, type: %d\n", + type); + return; + } + + sdk_info(hwdev->dev_hdl, "Hilink info report after %s\n", + hilink_info_report_type[type]); + + print_hilink_info(hwdev, type, info); +} + +static void __port_sfp_info_event(struct hifc_hwdev *hwdev, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_cmd_get_sfp_qsfp_info *sfp_info = buf_in; + struct hifc_port_routine_cmd *rt_cmd; + struct card_node *chip_node = hwdev->chip_node; + + if (in_size != sizeof(*sfp_info)) { + sdk_err(hwdev->dev_hdl, "Invalid sfp info cmd, length: %d, should be %ld\n", + in_size, sizeof(*sfp_info)); + return; + } + + if (sfp_info->port_id >= HIFC_MAX_PORT_ID) { + sdk_err(hwdev->dev_hdl, "Invalid sfp port id: %d, max port is %d\n", + sfp_info->port_id, HIFC_MAX_PORT_ID - 1); + return; + } + + if (!chip_node->rt_cmd) + return; + + rt_cmd = &chip_node->rt_cmd[sfp_info->port_id]; + mutex_lock(&chip_node->sfp_mutex); + memcpy(&rt_cmd->sfp_info, sfp_info, sizeof(rt_cmd->sfp_info)); + rt_cmd->up_send_sfp_info = true; + mutex_unlock(&chip_node->sfp_mutex); +} + +static void __port_sfp_abs_event(struct hifc_hwdev *hwdev, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_cmd_get_light_module_abs *sfp_abs = buf_in; + struct hifc_port_routine_cmd *rt_cmd; + struct card_node *chip_node = hwdev->chip_node; + + if (in_size != sizeof(*sfp_abs)) { + sdk_err(hwdev->dev_hdl, "Invalid sfp absent cmd, length: %d, should be %ld\n", + in_size, sizeof(*sfp_abs)); + return; + } + + if (sfp_abs->port_id >= HIFC_MAX_PORT_ID) { + sdk_err(hwdev->dev_hdl, "Invalid sfp port id: %d, max port is %d\n", + sfp_abs->port_id, HIFC_MAX_PORT_ID - 1); + return; + } + + if (!chip_node->rt_cmd) + return; + + rt_cmd = &chip_node->rt_cmd[sfp_abs->port_id]; + mutex_lock(&chip_node->sfp_mutex); + memcpy(&rt_cmd->abs, sfp_abs, sizeof(rt_cmd->abs)); + rt_cmd->up_send_sfp_abs = true; + mutex_unlock(&chip_node->sfp_mutex); +} + +static void mgmt_heartbeat_enhanced_event(struct hifc_hwdev *hwdev, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_heartbeat_event *hb_event = buf_in; + struct hifc_heartbeat_event *hb_event_out = buf_out; + struct hifc_hwdev *dev = hwdev; + + if (in_size != sizeof(*hb_event)) { + sdk_err(dev->dev_hdl, "Invalid data size from mgmt for heartbeat event: %d\n", + in_size); + return; + } + + if (dev->heartbeat_ehd.last_heartbeat != hb_event->heart) { + dev->heartbeat_ehd.last_update_jiffies = jiffies; + dev->heartbeat_ehd.last_heartbeat = hb_event->heart; + } + + hb_event_out->drv_heart = HEARTBEAT_DRV_MAGIC_ACK; + + hb_event_out->status = 0; + *out_size = sizeof(*hb_event_out); +} + +struct dev_event_handler { + u8 mod; + u8 cmd; + void (*handler)(struct hifc_hwdev *hwdev, void *buf_in, u16 in_size, + void *buf_out, u16 *out_size); +}; + +struct dev_event_handler dev_cmd_handler[] = { + { + .mod = HIFC_MOD_L2NIC, + .cmd = HIFC_PORT_CMD_GET_SFP_INFO, + .handler = __port_sfp_info_event, + }, + + { + .mod = HIFC_MOD_L2NIC, + .cmd = HIFC_PORT_CMD_GET_SFP_ABS, + .handler = __port_sfp_abs_event, + }, + + { + .mod = HIFC_MOD_HILINK, + .cmd = HIFC_HILINK_CMD_GET_LINK_INFO, + .handler = hifc_print_hilink_info, + }, + + { + .mod = HIFC_MOD_COMM, + .cmd = HIFC_MGMT_CMD_FAULT_REPORT, + .handler = fault_event_handler, + }, + + { + .mod = HIFC_MOD_L2NIC, + .cmd = HIFC_MGMT_CMD_HEART_LOST_REPORT, + .handler = heartbeat_lost_event_handler, + }, + + { + .mod = HIFC_MOD_COMM, + .cmd = HIFC_MGMT_CMD_WATCHDOG_INFO, + .handler = mgmt_watchdog_timeout_event_handler, + }, + + { + .mod = HIFC_MOD_L2NIC, + .cmd = HIFC_PORT_CMD_MGMT_RESET, + .handler = mgmt_reset_event_handler, + }, + + { + .mod = HIFC_MOD_COMM, + .cmd = HIFC_MGMT_CMD_FMW_ACT_NTC, + .handler = hifc_fmw_act_ntc_handler, + }, + + { + .mod = HIFC_MOD_COMM, + .cmd = HIFC_MGMT_CMD_PCIE_DFX_NTC, + .handler = hifc_pcie_dfx_event_handler, + }, + + { + .mod = HIFC_MOD_COMM, + .cmd = HIFC_MGMT_CMD_GET_HOST_INFO, + .handler = hifc_mctp_get_host_info_event_handler, + }, + + { + .mod = HIFC_MOD_COMM, + .cmd = HIFC_MGMT_CMD_HEARTBEAT_EVENT, + .handler = mgmt_heartbeat_enhanced_event, + }, +}; + +/* public process for this event: + * pf link change event + * pf heart lost event ,TBD + * pf fault report event + * vf link change event + * vf heart lost event, TBD + * vf fault report event, TBD + */ +static void _event_handler(struct hifc_hwdev *hwdev, enum hifc_mod_type mod, + u8 cmd, void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + u32 i, size = sizeof(dev_cmd_handler) / sizeof(dev_cmd_handler[0]); + + if (!hwdev) + return; + + *out_size = 0; + + for (i = 0; i < size; i++) { + if (cmd == dev_cmd_handler[i].cmd && + mod == dev_cmd_handler[i].mod) { + dev_cmd_handler[i].handler(hwdev, buf_in, in_size, + buf_out, out_size); + break; + } + } + + /* can't find this event cmd */ + if (i == size) + sdk_warn(hwdev->dev_hdl, "Unsupported mod(%d) event cmd(%d) to process\n", + mod, cmd); +} + +/* pf link change event */ +static void pf_nic_event_handler(void *hwdev, void *pri_handle, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + _event_handler(hwdev, HIFC_MOD_L2NIC, cmd, buf_in, in_size, + buf_out, out_size); +} + +static void pf_hilink_event_handler(void *hwdev, void *pri_handle, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + _event_handler(hwdev, HIFC_MOD_HILINK, cmd, buf_in, in_size, + buf_out, out_size); +} + +/* pf fault report event */ +void pf_fault_event_handler(void *hwdev, void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + _event_handler(hwdev, HIFC_MOD_COMM, HIFC_MGMT_CMD_FAULT_REPORT, + buf_in, in_size, buf_out, out_size); +} + +void mgmt_watchdog_event_handler(void *hwdev, void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + _event_handler(hwdev, HIFC_MOD_COMM, HIFC_MGMT_CMD_WATCHDOG_INFO, + buf_in, in_size, buf_out, out_size); +} + +void mgmt_fmw_act_event_handler(void *hwdev, void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + _event_handler(hwdev, HIFC_MOD_COMM, HIFC_MGMT_CMD_FMW_ACT_NTC, + buf_in, in_size, buf_out, out_size); +} + +void mgmt_pcie_dfx_event_handler(void *hwdev, void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + _event_handler(hwdev, HIFC_MOD_COMM, HIFC_MGMT_CMD_PCIE_DFX_NTC, + buf_in, in_size, buf_out, out_size); +} + +void mgmt_get_mctp_event_handler(void *hwdev, void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + _event_handler(hwdev, HIFC_MOD_COMM, HIFC_MGMT_CMD_GET_HOST_INFO, + buf_in, in_size, buf_out, out_size); +} + +void mgmt_heartbeat_event_handler(void *hwdev, void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + _event_handler(hwdev, HIFC_MOD_COMM, HIFC_MGMT_CMD_HEARTBEAT_EVENT, + buf_in, in_size, buf_out, out_size); +} + +static void pf_event_register(struct hifc_hwdev *hwdev) +{ + if (hifc_is_hwdev_mod_inited(hwdev, HIFC_HWDEV_MGMT_INITED)) { + hifc_register_mgmt_msg_cb(hwdev, HIFC_MOD_L2NIC, + hwdev, pf_nic_event_handler); + hifc_register_mgmt_msg_cb(hwdev, HIFC_MOD_HILINK, + hwdev, + pf_hilink_event_handler); + hifc_comm_recv_mgmt_self_cmd_reg(hwdev, + HIFC_MGMT_CMD_FAULT_REPORT, + pf_fault_event_handler); + + hifc_comm_recv_mgmt_self_cmd_reg(hwdev, + HIFC_MGMT_CMD_WATCHDOG_INFO, + mgmt_watchdog_event_handler); + + hifc_comm_recv_mgmt_self_cmd_reg(hwdev, + HIFC_MGMT_CMD_FMW_ACT_NTC, + mgmt_fmw_act_event_handler); + hifc_comm_recv_mgmt_self_cmd_reg(hwdev, + HIFC_MGMT_CMD_PCIE_DFX_NTC, + mgmt_pcie_dfx_event_handler); + hifc_comm_recv_mgmt_self_cmd_reg(hwdev, + HIFC_MGMT_CMD_GET_HOST_INFO, + mgmt_get_mctp_event_handler); + } +} + +void hifc_event_register(void *dev, void *pri_handle, + hifc_event_handler callback) +{ + struct hifc_hwdev *hwdev = dev; + + if (!dev) { + pr_err("Hwdev pointer is NULL for register event\n"); + return; + } + + hwdev->event_callback = callback; + hwdev->event_pri_handle = pri_handle; + + pf_event_register(hwdev); +} + +void hifc_event_unregister(void *dev) +{ + struct hifc_hwdev *hwdev = dev; + + hwdev->event_callback = NULL; + hwdev->event_pri_handle = NULL; + + hifc_unregister_mgmt_msg_cb(hwdev, HIFC_MOD_L2NIC); + hifc_unregister_mgmt_msg_cb(hwdev, HIFC_MOD_HILINK); + hifc_comm_recv_up_self_cmd_unreg(hwdev, + HIFC_MGMT_CMD_FAULT_REPORT); + hifc_comm_recv_up_self_cmd_unreg(hwdev, + HIFC_MGMT_CMD_WATCHDOG_INFO); + hifc_comm_recv_up_self_cmd_unreg(hwdev, + HIFC_MGMT_CMD_FMW_ACT_NTC); + hifc_comm_recv_up_self_cmd_unreg(hwdev, + HIFC_MGMT_CMD_PCIE_DFX_NTC); + hifc_comm_recv_up_self_cmd_unreg(hwdev, + HIFC_MGMT_CMD_GET_HOST_INFO); +} + +/* 0 - heartbeat lost, 1 - normal */ +static u8 hifc_get_heartbeat_status(struct hifc_hwdev *hwdev) +{ + struct hifc_hwif *hwif = hwdev->hwif; + u32 attr1; + + /* suprise remove should be set 1 */ + if (!hifc_get_chip_present_flag(hwdev)) + return 1; + + attr1 = hifc_hwif_read_reg(hwif, HIFC_CSR_FUNC_ATTR1_ADDR); + if (attr1 == HIFC_PCIE_LINK_DOWN) { + sdk_err(hwdev->dev_hdl, "Detect pcie is link down\n"); + hifc_set_chip_absent(hwdev); + hifc_force_complete_all(hwdev); + /* should notify chiperr to pangea + * when detecting pcie link down + */ + return 1; + } + + return HIFC_AF1_GET(attr1, MGMT_INIT_STATUS); +} + +static void hifc_heartbeat_event_handler(struct work_struct *work) +{ + struct hifc_hwdev *hwdev = + container_of(work, struct hifc_hwdev, timer_work); + u16 out = 0; + + _event_handler(hwdev, HIFC_MOD_L2NIC, HIFC_MGMT_CMD_HEART_LOST_REPORT, + NULL, 0, &out, &out); +} + +static bool __detect_heartbeat_ehd_lost(struct hifc_hwdev *hwdev) +{ + struct hifc_heartbeat_enhanced *hb_ehd = &hwdev->heartbeat_ehd; + u64 update_time; + bool hb_ehd_lost = false; + + if (!hb_ehd->en) + return false; + + if (time_after(jiffies, hb_ehd->start_detect_jiffies)) { + update_time = jiffies_to_msecs(jiffies - + hb_ehd->last_update_jiffies); + if (update_time > HIFC_HEARBEAT_ENHANCED_LOST) { + sdk_warn(hwdev->dev_hdl, "Heartbeat enhanced lost for %d millisecond\n", + (u32)update_time); + hb_ehd_lost = true; + } + } else { + /* mgmt may not report heartbeart enhanced event and won't + * update last_update_jiffies + */ + hb_ehd->last_update_jiffies = jiffies; + } + + return hb_ehd_lost; +} + +static void hifc_heartbeat_timer_handler(struct timer_list *t) +{ + struct hifc_hwdev *hwdev = from_timer(hwdev, t, heartbeat_timer); + + if (__detect_heartbeat_ehd_lost(hwdev) || + !hifc_get_heartbeat_status(hwdev)) { + hwdev->heartbeat_lost = 1; + queue_work(hwdev->workq, &hwdev->timer_work); + } else { + mod_timer(&hwdev->heartbeat_timer, + jiffies + msecs_to_jiffies(HIFC_HEARTBEAT_PERIOD)); + } +} + +void add_to_timer(struct timer_list *timer, long period) +{ + if (!timer) + return; + + add_timer(timer); +} + +void delete_timer(struct timer_list *timer) +{ + if (!timer) + return; + + del_timer_sync(timer); +} + +void hifc_init_heartbeat(struct hifc_hwdev *hwdev) +{ + timer_setup(&hwdev->heartbeat_timer, hifc_heartbeat_timer_handler, 0); + hwdev->heartbeat_timer.expires = + jiffies + msecs_to_jiffies(HIFC_HEARTBEAT_START_EXPIRE); + + add_to_timer(&hwdev->heartbeat_timer, HIFC_HEARTBEAT_PERIOD); + + INIT_WORK(&hwdev->timer_work, hifc_heartbeat_event_handler); +} + +void hifc_destroy_heartbeat(struct hifc_hwdev *hwdev) +{ + delete_timer(&hwdev->heartbeat_timer); +} + +u8 hifc_nic_sw_aeqe_handler(void *handle, u8 event, u64 data) +{ + struct hifc_hwdev *hwdev = (struct hifc_hwdev *)handle; + u8 event_level = FAULT_LEVEL_MAX; + + switch (event) { + case HIFC_INTERNAL_TSO_FATAL_ERROR: + case HIFC_INTERNAL_LRO_FATAL_ERROR: + case HIFC_INTERNAL_TX_FATAL_ERROR: + case HIFC_INTERNAL_RX_FATAL_ERROR: + case HIFC_INTERNAL_OTHER_FATAL_ERROR: + atomic_inc(&hwdev->hw_stats.nic_ucode_event_stats[event]); + sdk_err(hwdev->dev_hdl, "SW aeqe event type: 0x%x, data: 0x%llx\n", + event, data); + event_level = FAULT_LEVEL_FATAL; + break; + default: + sdk_err(hwdev->dev_hdl, "Unsupported sw event %d to process.\n", + event); + } + + return event_level; +} + +void hifc_set_pcie_order_cfg(void *handle) +{ + struct hifc_hwdev *hwdev = handle; + u32 val; + + if (!hwdev) + return; + + val = hifc_hwif_read_reg(hwdev->hwif, + HIFC_GLB_DMA_SO_RO_REPLACE_ADDR); + + if (HIFC_GLB_DMA_SO_RO_GET(val, SO_RO_CFG)) { + val = HIFC_GLB_DMA_SO_R0_CLEAR(val, SO_RO_CFG); + val |= HIFC_GLB_DMA_SO_R0_SET(HIFC_DISABLE_ORDER, SO_RO_CFG); + hifc_hwif_write_reg(hwdev->hwif, + HIFC_GLB_DMA_SO_RO_REPLACE_ADDR, val); + } +} + +int hifc_get_board_info(void *hwdev, struct hifc_board_info *info) +{ + struct hifc_comm_board_info board_info = {0}; + u16 out_size = sizeof(board_info); + int err; + + if (!hwdev || !info) + return -EINVAL; + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_GET_BOARD_INFO, + &board_info, sizeof(board_info), + &board_info, &out_size, 0); + if (err || board_info.status || !out_size) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Failed to get board info, err: %d, status: 0x%x, out size: 0x%x\n", + err, board_info.status, out_size); + return -EFAULT; + } + + memcpy(info, &board_info.info, sizeof(*info)); + + return 0; +} + +int hifc_get_phy_init_status(void *hwdev, + enum phy_init_status_type *init_status) +{ + struct hifc_phy_init_status phy_info = {0}; + u16 out_size = sizeof(phy_info); + int err; + + if (!hwdev || !init_status) + return -EINVAL; + + err = hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_GET_PHY_INIT_STATUS, + &phy_info, sizeof(phy_info), + &phy_info, &out_size, 0); + if ((phy_info.status != HIFC_MGMT_CMD_UNSUPPORTED && + phy_info.status) || err || !out_size) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Failed to get phy info, err: %d, status: 0x%x, out size: 0x%x\n", + err, phy_info.status, out_size); + return -EFAULT; + } + + *init_status = phy_info.init_status; + + return phy_info.status; +} + +int hifc_phy_init_status_judge(void *hwdev) +{ + enum phy_init_status_type init_status; + int ret; + unsigned long end; + + /* It's not a phy, so don't judge phy status */ + if (!HIFC_BOARD_IS_PHY((struct hifc_hwdev *)hwdev)) + return 0; + + end = jiffies + msecs_to_jiffies(PHY_DOING_INIT_TIMEOUT); + do { + ret = hifc_get_phy_init_status(hwdev, &init_status); + if (ret == HIFC_MGMT_CMD_UNSUPPORTED) + return 0; + else if (ret) + return -EFAULT; + + switch (init_status) { + case PHY_INIT_SUCCESS: + sdk_info(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Phy init is success\n"); + return 0; + case PHY_NONSUPPORT: + sdk_info(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Phy init is nonsupport\n"); + return 0; + case PHY_INIT_FAIL: + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Phy init is failed\n"); + return -EIO; + case PHY_INIT_DOING: + msleep(250); + break; + default: + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Phy init is invalid, init_status: %d\n", + init_status); + return -EINVAL; + } + } while (time_before(jiffies, end)); + + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Phy init is timeout\n"); + + return -ETIMEDOUT; +} + +int hifc_get_mgmt_channel_status(void *handle) +{ + struct hifc_hwdev *hwdev = handle; + u32 val; + + if (!hwdev) + return true; + + if (hifc_func_type(hwdev) == TYPE_VF || + !(hwdev->feature_cap & HIFC_FUNC_SUPP_DFX_REG)) + return false; + + val = hifc_hwif_read_reg(hwdev->hwif, HIFC_ICPL_RESERVD_ADDR); + + return HIFC_GET_MGMT_CHANNEL_STATUS(val, MGMT_CHANNEL_STATUS); +} + +#define HIFC_RED_REG_TIME_OUT 3000 + +int hifc_read_reg(void *hwdev, u32 reg_addr, u32 *val) +{ + struct hifc_reg_info reg_info = {0}; + u16 out_size = sizeof(reg_info); + int err; + + if (!hwdev || !val) + return -EINVAL; + + reg_info.reg_addr = reg_addr; + reg_info.val_length = sizeof(u32); + + err = hifc_pf_msg_to_mgmt_sync(hwdev, HIFC_MOD_COMM, + HIFC_MGMT_CMD_REG_READ, + &reg_info, sizeof(reg_info), + &reg_info, &out_size, + HIFC_RED_REG_TIME_OUT); + if (reg_info.status || err || !out_size) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Failed to read reg, err: %d, status: 0x%x, out size: 0x%x\n", + err, reg_info.status, out_size); + return -EFAULT; + } + + *val = reg_info.data[0]; + + return 0; +} + +void hifc_swe_fault_handler(struct hifc_hwdev *hwdev, u8 level, + u8 event, u64 val) +{ + struct hifc_fault_info_node *fault_node; + + if (level < FAULT_LEVEL_MAX) { + fault_node = kzalloc(sizeof(*fault_node), GFP_KERNEL); + if (!fault_node) { + sdk_err(hwdev->dev_hdl, "Malloc fault node memory failed\n"); + return; + } + + fault_node->info.fault_src = HIFC_FAULT_SRC_SW_MGMT_UCODE; + fault_node->info.fault_lev = level; + fault_node->info.fault_data.sw_mgmt.event_id = event; + fault_node->info.fault_data.sw_mgmt.event_data = val; + hifc_refresh_history_fault(hwdev, &fault_node->info); + + down(&hwdev->fault_list_sem); + kfree(fault_node); + up(&hwdev->fault_list_sem); + } +} + +void hifc_set_func_deinit_flag(void *hwdev) +{ + struct hifc_hwdev *dev = hwdev; + + set_bit(HIFC_HWDEV_FUNC_DEINIT, &dev->func_state); +} + +int hifc_get_card_present_state(void *hwdev, bool *card_present_state) +{ + u32 addr, attr1; + + if (!hwdev || !card_present_state) + return -EINVAL; + + addr = HIFC_CSR_FUNC_ATTR1_ADDR; + attr1 = hifc_hwif_read_reg(((struct hifc_hwdev *)hwdev)->hwif, addr); + if (attr1 == HIFC_PCIE_LINK_DOWN) { + sdk_warn(((struct hifc_hwdev *)hwdev)->dev_hdl, "Card is not present\n"); + *card_present_state = (bool)0; + } else { + *card_present_state = (bool)1; + } + + return 0; +} + +void hifc_disable_mgmt_msg_report(void *hwdev) +{ + struct hifc_hwdev *hw_dev = (struct hifc_hwdev *)hwdev; + + hifc_set_pf_status(hw_dev->hwif, HIFC_PF_STATUS_INIT); +} + diff --git a/drivers/scsi/huawei/hifc/hifc_hwdev.h b/drivers/scsi/huawei/hifc/hifc_hwdev.h new file mode 100644 index 000000000000..6ebf59b31fb8 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_hwdev.h @@ -0,0 +1,456 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#ifndef HIFC_HWDEV_H_ +#define HIFC_HWDEV_H_ + +/* to use 0-level CLA, page size must be: 64B(wqebb) * 4096(max_q_depth) */ +#define HIFC_DEFAULT_WQ_PAGE_SIZE 0x40000 +#define HIFC_HW_WQ_PAGE_SIZE 0x1000 + +#define HIFC_MSG_TO_MGMT_MAX_LEN 2016 + +#define HIFC_MGMT_STATUS_ERR_OK 0 /* Ok */ +#define HIFC_MGMT_STATUS_ERR_PARAM 1 /* Invalid parameter */ +#define HIFC_MGMT_STATUS_ERR_FAILED 2 /* Operation failed */ +#define HIFC_MGMT_STATUS_ERR_PORT 3 /* Invalid port */ +#define HIFC_MGMT_STATUS_ERR_TIMEOUT 4 /* Operation time out */ +#define HIFC_MGMT_STATUS_ERR_NOMATCH 5 /* Version not match */ +#define HIFC_MGMT_STATUS_ERR_EXIST 6 /* Entry exists */ +#define HIFC_MGMT_STATUS_ERR_NOMEM 7 /* Out of memory */ +#define HIFC_MGMT_STATUS_ERR_INIT 8 /* Feature not initialized */ +#define HIFC_MGMT_STATUS_ERR_FAULT 9 /* Invalid address */ +#define HIFC_MGMT_STATUS_ERR_PERM 10 /* Operation not permitted */ +#define HIFC_MGMT_STATUS_ERR_EMPTY 11 /* Table empty */ +#define HIFC_MGMT_STATUS_ERR_FULL 12 /* Table full */ +#define HIFC_MGMT_STATUS_ERR_NOT_FOUND 13 /* Not found */ +#define HIFC_MGMT_STATUS_ERR_BUSY 14 /* Device or resource busy */ +#define HIFC_MGMT_STATUS_ERR_RESOURCE 15 /* No resources for operation */ +#define HIFC_MGMT_STATUS_ERR_CONFIG 16 /* Invalid configuration */ +#define HIFC_MGMT_STATUS_ERR_UNAVAIL 17 /* Feature unavailable */ +#define HIFC_MGMT_STATUS_ERR_CRC 18 /* CRC check failed */ +#define HIFC_MGMT_STATUS_ERR_NXIO 19 /* No such device or address */ +#define HIFC_MGMT_STATUS_ERR_ROLLBACK 20 /* Chip rollback fail */ +#define HIFC_MGMT_STATUS_ERR_LEN 32 /* Length too short or too long */ +#define HIFC_MGMT_STATUS_ERR_UNSUPPORT 0xFF /* Feature not supported*/ +/* Qe buffer relates define */ + +enum hifc_rx_buf_size { + HIFC_RX_BUF_SIZE_32B = 0x20, + HIFC_RX_BUF_SIZE_64B = 0x40, + HIFC_RX_BUF_SIZE_96B = 0x60, + HIFC_RX_BUF_SIZE_128B = 0x80, + HIFC_RX_BUF_SIZE_192B = 0xC0, + HIFC_RX_BUF_SIZE_256B = 0x100, + HIFC_RX_BUF_SIZE_384B = 0x180, + HIFC_RX_BUF_SIZE_512B = 0x200, + HIFC_RX_BUF_SIZE_768B = 0x300, + HIFC_RX_BUF_SIZE_1K = 0x400, + HIFC_RX_BUF_SIZE_1_5K = 0x600, + HIFC_RX_BUF_SIZE_2K = 0x800, + HIFC_RX_BUF_SIZE_3K = 0xC00, + HIFC_RX_BUF_SIZE_4K = 0x1000, + HIFC_RX_BUF_SIZE_8K = 0x2000, + HIFC_RX_BUF_SIZE_16K = 0x4000, +}; + +enum hifc_res_state { + HIFC_RES_CLEAN = 0, + HIFC_RES_ACTIVE = 1, +}; + +enum ppf_tmr_status { + HIFC_PPF_TMR_FLAG_STOP, + HIFC_PPF_TMR_FLAG_START, +}; + +struct cfg_mgmt_info; +struct hifc_hwif; +struct hifc_wqs; +struct hifc_aeqs; +struct hifc_ceqs; +struct hifc_msg_pf_to_mgmt; +struct hifc_cmdqs; + +struct hifc_root_ctxt { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_idx; + u16 rsvd1; + u8 set_cmdq_depth; + u8 cmdq_depth; + u8 lro_en; + u8 rsvd2; + u8 ppf_idx; + u8 rsvd3; + u16 rq_depth; + u16 rx_buf_sz; + u16 sq_depth; +}; + +struct hifc_page_addr { + void *virt_addr; + u64 phys_addr; +}; + +#define HIFC_PCIE_LINK_DOWN 0xFFFFFFFF + +#define HIFC_DEV_ACTIVE_FW_TIMEOUT (35 * 1000) +#define HIFC_DEV_BUSY_ACTIVE_FW 0xFE + +#define HIFC_HW_WQ_NAME "hifc_hardware" +#define HIFC_HEARTBEAT_PERIOD 1000 +#define HIFC_HEARTBEAT_START_EXPIRE 5000 + +#define HIFC_CHIP_ERROR_TYPE_MAX 1024 +#define HIFC_CHIP_FAULT_SIZE \ + (HIFC_NODE_ID_MAX * FAULT_LEVEL_MAX * HIFC_CHIP_ERROR_TYPE_MAX) + +#define HIFC_CSR_DMA_ATTR_TBL_BASE 0xC80 +#define HIFC_CSR_DMA_ATTR_TBL_STRIDE 0x4 +#define HIFC_CSR_DMA_ATTR_TBL_ADDR(idx) \ + (HIFC_CSR_DMA_ATTR_TBL_BASE \ + + (idx) * HIFC_CSR_DMA_ATTR_TBL_STRIDE) + +/* MSI-X registers */ +#define HIFC_CSR_MSIX_CNT_BASE 0x2004 +#define HIFC_CSR_MSIX_STRIDE 0x8 + +#define HIFC_CSR_MSIX_CNT_ADDR(idx) \ + (HIFC_CSR_MSIX_CNT_BASE + (idx) * HIFC_CSR_MSIX_STRIDE) + +enum hifc_node_id { + HIFC_NODE_ID_IPSU = 4, + HIFC_NODE_ID_MGMT_HOST = 21, /*Host CPU send API to uP */ + HIFC_NODE_ID_MAX = 22 +}; + +#define HIFC_HWDEV_INIT_MODES_MASK ((1UL << HIFC_HWDEV_ALL_INITED) - 1) + +enum hifc_hwdev_func_state { + HIFC_HWDEV_FUNC_INITED = HIFC_HWDEV_ALL_INITED, + HIFC_HWDEV_FUNC_DEINIT, + HIFC_HWDEV_STATE_BUSY = 31, +}; + +struct hifc_cqm_stats { + atomic_t cqm_cmd_alloc_cnt; + atomic_t cqm_cmd_free_cnt; + atomic_t cqm_send_cmd_box_cnt; + atomic_t cqm_db_addr_alloc_cnt; + atomic_t cqm_db_addr_free_cnt; + atomic_t cqm_fc_srq_create_cnt; + atomic_t cqm_qpc_mpt_create_cnt; + atomic_t cqm_nonrdma_queue_create_cnt; + atomic_t cqm_qpc_mpt_delete_cnt; + atomic_t cqm_nonrdma_queue_delete_cnt; + atomic_t cqm_aeq_callback_cnt[112]; +}; + +struct hifc_link_event_stats { + atomic_t link_down_stats; + atomic_t link_up_stats; +}; + +struct hifc_fault_event_stats { + atomic_t chip_fault_stats[HIFC_NODE_ID_MAX][FAULT_LEVEL_MAX]; + atomic_t fault_type_stat[FAULT_TYPE_MAX]; + atomic_t pcie_fault_stats; +}; + +struct hifc_hw_stats { + atomic_t heart_lost_stats; + atomic_t nic_ucode_event_stats[HIFC_NIC_FATAL_ERROR_MAX]; + struct hifc_cqm_stats cqm_stats; + struct hifc_link_event_stats link_event_stats; + struct hifc_fault_event_stats fault_event_stats; +}; + +struct hifc_fault_info_node { + struct list_head list; + struct hifc_hwdev *hwdev; + struct hifc_fault_recover_info info; +}; + +enum heartbeat_support_state { + HEARTBEAT_NOT_SUPPORT = 0, + HEARTBEAT_SUPPORT, +}; + +/* 25s for max 5 heartbeat event lost */ +#define HIFC_HEARBEAT_ENHANCED_LOST 25000 +struct hifc_heartbeat_enhanced { + bool en; /* enable enhanced heartbeat or not */ + + unsigned long last_update_jiffies; + u32 last_heartbeat; + + unsigned long start_detect_jiffies; +}; + +#define HIFC_CMD_VER_FUNC_ID 2 +#define HIFC_GLB_DMA_SO_RO_REPLACE_ADDR 0x488C +#define HIFC_ICPL_RESERVD_ADDR 0x9204 + +#define l2nic_msg_to_mgmt_sync(hwdev, cmd, buf_in, in_size, buf_out, out_size)\ + hifc_msg_to_mgmt_sync(hwdev, HIFC_MOD_L2NIC, cmd, \ + buf_in, in_size, \ + buf_out, out_size, 0) + +struct hifc_hwdev { + void *adapter_hdl; /* pointer to hifc_pcidev or NDIS_Adapter */ + void *pcidev_hdl; /* pointer to pcidev or Handler */ + void *dev_hdl; /* pointer to pcidev->dev or Handler, for + * sdk_err() or dma_alloc() + */ + u32 wq_page_size; + + void *cqm_hdl; + void *chip_node; + + struct hifc_hwif *hwif; /* include void __iomem *bar */ + struct cfg_mgmt_info *cfg_mgmt; + struct hifc_wqs *wqs; /* for FC slq */ + + struct hifc_aeqs *aeqs; + struct hifc_ceqs *ceqs; + + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + struct hifc_clp_pf_to_mgmt *clp_pf_to_mgmt; + + struct hifc_cmdqs *cmdqs; + + struct hifc_page_addr page_pa0; + struct hifc_page_addr page_pa1; + + hifc_event_handler event_callback; + void *event_pri_handle; + bool history_fault_flag; + struct hifc_fault_recover_info history_fault; + struct semaphore fault_list_sem; + + struct work_struct timer_work; + struct workqueue_struct *workq; + struct timer_list heartbeat_timer; + /* true represent heartbeat lost, false represent heartbeat restore */ + u32 heartbeat_lost; + int chip_present_flag; + struct hifc_heartbeat_enhanced heartbeat_ehd; + struct hifc_hw_stats hw_stats; + u8 *chip_fault_stats; + + u32 statufull_ref_cnt; + ulong func_state; + + u64 feature_cap; /* enum hifc_func_cap */ + + /* In bmgw x86 host, driver can't send message to mgmt cpu directly, + * need to trasmit message ppf mbox to bmgw arm host. + */ + + struct hifc_board_info board_info; +}; + +int hifc_init_comm_ch(struct hifc_hwdev *hwdev); +void hifc_uninit_comm_ch(struct hifc_hwdev *hwdev); + +enum hifc_set_arm_type { + HIFC_SET_ARM_CMDQ, + HIFC_SET_ARM_SQ, + HIFC_SET_ARM_TYPE_NUM, +}; + +/* up to driver event */ +#define HIFC_PORT_CMD_MGMT_RESET 0x0 +struct hifc_vport_state { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 rsvd1; + u8 state; + u8 rsvd2[3]; +}; + +struct hifc_l2nic_reset { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 reset_flag; +}; + +/* HILINK module interface */ + +/* cmd of mgmt CPU message for HILINK module */ +enum hifc_hilink_cmd { + HIFC_HILINK_CMD_GET_LINK_INFO = 0x3, + HIFC_HILINK_CMD_SET_LINK_SETTINGS = 0x8, +}; + +enum hilink_info_print_event { + HILINK_EVENT_LINK_UP = 1, + HILINK_EVENT_LINK_DOWN, + HILINK_EVENT_CABLE_PLUGGED, + HILINK_EVENT_MAX_TYPE, +}; + +enum hifc_link_port_type { + LINK_PORT_FIBRE = 1, + LINK_PORT_ELECTRIC, + LINK_PORT_COPPER, + LINK_PORT_AOC, + LINK_PORT_BACKPLANE, + LINK_PORT_BASET, + LINK_PORT_MAX_TYPE, +}; + +enum hilink_fibre_subtype { + FIBRE_SUBTYPE_SR = 1, + FIBRE_SUBTYPE_LR, + FIBRE_SUBTYPE_MAX, +}; + +enum hilink_fec_type { + HILINK_FEC_RSFEC, + HILINK_FEC_BASEFEC, + HILINK_FEC_NOFEC, + HILINK_FEC_MAX_TYPE, +}; + +/* cmd of mgmt CPU message */ +enum hifc_port_cmd { + HIFC_PORT_CMD_SET_MAC = 0x9, + HIFC_PORT_CMD_GET_AUTONEG_CAP = 0xf, + HIFC_PORT_CMD_SET_VPORT_ENABLE = 0x5d, + HIFC_PORT_CMD_UPDATE_MAC = 0xa4, + HIFC_PORT_CMD_GET_SFP_INFO = 0xad, + HIFC_PORT_CMD_GET_STD_SFP_INFO = 0xF0, + HIFC_PORT_CMD_GET_SFP_ABS = 0xFB, +}; + +struct hi30_ffe_data { + u8 PRE2; + u8 PRE1; + u8 POST1; + u8 POST2; + u8 MAIN; +}; + +struct hi30_ctle_data { + u8 ctlebst[3]; + u8 ctlecmband[3]; + u8 ctlermband[3]; + u8 ctleza[3]; + u8 ctlesqh[3]; + u8 ctleactgn[3]; + u8 ctlepassgn; +}; + +#define HILINK_MAX_LANE 4 + +struct hilink_lane { + u8 lane_used; + u8 hi30_ffe[5]; + u8 hi30_ctle[19]; + u8 hi30_dfe[14]; + u8 rsvd4; +}; + +struct hifc_link_info { + u8 vendor_name[16]; + /* port type: + * 1 - fiber; 2 - electric; 3 - copper; 4 - AOC; 5 - backplane; + * 6 - baseT; 0xffff - unknown + * + * port subtype: + * Only when port_type is fiber: + * 1 - SR; 2 - LR + */ + u32 port_type; + u32 port_sub_type; + u32 cable_length; + u8 cable_temp; + u8 cable_max_speed; /* 1(G)/10(G)/25(G)... */ + u8 sfp_type; /* 0 - qsfp; 1 - sfp */ + u8 rsvd0; + u32 power[4]; /* uW; if is sfp, only power[2] is valid */ + + u8 an_state; /* 0 - off; 1 - on */ + u8 fec; /* 0 - RSFEC; 1 - BASEFEC; 2 - NOFEC */ + u16 speed; /* 1(G)/10(G)/25(G)... */ + + u8 cable_absent; /* 0 - cable present; 1 - cable unpresent */ + u8 alos; /* 0 - yes; 1 - no */ + u8 rx_los; /* 0 - yes; 1 - no */ + u8 pma_status; + u32 pma_dbg_info_reg; /* pma debug info: */ + u32 pma_signal_ok_reg; /* signal ok: */ + + u32 pcs_err_blk_cnt_reg; /* error block counter: */ + u32 rf_lf_status_reg; /* RF/LF status: */ + u8 pcs_link_reg; /* pcs link: */ + u8 mac_link_reg; /* mac link: */ + u8 mac_tx_en; + u8 mac_rx_en; + u32 pcs_err_cnt; + + /* struct hifc_hilink_lane: 40 bytes */ + u8 lane1[40]; /* 25GE lane in old firmware */ + + u8 rsvd1[266]; /* hilink machine state */ + + u8 lane2[HILINK_MAX_LANE * 40]; /* max 4 lane for 40GE/100GE */ + + u8 rsvd2[2]; +}; + +struct hifc_hilink_link_info { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 port_id; + u8 info_type; /* 1: link up 2: link down 3 cable plugged */ + u8 rsvd1; + + struct hifc_link_info info; + + u8 rsvd2[352]; +}; + +int hifc_set_arm_bit(void *hwdev, enum hifc_set_arm_type q_type, u16 q_id); +void hifc_set_chip_present(void *hwdev); +void hifc_force_complete_all(void *hwdev); +void hifc_init_heartbeat(struct hifc_hwdev *hwdev); +void hifc_destroy_heartbeat(struct hifc_hwdev *hwdev); +u8 hifc_nic_sw_aeqe_handler(void *handle, u8 event, u64 data); +int hifc_l2nic_reset_base(struct hifc_hwdev *hwdev, u16 reset_flag); +int hifc_pf_msg_to_mgmt_sync(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size, u32 timeout); +void hifc_swe_fault_handler(struct hifc_hwdev *hwdev, u8 level, + u8 event, u64 val); +bool hifc_mgmt_event_ack_first(u8 mod, u8 cmd); +int hifc_phy_init_status_judge(void *hwdev); +int hifc_api_csr_rd32(void *hwdev, u8 dest, u32 addr, u32 *val); +int hifc_api_csr_wr32(void *hwdev, u8 dest, u32 addr, u32 val); +void mgmt_heartbeat_event_handler(void *hwdev, void *buf_in, u16 in_size, + void *buf_out, u16 *out_size); +struct hifc_sge { + u32 hi_addr; + u32 lo_addr; + u32 len; +}; + +void hifc_cpu_to_be32(void *data, int len); +void hifc_be32_to_cpu(void *data, int len); +void hifc_set_sge(struct hifc_sge *sge, dma_addr_t addr, u32 len); +#endif diff --git a/drivers/scsi/huawei/hifc/hifc_hwif.c b/drivers/scsi/huawei/hifc/hifc_hwif.c new file mode 100644 index 000000000000..ec84c9bc2f2f --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_hwif.c @@ -0,0 +1,630 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ +#define pr_fmt(fmt) KBUILD_MODNAME ": [COMM]" fmt + +#include <linux/types.h> +#include <linux/pci.h> +#include <linux/delay.h> +#include <linux/module.h> +#include <linux/io-mapping.h> + +#include "hifc_knl_adp.h" +#include "hifc_hw.h" +#include "hifc_hwdev.h" +#include "hifc_hwif.h" +#include "hifc_api_cmd.h" +#include "hifc_mgmt.h" +#include "hifc_eqs.h" + +#define WAIT_HWIF_READY_TIMEOUT 10000 +#define HIFC_SELFTEST_RESULT 0x883C + +u32 hifc_hwif_read_reg(struct hifc_hwif *hwif, u32 reg) +{ + return be32_to_cpu(readl(hwif->cfg_regs_base + reg)); +} + +void hifc_hwif_write_reg(struct hifc_hwif *hwif, u32 reg, u32 val) +{ + writel(cpu_to_be32(val), hwif->cfg_regs_base + reg); +} + +/** + * hwif_ready - test if the HW initialization passed + * @hwdev: the pointer to hw device + * Return: 0 - success, negative - failure + **/ +static int hwif_ready(struct hifc_hwdev *hwdev) +{ + u32 addr, attr1; + + addr = HIFC_CSR_FUNC_ATTR1_ADDR; + attr1 = hifc_hwif_read_reg(hwdev->hwif, addr); + + if (attr1 == HIFC_PCIE_LINK_DOWN) + return -EBUSY; + + if (!HIFC_AF1_GET(attr1, MGMT_INIT_STATUS)) + return -EBUSY; + + return 0; +} + +static int wait_hwif_ready(struct hifc_hwdev *hwdev) +{ + ulong timeout = 0; + + do { + if (!hwif_ready(hwdev)) + return 0; + + usleep_range(999, 1000); + timeout++; + } while (timeout <= WAIT_HWIF_READY_TIMEOUT); + + sdk_err(hwdev->dev_hdl, "Wait for hwif timeout\n"); + return -EBUSY; +} + +/** + * set_hwif_attr - set the attributes as members in hwif + * @hwif: the hardware interface of a pci function device + * @attr0: the first attribute that was read from the hw + * @attr1: the second attribute that was read from the hw + * @attr2: the third attribute that was read from the hw + **/ +static void set_hwif_attr(struct hifc_hwif *hwif, u32 attr0, u32 attr1, + u32 attr2) +{ + hwif->attr.func_global_idx = HIFC_AF0_GET(attr0, FUNC_GLOBAL_IDX); + hwif->attr.port_to_port_idx = HIFC_AF0_GET(attr0, P2P_IDX); + hwif->attr.pci_intf_idx = HIFC_AF0_GET(attr0, PCI_INTF_IDX); + hwif->attr.vf_in_pf = HIFC_AF0_GET(attr0, VF_IN_PF); + hwif->attr.func_type = HIFC_AF0_GET(attr0, FUNC_TYPE); + + hwif->attr.ppf_idx = HIFC_AF1_GET(attr1, PPF_IDX); + + hwif->attr.num_aeqs = BIT(HIFC_AF1_GET(attr1, AEQS_PER_FUNC)); + hwif->attr.num_ceqs = BIT(HIFC_AF1_GET(attr1, CEQS_PER_FUNC)); + hwif->attr.num_irqs = BIT(HIFC_AF1_GET(attr1, IRQS_PER_FUNC)); + hwif->attr.num_dma_attr = BIT(HIFC_AF1_GET(attr1, DMA_ATTR_PER_FUNC)); +} + +/** + * get_hwif_attr - read and set the attributes as members in hwif + * @hwif: the hardware interface of a pci function device + **/ +static void get_hwif_attr(struct hifc_hwif *hwif) +{ + u32 addr, attr0, attr1, attr2; + + addr = HIFC_CSR_FUNC_ATTR0_ADDR; + attr0 = hifc_hwif_read_reg(hwif, addr); + + addr = HIFC_CSR_FUNC_ATTR1_ADDR; + attr1 = hifc_hwif_read_reg(hwif, addr); + + addr = HIFC_CSR_FUNC_ATTR2_ADDR; + attr2 = hifc_hwif_read_reg(hwif, addr); + + set_hwif_attr(hwif, attr0, attr1, attr2); +} + +void hifc_set_pf_status(struct hifc_hwif *hwif, enum hifc_pf_status status) +{ + u32 attr5 = HIFC_AF5_SET(status, PF_STATUS); + u32 addr = HIFC_CSR_FUNC_ATTR5_ADDR; + + hifc_hwif_write_reg(hwif, addr, attr5); +} + +enum hifc_pf_status hifc_get_pf_status(struct hifc_hwif *hwif) +{ + u32 attr5 = hifc_hwif_read_reg(hwif, HIFC_CSR_FUNC_ATTR5_ADDR); + + return HIFC_AF5_GET(attr5, PF_STATUS); +} + +enum hifc_doorbell_ctrl hifc_get_doorbell_ctrl_status(struct hifc_hwif *hwif) +{ + u32 attr4 = hifc_hwif_read_reg(hwif, HIFC_CSR_FUNC_ATTR4_ADDR); + + return HIFC_AF4_GET(attr4, DOORBELL_CTRL); +} + +enum hifc_outbound_ctrl hifc_get_outbound_ctrl_status(struct hifc_hwif *hwif) +{ + u32 attr4 = hifc_hwif_read_reg(hwif, HIFC_CSR_FUNC_ATTR4_ADDR); + + return HIFC_AF4_GET(attr4, OUTBOUND_CTRL); +} + +void hifc_enable_doorbell(struct hifc_hwif *hwif) +{ + u32 addr, attr4; + + addr = HIFC_CSR_FUNC_ATTR4_ADDR; + attr4 = hifc_hwif_read_reg(hwif, addr); + + attr4 = HIFC_AF4_CLEAR(attr4, DOORBELL_CTRL); + attr4 |= HIFC_AF4_SET(ENABLE_DOORBELL, DOORBELL_CTRL); + + hifc_hwif_write_reg(hwif, addr, attr4); +} + +void hifc_disable_doorbell(struct hifc_hwif *hwif) +{ + u32 addr, attr4; + + addr = HIFC_CSR_FUNC_ATTR4_ADDR; + attr4 = hifc_hwif_read_reg(hwif, addr); + + attr4 = HIFC_AF4_CLEAR(attr4, DOORBELL_CTRL); + attr4 |= HIFC_AF4_SET(DISABLE_DOORBELL, DOORBELL_CTRL); + + hifc_hwif_write_reg(hwif, addr, attr4); +} + +/** + * set_ppf - try to set hwif as ppf and set the type of hwif in this case + * @hwif: the hardware interface of a pci function device + **/ +static void set_ppf(struct hifc_hwif *hwif) +{ + struct hifc_func_attr *attr = &hwif->attr; + u32 addr, val, ppf_election; + + /* Read Modify Write */ + addr = HIFC_CSR_PPF_ELECTION_ADDR; + + val = hifc_hwif_read_reg(hwif, addr); + val = HIFC_PPF_ELECTION_CLEAR(val, IDX); + + ppf_election = HIFC_PPF_ELECTION_SET(attr->func_global_idx, IDX); + val |= ppf_election; + + hifc_hwif_write_reg(hwif, addr, val); + + /* Check PPF */ + val = hifc_hwif_read_reg(hwif, addr); + + attr->ppf_idx = HIFC_PPF_ELECTION_GET(val, IDX); + if (attr->ppf_idx == attr->func_global_idx) + attr->func_type = TYPE_PPF; +} + +/** + * get_mpf - get the mpf index into the hwif + * @hwif: the hardware interface of a pci function device + **/ +static void get_mpf(struct hifc_hwif *hwif) +{ + struct hifc_func_attr *attr = &hwif->attr; + u32 mpf_election, addr; + + addr = HIFC_CSR_GLOBAL_MPF_ELECTION_ADDR; + + mpf_election = hifc_hwif_read_reg(hwif, addr); + attr->mpf_idx = HIFC_MPF_ELECTION_GET(mpf_election, IDX); +} + +/** + * set_mpf - try to set hwif as mpf and set the mpf idx in hwif + * @hwif: the hardware interface of a pci function device + **/ +static void set_mpf(struct hifc_hwif *hwif) +{ + struct hifc_func_attr *attr = &hwif->attr; + u32 addr, val, mpf_election; + + /* Read Modify Write */ + addr = HIFC_CSR_GLOBAL_MPF_ELECTION_ADDR; + + val = hifc_hwif_read_reg(hwif, addr); + + val = HIFC_MPF_ELECTION_CLEAR(val, IDX); + mpf_election = HIFC_MPF_ELECTION_SET(attr->func_global_idx, IDX); + + val |= mpf_election; + hifc_hwif_write_reg(hwif, addr, val); +} + +static void init_db_area_idx(struct hifc_free_db_area *free_db_area) +{ + u32 i; + + for (i = 0; i < HIFC_DB_MAX_AREAS; i++) + free_db_area->db_idx[i] = i; + + free_db_area->num_free = HIFC_DB_MAX_AREAS; + + spin_lock_init(&free_db_area->idx_lock); +} + +static int get_db_idx(struct hifc_hwif *hwif, u32 *idx) +{ + struct hifc_free_db_area *free_db_area = &hwif->free_db_area; + u32 pos; + u32 pg_idx; + + spin_lock(&free_db_area->idx_lock); + +retry: + if (free_db_area->num_free == 0) { + spin_unlock(&free_db_area->idx_lock); + return -ENOMEM; + } + + free_db_area->num_free--; + + pos = free_db_area->alloc_pos++; + pos &= HIFC_DB_MAX_AREAS - 1; + + pg_idx = free_db_area->db_idx[pos]; + + free_db_area->db_idx[pos] = 0xFFFFFFFF; + + /* pg_idx out of range */ + if (pg_idx >= HIFC_DB_MAX_AREAS) + goto retry; + + spin_unlock(&free_db_area->idx_lock); + + *idx = pg_idx; + + return 0; +} + +static void free_db_idx(struct hifc_hwif *hwif, u32 idx) +{ + struct hifc_free_db_area *free_db_area = &hwif->free_db_area; + u32 pos; + + if (idx >= HIFC_DB_MAX_AREAS) + return; + + spin_lock(&free_db_area->idx_lock); + + pos = free_db_area->return_pos++; + pos &= HIFC_DB_MAX_AREAS - 1; + + free_db_area->db_idx[pos] = idx; + + free_db_area->num_free++; + + spin_unlock(&free_db_area->idx_lock); +} + +void hifc_free_db_addr(void *hwdev, void __iomem *db_base, + void __iomem *dwqe_base) +{ + struct hifc_hwif *hwif; + u32 idx; + + if (!hwdev || !db_base) + return; + + hwif = ((struct hifc_hwdev *)hwdev)->hwif; + idx = DB_IDX(db_base, hwif->db_base); + +#if defined(__aarch64__) + /* No need to unmap */ +#else + if (dwqe_base) + io_mapping_unmap(dwqe_base); +#endif + + free_db_idx(hwif, idx); +} + +int hifc_alloc_db_addr(void *hwdev, void __iomem **db_base, + void __iomem **dwqe_base) +{ + struct hifc_hwif *hwif; + u64 offset; + u32 idx; + int err; + + if (!hwdev || !db_base) + return -EINVAL; + + hwif = ((struct hifc_hwdev *)hwdev)->hwif; + + err = get_db_idx(hwif, &idx); + if (err) + return -EFAULT; + + *db_base = hwif->db_base + idx * HIFC_DB_PAGE_SIZE; + + if (!dwqe_base) + return 0; + + offset = ((u64)idx) << PAGE_SHIFT; + +#if defined(__aarch64__) + *dwqe_base = hwif->dwqe_mapping + offset; +#else + *dwqe_base = io_mapping_map_wc(hwif->dwqe_mapping, offset, + HIFC_DB_PAGE_SIZE); +#endif + + if (!(*dwqe_base)) { + hifc_free_db_addr(hwdev, *db_base, NULL); + return -EFAULT; + } + + return 0; +} + +void hifc_set_msix_state(void *hwdev, u16 msix_idx, enum hifc_msix_state flag) +{ + struct hifc_hwif *hwif; + u32 offset = msix_idx * HIFC_PCI_MSIX_ENTRY_SIZE + + HIFC_PCI_MSIX_ENTRY_VECTOR_CTRL; + u32 mask_bits; + + if (!hwdev) + return; + + hwif = ((struct hifc_hwdev *)hwdev)->hwif; + + mask_bits = readl(hwif->intr_regs_base + offset); + mask_bits &= ~HIFC_PCI_MSIX_ENTRY_CTRL_MASKBIT; + if (flag) + mask_bits |= HIFC_PCI_MSIX_ENTRY_CTRL_MASKBIT; + + writel(mask_bits, hwif->intr_regs_base + offset); +} + +static void disable_all_msix(struct hifc_hwdev *hwdev) +{ + u16 num_irqs = hwdev->hwif->attr.num_irqs; + u16 i; + + for (i = 0; i < num_irqs; i++) + hifc_set_msix_state(hwdev, i, HIFC_MSIX_DISABLE); +} + +static int wait_until_doorbell_and_outbound_enabled(struct hifc_hwif *hwif) +{ + enum hifc_doorbell_ctrl db_ctrl; + enum hifc_outbound_ctrl outbound_ctrl; + u32 cnt = 0; + + while (cnt < HIFC_WAIT_DOORBELL_AND_OUTBOUND_TIMEOUT) { + db_ctrl = hifc_get_doorbell_ctrl_status(hwif); + outbound_ctrl = hifc_get_outbound_ctrl_status(hwif); + + if (outbound_ctrl == ENABLE_OUTBOUND && + db_ctrl == ENABLE_DOORBELL) + return 0; + + usleep_range(900, 1000); + cnt++; + } + + return -EFAULT; +} + +static void __print_selftest_reg(struct hifc_hwdev *hwdev) +{ + u32 addr, attr0, attr1; + + addr = HIFC_CSR_FUNC_ATTR1_ADDR; + attr1 = hifc_hwif_read_reg(hwdev->hwif, addr); + + if (attr1 == HIFC_PCIE_LINK_DOWN) { + sdk_err(hwdev->dev_hdl, "PCIE is link down\n"); + return; + } + + addr = HIFC_CSR_FUNC_ATTR0_ADDR; + attr0 = hifc_hwif_read_reg(hwdev->hwif, addr); + if (HIFC_AF0_GET(attr0, FUNC_TYPE) != TYPE_VF && + !HIFC_AF0_GET(attr0, PCI_INTF_IDX)) + sdk_err(hwdev->dev_hdl, "Selftest reg: 0x%08x\n", + hifc_hwif_read_reg(hwdev->hwif, + HIFC_SELFTEST_RESULT)); +} + +/** + * hifc_init_hwif - initialize the hw interface + * @hwdev: the pointer to hw device + * @cfg_reg_base: configuration base address + * Return: 0 - success, negative - failure + **/ +int hifc_init_hwif(struct hifc_hwdev *hwdev, void *cfg_reg_base, + void *intr_reg_base, u64 db_base_phy, + void *db_base, void *dwqe_mapping) +{ + struct hifc_hwif *hwif; + int err; + + hwif = kzalloc(sizeof(*hwif), GFP_KERNEL); + if (!hwif) + return -ENOMEM; + + hwdev->hwif = hwif; + hwif->pdev = hwdev->pcidev_hdl; + + hwif->cfg_regs_base = cfg_reg_base; + hwif->intr_regs_base = intr_reg_base; + + hwif->db_base_phy = db_base_phy; + hwif->db_base = db_base; + hwif->dwqe_mapping = dwqe_mapping; + init_db_area_idx(&hwif->free_db_area); + + err = wait_hwif_ready(hwdev); + if (err) { + sdk_err(hwdev->dev_hdl, "Chip status is not ready\n"); + __print_selftest_reg(hwdev); + goto hwif_ready_err; + } + + get_hwif_attr(hwif); + + err = wait_until_doorbell_and_outbound_enabled(hwif); + if (err) { + sdk_err(hwdev->dev_hdl, "Hw doorbell/outbound is disabled\n"); + goto hwif_ready_err; + } + + set_ppf(hwif); + + if (HIFC_IS_PPF(hwdev)) + set_mpf(hwif); + + get_mpf(hwif); + + disable_all_msix(hwdev); + /* disable mgmt cpu report any event */ + hifc_set_pf_status(hwdev->hwif, HIFC_PF_STATUS_INIT); + + pr_info("global_func_idx: %d, func_type: %d, host_id: %d, ppf: %d, mpf: %d\n", + hwif->attr.func_global_idx, hwif->attr.func_type, + hwif->attr.pci_intf_idx, hwif->attr.ppf_idx, + hwif->attr.mpf_idx); + + return 0; + +hwif_ready_err: + kfree(hwif); + + return err; +} + +/** + * hifc_free_hwif - free the hw interface + * @hwdev: the pointer to hw device + **/ +void hifc_free_hwif(struct hifc_hwdev *hwdev) +{ + kfree(hwdev->hwif); +} + +int hifc_dma_zalloc_coherent_align(void *dev_hdl, u64 size, u64 align, + unsigned flag, + struct hifc_dma_addr_align *mem_align) +{ + void *vaddr, *align_vaddr; + dma_addr_t paddr, align_paddr; + u64 real_size = size; + + vaddr = dma_zalloc_coherent(dev_hdl, real_size, &paddr, flag); + if (!vaddr) + return -ENOMEM; + + align_paddr = ALIGN(paddr, align); + /* align */ + if (align_paddr == paddr) { + align_vaddr = vaddr; + goto out; + } + + dma_free_coherent(dev_hdl, real_size, vaddr, paddr); + + /* realloc memory for align */ + real_size = size + align; + vaddr = dma_zalloc_coherent(dev_hdl, real_size, &paddr, flag); + if (!vaddr) + return -ENOMEM; + + align_paddr = ALIGN(paddr, align); + align_vaddr = (void *)((u64)vaddr + (align_paddr - paddr)); + +out: + mem_align->real_size = (u32)real_size; + mem_align->ori_vaddr = vaddr; + mem_align->ori_paddr = paddr; + mem_align->align_vaddr = align_vaddr; + mem_align->align_paddr = align_paddr; + + return 0; +} + +void hifc_dma_free_coherent_align(void *dev_hdl, + struct hifc_dma_addr_align *mem_align) +{ + dma_free_coherent(dev_hdl, mem_align->real_size, + mem_align->ori_vaddr, mem_align->ori_paddr); +} + +u16 hifc_global_func_id(void *hwdev) +{ + struct hifc_hwif *hwif; + + if (!hwdev) + return 0; + + hwif = ((struct hifc_hwdev *)hwdev)->hwif; + + return hwif->attr.func_global_idx; +} + +/** + * get function id from register,used by sriov hot migration process + * @hwdev: the pointer to hw device + **/ +u16 hifc_global_func_id_hw(void *hwdev) +{ + u32 addr, attr0; + struct hifc_hwdev *dev; + + dev = (struct hifc_hwdev *)hwdev; + addr = HIFC_CSR_FUNC_ATTR0_ADDR; + attr0 = hifc_hwif_read_reg(dev->hwif, addr); + + return HIFC_AF0_GET(attr0, FUNC_GLOBAL_IDX); +} + +/** + * get function id, used by sriov hot migratition process. + * @hwdev: the pointer to hw device + * @func_id: function id + **/ +int hifc_global_func_id_get(void *hwdev, u16 *func_id) +{ + *func_id = hifc_global_func_id(hwdev); + return 0; +} + +u8 hifc_pcie_itf_id(void *hwdev) +{ + struct hifc_hwif *hwif; + + if (!hwdev) + return 0; + + hwif = ((struct hifc_hwdev *)hwdev)->hwif; + + return hwif->attr.pci_intf_idx; +} +EXPORT_SYMBOL(hifc_pcie_itf_id); + +enum func_type hifc_func_type(void *hwdev) +{ + struct hifc_hwif *hwif; + + if (!hwdev) + return 0; + + hwif = ((struct hifc_hwdev *)hwdev)->hwif; + + return hwif->attr.func_type; +} + +u8 hifc_ppf_idx(void *hwdev) +{ + struct hifc_hwif *hwif; + + if (!hwdev) + return 0; + + hwif = ((struct hifc_hwdev *)hwdev)->hwif; + + return hwif->attr.ppf_idx; +} diff --git a/drivers/scsi/huawei/hifc/hifc_hwif.h b/drivers/scsi/huawei/hifc/hifc_hwif.h new file mode 100644 index 000000000000..da72253dcf5f --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_hwif.h @@ -0,0 +1,243 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#ifndef HIFC_HWIF_H +#define HIFC_HWIF_H + +#include "hifc_hwdev.h" + +#define HIFC_WAIT_DOORBELL_AND_OUTBOUND_TIMEOUT 60000 +#define HIFC_CSR_GLOBAL_BASE_ADDR 0x4000 +/* HW interface registers */ +#define HIFC_CSR_FUNC_ATTR0_ADDR 0x0 +#define HIFC_CSR_FUNC_ATTR1_ADDR 0x4 +#define HIFC_CSR_FUNC_ATTR2_ADDR 0x8 +#define HIFC_CSR_FUNC_ATTR4_ADDR 0x10 + +#define HIFC_CSR_FUNC_ATTR5_ADDR 0x14 +#define HIFC_PCI_MSIX_ENTRY_SIZE 16 +#define HIFC_PCI_MSIX_ENTRY_VECTOR_CTRL 12 +#define HIFC_PCI_MSIX_ENTRY_CTRL_MASKBIT 1 + +/* total doorbell or direct wqe size is 512kB, db num: 128, dwqe: 128*/ +#define HIFC_DB_DWQE_SIZE 0x00080000 +/* db/dwqe page size: 4K */ +#define HIFC_DB_PAGE_SIZE 0x00001000ULL +#define HIFC_DB_MAX_AREAS (HIFC_DB_DWQE_SIZE / HIFC_DB_PAGE_SIZE) + +#define HIFC_ELECTION_BASE 0x200 +#define HIFC_PPF_ELECTION_STRIDE 0x4 +#define HIFC_CSR_MAX_PORTS 4 +#define HIFC_CSR_PPF_ELECTION_ADDR \ + (HIFC_CSR_GLOBAL_BASE_ADDR + HIFC_ELECTION_BASE) + +#define HIFC_CSR_GLOBAL_MPF_ELECTION_ADDR \ + (HIFC_CSR_GLOBAL_BASE_ADDR + HIFC_ELECTION_BASE + \ + HIFC_CSR_MAX_PORTS * HIFC_PPF_ELECTION_STRIDE) +#define DB_IDX(db, db_base) \ + ((u32)(((ulong)(db) - (ulong)(db_base)) / \ + HIFC_DB_PAGE_SIZE)) + +#define HIFC_AF0_FUNC_GLOBAL_IDX_SHIFT 0 +#define HIFC_AF0_P2P_IDX_SHIFT 10 +#define HIFC_AF0_PCI_INTF_IDX_SHIFT 14 +#define HIFC_AF0_VF_IN_PF_SHIFT 16 +#define HIFC_AF0_FUNC_TYPE_SHIFT 24 +#define HIFC_AF0_FUNC_GLOBAL_IDX_MASK 0x3FF +#define HIFC_AF0_P2P_IDX_MASK 0xF +#define HIFC_AF0_PCI_INTF_IDX_MASK 0x3 +#define HIFC_AF0_VF_IN_PF_MASK 0xFF +#define HIFC_AF0_FUNC_TYPE_MASK 0x1 + +#define HIFC_AF0_GET(val, member) \ + (((val) >> HIFC_AF0_##member##_SHIFT) & HIFC_AF0_##member##_MASK) + +#define HIFC_AF1_PPF_IDX_SHIFT 0 +#define HIFC_AF1_AEQS_PER_FUNC_SHIFT 8 +#define HIFC_AF1_CEQS_PER_FUNC_SHIFT 12 +#define HIFC_AF1_IRQS_PER_FUNC_SHIFT 20 +#define HIFC_AF1_DMA_ATTR_PER_FUNC_SHIFT 24 +#define HIFC_AF1_MGMT_INIT_STATUS_SHIFT 30 +#define HIFC_AF1_PF_INIT_STATUS_SHIFT 31 + +#define HIFC_AF1_PPF_IDX_MASK 0x1F +#define HIFC_AF1_AEQS_PER_FUNC_MASK 0x3 +#define HIFC_AF1_CEQS_PER_FUNC_MASK 0x7 +#define HIFC_AF1_IRQS_PER_FUNC_MASK 0xF +#define HIFC_AF1_DMA_ATTR_PER_FUNC_MASK 0x7 +#define HIFC_AF1_MGMT_INIT_STATUS_MASK 0x1 +#define HIFC_AF1_PF_INIT_STATUS_MASK 0x1 + +#define HIFC_AF1_GET(val, member) \ + (((val) >> HIFC_AF1_##member##_SHIFT) & HIFC_AF1_##member##_MASK) + +#define HIFC_AF4_OUTBOUND_CTRL_SHIFT 0 +#define HIFC_AF4_DOORBELL_CTRL_SHIFT 1 +#define HIFC_AF4_OUTBOUND_CTRL_MASK 0x1 +#define HIFC_AF4_DOORBELL_CTRL_MASK 0x1 + +#define HIFC_AF4_GET(val, member) \ + (((val) >> HIFC_AF4_##member##_SHIFT) & HIFC_AF4_##member##_MASK) + +#define HIFC_AF4_SET(val, member) \ + (((val) & HIFC_AF4_##member##_MASK) << HIFC_AF4_##member##_SHIFT) + +#define HIFC_AF4_CLEAR(val, member) \ + ((val) & (~(HIFC_AF4_##member##_MASK << \ + HIFC_AF4_##member##_SHIFT))) + +#define HIFC_AF5_PF_STATUS_SHIFT 0 +#define HIFC_AF5_PF_STATUS_MASK 0xFFFF + +#define HIFC_AF5_SET(val, member) \ + (((val) & HIFC_AF5_##member##_MASK) << HIFC_AF5_##member##_SHIFT) + +#define HIFC_AF5_GET(val, member) \ + (((val) >> HIFC_AF5_##member##_SHIFT) & HIFC_AF5_##member##_MASK) + +#define HIFC_PPF_ELECTION_IDX_SHIFT 0 +#define HIFC_PPF_ELECTION_IDX_MASK 0x1F + +#define HIFC_PPF_ELECTION_SET(val, member) \ + (((val) & HIFC_PPF_ELECTION_##member##_MASK) << \ + HIFC_PPF_ELECTION_##member##_SHIFT) + +#define HIFC_PPF_ELECTION_GET(val, member) \ + (((val) >> HIFC_PPF_ELECTION_##member##_SHIFT) & \ + HIFC_PPF_ELECTION_##member##_MASK) + +#define HIFC_PPF_ELECTION_CLEAR(val, member) \ + ((val) & (~(HIFC_PPF_ELECTION_##member##_MASK \ + << HIFC_PPF_ELECTION_##member##_SHIFT))) + +#define HIFC_MPF_ELECTION_IDX_SHIFT 0 +#define HIFC_MPF_ELECTION_IDX_MASK 0x1F + +#define HIFC_MPF_ELECTION_SET(val, member) \ + (((val) & HIFC_MPF_ELECTION_##member##_MASK) << \ + HIFC_MPF_ELECTION_##member##_SHIFT) + +#define HIFC_MPF_ELECTION_GET(val, member) \ + (((val) >> HIFC_MPF_ELECTION_##member##_SHIFT) & \ + HIFC_MPF_ELECTION_##member##_MASK) + +#define HIFC_MPF_ELECTION_CLEAR(val, member) \ + ((val) & (~(HIFC_MPF_ELECTION_##member##_MASK \ + << HIFC_MPF_ELECTION_##member##_SHIFT))) + +#define HIFC_HWIF_NUM_AEQS(hwif) ((hwif)->attr.num_aeqs) +#define HIFC_HWIF_NUM_CEQS(hwif) ((hwif)->attr.num_ceqs) +#define HIFC_HWIF_PPF_IDX(hwif) ((hwif)->attr.ppf_idx) +#define HIFC_PCI_INTF_IDX(hwif) ((hwif)->attr.pci_intf_idx) + +#define HIFC_FUNC_TYPE(dev) ((dev)->hwif->attr.func_type) +#define HIFC_IS_PPF(dev) (HIFC_FUNC_TYPE(dev) == TYPE_PPF) + +enum hifc_pcie_nosnoop { + HIFC_PCIE_SNOOP = 0, + HIFC_PCIE_NO_SNOOP = 1, +}; + +enum hifc_pcie_tph { + HIFC_PCIE_TPH_DISABLE = 0, + HIFC_PCIE_TPH_ENABLE = 1, +}; + +enum hifc_pf_status { + HIFC_PF_STATUS_INIT = 0X0, + HIFC_PF_STATUS_ACTIVE_FLAG = 0x11, + HIFC_PF_STATUS_FLR_START_FLAG = 0x12, + HIFC_PF_STATUS_FLR_FINISH_FLAG = 0x13, +}; + +enum hifc_outbound_ctrl { + ENABLE_OUTBOUND = 0x0, + DISABLE_OUTBOUND = 0x1, +}; + +enum hifc_doorbell_ctrl { + ENABLE_DOORBELL = 0x0, + DISABLE_DOORBELL = 0x1, +}; + +struct hifc_free_db_area { + u32 db_idx[HIFC_DB_MAX_AREAS]; + u32 num_free; + u32 alloc_pos; + u32 return_pos; + /* spinlock for allocating doorbell area */ + spinlock_t idx_lock; +}; + +enum func_type { + TYPE_PF, + TYPE_VF, + TYPE_PPF, + TYPE_UNKNOWN, +}; + +struct hifc_func_attr { + u16 func_global_idx; + u8 port_to_port_idx; + u8 pci_intf_idx; + u8 vf_in_pf; + enum func_type func_type; + + u8 mpf_idx; + + u8 ppf_idx; + + u16 num_irqs; /* max: 2 ^ 15 */ + u8 num_aeqs; /* max: 2 ^ 3 */ + u8 num_ceqs; /* max: 2 ^ 7 */ + + u8 num_dma_attr; /* max: 2 ^ 6 */ +}; + +struct hifc_hwif { + u8 __iomem *cfg_regs_base; + u8 __iomem *intr_regs_base; + u64 db_base_phy; + u8 __iomem *db_base; + +#if defined(__aarch64__) + void __iomem *dwqe_mapping; +#else + struct io_mapping *dwqe_mapping; +#endif + struct hifc_free_db_area free_db_area; + struct hifc_func_attr attr; + void *pdev; +}; + +struct hifc_dma_addr_align { + u32 real_size; + void *ori_vaddr; + dma_addr_t ori_paddr; + void *align_vaddr; + dma_addr_t align_paddr; +}; + +u32 hifc_hwif_read_reg(struct hifc_hwif *hwif, u32 reg); +void hifc_hwif_write_reg(struct hifc_hwif *hwif, u32 reg, u32 val); +void hifc_set_pf_status(struct hifc_hwif *hwif, enum hifc_pf_status status); +enum hifc_pf_status hifc_get_pf_status(struct hifc_hwif *hwif); +enum hifc_doorbell_ctrl + hifc_get_doorbell_ctrl_status(struct hifc_hwif *hwif); +enum hifc_outbound_ctrl + hifc_get_outbound_ctrl_status(struct hifc_hwif *hwif); +void hifc_enable_doorbell(struct hifc_hwif *hwif); +void hifc_disable_doorbell(struct hifc_hwif *hwif); +int hifc_init_hwif(struct hifc_hwdev *hwdev, void *cfg_reg_base, + void *intr_reg_base, u64 db_base_phy, + void *db_base, void *dwqe_mapping); +void hifc_free_hwif(struct hifc_hwdev *hwdev); +int hifc_dma_zalloc_coherent_align(void *dev_hdl, u64 size, u64 align, + unsigned flag, + struct hifc_dma_addr_align *mem_align); +void hifc_dma_free_coherent_align(void *dev_hdl, + struct hifc_dma_addr_align *mem_align); +#endif diff --git a/drivers/scsi/huawei/hifc/hifc_mgmt.c b/drivers/scsi/huawei/hifc/hifc_mgmt.c new file mode 100644 index 000000000000..3f4818898e8d --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_mgmt.c @@ -0,0 +1,1426 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": [COMM]" fmt + +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/pci.h> +#include <linux/device.h> +#include <linux/spinlock.h> +#include <linux/completion.h> +#include <linux/slab.h> +#include <linux/module.h> +#include <linux/interrupt.h> +#include <linux/semaphore.h> + +#include "hifc_knl_adp.h" +#include "hifc_hw.h" +#include "hifc_hwdev.h" +#include "hifc_hwif.h" +#include "hifc_api_cmd.h" +#include "hifc_mgmt.h" +#include "hifc_eqs.h" + +#define BUF_OUT_DEFAULT_SIZE 1 +#define SEGMENT_LEN 48 +#define MGMT_MSG_MAX_SEQ_ID (ALIGN(HIFC_MSG_TO_MGMT_MAX_LEN, \ + SEGMENT_LEN) / SEGMENT_LEN) + +#define MAX_PF_MGMT_BUF_SIZE 2048UL +#define MGMT_MSG_SIZE_MIN 20 +#define MGMT_MSG_SIZE_STEP 16 +#define MGMT_MSG_RSVD_FOR_DEV 8 +#define MGMT_MSG_TIMEOUT 5000 /* millisecond */ +#define SYNC_MSG_ID_MASK 0x1FF +#define ASYNC_MSG_ID_MASK 0x1FF +#define ASYNC_MSG_FLAG 0x200 +#define MSG_NO_RESP 0xFFFF +#define MAX_MSG_SZ 2016 + +#define MSG_SZ_IS_VALID(in_size) ((in_size) <= MAX_MSG_SZ) + +#define SYNC_MSG_ID(pf_to_mgmt) ((pf_to_mgmt)->sync_msg_id) + +#define SYNC_MSG_ID_INC(pf_to_mgmt) (SYNC_MSG_ID(pf_to_mgmt) = \ + (SYNC_MSG_ID(pf_to_mgmt) + 1) & SYNC_MSG_ID_MASK) + +#define ASYNC_MSG_ID(pf_to_mgmt) ((pf_to_mgmt)->async_msg_id) + +#define ASYNC_MSG_ID_INC(pf_to_mgmt) (ASYNC_MSG_ID(pf_to_mgmt) = \ + ((ASYNC_MSG_ID(pf_to_mgmt) + 1) & ASYNC_MSG_ID_MASK) \ + | ASYNC_MSG_FLAG) + +static void pf_to_mgmt_send_event_set(struct hifc_msg_pf_to_mgmt *pf_to_mgmt, + int event_flag) +{ + spin_lock(&pf_to_mgmt->sync_event_lock); + pf_to_mgmt->event_flag = event_flag; + spin_unlock(&pf_to_mgmt->sync_event_lock); +} + +/** + * hifc_register_mgmt_msg_cb - register sync msg handler for a module + * @hwdev: the pointer to hw device + * @mod: module in the chip that this handler will handle its sync messages + * @pri_handle: pri handle function + * @callback: the handler for a sync message that will handle messages + * Return: 0 - success, negative - failure + **/ +int hifc_register_mgmt_msg_cb(void *hwdev, enum hifc_mod_type mod, + void *pri_handle, hifc_mgmt_msg_cb callback) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + + if (mod >= HIFC_MOD_HW_MAX || !hwdev) + return -EFAULT; + + pf_to_mgmt = ((struct hifc_hwdev *)hwdev)->pf_to_mgmt; + if (!pf_to_mgmt) + return -EINVAL; + + pf_to_mgmt->recv_mgmt_msg_cb[mod] = callback; + pf_to_mgmt->recv_mgmt_msg_data[mod] = pri_handle; + + set_bit(HIFC_MGMT_MSG_CB_REG, &pf_to_mgmt->mgmt_msg_cb_state[mod]); + + return 0; +} + +/** + * hifc_unregister_mgmt_msg_cb - unregister sync msg handler for a module + * @hwdev: the pointer to hw device + * @mod: module in the chip that this handler will handle its sync messages + **/ +void hifc_unregister_mgmt_msg_cb(void *hwdev, enum hifc_mod_type mod) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + + if (!hwdev || mod >= HIFC_MOD_HW_MAX) + return; + + pf_to_mgmt = ((struct hifc_hwdev *)hwdev)->pf_to_mgmt; + if (!pf_to_mgmt) + return; + + clear_bit(HIFC_MGMT_MSG_CB_REG, &pf_to_mgmt->mgmt_msg_cb_state[mod]); + + while (test_bit(HIFC_MGMT_MSG_CB_RUNNING, + &pf_to_mgmt->mgmt_msg_cb_state[mod])) + usleep_range(900, 1000); + + pf_to_mgmt->recv_mgmt_msg_cb[mod] = NULL; + pf_to_mgmt->recv_mgmt_msg_data[mod] = NULL; +} + +void hifc_comm_recv_mgmt_self_cmd_reg(void *hwdev, u8 cmd, + comm_up_self_msg_proc proc) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + u8 cmd_idx; + + if (!hwdev || !proc) + return; + + pf_to_mgmt = ((struct hifc_hwdev *)hwdev)->pf_to_mgmt; + if (!pf_to_mgmt) + return; + + cmd_idx = pf_to_mgmt->proc.cmd_num; + if (cmd_idx >= HIFC_COMM_SELF_CMD_MAX) { + sdk_err(pf_to_mgmt->hwdev->dev_hdl, + "Register recv up process failed(cmd=0x%x)\r\n", cmd); + return; + } + + pf_to_mgmt->proc.info[cmd_idx].cmd = cmd; + pf_to_mgmt->proc.info[cmd_idx].proc = proc; + + pf_to_mgmt->proc.cmd_num++; +} + +void hifc_comm_recv_up_self_cmd_unreg(void *hwdev, u8 cmd) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + u8 cmd_idx; + + if (!hwdev) + return; + + pf_to_mgmt = ((struct hifc_hwdev *)hwdev)->pf_to_mgmt; + if (!pf_to_mgmt) + return; + + cmd_idx = pf_to_mgmt->proc.cmd_num; + if (cmd_idx >= HIFC_COMM_SELF_CMD_MAX) { + sdk_err(pf_to_mgmt->hwdev->dev_hdl, + "Unregister recv up process failed(cmd=0x%x)\r\n", cmd); + return; + } + + for (cmd_idx = 0; cmd_idx < HIFC_COMM_SELF_CMD_MAX; cmd_idx++) { + if (cmd == pf_to_mgmt->proc.info[cmd_idx].cmd) { + pf_to_mgmt->proc.info[cmd_idx].cmd = 0; + pf_to_mgmt->proc.info[cmd_idx].proc = NULL; + pf_to_mgmt->proc.cmd_num--; + } + } +} + +/** + * mgmt_msg_len - calculate the total message length + * @msg_data_len: the length of the message data + * Return: the total message length + **/ +static u16 mgmt_msg_len(u16 msg_data_len) +{ + /* u64 - the size of the header */ + u16 msg_size; + + msg_size = (u16)(MGMT_MSG_RSVD_FOR_DEV + sizeof(u64) + msg_data_len); + + if (msg_size > MGMT_MSG_SIZE_MIN) + msg_size = MGMT_MSG_SIZE_MIN + + ALIGN((msg_size - MGMT_MSG_SIZE_MIN), + MGMT_MSG_SIZE_STEP); + else + msg_size = MGMT_MSG_SIZE_MIN; + + return msg_size; +} + +/** + * prepare_header - prepare the header of the message + * @pf_to_mgmt: PF to MGMT channel + * @header: pointer of the header to prepare + * @msg_len: the length of the message + * @mod: module in the chip that will get the message + * @ack_type: message ack type + * @direction: the direction of the original message + * @cmd: vmd type + * @msg_id: message id + **/ +static void prepare_header(struct hifc_msg_pf_to_mgmt *pf_to_mgmt, + u64 *header, u16 msg_len, enum hifc_mod_type mod, + enum hifc_msg_ack_type ack_type, + enum hifc_msg_direction_type direction, + enum hifc_mgmt_cmd cmd, u32 msg_id) +{ + struct hifc_hwif *hwif = pf_to_mgmt->hwdev->hwif; + + *header = HIFC_MSG_HEADER_SET(msg_len, MSG_LEN) | + HIFC_MSG_HEADER_SET(mod, MODULE) | + HIFC_MSG_HEADER_SET(msg_len, SEG_LEN) | + HIFC_MSG_HEADER_SET(ack_type, NO_ACK) | + HIFC_MSG_HEADER_SET(0, ASYNC_MGMT_TO_PF) | + HIFC_MSG_HEADER_SET(0, SEQID) | + HIFC_MSG_HEADER_SET(LAST_SEGMENT, LAST) | + HIFC_MSG_HEADER_SET(direction, DIRECTION) | + HIFC_MSG_HEADER_SET(cmd, CMD) | + HIFC_MSG_HEADER_SET(HIFC_PCI_INTF_IDX(hwif), PCI_INTF_IDX) | + HIFC_MSG_HEADER_SET(hwif->attr.port_to_port_idx, P2P_IDX) | + HIFC_MSG_HEADER_SET(msg_id, MSG_ID); +} + +static void clp_prepare_header(struct hifc_hwdev *hwdev, + u64 *header, u16 msg_len, enum hifc_mod_type mod, + enum hifc_msg_ack_type ack_type, + enum hifc_msg_direction_type direction, + enum hifc_mgmt_cmd cmd, u32 msg_id) +{ + struct hifc_hwif *hwif = hwdev->hwif; + + *header = HIFC_MSG_HEADER_SET(msg_len, MSG_LEN) | + HIFC_MSG_HEADER_SET(mod, MODULE) | + HIFC_MSG_HEADER_SET(msg_len, SEG_LEN) | + HIFC_MSG_HEADER_SET(ack_type, NO_ACK) | + HIFC_MSG_HEADER_SET(0, ASYNC_MGMT_TO_PF) | + HIFC_MSG_HEADER_SET(0, SEQID) | + HIFC_MSG_HEADER_SET(LAST_SEGMENT, LAST) | + HIFC_MSG_HEADER_SET(direction, DIRECTION) | + HIFC_MSG_HEADER_SET(cmd, CMD) | + HIFC_MSG_HEADER_SET(HIFC_PCI_INTF_IDX(hwif), PCI_INTF_IDX) | + HIFC_MSG_HEADER_SET(hwif->attr.port_to_port_idx, P2P_IDX) | + HIFC_MSG_HEADER_SET(msg_id, MSG_ID); +} + +/** + * prepare_mgmt_cmd - prepare the mgmt command + * @mgmt_cmd: pointer to the command to prepare + * @header: pointer of the header to prepare + * @msg: the data of the message + * @msg_len: the length of the message + **/ +static void prepare_mgmt_cmd(u8 *mgmt_cmd, u64 *header, const void *msg, + int msg_len) +{ + memset(mgmt_cmd, 0, MGMT_MSG_RSVD_FOR_DEV); + + mgmt_cmd += MGMT_MSG_RSVD_FOR_DEV; + memcpy(mgmt_cmd, header, sizeof(*header)); + + mgmt_cmd += sizeof(*header); + memcpy(mgmt_cmd, msg, msg_len); +} + +/** + * send_msg_to_mgmt_async - send async message + * @pf_to_mgmt: PF to MGMT channel + * @mod: module in the chip that will get the message + * @cmd: command of the message + * @msg: the data of the message + * @msg_len: the length of the message + * @direction: the direction of the original message + * @resp_msg_id: msg id to response for + * Return: 0 - success, negative - failure + **/ +static int send_msg_to_mgmt_async(struct hifc_msg_pf_to_mgmt *pf_to_mgmt, + enum hifc_mod_type mod, u8 cmd, + void *msg, u16 msg_len, + enum hifc_msg_direction_type direction, + u16 resp_msg_id) +{ + void *mgmt_cmd = pf_to_mgmt->async_msg_buf; + struct hifc_api_cmd_chain *chain; + u64 header; + u16 cmd_size = mgmt_msg_len(msg_len); + + if (!hifc_get_chip_present_flag(pf_to_mgmt->hwdev)) + return -EFAULT; + + if (direction == HIFC_MSG_RESPONSE) + prepare_header(pf_to_mgmt, &header, msg_len, mod, HIFC_MSG_ACK, + direction, cmd, resp_msg_id); + else + prepare_header(pf_to_mgmt, &header, msg_len, mod, HIFC_MSG_ACK, + direction, cmd, ASYNC_MSG_ID(pf_to_mgmt)); + + prepare_mgmt_cmd((u8 *)mgmt_cmd, &header, msg, msg_len); + + chain = pf_to_mgmt->cmd_chain[HIFC_API_CMD_WRITE_ASYNC_TO_MGMT_CPU]; + + return hifc_api_cmd_write(chain, HIFC_NODE_ID_MGMT_HOST, mgmt_cmd, + cmd_size); +} + +int hifc_pf_to_mgmt_async(void *hwdev, enum hifc_mod_type mod, + u8 cmd, void *buf_in, u16 in_size) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + void *dev = ((struct hifc_hwdev *)hwdev)->dev_hdl; + int err; + + pf_to_mgmt = ((struct hifc_hwdev *)hwdev)->pf_to_mgmt; + + /* Lock the async_msg_buf */ + spin_lock_bh(&pf_to_mgmt->async_msg_lock); + ASYNC_MSG_ID_INC(pf_to_mgmt); + + err = send_msg_to_mgmt_async(pf_to_mgmt, mod, cmd, buf_in, in_size, + HIFC_MSG_DIRECT_SEND, MSG_NO_RESP); + spin_unlock_bh(&pf_to_mgmt->async_msg_lock); + + if (err) { + sdk_err(dev, "Failed to send async mgmt msg\n"); + return err; + } + + return 0; +} + +/** + * send_msg_to_mgmt_sync - send async message + * @pf_to_mgmt: PF to MGMT channel + * @mod: module in the chip that will get the message + * @cmd: command of the message + * @msg: the msg data + * @msg_len: the msg data length + * @direction: the direction of the original message + * @resp_msg_id: msg id to response for + * Return: 0 - success, negative - failure + **/ +static int send_msg_to_mgmt_sync(struct hifc_msg_pf_to_mgmt *pf_to_mgmt, + enum hifc_mod_type mod, u8 cmd, + void *msg, u16 msg_len, + enum hifc_msg_ack_type ack_type, + enum hifc_msg_direction_type direction, + u16 resp_msg_id) +{ + void *mgmt_cmd = pf_to_mgmt->sync_msg_buf; + struct hifc_api_cmd_chain *chain; + u64 header; + u16 cmd_size = mgmt_msg_len(msg_len); + + if (!hifc_get_chip_present_flag(pf_to_mgmt->hwdev)) + return -EFAULT; + + if (direction == HIFC_MSG_RESPONSE) + prepare_header(pf_to_mgmt, &header, msg_len, mod, ack_type, + direction, cmd, resp_msg_id); + else + prepare_header(pf_to_mgmt, &header, msg_len, mod, ack_type, + direction, cmd, SYNC_MSG_ID_INC(pf_to_mgmt)); + + if (ack_type == HIFC_MSG_ACK) + pf_to_mgmt_send_event_set(pf_to_mgmt, SEND_EVENT_START); + + prepare_mgmt_cmd((u8 *)mgmt_cmd, &header, msg, msg_len); + + chain = pf_to_mgmt->cmd_chain[HIFC_API_CMD_WRITE_TO_MGMT_CPU]; + + return hifc_api_cmd_write(chain, HIFC_NODE_ID_MGMT_HOST, mgmt_cmd, + cmd_size); +} + +static inline void msg_to_mgmt_pre(enum hifc_mod_type mod, void *buf_in) +{ + struct hifc_msg_head *msg_head; + + /* set aeq fix num to 3, need to ensure response aeq id < 3*/ + if (mod == HIFC_MOD_COMM || mod == HIFC_MOD_L2NIC) { + msg_head = buf_in; + + if (msg_head->resp_aeq_num >= HIFC_MAX_AEQS) + msg_head->resp_aeq_num = 0; + } +} + +int hifc_pf_to_mgmt_sync(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, void *buf_out, + u16 *out_size, u32 timeout) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + void *dev = ((struct hifc_hwdev *)hwdev)->dev_hdl; + struct hifc_recv_msg *recv_msg; + struct completion *recv_done; + ulong timeo; + int err; + ulong ret; + + msg_to_mgmt_pre(mod, buf_in); + + pf_to_mgmt = ((struct hifc_hwdev *)hwdev)->pf_to_mgmt; + + /* Lock the sync_msg_buf */ + down(&pf_to_mgmt->sync_msg_lock); + recv_msg = &pf_to_mgmt->recv_resp_msg_from_mgmt; + recv_done = &recv_msg->recv_done; + + init_completion(recv_done); + + err = send_msg_to_mgmt_sync(pf_to_mgmt, mod, cmd, buf_in, in_size, + HIFC_MSG_ACK, HIFC_MSG_DIRECT_SEND, + MSG_NO_RESP); + if (err) { + sdk_err(dev, "Failed to send sync msg to mgmt, sync_msg_id: %d\n", + pf_to_mgmt->sync_msg_id); + pf_to_mgmt_send_event_set(pf_to_mgmt, SEND_EVENT_FAIL); + goto unlock_sync_msg; + } + + timeo = msecs_to_jiffies(timeout ? timeout : MGMT_MSG_TIMEOUT); + + ret = wait_for_completion_timeout(recv_done, timeo); + if (!ret) { + sdk_err(dev, "Mgmt response sync cmd timeout, sync_msg_id: %d\n", + pf_to_mgmt->sync_msg_id); + hifc_dump_aeq_info((struct hifc_hwdev *)hwdev); + err = -ETIMEDOUT; + pf_to_mgmt_send_event_set(pf_to_mgmt, SEND_EVENT_TIMEOUT); + goto unlock_sync_msg; + } + pf_to_mgmt_send_event_set(pf_to_mgmt, SEND_EVENT_END); + + if (!(((struct hifc_hwdev *)hwdev)->chip_present_flag)) { + up(&pf_to_mgmt->sync_msg_lock); + return -ETIMEDOUT; + } + + if (buf_out && out_size) { + if (*out_size < recv_msg->msg_len) { + sdk_err(dev, "Invalid response message length: %d for mod %d cmd %d from mgmt, should less than: %d\n", + recv_msg->msg_len, mod, cmd, *out_size); + err = -EFAULT; + goto unlock_sync_msg; + } + + if (recv_msg->msg_len) + memcpy(buf_out, recv_msg->msg, recv_msg->msg_len); + + *out_size = recv_msg->msg_len; + } + +unlock_sync_msg: + up(&pf_to_mgmt->sync_msg_lock); + + return err; +} + +static int __get_clp_reg(void *hwdev, enum clp_data_type data_type, + enum clp_reg_type reg_type, u32 *reg_addr) +{ + struct hifc_hwdev *dev = hwdev; + u32 offset; + + offset = HIFC_CLP_REG_GAP * hifc_pcie_itf_id(dev); + + switch (reg_type) { + case HIFC_CLP_BA_HOST: + *reg_addr = (data_type == HIFC_CLP_REQ_HOST) ? + HIFC_CLP_REG(REQ_SRAM_BA) : + HIFC_CLP_REG(RSP_SRAM_BA); + break; + + case HIFC_CLP_SIZE_HOST: + *reg_addr = HIFC_CLP_REG(SRAM_SIZE); + break; + + case HIFC_CLP_LEN_HOST: + *reg_addr = (data_type == HIFC_CLP_REQ_HOST) ? + HIFC_CLP_REG(REQ) : HIFC_CLP_REG(RSP); + break; + + case HIFC_CLP_START_REQ_HOST: + *reg_addr = HIFC_CLP_REG(REQ); + break; + + case HIFC_CLP_READY_RSP_HOST: + *reg_addr = HIFC_CLP_REG(RSP); + break; + + default: + *reg_addr = 0; + break; + } + if (*reg_addr == 0) + return -EINVAL; + + *reg_addr += offset; + + return 0; +} + +static inline int clp_param_valid(struct hifc_hwdev *hwdev, + enum clp_data_type data_type, + enum clp_reg_type reg_type) +{ + if (data_type == HIFC_CLP_REQ_HOST && + reg_type == HIFC_CLP_READY_RSP_HOST) + return -EINVAL; + + if (data_type == HIFC_CLP_RSP_HOST && + reg_type == HIFC_CLP_START_REQ_HOST) + return -EINVAL; + + return 0; +} + +static u32 get_clp_reg_value(struct hifc_hwdev *hwdev, + enum clp_reg_type reg_type, u32 reg_addr) +{ + u32 reg_value; + + reg_value = hifc_hwif_read_reg(hwdev->hwif, reg_addr); + + switch (reg_type) { + case HIFC_CLP_BA_HOST: + reg_value = ((reg_value >> + HIFC_CLP_OFFSET(SRAM_BASE)) & + HIFC_CLP_MASK(SRAM_BASE)); + break; + + case HIFC_CLP_SIZE_HOST: + reg_value = ((reg_value >> + HIFC_CLP_OFFSET(SRAM_SIZE)) & + HIFC_CLP_MASK(SRAM_SIZE)); + break; + + case HIFC_CLP_LEN_HOST: + reg_value = ((reg_value >> HIFC_CLP_OFFSET(LEN)) & + HIFC_CLP_MASK(LEN)); + break; + + case HIFC_CLP_START_REQ_HOST: + reg_value = ((reg_value >> HIFC_CLP_OFFSET(START)) & + HIFC_CLP_MASK(START)); + break; + + case HIFC_CLP_READY_RSP_HOST: + reg_value = ((reg_value >> HIFC_CLP_OFFSET(READY)) & + HIFC_CLP_MASK(READY)); + break; + + default: + break; + } + + return reg_value; +} + +static int hifc_read_clp_reg(struct hifc_hwdev *hwdev, + enum clp_data_type data_type, + enum clp_reg_type reg_type, u32 *read_value) +{ + u32 reg_addr; + int err; + + err = clp_param_valid(hwdev, data_type, reg_type); + if (err) + return err; + + err = __get_clp_reg(hwdev, data_type, reg_type, &reg_addr); + if (err) + return err; + + *read_value = get_clp_reg_value(hwdev, reg_type, reg_addr); + + return 0; +} + +static int __check_data_type(enum clp_data_type data_type, + enum clp_reg_type reg_type) +{ + if (data_type == HIFC_CLP_REQ_HOST && + reg_type == HIFC_CLP_READY_RSP_HOST) + return -EINVAL; + if (data_type == HIFC_CLP_RSP_HOST && + reg_type == HIFC_CLP_START_REQ_HOST) + return -EINVAL; + + return 0; +} + +static int __check_reg_value(enum clp_reg_type reg_type, u32 value) +{ + if (reg_type == HIFC_CLP_BA_HOST && + value > HIFC_CLP_SRAM_BASE_REG_MAX) + return -EINVAL; + + if (reg_type == HIFC_CLP_SIZE_HOST && + value > HIFC_CLP_SRAM_SIZE_REG_MAX) + return -EINVAL; + + if (reg_type == HIFC_CLP_LEN_HOST && + value > HIFC_CLP_LEN_REG_MAX) + return -EINVAL; + + if ((reg_type == HIFC_CLP_START_REQ_HOST || + reg_type == HIFC_CLP_READY_RSP_HOST) && + value > HIFC_CLP_START_OR_READY_REG_MAX) + return -EINVAL; + + return 0; +} + +static void hifc_write_clp_reg(struct hifc_hwdev *hwdev, + enum clp_data_type data_type, + enum clp_reg_type reg_type, u32 value) +{ + u32 reg_addr, reg_value; + + if (__check_data_type(data_type, reg_type)) + return; + + if (__check_reg_value(reg_type, value)) + return; + + if (__get_clp_reg(hwdev, data_type, reg_type, &reg_addr)) + return; + + reg_value = hifc_hwif_read_reg(hwdev->hwif, reg_addr); + + switch (reg_type) { + case HIFC_CLP_LEN_HOST: + reg_value = reg_value & + (~(HIFC_CLP_MASK(LEN) << HIFC_CLP_OFFSET(LEN))); + reg_value = reg_value | (value << HIFC_CLP_OFFSET(LEN)); + break; + + case HIFC_CLP_START_REQ_HOST: + reg_value = reg_value & + (~(HIFC_CLP_MASK(START) << + HIFC_CLP_OFFSET(START))); + reg_value = reg_value | (value << HIFC_CLP_OFFSET(START)); + break; + + case HIFC_CLP_READY_RSP_HOST: + reg_value = reg_value & + (~(HIFC_CLP_MASK(READY) << + HIFC_CLP_OFFSET(READY))); + reg_value = reg_value | (value << HIFC_CLP_OFFSET(READY)); + break; + + default: + return; + } + + hifc_hwif_write_reg(hwdev->hwif, reg_addr, reg_value); +} + +static int hifc_read_clp_data(struct hifc_hwdev *hwdev, + void *buf_out, u16 *out_size) +{ + int err; + u32 reg = HIFC_CLP_DATA(RSP); + u32 ready, delay_cnt; + u32 *ptr = (u32 *)buf_out; + u32 temp_out_size = 0; + + err = hifc_read_clp_reg(hwdev, HIFC_CLP_RSP_HOST, + HIFC_CLP_READY_RSP_HOST, &ready); + if (err) + return err; + + delay_cnt = 0; + while (ready == 0) { + usleep_range(9000, 10000); + delay_cnt++; + err = hifc_read_clp_reg(hwdev, HIFC_CLP_RSP_HOST, + HIFC_CLP_READY_RSP_HOST, &ready); + if (err || delay_cnt > HIFC_CLP_DELAY_CNT_MAX) { + sdk_err(hwdev->dev_hdl, "timeout with delay_cnt:%d\n", + delay_cnt); + return -EINVAL; + } + } + + err = hifc_read_clp_reg(hwdev, HIFC_CLP_RSP_HOST, + HIFC_CLP_LEN_HOST, &temp_out_size); + if (err) + return err; + + if (temp_out_size > HIFC_CLP_SRAM_SIZE_REG_MAX || !temp_out_size) { + sdk_err(hwdev->dev_hdl, "invalid temp_out_size:%d\n", + temp_out_size); + return -EINVAL; + } + + *out_size = (u16)(temp_out_size & 0xffff); + for (; temp_out_size > 0; temp_out_size--) { + *ptr = hifc_hwif_read_reg(hwdev->hwif, reg); + ptr++; + reg = reg + 4; + } + + hifc_write_clp_reg(hwdev, HIFC_CLP_RSP_HOST, + HIFC_CLP_READY_RSP_HOST, (u32)0x0); + hifc_write_clp_reg(hwdev, HIFC_CLP_RSP_HOST, + HIFC_CLP_LEN_HOST, (u32)0x0); + + return 0; +} + +static int hifc_write_clp_data(struct hifc_hwdev *hwdev, + void *buf_in, u16 in_size) +{ + int err; + u32 reg = HIFC_CLP_DATA(REQ); + u32 start = 1; + u32 delay_cnt = 0; + u32 *ptr = (u32 *)buf_in; + + err = hifc_read_clp_reg(hwdev, HIFC_CLP_REQ_HOST, + HIFC_CLP_START_REQ_HOST, &start); + if (err) + return err; + + while (start == 1) { + usleep_range(9000, 10000); + delay_cnt++; + err = hifc_read_clp_reg(hwdev, HIFC_CLP_REQ_HOST, + HIFC_CLP_START_REQ_HOST, &start); + if (err || delay_cnt > HIFC_CLP_DELAY_CNT_MAX) + return -EINVAL; + } + + hifc_write_clp_reg(hwdev, HIFC_CLP_REQ_HOST, + HIFC_CLP_LEN_HOST, in_size); + hifc_write_clp_reg(hwdev, HIFC_CLP_REQ_HOST, + HIFC_CLP_START_REQ_HOST, (u32)0x1); + + for (; in_size > 0; in_size--) { + hifc_hwif_write_reg(hwdev->hwif, reg, *ptr); + ptr++; + reg = reg + 4; + } + + return 0; +} + +static int hifc_check_clp_init_status(struct hifc_hwdev *hwdev) +{ + int err; + u32 reg_value = 0; + + err = hifc_read_clp_reg(hwdev, HIFC_CLP_REQ_HOST, + HIFC_CLP_BA_HOST, &reg_value); + if (err || !reg_value) { + sdk_err(hwdev->dev_hdl, "Wrong req ba value:0x%x\n", reg_value); + return -EINVAL; + } + + err = hifc_read_clp_reg(hwdev, HIFC_CLP_RSP_HOST, + HIFC_CLP_BA_HOST, &reg_value); + if (err || !reg_value) { + sdk_err(hwdev->dev_hdl, "Wrong rsp ba value:0x%x\n", reg_value); + return -EINVAL; + } + + err = hifc_read_clp_reg(hwdev, HIFC_CLP_REQ_HOST, + HIFC_CLP_SIZE_HOST, &reg_value); + if (err || !reg_value) { + sdk_err(hwdev->dev_hdl, "Wrong req size\n"); + return -EINVAL; + } + + err = hifc_read_clp_reg(hwdev, HIFC_CLP_RSP_HOST, + HIFC_CLP_SIZE_HOST, &reg_value); + if (err || !reg_value) { + sdk_err(hwdev->dev_hdl, "Wrong rsp size\n"); + return -EINVAL; + } + + return 0; +} + +static void hifc_clear_clp_data(struct hifc_hwdev *hwdev, + enum clp_data_type data_type) +{ + u32 reg = (data_type == HIFC_CLP_REQ_HOST) ? + HIFC_CLP_DATA(REQ) : HIFC_CLP_DATA(RSP); + u32 count = HIFC_CLP_INPUT_BUFFER_LEN_HOST / HIFC_CLP_DATA_UNIT_HOST; + + for (; count > 0; count--) { + hifc_hwif_write_reg(hwdev->hwif, reg, 0x0); + reg = reg + 4; + } +} + +int hifc_pf_clp_to_mgmt(void *hwdev, enum hifc_mod_type mod, u8 cmd, + const void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hifc_clp_pf_to_mgmt *clp_pf_to_mgmt; + struct hifc_hwdev *dev = hwdev; + u64 header; + u16 real_size; + u8 *clp_msg_buf; + int err; + + clp_pf_to_mgmt = ((struct hifc_hwdev *)hwdev)->clp_pf_to_mgmt; + clp_msg_buf = clp_pf_to_mgmt->clp_msg_buf; + + /*4 bytes alignment*/ + if (in_size % HIFC_CLP_DATA_UNIT_HOST) + real_size = (in_size + (u16)sizeof(header) + + HIFC_CLP_DATA_UNIT_HOST); + else + real_size = in_size + (u16)sizeof(header); + real_size = real_size / HIFC_CLP_DATA_UNIT_HOST; + + if (real_size > + (HIFC_CLP_INPUT_BUFFER_LEN_HOST / HIFC_CLP_DATA_UNIT_HOST)) { + sdk_err(dev->dev_hdl, "Invalid real_size:%d\n", real_size); + return -EINVAL; + } + down(&clp_pf_to_mgmt->clp_msg_lock); + + err = hifc_check_clp_init_status(dev); + if (err) { + sdk_err(dev->dev_hdl, "Check clp init status failed\n"); + up(&clp_pf_to_mgmt->clp_msg_lock); + return err; + } + + hifc_clear_clp_data(dev, HIFC_CLP_RSP_HOST); + hifc_write_clp_reg(dev, HIFC_CLP_RSP_HOST, + HIFC_CLP_READY_RSP_HOST, 0x0); + + /*Send request*/ + memset(clp_msg_buf, 0x0, HIFC_CLP_INPUT_BUFFER_LEN_HOST); + clp_prepare_header(dev, &header, in_size, mod, 0, 0, cmd, 0); + + memcpy(clp_msg_buf, &header, sizeof(header)); + clp_msg_buf += sizeof(header); + memcpy(clp_msg_buf, buf_in, in_size); + + clp_msg_buf = clp_pf_to_mgmt->clp_msg_buf; + + hifc_clear_clp_data(dev, HIFC_CLP_REQ_HOST); + err = hifc_write_clp_data(hwdev, + clp_pf_to_mgmt->clp_msg_buf, real_size); + if (err) { + sdk_err(dev->dev_hdl, "Send clp request failed\n"); + up(&clp_pf_to_mgmt->clp_msg_lock); + return -EINVAL; + } + + /*Get response*/ + clp_msg_buf = clp_pf_to_mgmt->clp_msg_buf; + memset(clp_msg_buf, 0x0, HIFC_CLP_INPUT_BUFFER_LEN_HOST); + err = hifc_read_clp_data(hwdev, clp_msg_buf, &real_size); + hifc_clear_clp_data(dev, HIFC_CLP_RSP_HOST); + if (err) { + sdk_err(dev->dev_hdl, "Read clp response failed\n"); + up(&clp_pf_to_mgmt->clp_msg_lock); + return -EINVAL; + } + + real_size = (u16)((real_size * HIFC_CLP_DATA_UNIT_HOST) & 0xffff); + if ((real_size <= sizeof(header)) || + (real_size > HIFC_CLP_INPUT_BUFFER_LEN_HOST)) { + sdk_err(dev->dev_hdl, "Invalid response size:%d", real_size); + up(&clp_pf_to_mgmt->clp_msg_lock); + return -EINVAL; + } + real_size = real_size - sizeof(header); + if (real_size != *out_size) { + sdk_err(dev->dev_hdl, "Invalid real_size:%d, out_size:%d\n", + real_size, *out_size); + up(&clp_pf_to_mgmt->clp_msg_lock); + return -EINVAL; + } + + memcpy(buf_out, (clp_msg_buf + sizeof(header)), real_size); + up(&clp_pf_to_mgmt->clp_msg_lock); + + return 0; +} + +/* This function is only used by txrx flush */ +int hifc_pf_to_mgmt_no_ack(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + void *dev = ((struct hifc_hwdev *)hwdev)->dev_hdl; + int err = -EINVAL; + + if (!hifc_is_hwdev_mod_inited(hwdev, HIFC_HWDEV_MGMT_INITED)) { + sdk_err(dev, "Mgmt module not initialized\n"); + return -EINVAL; + } + + pf_to_mgmt = ((struct hifc_hwdev *)hwdev)->pf_to_mgmt; + + if (!MSG_SZ_IS_VALID(in_size)) { + sdk_err(dev, "Mgmt msg buffer size: %d is not valid\n", + in_size); + return -EINVAL; + } + + if (!(((struct hifc_hwdev *)hwdev)->chip_present_flag)) + return -EPERM; + + /* Lock the sync_msg_buf */ + down(&pf_to_mgmt->sync_msg_lock); + + err = send_msg_to_mgmt_sync(pf_to_mgmt, mod, cmd, buf_in, in_size, + HIFC_MSG_NO_ACK, HIFC_MSG_DIRECT_SEND, + MSG_NO_RESP); + + up(&pf_to_mgmt->sync_msg_lock); + + return err; +} + +/** + * api cmd write or read bypass defaut use poll, if want to use aeq interrupt, + * please set wb_trigger_aeqe to 1 + **/ +int hifc_api_cmd_write_nack(void *hwdev, u8 dest, void *cmd, u16 size) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + struct hifc_api_cmd_chain *chain; + + if (!hwdev || !size || !cmd) + return -EINVAL; + + if (!hifc_is_hwdev_mod_inited(hwdev, HIFC_HWDEV_MGMT_INITED) || + hifc_get_mgmt_channel_status(hwdev)) + return -EPERM; + + pf_to_mgmt = ((struct hifc_hwdev *)hwdev)->pf_to_mgmt; + chain = pf_to_mgmt->cmd_chain[HIFC_API_CMD_POLL_WRITE]; + + if (!(((struct hifc_hwdev *)hwdev)->chip_present_flag)) + return -EPERM; + + return hifc_api_cmd_write(chain, dest, cmd, size); +} + +int hifc_api_cmd_read_ack(void *hwdev, u8 dest, void *cmd, u16 size, void *ack, + u16 ack_size) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + struct hifc_api_cmd_chain *chain; + + if (!hwdev || !cmd || (ack_size && !ack)) + return -EINVAL; + + if (!hifc_is_hwdev_mod_inited(hwdev, HIFC_HWDEV_MGMT_INITED) || + hifc_get_mgmt_channel_status(hwdev)) + return -EPERM; + + pf_to_mgmt = ((struct hifc_hwdev *)hwdev)->pf_to_mgmt; + chain = pf_to_mgmt->cmd_chain[HIFC_API_CMD_POLL_READ]; + + if (!(((struct hifc_hwdev *)hwdev)->chip_present_flag)) + return -EPERM; + + return hifc_api_cmd_read(chain, dest, cmd, size, ack, ack_size); +} + +static void __send_mgmt_ack(struct hifc_msg_pf_to_mgmt *pf_to_mgmt, + enum hifc_mod_type mod, u8 cmd, void *buf_in, + u16 in_size, u16 msg_id) +{ + u16 buf_size; + + if (!in_size) + buf_size = BUF_OUT_DEFAULT_SIZE; + else + buf_size = in_size; + + spin_lock_bh(&pf_to_mgmt->async_msg_lock); + /* MGMT sent sync msg, send the response */ + send_msg_to_mgmt_async(pf_to_mgmt, mod, cmd, + buf_in, buf_size, HIFC_MSG_RESPONSE, + msg_id); + spin_unlock_bh(&pf_to_mgmt->async_msg_lock); +} + +/** + * mgmt_recv_msg_handler - handler for message from mgmt cpu + * @pf_to_mgmt: PF to MGMT channel + * @recv_msg: received message details + **/ +static void mgmt_recv_msg_handler(struct hifc_msg_pf_to_mgmt *pf_to_mgmt, + enum hifc_mod_type mod, u8 cmd, void *buf_in, + u16 in_size, u16 msg_id, int need_resp) +{ + void *dev = pf_to_mgmt->hwdev->dev_hdl; + void *buf_out = pf_to_mgmt->mgmt_ack_buf; + enum hifc_mod_type tmp_mod = mod; + bool ack_first = false; + u16 out_size = 0; + + memset(buf_out, 0, MAX_PF_MGMT_BUF_SIZE); + + if (mod >= HIFC_MOD_HW_MAX) { + sdk_warn(dev, "Receive illegal message from mgmt cpu, mod = %d\n", + mod); + goto resp; + } + + set_bit(HIFC_MGMT_MSG_CB_RUNNING, + &pf_to_mgmt->mgmt_msg_cb_state[tmp_mod]); + + if (!pf_to_mgmt->recv_mgmt_msg_cb[mod] || + !test_bit(HIFC_MGMT_MSG_CB_REG, + &pf_to_mgmt->mgmt_msg_cb_state[tmp_mod])) { + sdk_warn(dev, "Receive mgmt callback is null, mod = %d\n", + mod); + clear_bit(HIFC_MGMT_MSG_CB_RUNNING, + &pf_to_mgmt->mgmt_msg_cb_state[tmp_mod]); + goto resp; + } + + ack_first = hifc_mgmt_event_ack_first(mod, cmd); + if (ack_first && need_resp) { + /* send ack to mgmt first to avoid command timeout in + * mgmt(100ms in mgmt); + * mgmt to host command don't need any response data from host, + * just need ack from host + */ + __send_mgmt_ack(pf_to_mgmt, mod, cmd, buf_out, in_size, msg_id); + } + + pf_to_mgmt->recv_mgmt_msg_cb[tmp_mod](pf_to_mgmt->hwdev, + pf_to_mgmt->recv_mgmt_msg_data[tmp_mod], + cmd, buf_in, in_size, + buf_out, &out_size); + + clear_bit(HIFC_MGMT_MSG_CB_RUNNING, + &pf_to_mgmt->mgmt_msg_cb_state[tmp_mod]); + +resp: + if (!ack_first && need_resp) + __send_mgmt_ack(pf_to_mgmt, mod, cmd, buf_out, out_size, + msg_id); +} + +/** + * mgmt_resp_msg_handler - handler for response message from mgmt cpu + * @pf_to_mgmt: PF to MGMT channel + * @recv_msg: received message details + **/ +static void mgmt_resp_msg_handler(struct hifc_msg_pf_to_mgmt *pf_to_mgmt, + struct hifc_recv_msg *recv_msg) +{ + void *dev = pf_to_mgmt->hwdev->dev_hdl; + + /* delete async msg */ + if (recv_msg->msg_id & ASYNC_MSG_FLAG) + return; + + spin_lock(&pf_to_mgmt->sync_event_lock); + if (recv_msg->msg_id == pf_to_mgmt->sync_msg_id && + pf_to_mgmt->event_flag == SEND_EVENT_START) { + complete(&recv_msg->recv_done); + } else if (recv_msg->msg_id != pf_to_mgmt->sync_msg_id) { + sdk_err(dev, "Send msg id(0x%x) recv msg id(0x%x) dismatch, event state=%d\n", + pf_to_mgmt->sync_msg_id, recv_msg->msg_id, + pf_to_mgmt->event_flag); + } else { + sdk_err(dev, "Wait timeout, send msg id(0x%x) recv msg id(0x%x), event state=%d!\n", + pf_to_mgmt->sync_msg_id, recv_msg->msg_id, + pf_to_mgmt->event_flag); + } + spin_unlock(&pf_to_mgmt->sync_event_lock); +} + +static void recv_mgmt_msg_work_handler(struct work_struct *work) +{ + struct hifc_mgmt_msg_handle_work *mgmt_work = + container_of(work, struct hifc_mgmt_msg_handle_work, work); + + mgmt_recv_msg_handler(mgmt_work->pf_to_mgmt, mgmt_work->mod, + mgmt_work->cmd, mgmt_work->msg, + mgmt_work->msg_len, mgmt_work->msg_id, + !mgmt_work->async_mgmt_to_pf); + + kfree(mgmt_work->msg); + kfree(mgmt_work); +} + +static bool check_mgmt_seq_id_and_seg_len(struct hifc_recv_msg *recv_msg, + u8 seq_id, u8 seg_len) +{ + if (seq_id > MGMT_MSG_MAX_SEQ_ID || seg_len > SEGMENT_LEN) + return false; + + if (seq_id == 0) { + recv_msg->seq_id = seq_id; + } else { + if (seq_id != recv_msg->seq_id + 1) + return false; + recv_msg->seq_id = seq_id; + } + + return true; +} + +/** + * recv_mgmt_msg_handler - handler a message from mgmt cpu + * @pf_to_mgmt: PF to MGMT channel + * @header: the header of the message + * @recv_msg: received message details + **/ +static void recv_mgmt_msg_handler(struct hifc_msg_pf_to_mgmt *pf_to_mgmt, + u8 *header, struct hifc_recv_msg *recv_msg) +{ + struct hifc_mgmt_msg_handle_work *mgmt_work; + u64 mbox_header = *((u64 *)header); + void *msg_body = header + sizeof(mbox_header); + u8 seq_id, seq_len; + u32 offset; + u64 dir; + + /* Don't need to get anything from hw when cmd is async */ + dir = HIFC_MSG_HEADER_GET(mbox_header, DIRECTION); + if (dir == HIFC_MSG_RESPONSE && + HIFC_MSG_HEADER_GET(mbox_header, MSG_ID) & ASYNC_MSG_FLAG) + return; + + seq_len = HIFC_MSG_HEADER_GET(mbox_header, SEG_LEN); + seq_id = HIFC_MSG_HEADER_GET(mbox_header, SEQID); + + if (!check_mgmt_seq_id_and_seg_len(recv_msg, seq_id, seq_len)) { + sdk_err(pf_to_mgmt->hwdev->dev_hdl, + "Mgmt msg sequence id and segment length check fail, front seq_id: 0x%x, current seq_id: 0x%x, seg len: 0x%x\n", + recv_msg->seq_id, seq_id, seq_len); + /* set seq_id to invalid seq_id */ + recv_msg->seq_id = MGMT_MSG_MAX_SEQ_ID; + return; + } + + offset = seq_id * SEGMENT_LEN; + memcpy((u8 *)recv_msg->msg + offset, msg_body, seq_len); + + if (!HIFC_MSG_HEADER_GET(mbox_header, LAST)) + return; + + recv_msg->cmd = HIFC_MSG_HEADER_GET(mbox_header, CMD); + recv_msg->mod = HIFC_MSG_HEADER_GET(mbox_header, MODULE); + recv_msg->async_mgmt_to_pf = HIFC_MSG_HEADER_GET(mbox_header, + ASYNC_MGMT_TO_PF); + recv_msg->msg_len = HIFC_MSG_HEADER_GET(mbox_header, MSG_LEN); + recv_msg->msg_id = HIFC_MSG_HEADER_GET(mbox_header, MSG_ID); + recv_msg->seq_id = MGMT_MSG_MAX_SEQ_ID; + + if (HIFC_MSG_HEADER_GET(mbox_header, DIRECTION) == + HIFC_MSG_RESPONSE) { + mgmt_resp_msg_handler(pf_to_mgmt, recv_msg); + return; + } + + mgmt_work = kzalloc(sizeof(*mgmt_work), GFP_KERNEL); + if (!mgmt_work) { + sdk_err(pf_to_mgmt->hwdev->dev_hdl, + "Allocate mgmt work memory failed\n"); + return; + } + + if (recv_msg->msg_len) { + mgmt_work->msg = kzalloc(recv_msg->msg_len, GFP_KERNEL); + if (!mgmt_work->msg) { + sdk_err(pf_to_mgmt->hwdev->dev_hdl, "Allocate mgmt msg memory failed\n"); + kfree(mgmt_work); + return; + } + } + + mgmt_work->pf_to_mgmt = pf_to_mgmt; + mgmt_work->msg_len = recv_msg->msg_len; + memcpy(mgmt_work->msg, recv_msg->msg, recv_msg->msg_len); + mgmt_work->msg_id = recv_msg->msg_id; + mgmt_work->mod = recv_msg->mod; + mgmt_work->cmd = recv_msg->cmd; + mgmt_work->async_mgmt_to_pf = recv_msg->async_mgmt_to_pf; + + INIT_WORK(&mgmt_work->work, recv_mgmt_msg_work_handler); + queue_work(pf_to_mgmt->workq, &mgmt_work->work); +} + +/** + * hifc_mgmt_msg_aeqe_handler - handler for a mgmt message event + * @hwdev: the pointer to hw device + * @header: the header of the message + * @size: unused + **/ +void hifc_mgmt_msg_aeqe_handler(void *hwdev, u8 *header, u8 size) +{ + struct hifc_hwdev *dev = (struct hifc_hwdev *)hwdev; + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + struct hifc_recv_msg *recv_msg; + bool is_send_dir = false; + + pf_to_mgmt = dev->pf_to_mgmt; + + is_send_dir = (HIFC_MSG_HEADER_GET(*(u64 *)header, DIRECTION) == + HIFC_MSG_DIRECT_SEND) ? true : false; + + recv_msg = is_send_dir ? &pf_to_mgmt->recv_msg_from_mgmt : + &pf_to_mgmt->recv_resp_msg_from_mgmt; + + recv_mgmt_msg_handler(pf_to_mgmt, header, recv_msg); +} + +/** + * alloc_recv_msg - allocate received message memory + * @recv_msg: pointer that will hold the allocated data + * Return: 0 - success, negative - failure + **/ +static int alloc_recv_msg(struct hifc_recv_msg *recv_msg) +{ + recv_msg->seq_id = MGMT_MSG_MAX_SEQ_ID; + + recv_msg->msg = kzalloc(MAX_PF_MGMT_BUF_SIZE, GFP_KERNEL); + if (!recv_msg->msg) + return -ENOMEM; + + return 0; +} + +/** + * free_recv_msg - free received message memory + * @recv_msg: pointer that holds the allocated data + **/ +static void free_recv_msg(struct hifc_recv_msg *recv_msg) +{ + kfree(recv_msg->msg); +} + +/** + * alloc_msg_buf - allocate all the message buffers of PF to MGMT channel + * @pf_to_mgmt: PF to MGMT channel + * Return: 0 - success, negative - failure + **/ +static int alloc_msg_buf(struct hifc_msg_pf_to_mgmt *pf_to_mgmt) +{ + int err; + void *dev = pf_to_mgmt->hwdev->dev_hdl; + + err = alloc_recv_msg(&pf_to_mgmt->recv_msg_from_mgmt); + if (err) { + sdk_err(dev, "Failed to allocate recv msg\n"); + return err; + } + + err = alloc_recv_msg(&pf_to_mgmt->recv_resp_msg_from_mgmt); + if (err) { + sdk_err(dev, "Failed to allocate resp recv msg\n"); + goto alloc_msg_for_resp_err; + } + + pf_to_mgmt->async_msg_buf = kzalloc(MAX_PF_MGMT_BUF_SIZE, GFP_KERNEL); + if (!pf_to_mgmt->async_msg_buf) { + err = -ENOMEM; + goto async_msg_buf_err; + } + + pf_to_mgmt->sync_msg_buf = kzalloc(MAX_PF_MGMT_BUF_SIZE, GFP_KERNEL); + if (!pf_to_mgmt->sync_msg_buf) { + err = -ENOMEM; + goto sync_msg_buf_err; + } + + pf_to_mgmt->mgmt_ack_buf = kzalloc(MAX_PF_MGMT_BUF_SIZE, GFP_KERNEL); + if (!pf_to_mgmt->mgmt_ack_buf) { + err = -ENOMEM; + goto ack_msg_buf_err; + } + + return 0; + +ack_msg_buf_err: + kfree(pf_to_mgmt->sync_msg_buf); + +sync_msg_buf_err: + kfree(pf_to_mgmt->async_msg_buf); + +async_msg_buf_err: + free_recv_msg(&pf_to_mgmt->recv_resp_msg_from_mgmt); + +alloc_msg_for_resp_err: + free_recv_msg(&pf_to_mgmt->recv_msg_from_mgmt); + return err; +} + +/** + * free_msg_buf - free all the message buffers of PF to MGMT channel + * @pf_to_mgmt: PF to MGMT channel + **/ +static void free_msg_buf(struct hifc_msg_pf_to_mgmt *pf_to_mgmt) +{ + kfree(pf_to_mgmt->mgmt_ack_buf); + kfree(pf_to_mgmt->sync_msg_buf); + kfree(pf_to_mgmt->async_msg_buf); + + free_recv_msg(&pf_to_mgmt->recv_resp_msg_from_mgmt); + free_recv_msg(&pf_to_mgmt->recv_msg_from_mgmt); +} + +/** + * hifc_pf_to_mgmt_init - initialize PF to MGMT channel + * @hwdev: the pointer to hw device + * Return: 0 - success, negative - failure + **/ +int hifc_pf_to_mgmt_init(struct hifc_hwdev *hwdev) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + void *dev = hwdev->dev_hdl; + int err; + + pf_to_mgmt = kzalloc(sizeof(*pf_to_mgmt), GFP_KERNEL); + if (!pf_to_mgmt) + return -ENOMEM; + + hwdev->pf_to_mgmt = pf_to_mgmt; + pf_to_mgmt->hwdev = hwdev; + spin_lock_init(&pf_to_mgmt->async_msg_lock); + spin_lock_init(&pf_to_mgmt->sync_event_lock); + sema_init(&pf_to_mgmt->sync_msg_lock, 1); + pf_to_mgmt->workq = create_singlethread_workqueue(HIFC_MGMT_WQ_NAME); + if (!pf_to_mgmt->workq) { + sdk_err(dev, "Failed to initialize MGMT workqueue\n"); + err = -ENOMEM; + goto create_mgmt_workq_err; + } + + err = alloc_msg_buf(pf_to_mgmt); + if (err) { + sdk_err(dev, "Failed to allocate msg buffers\n"); + goto alloc_msg_buf_err; + } + + err = hifc_api_cmd_init(hwdev, pf_to_mgmt->cmd_chain); + if (err) { + sdk_err(dev, "Failed to init the api cmd chains\n"); + goto api_cmd_init_err; + } + + return 0; + +api_cmd_init_err: + free_msg_buf(pf_to_mgmt); + +alloc_msg_buf_err: + destroy_workqueue(pf_to_mgmt->workq); + +create_mgmt_workq_err: + kfree(pf_to_mgmt); + + return err; +} + +/** + * hifc_pf_to_mgmt_free - free PF to MGMT channel + * @hwdev: the pointer to hw device + **/ +void hifc_pf_to_mgmt_free(struct hifc_hwdev *hwdev) +{ + struct hifc_msg_pf_to_mgmt *pf_to_mgmt = hwdev->pf_to_mgmt; + + /* destroy workqueue before free related pf_to_mgmt resources in case of + * illegal resource access + */ + destroy_workqueue(pf_to_mgmt->workq); + hifc_api_cmd_free(pf_to_mgmt->cmd_chain); + free_msg_buf(pf_to_mgmt); + kfree(pf_to_mgmt); +} + +void hifc_flush_mgmt_workq(void *hwdev) +{ + struct hifc_hwdev *dev = (struct hifc_hwdev *)hwdev; + + flush_workqueue(dev->aeqs->workq); + + if (hifc_func_type(dev) != TYPE_VF && + hifc_is_hwdev_mod_inited(hwdev, HIFC_HWDEV_MGMT_INITED)) + flush_workqueue(dev->pf_to_mgmt->workq); +} + +int hifc_clp_pf_to_mgmt_init(struct hifc_hwdev *hwdev) +{ + struct hifc_clp_pf_to_mgmt *clp_pf_to_mgmt; + + clp_pf_to_mgmt = kzalloc(sizeof(*clp_pf_to_mgmt), GFP_KERNEL); + if (!clp_pf_to_mgmt) + return -ENOMEM; + + clp_pf_to_mgmt->clp_msg_buf = kzalloc(HIFC_CLP_INPUT_BUFFER_LEN_HOST, + GFP_KERNEL); + if (!clp_pf_to_mgmt->clp_msg_buf) { + kfree(clp_pf_to_mgmt); + return -ENOMEM; + } + sema_init(&clp_pf_to_mgmt->clp_msg_lock, 1); + + hwdev->clp_pf_to_mgmt = clp_pf_to_mgmt; + + return 0; +} + +void hifc_clp_pf_to_mgmt_free(struct hifc_hwdev *hwdev) +{ + struct hifc_clp_pf_to_mgmt *clp_pf_to_mgmt = hwdev->clp_pf_to_mgmt; + + kfree(clp_pf_to_mgmt->clp_msg_buf); + kfree(clp_pf_to_mgmt); +} + diff --git a/drivers/scsi/huawei/hifc/hifc_mgmt.h b/drivers/scsi/huawei/hifc/hifc_mgmt.h new file mode 100644 index 000000000000..2adcfe2968c1 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_mgmt.h @@ -0,0 +1,407 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#ifndef HIFC_MGMT_H_ +#define HIFC_MGMT_H_ + +#define HIFC_MSG_HEADER_MSG_LEN_SHIFT 0 +#define HIFC_MSG_HEADER_MODULE_SHIFT 11 +#define HIFC_MSG_HEADER_SEG_LEN_SHIFT 16 +#define HIFC_MSG_HEADER_NO_ACK_SHIFT 22 +#define HIFC_MSG_HEADER_ASYNC_MGMT_TO_PF_SHIFT 23 +#define HIFC_MSG_HEADER_SEQID_SHIFT 24 +#define HIFC_MSG_HEADER_LAST_SHIFT 30 +#define HIFC_MSG_HEADER_DIRECTION_SHIFT 31 +#define HIFC_MSG_HEADER_CMD_SHIFT 32 +#define HIFC_MSG_HEADER_PCI_INTF_IDX_SHIFT 48 +#define HIFC_MSG_HEADER_P2P_IDX_SHIFT 50 +#define HIFC_MSG_HEADER_MSG_ID_SHIFT 54 + +#define HIFC_MSG_HEADER_MSG_LEN_MASK 0x7FF +#define HIFC_MSG_HEADER_MODULE_MASK 0x1F +#define HIFC_MSG_HEADER_SEG_LEN_MASK 0x3F +#define HIFC_MSG_HEADER_NO_ACK_MASK 0x1 +#define HIFC_MSG_HEADER_ASYNC_MGMT_TO_PF_MASK 0x1 +#define HIFC_MSG_HEADER_SEQID_MASK 0x3F +#define HIFC_MSG_HEADER_LAST_MASK 0x1 +#define HIFC_MSG_HEADER_DIRECTION_MASK 0x1 +#define HIFC_MSG_HEADER_CMD_MASK 0xFF +#define HIFC_MSG_HEADER_PCI_INTF_IDX_MASK 0x3 +#define HIFC_MSG_HEADER_P2P_IDX_MASK 0xF +#define HIFC_MSG_HEADER_MSG_ID_MASK 0x3FF + +#define HIFC_MSG_HEADER_GET(val, member) \ + (((val) >> HIFC_MSG_HEADER_##member##_SHIFT) & \ + HIFC_MSG_HEADER_##member##_MASK) + +#define HIFC_MSG_HEADER_SET(val, member) \ + ((u64)((val) & HIFC_MSG_HEADER_##member##_MASK) << \ + HIFC_MSG_HEADER_##member##_SHIFT) + +#define HIFC_MGMT_WQ_NAME "hifc_mgmt" + +/*CLP*/ +enum clp_data_type { + HIFC_CLP_REQ_HOST = 0, + HIFC_CLP_RSP_HOST = 1 +}; + +enum clp_reg_type { + HIFC_CLP_BA_HOST = 0, + HIFC_CLP_SIZE_HOST = 1, + HIFC_CLP_LEN_HOST = 2, + HIFC_CLP_START_REQ_HOST = 3, + HIFC_CLP_READY_RSP_HOST = 4 +}; + +/* cmd of mgmt CPU message for HW module */ +enum hifc_mgmt_cmd { + HIFC_MGMT_CMD_RESET_MGMT = 0x0, + HIFC_MGMT_CMD_START_FLR = 0x1, + HIFC_MGMT_CMD_FLUSH_DOORBELL = 0x2, + HIFC_MGMT_CMD_CMDQ_CTXT_SET = 0x10, + HIFC_MGMT_CMD_VAT_SET = 0x12, + HIFC_MGMT_CMD_L2NIC_SQ_CI_ATTR_SET = 0x14, + HIFC_MGMT_CMD_PPF_TMR_SET = 0x22, + HIFC_MGMT_CMD_PPF_HT_GPA_SET = 0x23, + HIFC_MGMT_CMD_RES_STATE_SET = 0x24, + HIFC_MGMT_CMD_FUNC_TMR_BITMAT_SET = 0x32, + HIFC_MGMT_CMD_CEQ_CTRL_REG_WR_BY_UP = 0x33, + HIFC_MGMT_CMD_MSI_CTRL_REG_WR_BY_UP, + HIFC_MGMT_CMD_MSI_CTRL_REG_RD_BY_UP, + HIFC_MGMT_CMD_FAULT_REPORT = 0x37, + HIFC_MGMT_CMD_HEART_LOST_REPORT = 0x38, + HIFC_MGMT_CMD_SYNC_TIME = 0x46, + HIFC_MGMT_CMD_REG_READ = 0x48, + HIFC_MGMT_CMD_L2NIC_RESET = 0x4b, + HIFC_MGMT_CMD_ACTIVATE_FW = 0x4F, + HIFC_MGMT_CMD_PAGESIZE_SET = 0x50, + HIFC_MGMT_CMD_GET_BOARD_INFO = 0x52, + HIFC_MGMT_CMD_WATCHDOG_INFO = 0x56, + HIFC_MGMT_CMD_FMW_ACT_NTC = 0x57, + HIFC_MGMT_CMD_PCIE_DFX_NTC = 0x65, + HIFC_MGMT_CMD_PCIE_DFX_GET = 0x66, + HIFC_MGMT_CMD_GET_HOST_INFO = 0x67, + HIFC_MGMT_CMD_GET_PHY_INIT_STATUS = 0x6A, + HIFC_MGMT_CMD_HEARTBEAT_EVENT = 0x6C, +}; + +#define HIFC_CLP_REG_GAP 0x20 +#define HIFC_CLP_INPUT_BUFFER_LEN_HOST 2048UL +#define HIFC_CLP_OUTPUT_BUFFER_LEN_HOST 2048UL +#define HIFC_CLP_DATA_UNIT_HOST 4UL +#define HIFC_BAR01_GLOABAL_CTL_OFFSET 0x4000 +#define HIFC_BAR01_CLP_OFFSET 0x5000 + +#define HIFC_CLP_SRAM_SIZE_REG (HIFC_BAR01_GLOABAL_CTL_OFFSET + 0x220) +#define HIFC_CLP_REQ_SRAM_BA_REG (HIFC_BAR01_GLOABAL_CTL_OFFSET + 0x224) +#define HIFC_CLP_RSP_SRAM_BA_REG (HIFC_BAR01_GLOABAL_CTL_OFFSET + 0x228) +#define HIFC_CLP_REQ_REG (HIFC_BAR01_GLOABAL_CTL_OFFSET + 0x22c) +#define HIFC_CLP_RSP_REG (HIFC_BAR01_GLOABAL_CTL_OFFSET + 0x230) +#define HIFC_CLP_REG(member) (HIFC_CLP_##member##_REG) + +#define HIFC_CLP_REQ_DATA (HIFC_BAR01_CLP_OFFSET) +#define HIFC_CLP_RSP_DATA (HIFC_BAR01_CLP_OFFSET + 0x1000) +#define HIFC_CLP_DATA(member) (HIFC_CLP_##member##_DATA) + +#define HIFC_CLP_SRAM_SIZE_OFFSET 16 +#define HIFC_CLP_SRAM_BASE_OFFSET 0 +#define HIFC_CLP_LEN_OFFSET 0 +#define HIFC_CLP_START_OFFSET 31 +#define HIFC_CLP_READY_OFFSET 31 +#define HIFC_CLP_OFFSET(member) (HIFC_CLP_##member##_OFFSET) + +#define HIFC_CLP_SRAM_SIZE_BIT_LEN 0x7ffUL +#define HIFC_CLP_SRAM_BASE_BIT_LEN 0x7ffffffUL +#define HIFC_CLP_LEN_BIT_LEN 0x7ffUL +#define HIFC_CLP_START_BIT_LEN 0x1UL +#define HIFC_CLP_READY_BIT_LEN 0x1UL +#define HIFC_CLP_MASK(member) (HIFC_CLP_##member##_BIT_LEN) + +#define HIFC_CLP_DELAY_CNT_MAX 200UL +#define HIFC_CLP_SRAM_SIZE_REG_MAX 0x3ff +#define HIFC_CLP_SRAM_BASE_REG_MAX 0x7ffffff +#define HIFC_CLP_LEN_REG_MAX 0x3ff +#define HIFC_CLP_START_OR_READY_REG_MAX 0x1 +#define HIFC_MGMT_CMD_UNSUPPORTED 0xFF + +enum hifc_msg_direction_type { + HIFC_MSG_DIRECT_SEND = 0, + HIFC_MSG_RESPONSE = 1 +}; + +enum hifc_msg_segment_type { + NOT_LAST_SEGMENT = 0, + LAST_SEGMENT = 1, +}; + +enum hifc_mgmt_msg_type { + ASYNC_MGMT_MSG = 0, + SYNC_MGMT_MSG = 1, +}; + +enum hifc_msg_ack_type { + HIFC_MSG_ACK = 0, + HIFC_MSG_NO_ACK = 1, +}; + +struct hifc_recv_msg { + void *msg; + + struct completion recv_done; + + u16 msg_len; + enum hifc_mod_type mod; + u8 cmd; + u8 seq_id; + u16 msg_id; + int async_mgmt_to_pf; +}; + +struct hifc_msg_head { + u8 status; + u8 version; + u8 resp_aeq_num; + u8 rsvd0[5]; +}; + +#define HIFC_COMM_SELF_CMD_MAX 8 + +struct comm_up_self_msg_sub_info { + u8 cmd; + comm_up_self_msg_proc proc; +}; + +struct comm_up_self_msg_info { + u8 cmd_num; + struct comm_up_self_msg_sub_info info[HIFC_COMM_SELF_CMD_MAX]; +}; + +enum comm_pf_to_mgmt_event_state { + SEND_EVENT_UNINIT = 0, + SEND_EVENT_START, + SEND_EVENT_FAIL, + SEND_EVENT_TIMEOUT, + SEND_EVENT_END, +}; + +enum hifc_mgmt_msg_cb_state { + HIFC_MGMT_MSG_CB_REG = 0, + HIFC_MGMT_MSG_CB_RUNNING, +}; + +struct hifc_clp_pf_to_mgmt { + struct semaphore clp_msg_lock; + void *clp_msg_buf; +}; + +struct hifc_msg_pf_to_mgmt { + struct hifc_hwdev *hwdev; + + /* Async cmd can not be scheduling */ + spinlock_t async_msg_lock; + struct semaphore sync_msg_lock; + + struct workqueue_struct *workq; + + void *async_msg_buf; + void *sync_msg_buf; + void *mgmt_ack_buf; + + struct hifc_recv_msg recv_msg_from_mgmt; + struct hifc_recv_msg recv_resp_msg_from_mgmt; + + u16 async_msg_id; + u16 sync_msg_id; + + struct hifc_api_cmd_chain *cmd_chain[HIFC_API_CMD_MAX]; + + hifc_mgmt_msg_cb recv_mgmt_msg_cb[HIFC_MOD_HW_MAX]; + void *recv_mgmt_msg_data[HIFC_MOD_HW_MAX]; + unsigned long mgmt_msg_cb_state[HIFC_MOD_HW_MAX]; + + struct comm_up_self_msg_info proc; + + /* lock when sending msg */ + spinlock_t sync_event_lock; + enum comm_pf_to_mgmt_event_state event_flag; +}; + +struct hifc_mgmt_msg_handle_work { + struct work_struct work; + struct hifc_msg_pf_to_mgmt *pf_to_mgmt; + void *msg; + u16 msg_len; + enum hifc_mod_type mod; + u8 cmd; + u16 msg_id; + int async_mgmt_to_pf; +}; + +/* show each drivers only such as nic_service_cap, + * toe_service_cap structure, but not show service_cap + */ +enum hifc_service_type { + SERVICE_T_NIC = 0, + + SERVICE_T_FC = 5, + + SERVICE_T_MAX, + + /* Only used for interruption resource management, + * mark the request module + */ + SERVICE_T_INTF = (1 << 15), + SERVICE_T_CQM = (1 << 16), +}; + +/* NIC service capability + * 1, The chip supports NIC RQ is 1K + * 2, PF/VF RQ specifications: + * disable RSS: + * disable VMDq: Each PF/VF at most 8 RQ + * enable the VMDq: Each PF/VF at most 1K RQ + * enable the RSS: + * disable VMDq: each PF at most 64 RQ, VF at most 32 RQ + * enable the VMDq: Each PF/VF at most 1K RQ + * + * 3, The chip supports NIC SQ is 1K + * 4, PF/VF SQ specifications: + * disable RSS: + * disable VMDq: Each PF/VF at most 8 SQ + * enable the VMDq: Each PF/VF at most 1K SQ + * enable the RSS: + * disable VMDq: each PF at most 64 SQ, VF at most 32 SQ + * enable the VMDq: Each PF/VF at most 1K SQ + */ +struct nic_service_cap { + /* PF resources*/ + u16 max_sqs; + u16 max_rqs; + + /* VF resources, vf obtain through the MailBox mechanism from + * according PF + */ + u16 vf_max_sqs; + u16 vf_max_rqs; + bool lro_en; /* LRO feature enable bit*/ + u8 lro_sz; /* LRO context space: n*16B */ + u8 tso_sz; /* TSO context space: n*16B */ + + u16 max_queue_allowed; +}; + +/* PF FC service resource structure defined*/ +struct dev_fc_svc_cap { + /* PF Parent QPC */ + u32 max_parent_qpc_num; /* max number is 2048*/ + + /* PF Child QPC */ + u32 max_child_qpc_num; /* max number is 2048*/ + + /* PF SCQ */ + u32 scq_num; /* 16 */ + + /* PF supports SRQ*/ + u32 srq_num; /* Number of SRQ is 2*/ + + u8 vp_id_start; + u8 vp_id_end; +}; + +/* FC services*/ +struct fc_service_cap { + struct dev_fc_svc_cap dev_fc_cap; + + /* Parent QPC */ + u32 parent_qpc_size; /* 256B */ + + /* Child QPC */ + u32 child_qpc_size; /* 256B */ + + /* SQ */ + u32 sqe_size; /* 128B(in linked list mode)*/ + + /* SCQ */ + u32 scqc_size; /* Size of the Context 32B*/ + u32 scqe_size; /* 64B */ + + /* SRQ */ + u32 srqc_size; /* Size of SRQ Context (64B)*/ + u32 srqe_size; /* 32B */ +}; + +bool hifc_support_fc(void *hwdev, struct fc_service_cap *cap); + +/* Service interface for obtaining service_cap public fields*/ +/* Obtain service_cap.host_oq_id_mask_val*/ +u8 hifc_host_oq_id_mask(void *hwdev); + +/* Obtain service_cap.dev_cap.max_sqs*/ +u16 hifc_func_max_qnum(void *hwdev); + +/* The following information is obtained from the bar space + * which is recorded by SDK layer. + * Here provide parameter query interface for service + */ +/* func_attr.glb_func_idx, global function index */ +u16 hifc_global_func_id(void *hwdev); +/* func_attr.intr_num, MSI-X table entry in function*/ +enum intr_type { + INTR_TYPE_MSIX, + INTR_TYPE_MSI, + INTR_TYPE_INT, + INTR_TYPE_NONE, +}; + +u8 hifc_pcie_itf_id(void *hwdev); /* func_attr.itf_idx, pcie interface index */ + +/* func_attr.func_type, 0-PF 1-VF 2-PPF */ +enum func_type hifc_func_type(void *hwdev); + +u8 hifc_ppf_idx(void *hwdev); + +enum hifc_msix_state { + HIFC_MSIX_ENABLE, + HIFC_MSIX_DISABLE, +}; + +void hifc_set_msix_state(void *hwdev, u16 msix_idx, + enum hifc_msix_state flag); + +/* Defines the IRQ information structure*/ +struct irq_info { + u16 msix_entry_idx; /* IRQ corresponding index number */ + u32 irq_id; /* the IRQ number from OS */ +}; + +int hifc_alloc_irqs(void *hwdev, enum hifc_service_type type, u16 req_num, + struct irq_info *irq_info_array, u16 *resp_num); +void hifc_free_irq(void *hwdev, enum hifc_service_type type, u32 irq_id); + +int hifc_sync_time(void *hwdev, u64 time); +void hifc_disable_mgmt_msg_report(void *hwdev); +void hifc_set_func_deinit_flag(void *hwdev); +void hifc_flush_mgmt_workq(void *hwdev); +int hifc_global_func_id_get(void *hwdev, u16 *func_id); +u16 hifc_global_func_id_hw(void *hwdev); +int hifc_pf_to_mgmt_no_ack(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size); +void hifc_mgmt_msg_aeqe_handler(void *handle, u8 *header, u8 size); +int hifc_pf_to_mgmt_init(struct hifc_hwdev *hwdev); +void hifc_pf_to_mgmt_free(struct hifc_hwdev *hwdev); +int hifc_pf_to_mgmt_sync(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size, void *buf_out, + u16 *out_size, u32 timeout); +int hifc_pf_to_mgmt_async(void *hwdev, enum hifc_mod_type mod, u8 cmd, + void *buf_in, u16 in_size); +int hifc_pf_clp_to_mgmt(void *hwdev, enum hifc_mod_type mod, u8 cmd, + const void *buf_in, u16 in_size, + void *buf_out, u16 *out_size); +int hifc_clp_pf_to_mgmt_init(struct hifc_hwdev *hwdev); +void hifc_clp_pf_to_mgmt_free(struct hifc_hwdev *hwdev); + +#endif diff --git a/drivers/scsi/huawei/hifc/hifc_sml.c b/drivers/scsi/huawei/hifc/hifc_sml.c new file mode 100644 index 000000000000..2d04ff6ed5ff --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_sml.c @@ -0,0 +1,361 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ +#define pr_fmt(fmt) KBUILD_MODNAME ": [COMM]" fmt + +#include <linux/types.h> +#include "hifc_knl_adp.h" +#include "hifc_hw.h" +#include "hifc_hwdev.h" +#include "hifc_sml.h" + +#ifndef HTONL +#define HTONL(x) \ + ((((x) & 0x000000ff) << 24) \ + | (((x) & 0x0000ff00) << 8) \ + | (((x) & 0x00ff0000) >> 8) \ + | (((x) & 0xff000000) >> 24)) +#endif + +static void sml_ctr_htonl_n(u32 *node, u32 len) +{ + u32 i; + + for (i = 0; i < len; i++) { + *node = HTONL(*node); + node++; + } +} + +static void hifc_sml_ctr_read_build_req(struct chipif_sml_ctr_rd_req_s *msg, + u8 instance_id, u8 op_id, + u8 ack, u32 ctr_id, u32 init_val) +{ + msg->head.value = 0; + msg->head.bs.instance = instance_id; + msg->head.bs.op_id = op_id; + msg->head.bs.ack = ack; + msg->head.value = HTONL(msg->head.value); + + msg->ctr_id = ctr_id; + msg->ctr_id = HTONL(msg->ctr_id); + + msg->initial = init_val; +} + +static void hifc_sml_ctr_write_build_req(struct chipif_sml_ctr_wr_req_s *msg, + u8 instance_id, u8 op_id, + u8 ack, u32 ctr_id, + u64 val1, u64 val2) +{ + msg->head.value = 0; + msg->head.bs.instance = instance_id; + msg->head.bs.op_id = op_id; + msg->head.bs.ack = ack; + msg->head.value = HTONL(msg->head.value); + + msg->ctr_id = ctr_id; + msg->ctr_id = HTONL(msg->ctr_id); + + msg->value1_h = val1 >> 32; + msg->value1_l = val1 & 0xFFFFFFFF; + + msg->value2_h = val2 >> 32; + msg->value2_l = val2 & 0xFFFFFFFF; +} + +/** + * hifc_sm_ctr_rd32 - small single 32 counter read + * @hwdev: the pointer to hw device + * @node: the node id + * @instance: instance value + * @ctr_id: counter id + * @value: read counter value ptr + * Return: 0 - success, negative - failure + */ +int hifc_sm_ctr_rd32(void *hwdev, u8 node, u8 instance, u32 ctr_id, u32 *value) +{ + struct chipif_sml_ctr_rd_req_s req; + union ctr_rd_rsp_u rsp; + int ret; + + if (!hwdev || !value) + return -EFAULT; + + hifc_sml_ctr_read_build_req(&req, instance, CHIPIF_SM_CTR_OP_READ, + CHIPIF_ACK, ctr_id, 0); + + ret = hifc_api_cmd_read_ack(hwdev, node, (u8 *)&req, + (unsigned short)sizeof(req), + (void *)&rsp, (unsigned short)sizeof(rsp)); + if (ret) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Sm 32bit counter read fail, err(%d)\n", ret); + return ret; + } + sml_ctr_htonl_n((u32 *)&rsp, 4); + *value = rsp.bs_ss32_rsp.value1; + + return 0; +} + +/** + * hifc_sm_ctr_rd32_clear - small single 32 counter read and clear to zero + * @hwdev: the pointer to hw device + * @node: the node id + * @instance: instance value + * @ctr_id: counter id + * @value: read counter value ptr + * Return: 0 - success, negative - failure + * according to ACN error code (ERR_OK, ERR_PARAM, ERR_FAILED...etc) + */ +int hifc_sm_ctr_rd32_clear(void *hwdev, u8 node, u8 instance, + u32 ctr_id, u32 *value) +{ + struct chipif_sml_ctr_rd_req_s req; + union ctr_rd_rsp_u rsp; + int ret; + + if (!hwdev || !value) + return -EFAULT; + + hifc_sml_ctr_read_build_req(&req, instance, + CHIPIF_SM_CTR_OP_READ_CLEAR, + CHIPIF_ACK, ctr_id, 0); + + ret = hifc_api_cmd_read_ack(hwdev, node, (u8 *)&req, + (unsigned short)sizeof(req), + (void *)&rsp, (unsigned short)sizeof(rsp)); + + if (ret) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Sm 32bit counter clear fail, err(%d)\n", ret); + return ret; + } + sml_ctr_htonl_n((u32 *)&rsp, 4); + *value = rsp.bs_ss32_rsp.value1; + + return 0; +} + +/** + * hifc_sm_ctr_wr32 - small single 32 counter write + * @hwdev: the pointer to hw device + * @node: the node id + * @instance: instance value + * @ctr_id: counter id + * @value: write counter value + * Return: 0 - success, negative - failure + */ +int hifc_sm_ctr_wr32(void *hwdev, u8 node, u8 instance, u32 ctr_id, u32 value) +{ + struct chipif_sml_ctr_wr_req_s req; + struct chipif_sml_ctr_wr_rsp_s rsp; + + if (!hwdev) + return -EFAULT; + + hifc_sml_ctr_write_build_req(&req, instance, CHIPIF_SM_CTR_OP_WRITE, + CHIPIF_NOACK, ctr_id, (u64)value, 0ULL); + + return hifc_api_cmd_read_ack(hwdev, node, (u8 *)&req, + (unsigned short)sizeof(req), (void *)&rsp, + (unsigned short)sizeof(rsp)); +} + +/** + * hifc_sm_ctr_rd64 - big counter 64 read + * @hwdev: the pointer to hw device + * @node: the node id + * @instance: instance value + * @ctr_id: counter id + * @value: read counter value ptr + * Return: 0 - success, negative - failure + */ +int hifc_sm_ctr_rd64(void *hwdev, u8 node, u8 instance, u32 ctr_id, u64 *value) +{ + struct chipif_sml_ctr_rd_req_s req; + union ctr_rd_rsp_u rsp; + int ret; + + if (!hwdev || !value) + return -EFAULT; + + hifc_sml_ctr_read_build_req(&req, instance, CHIPIF_SM_CTR_OP_READ, + CHIPIF_ACK, ctr_id, 0); + + ret = hifc_api_cmd_read_ack(hwdev, node, (u8 *)&req, + (unsigned short)sizeof(req), (void *)&rsp, + (unsigned short)sizeof(rsp)); + if (ret) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Sm 64bit counter read fail err(%d)\n", ret); + return ret; + } + sml_ctr_htonl_n((u32 *)&rsp, 4); + *value = ((u64)rsp.bs_bs64_rsp.value1 << 32) | rsp.bs_bs64_rsp.value2; + + return 0; +} + +/** + * hifc_sm_ctr_wr64 - big single 64 counter write + * @hwdev: the pointer to hw device + * @node: the node id + * @instance: instance value + * @ctr_id: counter id + * @value: write counter value + * Return: 0 - success, negative - failure + */ +int hifc_sm_ctr_wr64(void *hwdev, u8 node, u8 instance, u32 ctr_id, u64 value) +{ + struct chipif_sml_ctr_wr_req_s req; + struct chipif_sml_ctr_wr_rsp_s rsp; + + if (!hwdev) + return -EFAULT; + + hifc_sml_ctr_write_build_req(&req, instance, CHIPIF_SM_CTR_OP_WRITE, + CHIPIF_NOACK, ctr_id, value, 0ULL); + + return hifc_api_cmd_read_ack(hwdev, node, (u8 *)&req, + (unsigned short)sizeof(req), (void *)&rsp, + (unsigned short)sizeof(rsp)); +} + +/** + * hifc_sm_ctr_rd64_pair - big pair 128 counter read + * @hwdev: the pointer to hw device + * @node: the node id + * @instance: instance value + * @ctr_id: counter id + * @value1: read counter value ptr + * @value2: read counter value ptr + * Return: 0 - success, negative - failure + */ +int hifc_sm_ctr_rd64_pair(void *hwdev, u8 node, u8 instance, + u32 ctr_id, u64 *value1, u64 *value2) +{ + struct chipif_sml_ctr_rd_req_s req; + union ctr_rd_rsp_u rsp; + int ret; + + if (!hwdev || (0 != (ctr_id & 0x1)) || !value1 || !value2) { + pr_err("Hwdev(0x%p) or value1(0x%p) or value2(0x%p) is NULL or ctr_id(%d) is odd number\n", + hwdev, value1, value2, ctr_id); + return -EFAULT; + } + + hifc_sml_ctr_read_build_req(&req, instance, CHIPIF_SM_CTR_OP_READ, + CHIPIF_ACK, ctr_id, 0); + + ret = hifc_api_cmd_read_ack(hwdev, node, (u8 *)&req, + (unsigned short)sizeof(req), (void *)&rsp, + (unsigned short)sizeof(rsp)); + if (ret) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Sm 64 bit rd pair ret(%d)\n", ret); + return ret; + } + sml_ctr_htonl_n((u32 *)&rsp, 4); + *value1 = ((u64)rsp.bs_bp64_rsp.val1_h << 32) | rsp.bs_bp64_rsp.val1_l; + *value2 = ((u64)rsp.bs_bp64_rsp.val2_h << 32) | rsp.bs_bp64_rsp.val2_l; + + return 0; +} + +/** + * hifc_sm_ctr_wr64_pair - big pair 128 counter write + * @hwdev: the pointer to hw device + * @node: the node id + * @ctr_id: counter id + * @instance: instance value + * @value1: write counter value + * @value2: write counter value + * Return: 0 - success, negative - failure + */ +int hifc_sm_ctr_wr64_pair(void *hwdev, u8 node, u8 instance, + u32 ctr_id, u64 value1, u64 value2) +{ + struct chipif_sml_ctr_wr_req_s req; + struct chipif_sml_ctr_wr_rsp_s rsp; + + /* pair pattern ctr_id must be even number */ + if (!hwdev || (0 != (ctr_id & 0x1))) { + pr_err("Handle is NULL or ctr_id(%d) is odd number for write 64 bit pair\n", + ctr_id); + return -EFAULT; + } + + hifc_sml_ctr_write_build_req(&req, instance, CHIPIF_SM_CTR_OP_WRITE, + CHIPIF_NOACK, ctr_id, value1, value2); + return hifc_api_cmd_read_ack(hwdev, node, (u8 *)&req, + (unsigned short)sizeof(req), (void *)&rsp, + (unsigned short)sizeof(rsp)); +} + +int hifc_api_csr_rd32(void *hwdev, u8 dest, u32 addr, u32 *val) +{ + struct hifc_csr_request_api_data api_data = {0}; + u32 csr_val = 0; + u16 in_size = sizeof(api_data); + int ret; + + if (!hwdev || !val) + return -EFAULT; + + memset(&api_data, 0, sizeof(struct hifc_csr_request_api_data)); + api_data.dw0 = 0; + api_data.dw1.bits.operation_id = HIFC_CSR_OPERATION_READ_CSR; + api_data.dw1.bits.need_response = HIFC_CSR_NEED_RESP_DATA; + api_data.dw1.bits.data_size = HIFC_CSR_DATA_SZ_32; + api_data.dw1.val32 = cpu_to_be32(api_data.dw1.val32); + api_data.dw2.bits.csr_addr = addr; + api_data.dw2.val32 = cpu_to_be32(api_data.dw2.val32); + + ret = hifc_api_cmd_read_ack(hwdev, dest, (u8 *)(&api_data), + in_size, &csr_val, 4); + if (ret) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Read 32 bit csr fail, dest %d addr 0x%x, ret: 0x%x\n", + dest, addr, ret); + return ret; + } + + *val = csr_val; + + return 0; +} + +int hifc_api_csr_wr32(void *hwdev, u8 dest, u32 addr, u32 val) +{ + struct hifc_csr_request_api_data api_data; + u16 in_size = sizeof(api_data); + int ret; + + if (!hwdev) + return -EFAULT; + + memset(&api_data, 0, sizeof(struct hifc_csr_request_api_data)); + api_data.dw1.bits.operation_id = HIFC_CSR_OPERATION_WRITE_CSR; + api_data.dw1.bits.need_response = HIFC_CSR_NO_RESP_DATA; + api_data.dw1.bits.data_size = HIFC_CSR_DATA_SZ_32; + api_data.dw1.val32 = cpu_to_be32(api_data.dw1.val32); + api_data.dw2.bits.csr_addr = addr; + api_data.dw2.val32 = cpu_to_be32(api_data.dw2.val32); + api_data.csr_write_data_h = 0xffffffff; + api_data.csr_write_data_l = val; + + ret = hifc_api_cmd_write_nack(hwdev, dest, (u8 *)(&api_data), in_size); + if (ret) { + sdk_err(((struct hifc_hwdev *)hwdev)->dev_hdl, + "Write 32 bit csr fail! dest %d addr 0x%x val 0x%x\n", + dest, addr, val); + return ret; + } + + return 0; +} + diff --git a/drivers/scsi/huawei/hifc/hifc_sml.h b/drivers/scsi/huawei/hifc/hifc_sml.h new file mode 100644 index 000000000000..9fe2088f48a1 --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_sml.h @@ -0,0 +1,183 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#ifndef __CHIPIF_SML_COUNTER_H__ +#define __CHIPIF_SML_COUNTER_H__ + +#define CHIPIF_FUNC_PF 0 +#define CHIPIF_FUNC_VF 1 +#define CHIPIF_FUNC_PPF 2 + +#define CHIPIF_ACK 1 +#define CHIPIF_NOACK 0 + +#define CHIPIF_SM_CTR_OP_READ 0x2 +#define CHIPIF_SM_CTR_OP_READ_CLEAR 0x6 +#define CHIPIF_SM_CTR_OP_WRITE 0x3 + +#define SMALL_CNT_READ_RSP_SIZE 16 + +/* request head */ +union chipif_sml_ctr_req_head_u { + struct { + u32 pad:15; + u32 ack:1; + u32 op_id:5; + u32 instance:6; + u32 src:5; + } bs; + + u32 value; +}; + +/* counter read request struct */ +struct chipif_sml_ctr_rd_req_s { + u32 extra; + union chipif_sml_ctr_req_head_u head; + u32 ctr_id; + u32 initial; + u32 pad; +}; + +/* counter read response union */ +union ctr_rd_rsp_u { + struct { + u32 value1:16; + u32 pad0:16; + u32 pad1[3]; + } bs_ss16_rsp; + + struct { + u32 value1; + u32 pad[3]; + } bs_ss32_rsp; + + struct { + u32 value1:20; + u32 pad0:12; + u32 value2:12; + u32 pad1:20; + u32 pad2[2]; + } bs_sp_rsp; + + struct { + u32 value1; + u32 value2; + u32 pad[2]; + } bs_bs64_rsp; + + struct { + u32 val1_h; + u32 val1_l; + u32 val2_h; + u32 val2_l; + } bs_bp64_rsp; + +}; + +/* resopnse head */ +union sml_ctr_rsp_head_u { + struct { + u32 pad:30; /* reserve */ + u32 code:2; /* error code */ + } bs; + + u32 value; +}; + +/* counter write request struct */ +struct chipif_sml_ctr_wr_req_s { + u32 extra; + union chipif_sml_ctr_req_head_u head; + u32 ctr_id; + u32 rsv1; + u32 rsv2; + u32 value1_h; + u32 value1_l; + u32 value2_h; + u32 value2_l; +}; + +/* counter write response struct */ +struct chipif_sml_ctr_wr_rsp_s { + union sml_ctr_rsp_head_u head; + u32 pad[3]; +}; + +enum HIFC_CSR_API_DATA_OPERATION_ID { + HIFC_CSR_OPERATION_WRITE_CSR = 0x1E, + HIFC_CSR_OPERATION_READ_CSR = 0x1F +}; + +enum HIFC_CSR_API_DATA_NEED_RESPONSE_DATA { + HIFC_CSR_NO_RESP_DATA = 0, + HIFC_CSR_NEED_RESP_DATA = 1 +}; + +enum HIFC_CSR_API_DATA_DATA_SIZE { + HIFC_CSR_DATA_SZ_32 = 0, + HIFC_CSR_DATA_SZ_64 = 1 +}; + +struct hifc_csr_request_api_data { + u32 dw0; + + union { + struct { + u32 reserved1:13; + /* this field indicates the write/read data size: + * 2'b00: 32 bits + * 2'b01: 64 bits + * 2'b10~2'b11:reserved + */ + u32 data_size:2; + /* this field indicates that requestor expect receive a + * response data or not. + * 1'b0: expect not to receive a response data. + * 1'b1: expect to receive a response data. + */ + u32 need_response:1; + /* this field indicates the operation that the requestor + * expected. + * 5'b1_1110: write value to csr space. + * 5'b1_1111: read register from csr space. + */ + u32 operation_id:5; + u32 reserved2:6; + /* this field specifies the Src node ID for this API + * request message. + */ + u32 src_node_id:5; + } bits; + + u32 val32; + } dw1; + + union { + struct { + /* it specifies the CSR address. */ + u32 csr_addr:26; + u32 reserved3:6; + } bits; + + u32 val32; + } dw2; + + /* if data_size=2'b01, it is high 32 bits of write data. else, it is + * 32'hFFFF_FFFF. + */ + u32 csr_write_data_h; + /* the low 32 bits of write data. */ + u32 csr_write_data_l; +}; + +int hifc_sm_ctr_rd32(void *hwdev, u8 node, u8 instance, u32 ctr_id, u32 *value); +int hifc_sm_ctr_rd64(void *hwdev, u8 node, u8 instance, u32 ctr_id, u64 *value); +int hifc_sm_ctr_rd64_pair(void *hwdev, u8 node, u8 instance, + u32 ctr_id, u64 *value1, u64 *value2); + +#endif + diff --git a/drivers/scsi/huawei/hifc/hifc_wq.c b/drivers/scsi/huawei/hifc/hifc_wq.c new file mode 100644 index 000000000000..4e926d140b2c --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_wq.c @@ -0,0 +1,624 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ +#define pr_fmt(fmt) KBUILD_MODNAME ": [COMM]" fmt + +#include <linux/kernel.h> +#include <linux/pci.h> +#include <linux/dma-mapping.h> +#include <linux/device.h> +#include <linux/vmalloc.h> +#include <linux/types.h> +#include <linux/atomic.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/spinlock.h> + +#include "hifc_knl_adp.h" +#include "hifc_hw.h" +#include "hifc_hwif.h" +#include "hifc_wq.h" + +#define WQS_MAX_NUM_BLOCKS 128 +#define WQS_FREE_BLOCKS_SIZE(wqs) (WQS_MAX_NUM_BLOCKS * \ + sizeof((wqs)->free_blocks[0])) + +static void wqs_return_block(struct hifc_wqs *wqs, u32 page_idx, u32 block_idx) +{ + u32 pos; + + spin_lock(&wqs->alloc_blocks_lock); + + wqs->num_free_blks++; + + pos = wqs->return_blk_pos++; + pos &= WQS_MAX_NUM_BLOCKS - 1; + + wqs->free_blocks[pos].page_idx = page_idx; + wqs->free_blocks[pos].block_idx = block_idx; + + spin_unlock(&wqs->alloc_blocks_lock); +} + +static int wqs_next_block(struct hifc_wqs *wqs, u32 *page_idx, + u32 *block_idx) +{ + u32 pos; + + spin_lock(&wqs->alloc_blocks_lock); + + if (wqs->num_free_blks <= 0) { + spin_unlock(&wqs->alloc_blocks_lock); + return -ENOMEM; + } + wqs->num_free_blks--; + + pos = wqs->alloc_blk_pos++; + pos &= WQS_MAX_NUM_BLOCKS - 1; + + *page_idx = wqs->free_blocks[pos].page_idx; + *block_idx = wqs->free_blocks[pos].block_idx; + + wqs->free_blocks[pos].page_idx = 0xFFFFFFFF; + wqs->free_blocks[pos].block_idx = 0xFFFFFFFF; + + spin_unlock(&wqs->alloc_blocks_lock); + + return 0; +} + +static int queue_alloc_page(void *handle, u64 **vaddr, u64 *paddr, + u64 **shadow_vaddr, u64 page_sz) +{ + dma_addr_t dma_addr = 0; + + *vaddr = dma_zalloc_coherent(handle, page_sz, &dma_addr, + GFP_KERNEL); + if (!*vaddr) { + sdk_err(handle, "Failed to allocate dma to wqs page\n"); + return -ENOMEM; + } + + if (!ADDR_4K_ALIGNED(dma_addr)) { + sdk_err(handle, "Cla is not 4k aligned!\n"); + goto shadow_vaddr_err; + } + + *paddr = (u64)dma_addr; + + /* use vzalloc for big mem, shadow_vaddr only used at initialization */ + *shadow_vaddr = vzalloc(page_sz); + if (!*shadow_vaddr) { + sdk_err(handle, "Failed to allocate shadow page vaddr\n"); + goto shadow_vaddr_err; + } + + return 0; + +shadow_vaddr_err: + dma_free_coherent(handle, page_sz, *vaddr, dma_addr); + return -ENOMEM; +} + +static int wqs_allocate_page(struct hifc_wqs *wqs, u32 page_idx) +{ + return queue_alloc_page(wqs->dev_hdl, &wqs->page_vaddr[page_idx], + &wqs->page_paddr[page_idx], + &wqs->shadow_page_vaddr[page_idx], + WQS_PAGE_SIZE); +} + +static void wqs_free_page(struct hifc_wqs *wqs, u32 page_idx) +{ + dma_free_coherent(wqs->dev_hdl, WQS_PAGE_SIZE, + wqs->page_vaddr[page_idx], + (dma_addr_t)wqs->page_paddr[page_idx]); + vfree(wqs->shadow_page_vaddr[page_idx]); +} + +static int cmdq_allocate_page(struct hifc_cmdq_pages *cmdq_pages) +{ + return queue_alloc_page(cmdq_pages->dev_hdl, + &cmdq_pages->cmdq_page_vaddr, + &cmdq_pages->cmdq_page_paddr, + &cmdq_pages->cmdq_shadow_page_vaddr, + CMDQ_PAGE_SIZE); +} + +static void cmdq_free_page(struct hifc_cmdq_pages *cmdq_pages) +{ + dma_free_coherent(cmdq_pages->dev_hdl, CMDQ_PAGE_SIZE, + cmdq_pages->cmdq_page_vaddr, + (dma_addr_t)cmdq_pages->cmdq_page_paddr); + vfree(cmdq_pages->cmdq_shadow_page_vaddr); +} + +static int alloc_wqes_shadow(struct hifc_wq *wq) +{ + u64 size; + + /* if wq->max_wqe_size == 0, we don't need to alloc shadow */ + if (wq->max_wqe_size <= wq->wqebb_size) + return 0; + + size = (u64)wq->num_q_pages * wq->max_wqe_size; + wq->shadow_wqe = kzalloc(size, GFP_KERNEL); + if (!wq->shadow_wqe) { + pr_err("Failed to allocate shadow wqe\n"); + return -ENOMEM; + } + + size = wq->num_q_pages * sizeof(wq->prod_idx); + wq->shadow_idx = kzalloc(size, GFP_KERNEL); + if (!wq->shadow_idx) { + pr_err("Failed to allocate shadow index\n"); + goto shadow_idx_err; + } + + return 0; + +shadow_idx_err: + kfree(wq->shadow_wqe); + return -ENOMEM; +} + +static void free_wqes_shadow(struct hifc_wq *wq) +{ + if (wq->max_wqe_size <= wq->wqebb_size) + return; + + kfree(wq->shadow_idx); + kfree(wq->shadow_wqe); +} + +static void free_wq_pages(void *handle, struct hifc_wq *wq, + u32 num_q_pages) +{ + u32 i; + + for (i = 0; i < num_q_pages; i++) + hifc_dma_free_coherent_align(handle, &wq->mem_align[i]); + + free_wqes_shadow(wq); + + wq->block_vaddr = NULL; + wq->shadow_block_vaddr = NULL; + + kfree(wq->mem_align); +} + +static int alloc_wq_pages(void *dev_hdl, struct hifc_wq *wq) +{ + struct hifc_dma_addr_align *mem_align; + u64 *vaddr, *paddr; + u32 i, num_q_pages; + int err; + + vaddr = wq->shadow_block_vaddr; + paddr = wq->block_vaddr; + + num_q_pages = ALIGN(WQ_SIZE(wq), wq->wq_page_size) / wq->wq_page_size; + if (num_q_pages > WQ_MAX_PAGES) { + sdk_err(dev_hdl, "Number(%d) wq pages exceeds the limit\n", + num_q_pages); + return -EINVAL; + } + + if (num_q_pages & (num_q_pages - 1)) { + sdk_err(dev_hdl, "Wq num(%d) q pages must be power of 2\n", + num_q_pages); + return -EINVAL; + } + + wq->num_q_pages = num_q_pages; + + err = alloc_wqes_shadow(wq); + if (err) { + sdk_err(dev_hdl, "Failed to allocate wqe shadow\n"); + return err; + } + + wq->mem_align = kcalloc(wq->num_q_pages, sizeof(*wq->mem_align), + GFP_KERNEL); + if (!wq->mem_align) { + sdk_err(dev_hdl, "Failed to allocate mem_align\n"); + free_wqes_shadow(wq); + return -ENOMEM; + } + + for (i = 0; i < num_q_pages; i++) { + mem_align = &wq->mem_align[i]; + err = hifc_dma_zalloc_coherent_align(dev_hdl, wq->wq_page_size, + wq->wq_page_size, + GFP_KERNEL, mem_align); + if (err) { + sdk_err(dev_hdl, "Failed to allocate wq page\n"); + goto alloc_wq_pages_err; + } + + *paddr = cpu_to_be64(mem_align->align_paddr); + *vaddr = (u64)mem_align->align_vaddr; + + paddr++; + vaddr++; + } + + return 0; + +alloc_wq_pages_err: + free_wq_pages(dev_hdl, wq, i); + + return -ENOMEM; +} + +int hifc_wq_allocate(struct hifc_wqs *wqs, struct hifc_wq *wq, + u32 wqebb_size, u32 wq_page_size, u16 q_depth, + u32 max_wqe_size) +{ + u32 num_wqebbs_per_page; + int err; + + if (wqebb_size == 0) { + sdk_err(wqs->dev_hdl, "Wqebb_size must be >0\n"); + return -EINVAL; + } + + if (q_depth & (q_depth - 1)) { + sdk_err(wqs->dev_hdl, "Wq q_depth(%d) isn't power of 2\n", + q_depth); + return -EINVAL; + } + + if (wq_page_size & (wq_page_size - 1)) { + sdk_err(wqs->dev_hdl, "Wq page_size(%d) isn't power of 2\n", + wq_page_size); + return -EINVAL; + } + + num_wqebbs_per_page = ALIGN(wq_page_size, wqebb_size) / wqebb_size; + + if (num_wqebbs_per_page & (num_wqebbs_per_page - 1)) { + sdk_err(wqs->dev_hdl, "Num(%d) wqebbs per page isn't power of 2\n", + num_wqebbs_per_page); + return -EINVAL; + } + + err = wqs_next_block(wqs, &wq->page_idx, &wq->block_idx); + if (err) { + sdk_err(wqs->dev_hdl, "Failed to get free wqs next block\n"); + return err; + } + + wq->wqebb_size = wqebb_size; + wq->wq_page_size = wq_page_size; + wq->q_depth = q_depth; + wq->max_wqe_size = max_wqe_size; + wq->num_wqebbs_per_page = num_wqebbs_per_page; + + wq->wqebbs_per_page_shift = (u32)ilog2(num_wqebbs_per_page); + + wq->block_vaddr = WQ_BASE_VADDR(wqs, wq); + wq->shadow_block_vaddr = WQ_BASE_ADDR(wqs, wq); + wq->block_paddr = WQ_BASE_PADDR(wqs, wq); + + err = alloc_wq_pages(wqs->dev_hdl, wq); + if (err) { + sdk_err(wqs->dev_hdl, "Failed to allocate wq pages\n"); + goto alloc_wq_pages_err; + } + + atomic_set(&wq->delta, q_depth); + wq->cons_idx = 0; + wq->prod_idx = 0; + wq->mask = q_depth - 1; + + return 0; + +alloc_wq_pages_err: + wqs_return_block(wqs, wq->page_idx, wq->block_idx); + return err; +} + +void hifc_wq_free(struct hifc_wqs *wqs, struct hifc_wq *wq) +{ + free_wq_pages(wqs->dev_hdl, wq, wq->num_q_pages); + + wqs_return_block(wqs, wq->page_idx, wq->block_idx); +} + +static void init_wqs_blocks_arr(struct hifc_wqs *wqs) +{ + u32 page_idx, blk_idx, pos = 0; + + for (page_idx = 0; page_idx < wqs->num_pages; page_idx++) { + for (blk_idx = 0; blk_idx < WQS_BLOCKS_PER_PAGE; blk_idx++) { + wqs->free_blocks[pos].page_idx = page_idx; + wqs->free_blocks[pos].block_idx = blk_idx; + pos++; + } + } + + wqs->alloc_blk_pos = 0; + wqs->return_blk_pos = 0; + wqs->num_free_blks = WQS_MAX_NUM_BLOCKS; + spin_lock_init(&wqs->alloc_blocks_lock); +} + +void hifc_wq_wqe_pg_clear(struct hifc_wq *wq) +{ + u64 *block_vaddr; + u32 pg_idx; + + block_vaddr = wq->shadow_block_vaddr; + + atomic_set(&wq->delta, wq->q_depth); + wq->cons_idx = 0; + wq->prod_idx = 0; + + for (pg_idx = 0; pg_idx < wq->num_q_pages; pg_idx++) + memset((void *)(*(block_vaddr + pg_idx)), 0, wq->wq_page_size); +} + +int hifc_cmdq_alloc(struct hifc_cmdq_pages *cmdq_pages, + struct hifc_wq *wq, void *dev_hdl, + int cmdq_blocks, u32 wq_page_size, u32 wqebb_size, + u16 q_depth, u32 max_wqe_size) +{ + int i, j, err = -ENOMEM; + + if (q_depth & (q_depth - 1)) { + sdk_err(dev_hdl, "Cmdq q_depth(%d) isn't power of 2\n", + q_depth); + return -EINVAL; + } + + cmdq_pages->dev_hdl = dev_hdl; + + err = cmdq_allocate_page(cmdq_pages); + if (err) { + sdk_err(dev_hdl, "Failed to allocate CMDQ page\n"); + return err; + } + + for (i = 0; i < cmdq_blocks; i++) { + wq[i].page_idx = 0; + wq[i].block_idx = (u32)i; + wq[i].wqebb_size = wqebb_size; + wq[i].wq_page_size = wq_page_size; + wq[i].q_depth = q_depth; + wq[i].max_wqe_size = max_wqe_size; + wq[i].num_wqebbs_per_page = + ALIGN(wq_page_size, wqebb_size) / wqebb_size; + + wq[i].wqebbs_per_page_shift = + (u32)ilog2(wq[i].num_wqebbs_per_page); + + wq[i].block_vaddr = CMDQ_BASE_VADDR(cmdq_pages, &wq[i]); + wq[i].shadow_block_vaddr = CMDQ_BASE_ADDR(cmdq_pages, &wq[i]); + wq[i].block_paddr = CMDQ_BASE_PADDR(cmdq_pages, &wq[i]); + + err = alloc_wq_pages(cmdq_pages->dev_hdl, &wq[i]); + if (err) { + sdk_err(dev_hdl, "Failed to alloc CMDQ blocks\n"); + goto cmdq_block_err; + } + + atomic_set(&wq[i].delta, q_depth); + wq[i].cons_idx = 0; + wq[i].prod_idx = 0; + wq[i].mask = q_depth - 1; + } + + return 0; + +cmdq_block_err: + for (j = 0; j < i; j++) + free_wq_pages(cmdq_pages->dev_hdl, &wq[j], wq[j].num_q_pages); + + cmdq_free_page(cmdq_pages); + return err; +} + +void hifc_cmdq_free(struct hifc_cmdq_pages *cmdq_pages, + struct hifc_wq *wq, int cmdq_blocks) +{ + int i; + + for (i = 0; i < cmdq_blocks; i++) + free_wq_pages(cmdq_pages->dev_hdl, &wq[i], wq[i].num_q_pages); + + cmdq_free_page(cmdq_pages); +} + +static int alloc_page_addr(struct hifc_wqs *wqs) +{ + u64 size = wqs->num_pages * sizeof(*wqs->page_paddr); + + wqs->page_paddr = kzalloc(size, GFP_KERNEL); + if (!wqs->page_paddr) + return -ENOMEM; + + size = wqs->num_pages * sizeof(*wqs->page_vaddr); + wqs->page_vaddr = kzalloc(size, GFP_KERNEL); + if (!wqs->page_vaddr) + goto page_vaddr_err; + + size = wqs->num_pages * sizeof(*wqs->shadow_page_vaddr); + wqs->shadow_page_vaddr = kzalloc(size, GFP_KERNEL); + if (!wqs->shadow_page_vaddr) + goto page_shadow_vaddr_err; + + return 0; + +page_shadow_vaddr_err: + kfree(wqs->page_vaddr); + +page_vaddr_err: + kfree(wqs->page_paddr); + return -ENOMEM; +} + +static void free_page_addr(struct hifc_wqs *wqs) +{ + kfree(wqs->shadow_page_vaddr); + kfree(wqs->page_vaddr); + kfree(wqs->page_paddr); +} + +int hifc_wqs_alloc(struct hifc_wqs *wqs, int num_wqs, void *dev_hdl) +{ + u32 i, page_idx; + int err; + + wqs->dev_hdl = dev_hdl; + wqs->num_pages = WQ_NUM_PAGES(num_wqs); + + if (alloc_page_addr(wqs)) { + sdk_err(dev_hdl, "Failed to allocate mem for page addresses\n"); + return -ENOMEM; + } + + for (page_idx = 0; page_idx < wqs->num_pages; page_idx++) { + err = wqs_allocate_page(wqs, page_idx); + if (err) { + sdk_err(dev_hdl, "Failed wq page allocation\n"); + goto wq_allocate_page_err; + } + } + + wqs->free_blocks = kzalloc(WQS_FREE_BLOCKS_SIZE(wqs), GFP_KERNEL); + if (!wqs->free_blocks) { + err = -ENOMEM; + goto alloc_blocks_err; + } + + init_wqs_blocks_arr(wqs); + return 0; + +alloc_blocks_err: +wq_allocate_page_err: + for (i = 0; i < page_idx; i++) + wqs_free_page(wqs, i); + + free_page_addr(wqs); + return err; +} + +void hifc_wqs_free(struct hifc_wqs *wqs) +{ + u32 page_idx; + + for (page_idx = 0; page_idx < wqs->num_pages; page_idx++) + wqs_free_page(wqs, page_idx); + + free_page_addr(wqs); + kfree(wqs->free_blocks); +} + +static void copy_wqe_to_shadow(struct hifc_wq *wq, void *shadow_addr, + int num_wqebbs, u16 prod_idx) +{ + u8 *shadow_wqebb_addr, *wqe_page_addr, *wqebb_addr; + u32 i, offset; + u16 idx; + + for (i = 0; i < (u32)num_wqebbs; i++) { + offset = i * wq->wqebb_size; + shadow_wqebb_addr = (u8 *)shadow_addr + offset; + + idx = MASKED_WQE_IDX(wq, prod_idx + i); + wqe_page_addr = WQ_PAGE_ADDR(wq, idx); + wqebb_addr = wqe_page_addr + + WQE_PAGE_OFF(wq, MASKED_WQE_IDX(wq, idx)); + + memcpy(shadow_wqebb_addr, wqebb_addr, wq->wqebb_size); + } +} + +void *hifc_get_wqebb_addr(struct hifc_wq *wq, u16 index) +{ + return WQ_PAGE_ADDR(wq, index) + WQE_PAGE_OFF(wq, index); +} + +u64 hifc_get_first_wqe_page_addr(struct hifc_wq *wq) +{ + return be64_to_cpu(*wq->block_vaddr); +} + +void *hifc_get_wqe(struct hifc_wq *wq, int num_wqebbs, u16 *prod_idx) +{ + u32 curr_pg, end_pg; + u16 curr_prod_idx, end_prod_idx; + + if (atomic_sub_return(num_wqebbs, &wq->delta) < 0) { + atomic_add(num_wqebbs, &wq->delta); + return NULL; + } + + /* use original cur_pi and end_pi, no need queue depth mask as + * WQE_PAGE_NUM will do num_queue_pages mask + */ + curr_prod_idx = (u16)wq->prod_idx; + wq->prod_idx += num_wqebbs; + + /* end prod index should points to the last wqebb of wqe, + * therefore minus 1 + */ + end_prod_idx = (u16)wq->prod_idx - 1; + + curr_pg = WQE_PAGE_NUM(wq, curr_prod_idx); + end_pg = WQE_PAGE_NUM(wq, end_prod_idx); + + *prod_idx = MASKED_WQE_IDX(wq, curr_prod_idx); + + /* If we only have one page, still need to get shadown wqe when + * wqe rolling-over page + */ + if (curr_pg != end_pg || MASKED_WQE_IDX(wq, end_prod_idx) < *prod_idx) { + u32 offset = curr_pg * wq->max_wqe_size; + u8 *shadow_addr = wq->shadow_wqe + offset; + + wq->shadow_idx[curr_pg] = *prod_idx; + return shadow_addr; + } + + return WQ_PAGE_ADDR(wq, *prod_idx) + WQE_PAGE_OFF(wq, *prod_idx); +} + +void hifc_put_wqe(struct hifc_wq *wq, int num_wqebbs) +{ + atomic_add(num_wqebbs, &wq->delta); + wq->cons_idx += num_wqebbs; +} + +void *hifc_read_wqe(struct hifc_wq *wq, int num_wqebbs, u16 *cons_idx) +{ + u32 curr_pg, end_pg; + u16 curr_cons_idx, end_cons_idx; + + if ((atomic_read(&wq->delta) + num_wqebbs) > wq->q_depth) + return NULL; + + curr_cons_idx = (u16)wq->cons_idx; + + curr_cons_idx = MASKED_WQE_IDX(wq, curr_cons_idx); + end_cons_idx = MASKED_WQE_IDX(wq, curr_cons_idx + num_wqebbs - 1); + + curr_pg = WQE_PAGE_NUM(wq, curr_cons_idx); + end_pg = WQE_PAGE_NUM(wq, end_cons_idx); + + *cons_idx = curr_cons_idx; + + if (curr_pg != end_pg) { + u32 offset = curr_pg * wq->max_wqe_size; + u8 *shadow_addr = wq->shadow_wqe + offset; + + copy_wqe_to_shadow(wq, shadow_addr, num_wqebbs, *cons_idx); + + return shadow_addr; + } + + return WQ_PAGE_ADDR(wq, *cons_idx) + WQE_PAGE_OFF(wq, *cons_idx); +} diff --git a/drivers/scsi/huawei/hifc/hifc_wq.h b/drivers/scsi/huawei/hifc/hifc_wq.h new file mode 100644 index 000000000000..207d54191afa --- /dev/null +++ b/drivers/scsi/huawei/hifc/hifc_wq.h @@ -0,0 +1,165 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Huawei Hifc PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + * + */ + +#ifndef HIFC_WQ_H +#define HIFC_WQ_H + +#define WQS_BLOCKS_PER_PAGE 4 +#define WQ_SIZE(wq) (u32)((u64)(wq)->q_depth * (wq)->wqebb_size) + +#define WQE_PAGE_NUM(wq, idx) (((idx) >> ((wq)->wqebbs_per_page_shift)) & \ + ((wq)->num_q_pages - 1)) + +#define WQE_PAGE_OFF(wq, idx) ((u64)((wq)->wqebb_size) * \ + ((idx) & ((wq)->num_wqebbs_per_page - 1))) + +#define WQ_PAGE_ADDR_SIZE sizeof(u64) +#define WQ_PAGE_ADDR_SIZE_SHIFT 3 +#define WQ_PAGE_ADDR(wq, idx) \ + (u8 *)(*(u64 *)((u64)((wq)->shadow_block_vaddr) + \ + (WQE_PAGE_NUM(wq, idx) << WQ_PAGE_ADDR_SIZE_SHIFT))) + +#define WQ_BLOCK_SIZE 4096UL +#define WQS_PAGE_SIZE (WQS_BLOCKS_PER_PAGE * WQ_BLOCK_SIZE) +#define WQ_MAX_PAGES (WQ_BLOCK_SIZE >> WQ_PAGE_ADDR_SIZE_SHIFT) + +#define CMDQ_BLOCKS_PER_PAGE 8 +#define CMDQ_BLOCK_SIZE 512UL +#define CMDQ_PAGE_SIZE ALIGN((CMDQ_BLOCKS_PER_PAGE * \ + CMDQ_BLOCK_SIZE), PAGE_SIZE) + +#define ADDR_4K_ALIGNED(addr) (((addr) & 0xfff) == 0) + +#define WQ_BASE_VADDR(wqs, wq) \ + (u64 *)(((u64)((wqs)->page_vaddr[(wq)->page_idx])) \ + + (wq)->block_idx * WQ_BLOCK_SIZE) + +#define WQ_BASE_PADDR(wqs, wq) (((wqs)->page_paddr[(wq)->page_idx]) \ + + (u64)(wq)->block_idx * WQ_BLOCK_SIZE) + +#define WQ_BASE_ADDR(wqs, wq) \ + (u64 *)(((u64)((wqs)->shadow_page_vaddr[(wq)->page_idx])) \ + + (wq)->block_idx * WQ_BLOCK_SIZE) + +#define CMDQ_BASE_VADDR(cmdq_pages, wq) \ + (u64 *)(((u64)((cmdq_pages)->cmdq_page_vaddr)) \ + + (wq)->block_idx * CMDQ_BLOCK_SIZE) + +#define CMDQ_BASE_PADDR(cmdq_pages, wq) \ + (((u64)((cmdq_pages)->cmdq_page_paddr)) \ + + (u64)(wq)->block_idx * CMDQ_BLOCK_SIZE) + +#define CMDQ_BASE_ADDR(cmdq_pages, wq) \ + (u64 *)(((u64)((cmdq_pages)->cmdq_shadow_page_vaddr)) \ + + (wq)->block_idx * CMDQ_BLOCK_SIZE) + +#define MASKED_WQE_IDX(wq, idx) ((idx) & (wq)->mask) + +#define WQ_NUM_PAGES(num_wqs) \ + (ALIGN((u32)num_wqs, WQS_BLOCKS_PER_PAGE) / WQS_BLOCKS_PER_PAGE) + +#define MAX_WQE_SIZE(max_sge, wqebb_size) \ + ((max_sge <= 2) ? (wqebb_size) : \ + ((ALIGN(((max_sge) - 2), 4) / 4 + 1) * (wqebb_size))) + +struct hifc_free_block { + u32 page_idx; + u32 block_idx; +}; + +struct hifc_wq { + /* The addresses are 64 bit in the HW */ + u64 block_paddr; + u64 *shadow_block_vaddr; + u64 *block_vaddr; + + u32 wqebb_size; + u32 wq_page_size; + u16 q_depth; + u32 max_wqe_size; + u32 num_wqebbs_per_page; + + /* performance: replace mul/div as shift; + * num_wqebbs_per_page must be power of 2 + */ + u32 wqebbs_per_page_shift; + u32 page_idx; + u32 block_idx; + + u32 num_q_pages; + + struct hifc_dma_addr_align *mem_align; + + int cons_idx; + int prod_idx; + + atomic_t delta; + u16 mask; + + u8 *shadow_wqe; + u16 *shadow_idx; +}; + +struct hifc_cmdq_pages { + /* The addresses are 64 bit in the HW */ + u64 cmdq_page_paddr; + u64 *cmdq_page_vaddr; + u64 *cmdq_shadow_page_vaddr; + + void *dev_hdl; +}; + +struct hifc_wqs { + /* The addresses are 64 bit in the HW */ + u64 *page_paddr; + u64 **page_vaddr; + u64 **shadow_page_vaddr; + + struct hifc_free_block *free_blocks; + u32 alloc_blk_pos; + u32 return_blk_pos; + int num_free_blks; + + /* for allocate blocks */ + spinlock_t alloc_blocks_lock; + + u32 num_pages; + + void *dev_hdl; +}; + +void hifc_wq_wqe_pg_clear(struct hifc_wq *wq); + +int hifc_cmdq_alloc(struct hifc_cmdq_pages *cmdq_pages, + struct hifc_wq *wq, void *dev_hdl, + int cmdq_blocks, u32 wq_page_size, u32 wqebb_size, + u16 q_depth, u32 max_wqe_size); + +void hifc_cmdq_free(struct hifc_cmdq_pages *cmdq_pages, + struct hifc_wq *wq, int cmdq_blocks); + +int hifc_wqs_alloc(struct hifc_wqs *wqs, int num_wqs, void *dev_hdl); + +void hifc_wqs_free(struct hifc_wqs *wqs); + +int hifc_wq_allocate(struct hifc_wqs *wqs, struct hifc_wq *wq, + u32 wqebb_size, u32 wq_page_size, u16 q_depth, + u32 max_wqe_size); + +void hifc_wq_free(struct hifc_wqs *wqs, struct hifc_wq *wq); + +void *hifc_get_wqebb_addr(struct hifc_wq *wq, u16 index); + +u64 hifc_get_first_wqe_page_addr(struct hifc_wq *wq); + +void *hifc_get_wqe(struct hifc_wq *wq, int num_wqebbs, u16 *prod_idx); + +void hifc_put_wqe(struct hifc_wq *wq, int num_wqebbs); + +void *hifc_read_wqe(struct hifc_wq *wq, int num_wqebbs, u16 *cons_idx); + +#endif + -- 2.25.1

1 6

[PATCH 1/8] asm-generic/tlb: Provide a comment
by Yang Yingliang 16 Nov '20

16 Nov '20

From: Peter Zijlstra <peterz(a)infradead.org> mainline inclusion from mainline-v5.2-rc1 commit dea2434c23c102b3e7d320849ec1cfeb432edb60 category:feature bugzilla:NA CVE:NA ------------------- Write a comment explaining some of this.. Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Acked-by: Will Deacon <will.deacon(a)arm.com> Cc: Andrew Morton <akpm(a)linux-foundation.org> Cc: Andy Lutomirski <luto(a)kernel.org> Cc: Aneesh Kumar K.V <aneesh.kumar(a)linux.vnet.ibm.com> Cc: Borislav Petkov <bp(a)alien8.de> Cc: Dave Hansen <dave.hansen(a)linux.intel.com> Cc: H. Peter Anvin <hpa(a)zytor.com> Cc: Linus Torvalds <torvalds(a)linux-foundation.org> Cc: Nick Piggin <npiggin(a)gmail.com> Cc: Peter Zijlstra <peterz(a)infradead.org> Cc: Rik van Riel <riel(a)surriel.com> Cc: Thomas Gleixner <tglx(a)linutronix.de> Signed-off-by: Ingo Molnar <mingo(a)kernel.org> Signed-off-by: Chen Jun <chenjun102(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/asm-generic/tlb.h | 119 +++++++++++++++++++++++++++++++++++++- 1 file changed, 116 insertions(+), 3 deletions(-) diff --git a/include/asm-generic/tlb.h b/include/asm-generic/tlb.h index 147381aad7cc..632b1cdce357 100644 --- a/include/asm-generic/tlb.h +++ b/include/asm-generic/tlb.h @@ -22,6 +22,118 @@ #ifdef CONFIG_MMU +/* + * Generic MMU-gather implementation. + * + * The mmu_gather data structure is used by the mm code to implement the + * correct and efficient ordering of freeing pages and TLB invalidations. + * + * This correct ordering is: + * + * 1) unhook page + * 2) TLB invalidate page + * 3) free page + * + * That is, we must never free a page before we have ensured there are no live + * translations left to it. Otherwise it might be possible to observe (or + * worse, change) the page content after it has been reused. + * + * The mmu_gather API consists of: + * + * - tlb_gather_mmu() / tlb_finish_mmu(); start and finish a mmu_gather + * + * Finish in particular will issue a (final) TLB invalidate and free + * all (remaining) queued pages. + * + * - tlb_start_vma() / tlb_end_vma(); marks the start / end of a VMA + * + * Defaults to flushing at tlb_end_vma() to reset the range; helps when + * there's large holes between the VMAs. + * + * - tlb_remove_page() / __tlb_remove_page() + * - tlb_remove_page_size() / __tlb_remove_page_size() + * + * __tlb_remove_page_size() is the basic primitive that queues a page for + * freeing. __tlb_remove_page() assumes PAGE_SIZE. Both will return a + * boolean indicating if the queue is (now) full and a call to + * tlb_flush_mmu() is required. + * + * tlb_remove_page() and tlb_remove_page_size() imply the call to + * tlb_flush_mmu() when required and has no return value. + * + * - tlb_remove_check_page_size_change() + * + * call before __tlb_remove_page*() to set the current page-size; implies a + * possible tlb_flush_mmu() call. + * + * - tlb_flush_mmu() / tlb_flush_mmu_tlbonly() / tlb_flush_mmu_free() + * + * tlb_flush_mmu_tlbonly() - does the TLB invalidate (and resets + * related state, like the range) + * + * tlb_flush_mmu_free() - frees the queued pages; make absolutely + * sure no additional tlb_remove_page() + * calls happen between _tlbonly() and this. + * + * tlb_flush_mmu() - the above two calls. + * + * - mmu_gather::fullmm + * + * A flag set by tlb_gather_mmu() to indicate we're going to free + * the entire mm; this allows a number of optimizations. + * + * - We can ignore tlb_{start,end}_vma(); because we don't + * care about ranges. Everything will be shot down. + * + * - (RISC) architectures that use ASIDs can cycle to a new ASID + * and delay the invalidation until ASID space runs out. + * + * - mmu_gather::need_flush_all + * + * A flag that can be set by the arch code if it wants to force + * flush the entire TLB irrespective of the range. For instance + * x86-PAE needs this when changing top-level entries. + * + * And requires the architecture to provide and implement tlb_flush(). + * + * tlb_flush() may, in addition to the above mentioned mmu_gather fields, make + * use of: + * + * - mmu_gather::start / mmu_gather::end + * + * which provides the range that needs to be flushed to cover the pages to + * be freed. + * + * - mmu_gather::freed_tables + * + * set when we freed page table pages + * + * - tlb_get_unmap_shift() / tlb_get_unmap_size() + * + * returns the smallest TLB entry size unmapped in this range + * + * Additionally there are a few opt-in features: + * + * HAVE_RCU_TABLE_FREE + * + * This provides tlb_remove_table(), to be used instead of tlb_remove_page() + * for page directores (__p*_free_tlb()). This provides separate freeing of + * the page-table pages themselves in a semi-RCU fashion (see comment below). + * Useful if your architecture doesn't use IPIs for remote TLB invalidates + * and therefore doesn't naturally serialize with software page-table walkers. + * + * When used, an architecture is expected to provide __tlb_remove_table() + * which does the actual freeing of these pages. + * + * HAVE_RCU_TABLE_INVALIDATE + * + * This makes HAVE_RCU_TABLE_FREE call tlb_flush_mmu_tlbonly() before freeing + * the page-table pages. Required if you use HAVE_RCU_TABLE_FREE and your + * architecture uses the Linux page-tables natively. + * + */ +#define HAVE_GENERIC_MMU_GATHER + #ifdef CONFIG_HAVE_RCU_TABLE_FREE /* * Semi RCU freeing of the page directories. @@ -89,14 +201,17 @@ struct mmu_gather_batch { */ #define MAX_GATHER_BATCH_COUNT (10000UL/MAX_GATHER_BATCH) -/* struct mmu_gather is an opaque type used by the mm code for passing around +/* + * struct mmu_gather is an opaque type used by the mm code for passing around * any data needed by arch specific code for tlb_remove_page. */ struct mmu_gather { struct mm_struct *mm; + #ifdef CONFIG_HAVE_RCU_TABLE_FREE struct mmu_table_batch *batch; #endif + unsigned long start; unsigned long end; /* @@ -131,8 +246,6 @@ struct mmu_gather { int page_size; }; -#define HAVE_GENERIC_MMU_GATHER - void arch_tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm, unsigned long start, unsigned long end); void tlb_flush_mmu(struct mmu_gather *tlb); -- 2.25.1

1 7

[PATCH 1/3] mm: madvise: fix vma user-after-free
by Yang Yingliang 16 Nov '20

16 Nov '20

From: Yang Shi <shy828301(a)gmail.com> mainline inclusion from mainline-v5.9-rc4 commit 7867fd7cc44e63c6673cd0f8fea155456d34d0de category: bugfix bugzilla: 42216 CVE: NA ------------------------------------------------- The syzbot reported the below use-after-free: BUG: KASAN: use-after-free in madvise_willneed mm/madvise.c:293 [inline] BUG: KASAN: use-after-free in madvise_vma mm/madvise.c:942 [inline] BUG: KASAN: use-after-free in do_madvise.part.0+0x1c8b/0x1cf0 mm/madvise.c:1145 Read of size 8 at addr ffff8880a6163eb0 by task syz-executor.0/9996 CPU: 0 PID: 9996 Comm: syz-executor.0 Not tainted 5.9.0-rc1-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x18f/0x20d lib/dump_stack.c:118 print_address_description.constprop.0.cold+0xae/0x497 mm/kasan/report.c:383 __kasan_report mm/kasan/report.c:513 [inline] kasan_report.cold+0x1f/0x37 mm/kasan/report.c:530 madvise_willneed mm/madvise.c:293 [inline] madvise_vma mm/madvise.c:942 [inline] do_madvise.part.0+0x1c8b/0x1cf0 mm/madvise.c:1145 do_madvise mm/madvise.c:1169 [inline] __do_sys_madvise mm/madvise.c:1171 [inline] __se_sys_madvise mm/madvise.c:1169 [inline] __x64_sys_madvise+0xd9/0x110 mm/madvise.c:1169 do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Allocated by task 9992: kmem_cache_alloc+0x138/0x3a0 mm/slab.c:3482 vm_area_alloc+0x1c/0x110 kernel/fork.c:347 mmap_region+0x8e5/0x1780 mm/mmap.c:1743 do_mmap+0xcf9/0x11d0 mm/mmap.c:1545 vm_mmap_pgoff+0x195/0x200 mm/util.c:506 ksys_mmap_pgoff+0x43a/0x560 mm/mmap.c:1596 do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Freed by task 9992: kmem_cache_free.part.0+0x67/0x1f0 mm/slab.c:3693 remove_vma+0x132/0x170 mm/mmap.c:184 remove_vma_list mm/mmap.c:2613 [inline] __do_munmap+0x743/0x1170 mm/mmap.c:2869 do_munmap mm/mmap.c:2877 [inline] mmap_region+0x257/0x1780 mm/mmap.c:1716 do_mmap+0xcf9/0x11d0 mm/mmap.c:1545 vm_mmap_pgoff+0x195/0x200 mm/util.c:506 ksys_mmap_pgoff+0x43a/0x560 mm/mmap.c:1596 do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46 entry_SYSCALL_64_after_hwframe+0x44/0xa9 It is because vma is accessed after releasing mmap_lock, but someone else acquired the mmap_lock and the vma is gone. Releasing mmap_lock after accessing vma should fix the problem. Fixes: 692fe62433d4c ("mm: Handle MADV_WILLNEED through vfs_fadvise()") Reported-by: syzbot+b90df26038d1d5d85c97(a)syzkaller.appspotmail.com Signed-off-by: Yang Shi <shy828301(a)gmail.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Reviewed-by: Andrew Morton <akpm(a)linux-foundation.org> Reviewed-by: Jan Kara <jack(a)suse.cz> Cc: <stable(a)vger.kernel.org> [5.4+] Link: https://lkml.kernel.org/r/20200816141204.162624-1-shy828301@gmail.com Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Liu Shixin <liushixin2(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- mm/madvise.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/madvise.c b/mm/madvise.c index 464282e24a30..1369e6d062bc 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -308,9 +308,9 @@ static long madvise_willneed(struct vm_area_struct *vma, */ *prev = NULL; /* tell sys_madvise we drop mmap_sem */ get_file(file); - up_read(&current->mm->mmap_sem); offset = (loff_t)(start - vma->vm_start) + ((loff_t)vma->vm_pgoff << PAGE_SHIFT); + up_read(&current->mm->mmap_sem); vfs_fadvise(file, offset, end - start, POSIX_FADV_WILLNEED); fput(file); down_read(&current->mm->mmap_sem); -- 2.25.1

1 2

[PATCH 000/174] backport stable-4.19.155
by Yang Yingliang 16 Nov '20

16 Nov '20

Alain Volmat (1): cpufreq: sti-cpufreq: add stih418 support Aleksandr Nogikh (1): netem: fix zero division in tabledist Alex Hung (1): ACPI: video: use ACPI backlight for HP 635 Notebook Alexander Sverdlin (2): staging: octeon: repair "fixed-link" support staging: octeon: Drop on uncorrectable alignment or FCS error Alok Prasad (1): RDMA/qedr: Fix memory leak in iWARP CM Amit Cohen (1): mlxsw: core: Fix use-after-free in mlxsw_emad_trans_finish() Anand Jain (2): btrfs: fix replace of seed device btrfs: improve device scanning messages Anant Thazhemadam (2): net: 9p: initialize sun_server.sun_path to have addr's value only when addr is valid gfs2: add validation checks for size of superblock Andrew Donnellan (1): powerpc/rtas: Restrict RTAS requests from userspace Andrew Gabbasov (1): ravb: Fix bit fields checking in ravb_hwtstamp_get() Andy Shevchenko (2): device property: Keep secondary firmware node secondary by type device property: Don't clear secondary pointer for shared primary firmware node Aneesh Kumar K.V (1): powerpc/drmem: Make lmb_size 64 bit Antonio Borneo (1): drm/bridge/synopsys: dsi: add support for non-continuous HS clock Arjun Roy (1): tcp: Prevent low rmem stalls with SO_RCVLOWAT. Ashish Sangwan (1): NFS: fix nfs_path in case of a rename retry Badhri Jagan Sridharan (1): usb: typec: tcpm: During PR_SWAP, source caps should be sent only after tSwapSourceStart Bartosz Golaszewski (1): rtc: rx8010: don't modify the global rtc ops Ben Hutchings (1): ACPI / extlog: Check for RDMSR failure Chao Leng (1): nvme-rdma: fix crash when connect rejected Chao Yu (2): f2fs: fix uninit-value in f2fs_lookup f2fs: fix to check segment boundary during SIT page readahead Chris Lew (1): rpmsg: glink: Use complete_all for open states Chris Wilson (1): drm/i915: Force VT'd workarounds when running as a guest OS Chuck Lever (1): NFSD: Add missing NFSv2 .pc_func methods Dan Carpenter (1): memory: emif: Remove bogus debugfs error handling Daniel W. S. Almeida (1): media: uvcvideo: Fix dereference of out-of-bound list iterator Darrick J. Wong (2): xfs: fix realtime bitmap/summary file truncation when growing rt volume xfs: don't free rt blocks when we're doing a REMAP bunmapi call Dave Airlie (1): drm/ttm: fix eviction valuable range check. Denis Efremov (1): btrfs: use kvzalloc() to allocate clone_roots in btrfs_ioctl_send() Diana Craciun (1): bus/fsl_mc: Do not rely on caller to provide non NULL mc_io Dinghao Liu (1): ext4: fix error handling code in add_new_gdb Douglas Anderson (2): ARM: 8997/2: hw_breakpoint: Handle inexact watchpoint addresses kgdb: Make "kgdbcon" work properly with "kgdb_earlycon" Douglas Gilbert (1): sgl_alloc_order: fix memory leak Eric Biggers (7): fscrypt: return -EXDEV for incompatible rename or link into encrypted dir fscrypt: clean up and improve dentry revalidation fscrypt: fix race allowing rename() and link() of ciphertext dentries fs, fscrypt: clear DCACHE_ENCRYPTED_NAME when unaliasing directory fscrypt: only set dentry_operations on ciphertext dentries fscrypt: fix race where ->lookup() marks plaintext dentry as ciphertext ext4: fix leaking sysfs kobject after failed mount Fangzhi Zuo (1): drm/amd/display: HDMI remote sink need mode validation for Linux Filipe Manana (3): btrfs: reschedule if necessary when logging directory items btrfs: send, recompute reference path after orphanization of a directory btrfs: fix use-after-free on readahead extent after failure to create it Frank Wunderlich (1): arm: dts: mt7623: add missing pause for switchport Frederic Barrat (1): cxl: Rework error message for incompatible slots Geert Uytterhoeven (1): ata: sata_rcar: Fix DMA boundary mask Greg Kroah-Hartman (2): Revert "block: ratelimit handle_bad_sector() message" Linux 4.19.155 Gustavo A. R. Silva (1): mtd: lpddr: Fix bad logic in print_drs_error Hans Verkuil (2): media: videodev2.h: RGB BT2020 and HSV are always full range media: imx274: fix frame interval handling Hans de Goede (1): media: uvcvideo: Fix uvc_ctrl_fixup_xu_info() not having any effect Heiner Kallweit (1): r8169: fix issue with forced threading in combination with shared interrupts Helge Deller (2): scsi: mptfusion: Fix null pointer dereferences in mptscsih_remove() hil/parisc: Disable HIL driver when it gets stuck Ian Abbott (1): staging: comedi: cb_pcidas: Allow 2-channel commands for AO subdevice Ido Schimmel (1): mlxsw: core: Fix memory leak on module removal Ilya Dryomov (1): libceph: clear con->out_msg on Policy::stateful_server faults Jamie Iles (1): ACPI: debug: don't allow debugging when ACPI is disabled Jan Kara (2): ext4: Detect already used quota file early udf: Fix memory leak when mounting Jason Gerecke (1): HID: wacom: Avoid entering wacom_wac_pen_report for pad / battery Jason Gunthorpe (1): RDMA/addr: Fix race with netevent_callback()/rdma_addr_cancel() Jerome Brunet (1): usb: cdc-acm: fix cooldown mechanism Jia-Ju Bai (1): p54: avoid accessing the data mapped to streaming DMA Jiri Olsa (1): perf python scripting: Fix printable strings in python3 scripts Jiri Slaby (1): x86/unwind/orc: Fix inactive tasks with stack pointer in %sp on GCC 10 compiled kernels Jisheng Zhang (1): arm64: berlin: Select DW_APB_TIMER_OF Joel Stanley (1): powerpc: Warn about use of smt_snooze_delay Johannes Berg (1): um: change sigio_spinlock to a mutex John Ogness (1): printk: reduce LOG_BUF_SHIFT range for H8300 Jonathan Cameron (5): ACPI: Add out of bounds and numa_off protections to pxm_to_node() iio:light:si1145: Fix timestamp alignment and prevent data leak. iio:adc:ti-adc0832 Fix alignment issue with timestamp iio:adc:ti-adc12138 Fix alignment issue with timestamp iio:gyro:itg3200: Fix timestamp alignment and prevent data leak. Josef Bacik (1): btrfs: cleanup cow block on error Josh Poimboeuf (1): objtool: Support Clang non-section symbols in ORC generation Juergen Gross (2): x86/xen: disable Firmware First mode for correctable memory errors xen/events: block rogue events for some time Kim Phillips (3): arch/x86/amd/ibs: Fix re-arming IBS Fetch perf/x86/amd/ibs: Don't include randomized bits in get_ibs_op_count() perf/x86/amd/ibs: Fix raw sample data accumulation Krzysztof Kozlowski (8): power: supply: bq27xxx: report "not charging" on all types ARM: dts: s5pv210: remove DMA controller bus node name to fix dtschema warnings ARM: dts: s5pv210: move PMU node out of clock controller ARM: dts: s5pv210: remove dedicated 'audio-subsystem' node ia64: fix build error with !COREDUMP i2c: imx: Fix external abort on interrupt in exit paths ARM: samsung: fix PM debug build with DEBUG_LL but !MMU ARM: s3c24xx: fix missing system reset Lang Dai (1): uio: free uio id after uio file node is freed Li Jun (3): usb: dwc3: core: add phy cleanup for probe error handling usb: dwc3: core: don't trigger runtime pm when remove driver usb: typec: tcpm: reset hard_reset_count for any disconnect Luo Meng (1): ext4: fix invalid inode checksum Madhav Chauhan (1): drm/amdgpu: don't map BO in reserved region Madhuparna Bhowmik (2): mmc: via-sdmmc: Fix data race bug drivers: watchdog: rdc321x_wdt: Fix race condition bugs Mahesh Salgaonkar (1): powerpc/powernv/elog: Fix race while processing OPAL error log event. Marc Zyngier (2): arm64: Run ARCH_WORKAROUND_1 enabling code on all CPUs KVM: arm64: Fix AArch32 handling of DBGD{CCINT, SCRext} and DBGVCR Marek Behún (1): leds: bcm6328, bcm6358: use devres LED registering function Martin Fuzzey (1): w1: mxc_w1: Fix timeout resolution problem leading to bus error Masahiro Fujiwara (1): gtp: fix an use-before-init in gtp_newlink() Masami Hiramatsu (1): ia64: kprobes: Use generic kretprobe trampoline handler Mateusz Nosek (1): futex: Fix incorrect should_fail_futex() handling Matthew Wilcox (Oracle) (3): ceph: promote to unsigned long long before shifting 9P: Cast to loff_t before multiplying cachefiles: Handle readpage error correctly Michael Chan (1): bnxt_en: Log unknown link speed appropriately. Michael Neuling (1): powerpc: Fix undetected data corruption with P9N DD2.1 VSX CI load emulation Michael Schaller (1): efivarfs: Replace invalid slashes with exclamation marks in dentries. Miklos Szeredi (1): fuse: fix page dereference after free Nadezda Lutovinova (1): drm/brige/megachips: Add checking if ge_b850v3_lvds_init() is working correctly Nicholas Piggin (3): mm: fix exec activate_mm vs TLB shootdown and lazy tlb switching race powerpc: select ARCH_WANT_IRQS_OFF_ACTIVATE_MM sparc64: remove mm_cpumask clearing to fix kthread_use_mm race Nick Desaulniers (1): arm64: link with -z norelro regardless of CONFIG_RELOCATABLE Olga Kornievskaia (1): NFSv4.2: support EXCHGID4_FLAG_SUPP_FENCE_OPS 4.2 EXCHANGE_ID flag Oliver Neukum (1): USB: adutux: fix debugging Oliver O'Halloran (1): powerpc/powernv/smp: Fix spurious DBG() warning Paul Cercueil (1): dmaengine: dma-jz4780: Fix race in jz4780_dma_tx_status Peter Chen (1): usb: xhci: omit duplicate actions when suspending a runtime suspended host. Peter Zijlstra (1): serial: pl011: Fix lockdep splat when handling magic-sysrq interrupt Qiujun Huang (1): ring-buffer: Return 0 on success from ring_buffer_resize() Qu Wenruo (1): btrfs: qgroup: fix wrong qgroup metadata reserve for delayed inode Quinn Tran (1): scsi: qla2xxx: Fix crash on session cleanup with unload Raju Rangoju (1): cxgb4: set up filter action after rewrites Ran Wang (1): usb: host: fsl-mph-dr-of: check return of dma_set_mask() Randy Dunlap (1): x86/PCI: Fix intel_mid_pci.c build error when ACPI is not enabled Rasmus Villemoes (1): scripts/setlocalversion: make git describe output more reliable Raul E Rangel (1): mmc: sdhci-acpi: AMDI0040: Set SDHCI_QUIRK2_PRESET_VALUE_BROKEN Ronnie Sahlberg (1): cifs: handle -EINTR in cifs_setattr Sandeep Singh (1): usb: xhci: Workaround for S3 issue on AMD SNPS 3.0 xHC Sascha Hauer (1): ata: sata_nv: Fix retrieving of active qcs Sathishkumar Muruganandam (1): ath10k: fix VHT NSS calculation when STBC is enabled Song Liu (2): bpf: Fix comment for helper bpf_current_task_under_cgroup() md/raid5: fix oops during stripe resizing Stefano Garzarella (1): vringh: fix __vringh_iov() when riov and wiov are different Sven Schnelle (1): s390/stp: add locking to sysfs functions Takashi Iwai (1): drm/amd/display: Don't invoke kgdb_breakpoint() unconditionally Tero Kristo (1): clk: ti: clockdomain: fix static checker warning Thinh Nguyen (2): usb: dwc3: ep0: Fix ZLP for OUT ep0 requests usb: dwc3: gadget: Check MPS of the request length Tom Rix (2): video: fbdev: pvr2fb: initialize variables media: tw5864: check status of tw5864_frameinterval_get Tony Lindgren (1): ARM: dts: omap4: Fix sgx clock rate for 4430 Tung Nguyen (1): tipc: fix memory leak caused by tipc_buf_append() Valentin Schneider (1): arm64: topology: Stop using MPIDR for topology information Vinay Kumar Yadav (3): chelsio/chtls: fix deadlock issue chelsio/chtls: fix memory leaks in CPL handlers chelsio/chtls: fix tls record info to user Wei Huang (1): acpi-cpufreq: Honor _PSD table setting on new AMD CPUs Wen Gong (1): ath10k: start recovery process when payload length exceeds max htc length for sdio Xia Jiang (1): media: platform: Improve queue set up flow for bug fixing Xie He (1): drivers/net/wan/hdlc_fr: Correctly handle special skb->protocol values Xiongfeng Wang (1): power: supply: test_power: add missing newlines when printing parameters by sysfs Xiubo Li (1): nbd: make the config put is called before the notifying the waiter Yoshihiro Shimoda (1): arm64: dts: renesas: ulcb: add full-pwr-cycle-in-suspend into eMMC nodes Zhang Qilong (1): f2fs: add trace exit in exception path Zhao Heming (1): md/bitmap: md_bitmap_get_counter returns wrong blocks Zhengyuan Liu (1): arm64/mm: return cpu_all_mask when node is NUMA_NO_NODE Zhihao Cheng (2): ubifs: dent: Fix some potential memory leaks while iterating entries ubi: check kthread_should_stop() after the setting of task state Zong Li (1): riscv: Define AT_VECTOR_SIZE_ARCH for ARCH_DLINFO dmitry.torokhov(a)gmail.com (1): ACPI: button: fix handling lid state changes when input device closed Documentation/filesystems/fscrypt.rst | 12 +- .../media/uapi/v4l/colorspaces-defs.rst | 9 +- .../media/uapi/v4l/colorspaces-details.rst | 5 +- Makefile | 2 +- arch/Kconfig | 7 + arch/arm/Kconfig | 2 + arch/arm/boot/dts/mt7623n-bananapi-bpi-r2.dts | 1 + arch/arm/boot/dts/omap4.dtsi | 2 +- arch/arm/boot/dts/omap443x.dtsi | 10 ++ arch/arm/boot/dts/s5pv210.dtsi | 127 +++++++-------- arch/arm/kernel/hw_breakpoint.c | 100 ++++++++---- arch/arm/plat-samsung/Kconfig | 1 + arch/arm64/Kconfig.platforms | 1 + arch/arm64/Makefile | 4 +- arch/arm64/boot/dts/renesas/ulcb.dtsi | 1 + arch/arm64/include/asm/kvm_host.h | 1 + arch/arm64/include/asm/numa.h | 3 + arch/arm64/kernel/cpu_errata.c | 8 + arch/arm64/kernel/topology.c | 43 ++--- arch/arm64/kvm/sys_regs.c | 6 +- arch/arm64/mm/numa.c | 6 +- arch/ia64/kernel/Makefile | 2 +- arch/ia64/kernel/kprobes.c | 77 +-------- arch/powerpc/Kconfig | 14 ++ arch/powerpc/include/asm/drmem.h | 4 +- arch/powerpc/include/asm/mmu_context.h | 2 +- arch/powerpc/kernel/rtas.c | 153 ++++++++++++++++++ arch/powerpc/kernel/sysfs.c | 42 ++--- arch/powerpc/kernel/traps.c | 2 +- arch/powerpc/platforms/powernv/opal-elog.c | 33 +++- arch/powerpc/platforms/powernv/smp.c | 2 +- arch/riscv/include/uapi/asm/auxvec.h | 3 + arch/s390/kernel/time.c | 118 ++++++++++---- arch/sparc/kernel/smp_64.c | 65 ++------ arch/um/kernel/sigio.c | 6 +- arch/x86/events/amd/ibs.c | 53 ++++-- arch/x86/include/asm/msr-index.h | 1 + arch/x86/kernel/unwind_orc.c | 9 +- arch/x86/pci/intel_mid_pci.c | 1 + arch/x86/xen/enlighten_pv.c | 9 ++ block/blk-core.c | 9 +- drivers/acpi/acpi_dbg.c | 3 + drivers/acpi/acpi_extlog.c | 6 +- drivers/acpi/button.c | 13 +- drivers/acpi/numa.c | 2 +- drivers/acpi/video_detect.c | 9 ++ drivers/ata/sata_nv.c | 2 +- drivers/ata/sata_rcar.c | 2 +- drivers/base/core.c | 4 +- drivers/block/nbd.c | 2 +- drivers/bus/fsl-mc/mc-io.c | 7 +- drivers/clk/ti/clockdomain.c | 2 + drivers/cpufreq/acpi-cpufreq.c | 3 +- drivers/cpufreq/sti-cpufreq.c | 6 +- drivers/crypto/chelsio/chtls/chtls_cm.c | 29 ++-- drivers/crypto/chelsio/chtls/chtls_io.c | 7 +- drivers/dma/dma-jz4780.c | 7 +- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 10 ++ drivers/gpu/drm/amd/display/dc/core/dc_link.c | 2 +- drivers/gpu/drm/amd/display/dc/os_types.h | 2 +- .../bridge/megachips-stdpxxxx-ge-b850v3-fw.c | 12 +- drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c | 9 +- drivers/gpu/drm/i915/i915_drv.h | 6 +- drivers/gpu/drm/ttm/ttm_bo.c | 2 +- drivers/hid/wacom_wac.c | 4 +- drivers/i2c/busses/i2c-imx.c | 24 +-- drivers/iio/adc/ti-adc0832.c | 11 +- drivers/iio/adc/ti-adc12138.c | 13 +- drivers/iio/gyro/itg3200_buffer.c | 15 +- drivers/iio/light/si1145.c | 19 ++- drivers/infiniband/core/addr.c | 11 +- drivers/infiniband/hw/qedr/qedr_iw_cm.c | 1 + drivers/input/serio/hil_mlc.c | 21 ++- drivers/input/serio/hp_sdc_mlc.c | 8 +- drivers/leds/leds-bcm6328.c | 2 +- drivers/leds/leds-bcm6358.c | 2 +- drivers/md/md-bitmap.c | 2 +- drivers/md/raid5.c | 4 +- drivers/media/i2c/imx274.c | 8 +- drivers/media/pci/tw5864/tw5864-video.c | 6 + .../media/platform/mtk-jpeg/mtk_jpeg_core.c | 7 + drivers/media/usb/uvc/uvc_ctrl.c | 27 ++-- drivers/memory/emif.c | 33 +--- drivers/message/fusion/mptscsih.c | 13 +- drivers/misc/cxl/pci.c | 4 +- drivers/mmc/host/sdhci-acpi.c | 37 +++++ drivers/mmc/host/via-sdmmc.c | 3 + drivers/mtd/ubi/wl.c | 13 ++ drivers/net/ethernet/broadcom/bnxt/bnxt.c | 6 +- .../net/ethernet/chelsio/cxgb4/cxgb4_filter.c | 56 ++++--- drivers/net/ethernet/chelsio/cxgb4/t4_tcb.h | 4 + drivers/net/ethernet/mellanox/mlxsw/core.c | 5 + drivers/net/ethernet/realtek/r8169.c | 4 +- drivers/net/ethernet/renesas/ravb_main.c | 10 +- drivers/net/gtp.c | 16 +- drivers/net/wan/hdlc_fr.c | 98 +++++------ drivers/net/wireless/ath/ath10k/htt_rx.c | 8 +- drivers/net/wireless/ath/ath10k/sdio.c | 4 + drivers/net/wireless/intersil/p54/p54pci.c | 4 +- drivers/nvme/host/rdma.c | 1 - drivers/power/supply/bq27xxx_battery.c | 6 +- drivers/power/supply/test_power.c | 6 + drivers/rpmsg/qcom_glink_native.c | 6 +- drivers/rtc/rtc-rx8010.c | 24 ++- drivers/scsi/qla2xxx/qla_target.c | 13 +- drivers/staging/comedi/drivers/cb_pcidas.c | 1 + drivers/staging/octeon/ethernet-mdio.c | 6 - drivers/staging/octeon/ethernet-rx.c | 34 ++-- drivers/staging/octeon/ethernet.c | 9 ++ drivers/tty/serial/amba-pl011.c | 11 +- drivers/uio/uio.c | 4 +- drivers/usb/class/cdc-acm.c | 12 +- drivers/usb/class/cdc-acm.h | 3 +- drivers/usb/dwc3/core.c | 15 +- drivers/usb/dwc3/ep0.c | 11 +- drivers/usb/dwc3/gadget.c | 4 +- drivers/usb/host/fsl-mph-dr-of.c | 9 +- drivers/usb/host/xhci-pci.c | 17 ++ drivers/usb/host/xhci.c | 7 +- drivers/usb/host/xhci.h | 1 + drivers/usb/misc/adutux.c | 1 + drivers/usb/typec/tcpm.c | 8 +- drivers/vhost/vringh.c | 9 +- drivers/video/fbdev/pvr2fb.c | 2 + drivers/w1/masters/mxc_w1.c | 14 +- drivers/watchdog/rdc321x_wdt.c | 5 +- drivers/xen/events/events_base.c | 27 +++- drivers/xen/events/events_internal.h | 3 +- fs/9p/vfs_file.c | 4 +- fs/btrfs/ctree.c | 6 + fs/btrfs/delayed-inode.c | 3 +- fs/btrfs/dev-replace.c | 2 +- fs/btrfs/reada.c | 2 + fs/btrfs/send.c | 74 ++++++++- fs/btrfs/tree-log.c | 8 + fs/btrfs/volumes.c | 14 +- fs/cachefiles/rdwr.c | 3 +- fs/ceph/addr.c | 2 +- fs/cifs/inode.c | 13 +- fs/crypto/crypto.c | 58 +++---- fs/crypto/fname.c | 1 + fs/crypto/hooks.c | 34 ++-- fs/crypto/policy.c | 3 +- fs/dcache.c | 15 ++ fs/efivarfs/super.c | 3 + fs/exec.c | 15 +- fs/ext4/ext4.h | 62 +++++-- fs/ext4/inode.c | 11 +- fs/ext4/namei.c | 76 ++++++--- fs/ext4/resize.c | 4 +- fs/ext4/super.c | 6 + fs/f2fs/checkpoint.c | 8 +- fs/f2fs/dir.c | 8 +- fs/f2fs/namei.c | 17 +- fs/fuse/dev.c | 28 ++-- fs/gfs2/ops_fstype.c | 18 ++- fs/nfs/namespace.c | 12 +- fs/nfs/nfs4proc.c | 9 +- fs/nfsd/nfsproc.c | 16 ++ fs/ubifs/debug.c | 1 + fs/ubifs/dir.c | 8 +- fs/udf/super.c | 21 ++- fs/xfs/libxfs/xfs_bmap.c | 19 ++- fs/xfs/xfs_rtalloc.c | 10 +- include/linux/dcache.h | 2 +- include/linux/fscrypt.h | 34 ++-- include/linux/fscrypt_notsupp.h | 9 +- include/linux/fscrypt_supp.h | 6 +- include/linux/hil_mlc.h | 2 +- include/linux/mtd/pfow.h | 2 +- include/linux/usb/pd.h | 1 + include/uapi/linux/bpf.h | 4 +- include/uapi/linux/nfs4.h | 3 + include/uapi/linux/videodev2.h | 17 +- init/Kconfig | 3 +- kernel/debug/debug_core.c | 22 ++- kernel/futex.c | 4 +- kernel/trace/ring_buffer.c | 8 +- lib/scatterlist.c | 2 +- net/9p/trans_fd.c | 2 +- net/ceph/messenger.c | 5 + net/ipv4/tcp.c | 2 + net/ipv4/tcp_input.c | 3 +- net/sched/sch_netem.c | 9 +- net/tipc/msg.c | 5 +- scripts/setlocalversion | 21 ++- tools/include/uapi/linux/bpf.h | 4 +- tools/objtool/orc_gen.c | 33 +++- tools/perf/util/print_binary.c | 2 +- 189 files changed, 1746 insertions(+), 928 deletions(-) -- 2.25.1

1 174

[PATCH 1/5] partitions/efi: Fix partition name parsing in GUID partition entry
by Yang Yingliang 16 Nov '20

16 Nov '20

From: Nikolai Merinov <n.merinov(a)inango-systems.com> mainline inclusion from mainline-5.7-rc1 commit d5528d5e91041e68e8eab9792ce627705a0ed273 category: bugfix bugzilla: 32454 CVE: NA --------------------------- GUID partition entry defined to have a partition name as 36 UTF-16LE code units. This means that on big-endian platforms ASCII symbols would be read with 0xXX00 efi_char16_t character code. In order to correctly extract ASCII characters from a partition name field we should be converted from 16LE to CPU architecture. The problem exists on all big endian platforms. [ mingo: Minor edits. ] Fixes: eec7ecfede74 ("genhd, efi: add efi partition metadata to hd_structs") Reviewed-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Nikolai Merinov <n.merinov(a)inango-systems.com> Signed-off-by: Ard Biesheuvel <ardb(a)kernel.org> Signed-off-by: Ingo Molnar <mingo(a)kernel.org> Link: https://lore.kernel.org/r/20200308080859.21568-29-ardb@kernel.org Link: https://lore.kernel.org/r/797777312.1324734.1582544319435.JavaMail.zimbra@i… Signed-off-by: yangerkun <yangerkun(a)huawei.com> Reviewed-by: Yufen Yu <yuyufen(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- block/partitions/efi.c | 35 ++++++++++++++++++++++++++--------- block/partitions/efi.h | 2 +- 2 files changed, 27 insertions(+), 10 deletions(-) diff --git a/block/partitions/efi.c b/block/partitions/efi.c index 39f70d968754..b9beaa0a9b36 100644 --- a/block/partitions/efi.c +++ b/block/partitions/efi.c @@ -670,6 +670,31 @@ static int find_valid_gpt(struct parsed_partitions *state, gpt_header **gpt, return 0; } +/** + * utf16_le_to_7bit(): Naively converts a UTF-16LE string to 7-bit ASCII characters + * @in: input UTF-16LE string + * @size: size of the input string + * @out: output string ptr, should be capable to store @size+1 characters + * + * Description: Converts @size UTF16-LE symbols from @in string to 7-bit + * ASCII characters and stores them to @out. Adds trailing zero to @out array. + */ +static void utf16_le_to_7bit(const __le16 *in, unsigned int size, u8 *out) +{ + unsigned int i = 0; + + out[size] = 0; + + while (i < size) { + u8 c = le16_to_cpu(in[i]) & 0xff; + + if (c && !isprint(c)) + c = '!'; + out[i] = c; + i++; + } +} + /** * efi_partition(struct parsed_partitions *state) * @state: disk parsed partitions @@ -706,7 +731,6 @@ int efi_partition(struct parsed_partitions *state) for (i = 0; i < le32_to_cpu(gpt->num_partition_entries) && i < state->limit-1; i++) { struct partition_meta_info *info; - unsigned label_count = 0; unsigned label_max; u64 start = le64_to_cpu(ptes[i].starting_lba); u64 size = le64_to_cpu(ptes[i].ending_lba) - @@ -727,14 +751,7 @@ int efi_partition(struct parsed_partitions *state) /* Naively convert UTF16-LE to 7 bits. */ label_max = min(ARRAY_SIZE(info->volname) - 1, ARRAY_SIZE(ptes[i].partition_name)); - info->volname[label_max] = 0; - while (label_count < label_max) { - u8 c = ptes[i].partition_name[label_count] & 0xff; - if (c && !isprint(c)) - c = '!'; - info->volname[label_count] = c; - label_count++; - } + utf16_le_to_7bit(ptes[i].partition_name, label_max, info->volname); state->parts[i + 1].has_info = true; } kfree(ptes); diff --git a/block/partitions/efi.h b/block/partitions/efi.h index abd0b19288a6..42db2513ecfa 100644 --- a/block/partitions/efi.h +++ b/block/partitions/efi.h @@ -102,7 +102,7 @@ typedef struct _gpt_entry { __le64 starting_lba; __le64 ending_lba; gpt_entry_attributes attributes; - efi_char16_t partition_name[72 / sizeof (efi_char16_t)]; + __le16 partition_name[72/sizeof(__le16)]; } __packed gpt_entry; typedef struct _gpt_mbr_record { -- 2.25.1

1 4

[PATCH 1/2] evm: Reset IMA flags if EVM ignore mode is selected
by Roberto Sassu 14 Nov '20

14 Nov '20

hulk inclusion category: feature feature: digest-lists --------------------------- The EVM ignore mode works similarly to the metadata modification mode. They both allow an operation to be performed even if the operation causes metadata to become invalid. Currently, evm_reset_status() notifies to IMA that an operation modified metadata only when the metadata modification mode was chosen. With this patch, evm_reset_status() does the same also when the EVM ignore mode is selected. Signed-off-by: Roberto Sassu <roberto.sassu(a)huawei.com> --- security/integrity/evm/evm_main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/security/integrity/evm/evm_main.c b/security/integrity/evm/evm_main.c index 5155ff4c4ef2..2d3c1670d8d3 100644 --- a/security/integrity/evm/evm_main.c +++ b/security/integrity/evm/evm_main.c @@ -570,7 +570,8 @@ static void evm_reset_status(struct inode *inode, int bit) iint = integrity_iint_find(inode); if (iint) { - if (evm_initialized & EVM_ALLOW_METADATA_WRITES) + if ((evm_initialized & EVM_ALLOW_METADATA_WRITES) || + evm_ignoremode) set_bit(bit, &iint->atomic_flags); iint->evm_status = INTEGRITY_UNKNOWN; -- 2.27.GIT

1 1

openEuler kernel sig meeting
by openEuler conference 13 Nov '20

13 Nov '20

2 3

[PATCH] tty: make FONTX ioctl use the tty pointer they were actually passed
by Yang Yingliang 05 Nov '20

05 Nov '20

From: Linus Torvalds <torvalds(a)linux-foundation.org> mainline inclusion from mainline-v5.10-rc3 commit 90bfdeef83f1d6c696039b6a917190dcbbad3220 category: bugfix bugzilla: NA CVE: CVE-2020-25668 -------------------------------- Some of the font tty ioctl's always used the current foreground VC for their operations. Don't do that then. This fixes a data race on fg_console. Side note: both Michael Ellerman and Jiri Slaby point out that all these ioctls are deprecated, and should probably have been removed long ago, and everything seems to be using the KDFONTOP ioctl instead. In fact, Michael points out that it looks like busybox's loadfont program seems to have switched over to using KDFONTOP exactly _because_ of this bug (ahem.. 12 years ago ;-). Reported-by: Minh Yuan <yuanmingbuaa(a)gmail.com> Acked-by: Michael Ellerman <mpe(a)ellerman.id.au> Acked-by: Jiri Slaby <jirislaby(a)kernel.org> Cc: Greg KH <greg(a)kroah.com> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Conflicts: drivers/tty/vt/vt_ioctl.c [yyl: There is no vt_io_fontreset(), change the vc_cons to vc in do_fontx_ioctl()] Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/tty/vt/vt_ioctl.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) diff --git a/drivers/tty/vt/vt_ioctl.c b/drivers/tty/vt/vt_ioctl.c index 6a82030cf1ef..2e959563af53 100644 --- a/drivers/tty/vt/vt_ioctl.c +++ b/drivers/tty/vt/vt_ioctl.c @@ -244,7 +244,7 @@ int vt_waitactive(int n) static inline int -do_fontx_ioctl(int cmd, struct consolefontdesc __user *user_cfd, int perm, struct console_font_op *op) +do_fontx_ioctl(struct vc_data *vc, int cmd, struct consolefontdesc __user *user_cfd, int perm, struct console_font_op *op) { struct consolefontdesc cfdarg; int i; @@ -262,15 +262,16 @@ do_fontx_ioctl(int cmd, struct consolefontdesc __user *user_cfd, int perm, struc op->height = cfdarg.charheight; op->charcount = cfdarg.charcount; op->data = cfdarg.chardata; - return con_font_op(vc_cons[fg_console].d, op); - case GIO_FONTX: { + return con_font_op(vc, op); + + case GIO_FONTX: op->op = KD_FONT_OP_GET; op->flags = KD_FONT_FLAG_OLD; op->width = 8; op->height = cfdarg.charheight; op->charcount = cfdarg.charcount; op->data = cfdarg.chardata; - i = con_font_op(vc_cons[fg_console].d, op); + i = con_font_op(vc, op); if (i) return i; cfdarg.charheight = op->height; @@ -278,7 +279,6 @@ do_fontx_ioctl(int cmd, struct consolefontdesc __user *user_cfd, int perm, struc if (copy_to_user(user_cfd, &cfdarg, sizeof(struct consolefontdesc))) return -EFAULT; return 0; - } } return -EINVAL; } @@ -924,7 +924,7 @@ int vt_ioctl(struct tty_struct *tty, op.height = 0; op.charcount = 256; op.data = up; - ret = con_font_op(vc_cons[fg_console].d, &op); + ret = con_font_op(vc, &op); break; } @@ -935,7 +935,7 @@ int vt_ioctl(struct tty_struct *tty, op.height = 32; op.charcount = 256; op.data = up; - ret = con_font_op(vc_cons[fg_console].d, &op); + ret = con_font_op(vc, &op); break; } @@ -952,7 +952,7 @@ int vt_ioctl(struct tty_struct *tty, case PIO_FONTX: case GIO_FONTX: - ret = do_fontx_ioctl(cmd, up, perm, &op); + ret = do_fontx_ioctl(vc, cmd, up, perm, &op); break; case PIO_FONTRESET: @@ -969,11 +969,11 @@ int vt_ioctl(struct tty_struct *tty, { op.op = KD_FONT_OP_SET_DEFAULT; op.data = NULL; - ret = con_font_op(vc_cons[fg_console].d, &op); + ret = con_font_op(vc, &op); if (ret) break; console_lock(); - con_set_default_unimap(vc_cons[fg_console].d); + con_set_default_unimap(vc); console_unlock(); break; } @@ -1100,8 +1100,9 @@ struct compat_consolefontdesc { }; static inline int -compat_fontx_ioctl(int cmd, struct compat_consolefontdesc __user *user_cfd, - int perm, struct console_font_op *op) +compat_fontx_ioctl(struct vc_data *vc, int cmd, + struct compat_consolefontdesc __user *user_cfd, + int perm, struct console_font_op *op) { struct compat_consolefontdesc cfdarg; int i; @@ -1119,7 +1120,8 @@ compat_fontx_ioctl(int cmd, struct compat_consolefontdesc __user *user_cfd, op->height = cfdarg.charheight; op->charcount = cfdarg.charcount; op->data = compat_ptr(cfdarg.chardata); - return con_font_op(vc_cons[fg_console].d, op); + return con_font_op(vc, op); + case GIO_FONTX: op->op = KD_FONT_OP_GET; op->flags = KD_FONT_FLAG_OLD; @@ -1127,7 +1129,7 @@ compat_fontx_ioctl(int cmd, struct compat_consolefontdesc __user *user_cfd, op->height = cfdarg.charheight; op->charcount = cfdarg.charcount; op->data = compat_ptr(cfdarg.chardata); - i = con_font_op(vc_cons[fg_console].d, op); + i = con_font_op(vc, op); if (i) return i; cfdarg.charheight = op->height; @@ -1218,7 +1220,7 @@ long vt_compat_ioctl(struct tty_struct *tty, */ case PIO_FONTX: case GIO_FONTX: - ret = compat_fontx_ioctl(cmd, up, perm, &op); + ret = compat_fontx_ioctl(vc, cmd, up, perm, &op); break; case KDFONTOP: -- 2.25.1

1 0

openEuler Summit 2020 来了，欢迎申报申报 Summit 议题
by Liyongle (Fred) 05 Nov '20

05 Nov '20

openEuler Summit 是由 openEuler 社区举办的开发者交流会，首届线下 openEuler Summit 将于 12 月24-25日在北京举行。 openEuler Summit 广泛邀请操作系统生态的开发者、用户、社区贡献者、软件爱好者共同解读 openEuler 的最新版，探讨未来的技术路线，让技术、生态、商业在这里产生奇妙的化学反应。开源是一种态度、分享是一种精神。Call for Speaker、Call for Sponsor、Call for SIG、Call for Demo 现已全面开放报名。我们诚挚的邀请您报名演讲、演示 Demo、发表案例、参与社区建设。期待您的到来。 [cid:image001.png@01D6B34F.EDA0B950] [cid:image002.png@01D6B34F.EDA0B950] [cid:image003.png@01D6B34F.EDA0B950] [cid:image004.png@01D6B34F.EDA0B950] Fred 李永乐

1 0

[PATCH 1/2] vt: keyboard, simplify vt_kdgkbsent
by Yang Yingliang 04 Nov '20

04 Nov '20

From: Jiri Slaby <jslaby(a)suse.cz> mainline inclusion from mainline-v5.10-rc2 commit 6ca03f90527e499dd5e32d6522909e2ad390896b category: bugfix bugzilla: NA CVE: CVE-2020-25656 -------------------------------- Use 'strlen' of the string, add one for NUL terminator and simply do 'copy_to_user' instead of the explicit 'for' loop. This makes the KDGKBSENT case more compact. The only thing we need to take care about is NULL 'func_table[i]'. Use an empty string in that case. The original check for overflow could never trigger as the func_buf strings are always shorter or equal to 'struct kbsentry's. Cc: <stable(a)vger.kernel.org> Signed-off-by: Jiri Slaby <jslaby(a)suse.cz> Link: https://lore.kernel.org/r/20201019085517.10176-1-jslaby@suse.cz Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/tty/vt/keyboard.c | 28 +++++++++------------------- 1 file changed, 9 insertions(+), 19 deletions(-) diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c index a7455f8a4235..56315d23f34e 100644 --- a/drivers/tty/vt/keyboard.c +++ b/drivers/tty/vt/keyboard.c @@ -1994,9 +1994,7 @@ int vt_do_kdsk_ioctl(int cmd, struct kbentry __user *user_kbe, int perm, int vt_do_kdgkb_ioctl(int cmd, struct kbsentry __user *user_kdgkb, int perm) { struct kbsentry *kbs; - char *p; u_char *q; - u_char __user *up; int sz, fnw_sz; int delta; char *first_free, *fj, *fnw; @@ -2022,23 +2020,15 @@ int vt_do_kdgkb_ioctl(int cmd, struct kbsentry __user *user_kdgkb, int perm) i = kbs->kb_func; switch (cmd) { - case KDGKBSENT: - sz = sizeof(kbs->kb_string) - 1; /* sz should have been - a struct member */ - up = user_kdgkb->kb_string; - p = func_table[i]; - if(p) - for ( ; *p && sz; p++, sz--) - if (put_user(*p, up++)) { - ret = -EFAULT; - goto reterr; - } - if (put_user('\0', up)) { - ret = -EFAULT; - goto reterr; - } - kfree(kbs); - return ((p && *p) ? -EOVERFLOW : 0); + case KDGKBSENT: { + /* size should have been a struct member */ + unsigned char *from = func_table[i] ? : ""; + + ret = copy_to_user(user_kdgkb->kb_string, from, + strlen(from) + 1) ? -EFAULT : 0; + + goto reterr; + } case KDSKBSENT: if (!perm) { ret = -EPERM; -- 2.25.1

1 1

[PATCH] binder: fix UAF when releasing todo list
by Yang Yingliang 04 Nov '20

04 Nov '20

From: Todd Kjos <tkjos(a)google.com> stable inclusion from linux-4.19.153 commit 35cc2facc2a5ff52b9aa03f2dc81dcb000d97da3 CVE: CVE-2020-0423 -------------------------------- commit f3277cbfba763cd2826396521b9296de67cf1bbc upstream. When releasing a thread todo list when tearing down a binder_proc, the following race was possible which could result in a use-after-free: 1. Thread 1: enter binder_release_work from binder_thread_release 2. Thread 2: binder_update_ref_for_handle() -> binder_dec_node_ilocked() 3. Thread 2: dec nodeA --> 0 (will free node) 4. Thread 1: ACQ inner_proc_lock 5. Thread 2: block on inner_proc_lock 6. Thread 1: dequeue work (BINDER_WORK_NODE, part of nodeA) 7. Thread 1: REL inner_proc_lock 8. Thread 2: ACQ inner_proc_lock 9. Thread 2: todo list cleanup, but work was already dequeued 10. Thread 2: free node 11. Thread 2: REL inner_proc_lock 12. Thread 1: deref w->type (UAF) The problem was that for a BINDER_WORK_NODE, the binder_work element must not be accessed after releasing the inner_proc_lock while processing the todo list elements since another thread might be handling a deref on the node containing the binder_work element leading to the node being freed. Signed-off-by: Todd Kjos <tkjos(a)google.com> Link: https://lore.kernel.org/r/20201009232455.4054810-1-tkjos@google.com Cc: <stable(a)vger.kernel.org> # 4.14, 4.19, 5.4, 5.8 Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/android/binder.c | 35 ++++++++++------------------------- 1 file changed, 10 insertions(+), 25 deletions(-) diff --git a/drivers/android/binder.c b/drivers/android/binder.c index cf4367135a00..18c32637047f 100644 --- a/drivers/android/binder.c +++ b/drivers/android/binder.c @@ -285,7 +285,7 @@ struct binder_device { struct binder_work { struct list_head entry; - enum { + enum binder_work_type { BINDER_WORK_TRANSACTION = 1, BINDER_WORK_TRANSACTION_COMPLETE, BINDER_WORK_RETURN_ERROR, @@ -895,27 +895,6 @@ static struct binder_work *binder_dequeue_work_head_ilocked( return w; } -/** - * binder_dequeue_work_head() - Dequeues the item at head of list - * @proc: binder_proc associated with list - * @list: list to dequeue head - * - * Removes the head of the list if there are items on the list - * - * Return: pointer dequeued binder_work, NULL if list was empty - */ -static struct binder_work *binder_dequeue_work_head( - struct binder_proc *proc, - struct list_head *list) -{ - struct binder_work *w; - - binder_inner_proc_lock(proc); - w = binder_dequeue_work_head_ilocked(list); - binder_inner_proc_unlock(proc); - return w; -} - static void binder_defer_work(struct binder_proc *proc, enum binder_deferred_state defer); static void binder_free_thread(struct binder_thread *thread); @@ -4229,13 +4208,17 @@ static void binder_release_work(struct binder_proc *proc, struct list_head *list) { struct binder_work *w; + enum binder_work_type wtype; while (1) { - w = binder_dequeue_work_head(proc, list); + binder_inner_proc_lock(proc); + w = binder_dequeue_work_head_ilocked(list); + wtype = w ? w->type : 0; + binder_inner_proc_unlock(proc); if (!w) return; - switch (w->type) { + switch (wtype) { case BINDER_WORK_TRANSACTION: { struct binder_transaction *t; @@ -4269,9 +4252,11 @@ static void binder_release_work(struct binder_proc *proc, kfree(death); binder_stats_deleted(BINDER_STAT_DEATH); } break; + case BINDER_WORK_NODE: + break; default: pr_err("unexpected work type, %d, not freed\n", - w->type); + wtype); break; } } -- 2.25.1

1 0

[PATCH 001/110] mm: fix double page fault on arm64 if PTE_AF is cleared
by Yang Yingliang 04 Nov '20

04 Nov '20

From: Jia He <justin.he(a)arm.com> stable inclusion from linux-4.19.149 commit 8579a0440381353e0a71dd6a4d4371be8457eac4 -------------------------------- [ Upstream commit 83d116c53058d505ddef051e90ab27f57015b025 ] When we tested pmdk unit test [1] vmmalloc_fork TEST3 on arm64 guest, there will be a double page fault in __copy_from_user_inatomic of cow_user_page. To reproduce the bug, the cmd is as follows after you deployed everything: make -C src/test/vmmalloc_fork/ TEST_TIME=60m check Below call trace is from arm64 do_page_fault for debugging purpose: [ 110.016195] Call trace: [ 110.016826] do_page_fault+0x5a4/0x690 [ 110.017812] do_mem_abort+0x50/0xb0 [ 110.018726] el1_da+0x20/0xc4 [ 110.019492] __arch_copy_from_user+0x180/0x280 [ 110.020646] do_wp_page+0xb0/0x860 [ 110.021517] __handle_mm_fault+0x994/0x1338 [ 110.022606] handle_mm_fault+0xe8/0x180 [ 110.023584] do_page_fault+0x240/0x690 [ 110.024535] do_mem_abort+0x50/0xb0 [ 110.025423] el0_da+0x20/0x24 The pte info before __copy_from_user_inatomic is (PTE_AF is cleared): [ffff9b007000] pgd=000000023d4f8003, pud=000000023da9b003, pmd=000000023d4b3003, pte=360000298607bd3 As told by Catalin: "On arm64 without hardware Access Flag, copying from user will fail because the pte is old and cannot be marked young. So we always end up with zeroed page after fork() + CoW for pfn mappings. we don't always have a hardware-managed access flag on arm64." This patch fixes it by calling pte_mkyoung. Also, the parameter is changed because vmf should be passed to cow_user_page() Add a WARN_ON_ONCE when __copy_from_user_inatomic() returns error in case there can be some obscure use-case (by Kirill). [1] https://github.com/pmem/pmdk/tree/master/src/test/vmmalloc_fork Signed-off-by: Jia He <justin.he(a)arm.com> Reported-by: Yibo Cai <Yibo.Cai(a)arm.com> Reviewed-by: Catalin Marinas <catalin.marinas(a)arm.com> Acked-by: Kirill A. Shutemov <kirill.shutemov(a)linux.intel.com> Signed-off-by: Catalin Marinas <catalin.marinas(a)arm.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- mm/memory.c | 104 ++++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 89 insertions(+), 15 deletions(-) diff --git a/mm/memory.c b/mm/memory.c index c65d3993ac63..c4c667c0b89d 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -117,6 +117,18 @@ int randomize_va_space __read_mostly = 2; #endif +#ifndef arch_faults_on_old_pte +static inline bool arch_faults_on_old_pte(void) +{ + /* + * Those arches which don't have hw access flag feature need to + * implement their own helper. By default, "true" means pagefault + * will be hit on old pte. + */ + return true; +} +#endif + static int __init disable_randmaps(char *s) { randomize_va_space = 0; @@ -2089,32 +2101,82 @@ static inline int pte_unmap_same(struct mm_struct *mm, pmd_t *pmd, return same; } -static inline void cow_user_page(struct page *dst, struct page *src, unsigned long va, struct vm_area_struct *vma) +static inline bool cow_user_page(struct page *dst, struct page *src, + struct vm_fault *vmf) { + bool ret; + void *kaddr; + void __user *uaddr; + bool force_mkyoung; + struct vm_area_struct *vma = vmf->vma; + struct mm_struct *mm = vma->vm_mm; + unsigned long addr = vmf->address; + debug_dma_assert_idle(src); + if (likely(src)) { + copy_user_highpage(dst, src, addr, vma); + return true; + } + /* * If the source page was a PFN mapping, we don't have * a "struct page" for it. We do a best-effort copy by * just copying from the original user address. If that * fails, we just zero-fill it. Live with it. */ - if (unlikely(!src)) { - void *kaddr = kmap_atomic(dst); - void __user *uaddr = (void __user *)(va & PAGE_MASK); + kaddr = kmap_atomic(dst); + uaddr = (void __user *)(addr & PAGE_MASK); + + /* + * On architectures with software "accessed" bits, we would + * take a double page fault, so mark it accessed here. + */ + force_mkyoung = arch_faults_on_old_pte() && !pte_young(vmf->orig_pte); + if (force_mkyoung) { + pte_t entry; + + vmf->pte = pte_offset_map_lock(mm, vmf->pmd, addr, &vmf->ptl); + if (!likely(pte_same(*vmf->pte, vmf->orig_pte))) { + /* + * Other thread has already handled the fault + * and we don't need to do anything. If it's + * not the case, the fault will be triggered + * again on the same address. + */ + ret = false; + goto pte_unlock; + } + entry = pte_mkyoung(vmf->orig_pte); + if (ptep_set_access_flags(vma, addr, vmf->pte, entry, 0)) + update_mmu_cache(vma, addr, vmf->pte); + } + + /* + * This really shouldn't fail, because the page is there + * in the page tables. But it might just be unreadable, + * in which case we just give up and fill the result with + * zeroes. + */ + if (__copy_from_user_inatomic(kaddr, uaddr, PAGE_SIZE)) { /* - * This really shouldn't fail, because the page is there - * in the page tables. But it might just be unreadable, - * in which case we just give up and fill the result with - * zeroes. + * Give a warn in case there can be some obscure + * use-case */ - if (__copy_from_user_inatomic(kaddr, uaddr, PAGE_SIZE)) - clear_page(kaddr); - kunmap_atomic(kaddr); - flush_dcache_page(dst); - } else - copy_user_highpage(dst, src, va, vma); + WARN_ON_ONCE(1); + clear_page(kaddr); + } + + ret = true; + +pte_unlock: + if (force_mkyoung) + pte_unmap_unlock(vmf->pte, vmf->ptl); + kunmap_atomic(kaddr); + flush_dcache_page(dst); + + return ret; } static gfp_t __get_fault_gfp_mask(struct vm_area_struct *vma) @@ -2268,7 +2330,19 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf) vmf->address); if (!new_page) goto oom; - cow_user_page(new_page, old_page, vmf->address, vma); + + if (!cow_user_page(new_page, old_page, vmf)) { + /* + * COW failed, if the fault was solved by other, + * it's fine. If not, userspace would re-fault on + * the same address and we will handle the fault + * from the second attempt. + */ + put_page(new_page); + if (old_page) + put_page(old_page); + return 0; + } } if (mem_cgroup_try_charge_delay(new_page, mm, GFP_KERNEL, &memcg, false)) -- 2.25.1

1 109

[PATCH] iscsi: remove unnecessary __GENKSYMS__ define
by Yang Yingliang 04 Nov '20

04 Nov '20

hulk inclusion category: bugfix bugzilla: NA CVE: NA ------------------------------ remove unnecessary __GENKSYMS__ define Fixes: fc7b0914a6fa ("iscsi: use dynamic single thread workqueue...") Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/scsi/libiscsi.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h index a5f8094e3334..e03b9244499e 100644 --- a/include/scsi/libiscsi.h +++ b/include/scsi/libiscsi.h @@ -253,9 +253,7 @@ struct iscsi_conn { /* custom statistics */ uint32_t eh_abort_cnt; uint32_t fmr_unalign_cnt; -#ifndef __GENKSYMS__ - int intimate_cpu; /* offset:588, KABI is ok */ -#endif + int intimate_cpu; }; struct iscsi_pool { -- 2.25.1

1 0

[PATCH] arm64: fix abi change caused by ILP32
by Yang Yingliang 02 Nov '20

02 Nov '20

From: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> hulk inclusion category: bugfix bugzilla: NA CVE: NA ---------------------------- One of the ILP32 patchset rename 'compat_user_mode' and 'compat_thumb_mode' to 'a32_user_mode' and 'a32_user_mode'. But these two macros are used in some opensource userspace application. To keep compatibility, we redefine these two macros. Fixes: 23b2f00 ("arm64: rename functions that reference compat term") Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/include/asm/ptrace.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/arm64/include/asm/ptrace.h b/arch/arm64/include/asm/ptrace.h index 3fafb1af3d66..f1662df255ca 100644 --- a/arch/arm64/include/asm/ptrace.h +++ b/arch/arm64/include/asm/ptrace.h @@ -213,6 +213,8 @@ static inline void forget_syscall(struct pt_regs *regs) #define a32_thumb_mode(regs) (0) #endif +#define compat_thumb_mode(regs) a32_thumb_mode(regs) + #define user_mode(regs) \ (((regs)->pstate & PSR_MODE_MASK) == PSR_MODE_EL0t) @@ -220,6 +222,8 @@ static inline void forget_syscall(struct pt_regs *regs) (((regs)->pstate & (PSR_MODE32_BIT | PSR_MODE_MASK)) == \ (PSR_MODE32_BIT | PSR_MODE_EL0t)) +#define compat_user_mode(regs) a32_user_mode(regs) + #define processor_mode(regs) \ ((regs)->pstate & PSR_MODE_MASK) -- 2.25.1

1 0

[PATCH 1/7] xen/events: add a proper barrier to 2-level uevent unmasking
by Yang Yingliang 02 Nov '20

02 Nov '20

From: Juergen Gross <jgross(a)suse.com> mainline inclusion from mainline-v5.10 commit 4d3fe31bd993ef504350989786858aefdb877daa category: bugfix bugzilla: NA CVE: CVE-2020-27673 -------------------------------- A follow-up patch will require certain write to happen before an event channel is unmasked. While the memory barrier is not strictly necessary for all the callers, the main one will need it. In order to avoid an extra memory barrier when using fifo event channels, mandate evtchn_unmask() to provide write ordering. The 2-level event handling unmask operation is missing an appropriate barrier, so add it. Fifo event channels are fine in this regard due to using sync_cmpxchg(). This is part of XSA-332. Cc: stable(a)vger.kernel.org Suggested-by: Julien Grall <julien(a)xen.org> Signed-off-by: Juergen Gross <jgross(a)suse.com> Reviewed-by: Julien Grall <jgrall(a)amazon.com> Reviewed-by: Wei Liu <wl(a)xen.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/xen/events/events_2l.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/xen/events/events_2l.c b/drivers/xen/events/events_2l.c index 5478c761cd41..f026624898e7 100644 --- a/drivers/xen/events/events_2l.c +++ b/drivers/xen/events/events_2l.c @@ -91,6 +91,8 @@ static void evtchn_2l_unmask(unsigned port) BUG_ON(!irqs_disabled()); + smp_wmb(); /* All writes before unmask must be visible. */ + if (unlikely((cpu != cpu_from_evtchn(port)))) do_hypercall = 1; else { -- 2.25.1

1 6

[PATCH] sdei_watchdog: avoid possible false hardlockup
by Yang Yingliang 02 Nov '20

02 Nov '20

From: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> hulk inclusion category: bugfix bugzilla: NA CVE: NA ------------------------------ Firmware may not trigger SDEI event as required frequency. SDEI event may be triggered too soon, which cause false hardlockup in kernel. Check the time stamp in sdei_watchdog_callbak and skip the hardlockup check if it is invoked too soon. Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/kernel/watchdog_sdei.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/arch/arm64/kernel/watchdog_sdei.c b/arch/arm64/kernel/watchdog_sdei.c index 05b9a9a223a7..aa980b090598 100644 --- a/arch/arm64/kernel/watchdog_sdei.c +++ b/arch/arm64/kernel/watchdog_sdei.c @@ -23,6 +23,7 @@ static int sdei_watchdog_event_num; static bool disable_sdei_nmi_watchdog; static bool sdei_watchdog_registered; +static DEFINE_PER_CPU(ktime_t, last_check_time); int watchdog_sdei_enable(unsigned int cpu) { @@ -35,6 +36,7 @@ int watchdog_sdei_enable(unsigned int cpu) refresh_hld_last_timestamp(); #endif + __this_cpu_write(last_check_time, ktime_get_mono_fast_ns()); sdei_api_set_secure_timer_period(watchdog_thresh); ret = sdei_api_event_enable(sdei_watchdog_event_num); @@ -63,6 +65,23 @@ void watchdog_sdei_disable(unsigned int cpu) static int sdei_watchdog_callback(u32 event, struct pt_regs *regs, void *arg) { + ktime_t delta, now = ktime_get_mono_fast_ns(); + + delta = now - __this_cpu_read(last_check_time); + __this_cpu_write(last_check_time, now); + + /* + * Set delta to 4/5 of the actual watchdog threshold period so the + * hrtimer is guaranteed to fire at least once within the real + * watchdog threshold. + */ + if (delta < watchdog_thresh * (u64)NSEC_PER_SEC * 4 / 5) { + pr_err(FW_BUG "SDEI Watchdog event triggered too soon, " + "time to last check:%lld ns\n", delta); + WARN_ON(1); + return 0; + } + watchdog_hardlockup_check(regs); return 0; -- 2.25.1

1 0

[PATCH 000/119] backport stable-4.19.154
by Yang Yingliang 02 Nov '20

02 Nov '20

Abhishek Pandit-Subedi (1): Bluetooth: Only mark socket zapped after unlocking Adam Goode (1): media: uvcvideo: Ensure all probed info is returned to v4l2 Aditya Pakki (1): media: st-delta: Fix reference count leak in delta_run_work Adrian Hunter (1): perf intel-pt: Fix "context_switch event has no tid" error Al Grant (1): perf: correct SNOOPX field offset Alex Williamson (1): vfio/pci: Clear token on bypass registration failure Alexander Aring (1): fs: dlm: fix configfs memory leak Athira Rajeev (1): powerpc/perf: Exclude pmc5/6 from the irrelevant PMU group constraints Brooke Basile (1): ath9k: hif_usb: fix race condition between usb_get_urb() and usb_kill_anchored_urbs() Can Guo (1): scsi: ufs: ufs-qcom: Fix race conditions caused by ufs_qcom_testbus_config() Chris Chiu (1): rtl8xxxu: prevent potential memory leak Christian Eggers (1): eeprom: at25: set minimum read/write access stride to 1 Christoph Hellwig (1): PM: hibernate: remove the bogus call to get_gendisk() in software_resume() Claudiu Beznea (1): clk: at91: clk-main: update key before writing AT91_CKGR_MOR Colin Ian King (1): IB/rdmavt: Fix sizeof mismatch Cong Wang (1): ip_gre: set dev->hard_header_len and dev->needed_headroom properly Cristian Ciocaltea (1): ARM: dts: owl-s500: Fix incorrect PPI interrupt specifiers Dan Aloni (1): svcrdma: fix bounce buffers for unaligned offsets and multiple pages Dan Carpenter (3): rpmsg: smd: Fix a kobj leak in in qcom_smd_parse_edge() Input: imx6ul_tsc - clean up some errors in imx6ul_tsc_resume() memory: omap-gpmc: Fix a couple off by ones Daniel Thompson (1): kdb: Fix pager search for multi-line strings Darrick J. Wong (2): ext4: limit entries returned when counting fsmap records xfs: make sure the rt allocator doesn't run off the end Dinghao Liu (7): watchdog: Fix memleak in watchdog_cdev_register watchdog: Use put_device on error media: vsp1: Fix runtime PM imbalance on error media: platform: s3c-camif: Fix runtime PM imbalance on error media: platform: sti: hva: Fix runtime PM imbalance on error media: bdisp: Fix runtime PM imbalance on error media: venus: core: Fix runtime PM imbalance in venus_probe Dirk Behme (1): i2c: rcar: Auto select RESET_CONTROLLER Doug Horn (1): Fix use after free in get_capset_info callback. Eli Billauer (1): usb: core: Solve race condition in anchor cleanup functions Eric Biggers (1): reiserfs: only call unlock_new_inode() if I_NEW Finn Thain (2): powerpc/tau: Check processor type before enabling TAU interrupt powerpc/tau: Disable TAU between measurements Francesco Ruggeri (1): netfilter: conntrack: connection timeout after re-register Greg Kroah-Hartman (1): Linux 4.19.154 Guenter Roeck (1): watchdog: sp5100: Fix definition of EFCH_PM_DECODEEN3 Hamish Martin (1): usb: ohci: Default to per-port over-current protection Hans de Goede (1): i2c: core: Restore acpi_walk_dep_device_list() getting called after registering the ACPI i2c devs Hauke Mehrtens (1): pwm: img: Fix null pointer access in probe Horia Geantă (1): ARM: dts: imx6sl: fix rng node Jamie Iles (1): f2fs: wait for sysfs kobject removal before freeing f2fs_sb_info Jan Kara (3): udf: Limit sparing table size udf: Avoid accessing uninitialized data on failed inode read reiserfs: Fix memory leak in reiserfs_parse_options() Jason Gunthorpe (2): RDMA/cma: Remove dead code for kernel rdmacm multicast RDMA/cma: Consolidate the destruction of a cma_multicast in one place Jassi Brar (1): mailbox: avoid timer start from callback Jernej Skrabec (1): ARM: dts: sun8i: r40: bananapi-m2-ultra: Fix dcdc1 regulator Jing Xiangfeng (3): rapidio: fix the missed put_device() for rio_mport_add_riodev scsi: mvumi: Fix error return in mvumi_io_attach() scsi: ibmvfc: Fix error return in ibmvfc_probe() Joakim Zhang (1): can: flexcan: flexcan_chip_stop(): add error handling and propagate error value Johan Hovold (1): USB: cdc-acm: handle broken union descriptors Juri Lelli (1): sched/features: Fix !CONFIG_JUMP_LABEL case Kaige Li (1): NTB: hw: amd: fix an issue about leak system resources Kajol Jain (1): powerpc/perf/hv-gpci: Fix starting index value Keita Suzuki (2): misc: rtsx: Fix memory leak in rtsx_pci_probe brcmsmac: fix memory leak in wlc_phy_attach_lcnphy Krzysztof Kozlowski (5): Input: ep93xx_keypad - fix handling of platform_get_irq() error Input: omap4-keypad - fix handling of platform_get_irq() error Input: twl4030_keypad - fix handling of platform_get_irq() error Input: sun4i-ps2 - fix handling of platform_get_irq() error memory: fsl-corenet-cf: Fix handling of platform_get_irq() error Leon Romanovsky (1): overflow: Include header file with SIZE_MAX declaration Lijun Ou (1): RDMA/hns: Set the unsupported wr opcode Lorenzo Colitti (1): usb: gadget: f_ncm: allow using NCM in SuperSpeed Plus gadgets. Mark Tomlinson (1): PCI: iproc: Set affinity mask on MSI interrupts Martijn de Gouw (1): SUNRPC: fix copying of multiple pages in gss_read_proxy_verf() Matthew Wilcox (Oracle) (1): ramfs: fix nommu mmap with gaps in the page cache Mauro Carvalho Chehab (2): media: saa7134: avoid a shift overflow usb: dwc3: simple: add support for Hikey 970 Michal Simek (1): arm64: dts: zynqmp: Remove additional compatible string for i2c IPs Navid Emamdoost (1): clk: bcm2835: add missing release if devm_clk_hw_register fails Nicholas Piggin (1): powerpc/64s/radix: Fix mm_cpumask trimming race vs kthread_use_mm Nilesh Javali (2): scsi: qedi: Protect active command list to avoid list corruption scsi: qedi: Fix list_del corruption while removing active I/O Oliver Neukum (2): media: ati_remote: sanity check for both endpoints USB: cdc-wdm: Make wdm_flush() interruptible and add wdm_fsync(). Pablo Neira Ayuso (1): netfilter: nf_fwd_netdev: clear timestamp in forwarding path Pali Rohár (1): mmc: sdio: Check for CISTPL_VERS_1 buffer size Pavel Machek (2): crypto: ccp - fix error handling media: firewire: fix memory leak Peilin Ye (1): ipvs: Fix uninit-value in do_ip_vs_set_ctl() Peng Fan (1): tty: serial: fsl_lpuart: fix lpuart32_poll_get_char Qiushi Wu (4): media: sti: Fix reference count leaks media: exynos4-is: Fix several reference count leaks due to pm_runtime_get_sync media: exynos4-is: Fix a reference count leak due to pm_runtime_get_sync media: exynos4-is: Fix a reference count leak Robert Hoo (1): KVM: x86: emulating RDPID failure shall return #UD rather than #GP Roman Bolshakov (1): scsi: target: core: Add CONTROL field for trace events Rustam Kovhaev (1): ntfs: add check for mft record size in superblock Sherry Sun (2): mic: vop: copy data to kernel space then write to io memory misc: vop: add round_up(x,4) for vring_size to avoid kernel panic Souptick Joarder (1): rapidio: fix error handling path Srikar Dronamraju (1): cpufreq: powernv: Fix frame-size-overflow in powernv_cpufreq_reboot_notifier Stephan Gerhold (2): arm64: dts: qcom: pm8916: Remove invalid reg size from wcd_codec arm64: dts: qcom: msm8916: Fix MDP/DSI interrupts Stephen Boyd (1): clk: rockchip: Initialize hw to error to avoid undefined behavior Tetsuo Handa (2): block: ratelimit handle_bad_sector() message mwifiex: don't call del_timer_sync() on uninitialized timer Thomas Pedersen (1): mac80211: handle lack of sband->bitrates in rates Tobias Jordan (1): lib/crc32.c: fix trivial typo in preprocessor condition Tong Zhang (1): tty: ipwireless: fix error handling Valentin Vidic (1): net: korina: cast KSEG0 address to pointer in kfree Vasant Hegde (1): powerpc/powernv/dump: Fix race while processing OPAL dump Vincent Mailhol (1): usb: cdc-acm: add quirk to blacklist ETAS ES58X devices Wang Yufen (1): brcm80211: fix possible memleak in brcmf_proto_msgbuf_attach Weihang Li (1): RDMA/hns: Fix missing sq_sig_type when querying QP Xiaolong Huang (1): media: media/pci: prevent memory leak in bttv_probe Xiaoyang Xu (1): vfio iommu type1: Fix memory leak in vfio_iommu_type1_pin_pages YueHaibing (2): Input: stmfts - fix a & vs && typo memory: omap-gpmc: Fix build error without CONFIG_OF Zekun Shen (1): ath10k: check idx validity in __ath10k_htt_rx_ring_fill_n() Zqiang (1): usb: gadget: function: printer: fix use-after-free in __lock_acquire zhenwei pi (1): nvmet: fix uninitialized work for zero kato Makefile | 2 +- arch/arm/boot/dts/imx6sl.dtsi | 2 + arch/arm/boot/dts/owl-s500.dtsi | 6 +- .../boot/dts/sun8i-r40-bananapi-m2-ultra.dts | 10 +- arch/arm64/boot/dts/qcom/msm8916.dtsi | 4 +- arch/arm64/boot/dts/qcom/pm8916.dtsi | 2 +- arch/arm64/boot/dts/xilinx/zynqmp.dtsi | 4 +- arch/powerpc/include/asm/tlb.h | 13 --- arch/powerpc/kernel/tau_6xx.c | 96 +++++++------------ arch/powerpc/mm/tlb-radix.c | 23 +++-- arch/powerpc/perf/hv-gpci-requests.h | 6 +- arch/powerpc/perf/isa207-common.c | 10 ++ arch/powerpc/platforms/Kconfig | 14 +-- arch/powerpc/platforms/powernv/opal-dump.c | 41 +++++--- arch/x86/kvm/emulate.c | 2 +- block/blk-core.c | 9 +- drivers/clk/at91/clk-main.c | 11 ++- drivers/clk/bcm/clk-bcm2835.c | 4 +- drivers/clk/rockchip/clk-half-divider.c | 2 +- drivers/cpufreq/powernv-cpufreq.c | 9 +- drivers/crypto/ccp/ccp-ops.c | 2 +- drivers/gpu/drm/virtio/virtgpu_kms.c | 2 + drivers/gpu/drm/virtio/virtgpu_vq.c | 10 +- drivers/i2c/busses/Kconfig | 1 + drivers/i2c/i2c-core-acpi.c | 11 ++- drivers/infiniband/core/cma.c | 84 +++++++--------- drivers/infiniband/hw/hns/hns_roce_hw_v1.c | 1 - drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 1 + drivers/infiniband/sw/rdmavt/vt.c | 4 +- drivers/input/keyboard/ep93xx_keypad.c | 4 +- drivers/input/keyboard/omap4-keypad.c | 6 +- drivers/input/keyboard/twl4030_keypad.c | 8 +- drivers/input/serio/sun4i-ps2.c | 9 +- drivers/input/touchscreen/imx6ul_tsc.c | 27 +++--- drivers/input/touchscreen/stmfts.c | 2 +- drivers/mailbox/mailbox.c | 12 ++- drivers/media/firewire/firedtv-fw.c | 6 +- drivers/media/pci/bt8xx/bttv-driver.c | 13 ++- drivers/media/pci/saa7134/saa7134-tvaudio.c | 3 +- drivers/media/platform/exynos4-is/fimc-isp.c | 4 +- drivers/media/platform/exynos4-is/fimc-lite.c | 2 +- drivers/media/platform/exynos4-is/media-dev.c | 4 +- drivers/media/platform/exynos4-is/mipi-csis.c | 4 +- drivers/media/platform/qcom/venus/core.c | 5 +- drivers/media/platform/s3c-camif/camif-core.c | 5 +- drivers/media/platform/sti/bdisp/bdisp-v4l2.c | 3 +- drivers/media/platform/sti/delta/delta-v4l2.c | 4 +- drivers/media/platform/sti/hva/hva-hw.c | 4 +- drivers/media/platform/vsp1/vsp1_drv.c | 11 ++- drivers/media/rc/ati_remote.c | 4 + drivers/media/usb/uvc/uvc_v4l2.c | 30 ++++++ drivers/memory/fsl-corenet-cf.c | 6 +- drivers/memory/omap-gpmc.c | 8 +- drivers/misc/cardreader/rtsx_pcr.c | 4 +- drivers/misc/eeprom/at25.c | 2 +- drivers/misc/mic/vop/vop_main.c | 2 +- drivers/misc/mic/vop/vop_vringh.c | 24 +++-- drivers/mmc/core/sdio_cis.c | 3 + drivers/net/can/flexcan.c | 34 +++++-- drivers/net/ethernet/korina.c | 4 +- drivers/net/wireless/ath/ath10k/htt_rx.c | 8 ++ drivers/net/wireless/ath/ath9k/hif_usb.c | 19 ++++ .../broadcom/brcm80211/brcmfmac/msgbuf.c | 2 + .../broadcom/brcm80211/brcmsmac/phy/phy_lcn.c | 4 +- drivers/net/wireless/marvell/mwifiex/usb.c | 3 +- .../wireless/realtek/rtl8xxxu/rtl8xxxu_core.c | 10 +- drivers/ntb/hw/amd/ntb_hw_amd.c | 1 + drivers/nvme/target/core.c | 3 +- drivers/pci/controller/pcie-iproc-msi.c | 13 ++- drivers/pwm/pwm-img.c | 3 +- drivers/rapidio/devices/rio_mport_cdev.c | 18 ++-- drivers/rpmsg/qcom_smd.c | 32 +++++-- drivers/scsi/ibmvscsi/ibmvfc.c | 1 + drivers/scsi/mvumi.c | 1 + drivers/scsi/qedi/qedi_fw.c | 23 ++++- drivers/scsi/qedi/qedi_iscsi.c | 2 + drivers/scsi/ufs/ufs-qcom.c | 5 - drivers/tty/ipwireless/network.c | 4 +- drivers/tty/ipwireless/tty.c | 2 +- drivers/tty/serial/fsl_lpuart.c | 2 +- drivers/usb/class/cdc-acm.c | 23 +++++ drivers/usb/class/cdc-wdm.c | 72 ++++++++++---- drivers/usb/core/urb.c | 89 ++++++++++------- drivers/usb/dwc3/dwc3-of-simple.c | 1 + drivers/usb/gadget/function/f_ncm.c | 2 +- drivers/usb/gadget/function/f_printer.c | 16 +++- drivers/usb/host/ohci-hcd.c | 16 ++-- drivers/vfio/pci/vfio_pci_intrs.c | 4 +- drivers/vfio/vfio_iommu_type1.c | 3 +- drivers/watchdog/sp5100_tco.h | 2 +- drivers/watchdog/watchdog_dev.c | 6 +- fs/dlm/config.c | 3 + fs/ext4/fsmap.c | 3 + fs/f2fs/sysfs.c | 1 + fs/ntfs/inode.c | 6 ++ fs/ramfs/file-nommu.c | 2 +- fs/reiserfs/inode.c | 3 +- fs/reiserfs/super.c | 8 +- fs/udf/inode.c | 25 ++--- fs/udf/super.c | 6 ++ fs/xfs/xfs_rtalloc.c | 11 +++ include/linux/overflow.h | 1 + include/scsi/scsi_common.h | 7 ++ include/trace/events/target.h | 12 +-- include/uapi/linux/perf_event.h | 2 +- kernel/debug/kdb/kdb_io.c | 8 +- kernel/power/hibernate.c | 11 --- kernel/sched/core.c | 2 +- kernel/sched/sched.h | 13 ++- lib/crc32.c | 2 +- net/bluetooth/l2cap_sock.c | 7 +- net/ipv4/ip_gre.c | 15 ++- net/mac80211/cfg.c | 3 +- net/mac80211/sta_info.c | 4 + net/netfilter/ipvs/ip_vs_ctl.c | 7 +- net/netfilter/nf_conntrack_proto_tcp.c | 19 ++-- net/netfilter/nf_dup_netdev.c | 1 + net/netfilter/nft_fwd_netdev.c | 1 + net/sunrpc/auth_gss/svcauth_gss.c | 27 ++++-- net/sunrpc/xprtrdma/svc_rdma_sendto.c | 3 +- samples/mic/mpssd/mpssd.c | 4 +- tools/perf/util/intel-pt.c | 8 +- 122 files changed, 808 insertions(+), 450 deletions(-) -- 2.25.1

1 119

[PATCH 000/145] backport stable-4.19.153
by Yang Yingliang 30 Oct '20

30 Oct '20

Alex Dewar (1): VMCI: check return value of get_user_pages_fast() for errors Arnd Bergmann (1): mtd: lpddr: fix excessive stack usage with clang Artem Savkov (1): pty: do tty_flip_buffer_push without port->lock in pty_write Arvind Sankar (1): x86/fpu: Allow multiple bits in clearcpuid= parameter Bryan O'Donoghue (1): wcn36xx: Fix reported 802.11n rx_highest rate wcn3660/wcn3680 Christophe JAILLET (5): crypto: ixp4xx - Fix the size used in a 'dma_free_coherent()' call ath10k: Fix the size used in a 'dma_free_coherent()' call in an error handling path mwifiex: Do not use GFP_KERNEL in atomic context staging: rtl8192u: Do not use GFP_KERNEL in atomic context scsi: qla4xxx: Fix an error handling path in 'qla4xxx_get_host_stats()' Colin Ian King (3): x86/events/amd/iommu: Fix sizeof mismatch video: fbdev: vga16fb: fix setting of pixclock because a pass-by-value error qtnfmac: fix resource leaks on unsupported iftype error return path Cong Wang (1): tipc: fix the skb_unshare() in tipc_buf_append() Dan Carpenter (8): ALSA: bebob: potential info leak in hwdep_read() cifs: remove bogus debug code ath6kl: prevent potential array overflow in ath6kl_add_new_sta() ath9k: Fix potential out of bounds in ath9k_htc_txcompletion_cb() HID: roccat: add bounds checking in kone_sysfs_write_settings() ath6kl: wmi: prevent a shift wrapping bug in ath6kl_wmi_delete_pstream_cmd() mfd: sm501: Fix leaks in probe() scsi: be2iscsi: Fix a theoretical leak in beiscsi_create_eqs() Darrick J. Wong (2): xfs: limit entries returned when counting fsmap records xfs: fix high key handling in the rt allocator's query_range function David Ahern (1): ipv4: Restore flowi4_oif update before call to xfrm_lookup_route David Wilder (2): ibmveth: Switch order of ibmveth_helper calls. ibmveth: Identify ingress large send packets. Davide Caratti (1): net/sched: act_tunnel_key: fix OOB write in case of IPv6 ERSPAN tunnels Defang Bo (1): nfc: Ensure presence of NFC_ATTR_FIRMWARE_NAME attribute in nfc_genl_fw_download() Dinghao Liu (4): EDAC/i5100: Fix error handling order in i5100_init_one() media: omap3isp: Fix memleak in isp_probe media: mx2_emmaprp: Fix memleak in emmaprp_probe video: fbdev: radeon: Fix memleak in radeonfb_pci_register Dmitry Torokhov (1): HID: hid-input: fix stylus battery reporting Emmanuel Grumbach (1): iwlwifi: mvm: split a print to avoid a WARNING in ROC Eran Ben Elisha (1): net/mlx5: Don't call timecounter cyc2time directly from 1PPS flow Eric Dumazet (2): icmp: randomize the global rate limiter quota: clear padding in v2r1_mem2diskdqb() Finn Thain (3): powerpc/tau: Use appropriate temperature sample interval powerpc/tau: Convert from timer to workqueue powerpc/tau: Remove duplicated set_thresholds() call Greg Kroah-Hartman (1): Linux 4.19.153 Guenter Roeck (1): hwmon: (pmbus/max34440) Fix status register reads for MAX344{51,60,61} Guillaume Tucker (1): ARM: 9007/1: l2c: fix prefetch bits init in L2X0_AUX_CTRL using DT values Hans de Goede (2): pwm: lpss: Fix off by one error in base_unit math in pwm_lpss_prepare() pwm: lpss: Add range limit check for the base_unit register value Heiner Kallweit (2): r8169: fix data corruption issue on RTL8402 r8169: fix operation under forced interrupt threading Herbert Xu (2): crypto: algif_aead - Do not set MAY_BACKLOG on the async path crypto: algif_skcipher - EBUSY on aio should be an error Håkon Bugge (2): IB/mlx4: Fix starvation in paravirt mux/demux IB/mlx4: Adjust delayed work when a dup is observed Jason Gunthorpe (2): RDMA/ucma: Fix locking for ctx->events_reported RDMA/ucma: Add missing locking around rdma_leave_multicast() Jian-Hong Pan (1): ALSA: hda/realtek: Enable audio jacks of ASUS D700SA with ALC887 Johannes Berg (1): nl80211: fix non-split wiphy information John Donnelly (1): scsi: target: tcmu: Fix warning: 'page' may be used uninitialized Jonathan Lemon (1): mlx4: handle non-napi callers to napi_poll Julian Anastasov (1): ipvs: clear skb->tstamp in forwarding path Karsten Graul (1): net/smc: fix valid DMBE buffer sizes Krzysztof Kozlowski (1): EDAC/ti: Fix handling of platform_get_irq() error Laurent Pinchart (2): media: uvcvideo: Set media controller entity functions media: uvcvideo: Silence shift-out-of-bounds warning Libing Zhou (1): x86/nmi: Fix nmi_handle() duration miscalculation Linus Walleij (4): net: dsa: rtl8366: Check validity of passed VLANs net: dsa: rtl8366: Refactor VLAN/PVID init net: dsa: rtl8366: Skip PVID setting if not requested net: dsa: rtl8366rb: Support all 4096 VLANs Lorenzo Colitti (2): usb: gadget: f_ncm: fix ncm_bitrate for SuperSpeed and above. usb: gadget: u_ether: enable qmult on SuperSpeed Plus as well Maciej Żenczykowski (1): net/ipv4: always honour route mtu during forwarding Madhuparna Bhowmik (1): crypto: picoxcell - Fix potential race condition bug Marek Vasut (2): net: fec: Fix phy_device lookup for phy_reset_after_clk_enable() net: fec: Fix PHY init after phy_reset_after_clk_enable() Mark Salter (1): drivers/perf: xgene_pmu: Fix uninitialized resource struct Mark Tomlinson (1): mtd: mtdoops: Don't write panic data twice Michal Kalderon (2): RDMA/qedr: Fix use of uninitialized field RDMA/qedr: Fix inline size returned for iWARP Michał Mirosław (1): regulator: resolve supply after creating regulator Minas Harutyunyan (1): usb: dwc2: Fix INTR OUT transfers in DDMA mode. Nathan Chancellor (1): usb: dwc2: Fix parameter type in function pointer prototype Nathan Lynch (1): powerpc/pseries: explicitly reschedule during drmem_lmb list traversal Neal Cardwell (1): tcp: fix to update snd_wl1 in bulk receiver fast path Necip Fazil Yildiran (2): pinctrl: bcm: fix kconfig dependency warning when !GPIOLIB arc: plat-hsdk: fix kconfig dependency warning when !RESET_CONTROLLER Nicholas Mc Guire (2): powerpc/pseries: Fix missing of_node_put() in rng_init() powerpc/icp-hv: Fix missing of_node_put() in success path Ong Boon Leong (1): net: stmmac: use netif_tx_start|stop_all_queues() function Pablo Neira Ayuso (1): netfilter: nf_log: missing vlan offload tag and proto Pali Rohár (1): cpufreq: armada-37xx: Add missing MODULE_DEVICE_TABLE Qiushi Wu (7): media: rcar-vin: Fix a reference count leak. media: rockchip/rga: Fix a reference count leak. media: platform: fcp: Fix a reference count leak. media: camss: Fix a reference count leak. media: s5p-mfc: Fix a reference count leak media: stm32-dcmi: Fix a reference count leak media: ti-vpe: Fix a missing check and reference count leak Ralph Campbell (1): mm/memcg: fix device private memcg accounting Rohit Maheshwari (1): net/tls: sendfile fails with ktls offload Rohit kumar (2): ASoC: qcom: lpass-platform: fix memory leak ASoC: qcom: lpass-cpu: fix concurrency issue Samuel Holland (1): Bluetooth: hci_uart: Cancel init work before unregistering Sean Christopherson (1): KVM: x86/mmu: Commit zap of remaining invalid pages when recovering lpages Shyam Prasad N (1): cifs: Return the error from crypt_message when enc/dec key not found. Souptick Joarder (2): drivers/virt/fsl_hypervisor: Fix error handling path misc: mic: scif: Fix error handling path Srinivas Kandagatla (3): slimbus: core: check get_addr before removing laddr ida slimbus: core: do not enter to clock pause mode in core slimbus: qcom-ngd-ctrl: disable ngd in qmi server down callback Suravee Suthikulpanit (1): KVM: SVM: Initialize prev_ga_tag before use Suren Baghdasaryan (1): mm, oom_adj: don't loop through tasks in __set_oom_adj when not necessary Sylwester Nawrocki (1): media: Revert "media: exynos4-is: Add missed check for pinctrl_lookup_state()" Takashi Iwai (1): ALSA: seq: oss: Avoid mutex lock for a long-time ioctl Tero Kristo (1): crypto: omap-sham - fix digcnt register handling with export/import Thomas Gleixner (1): net: enic: Cure the enic api locking trainwreck Thomas Preston (2): pinctrl: mcp23s08: Fix mcp23x17_regmap initialiser pinctrl: mcp23s08: Fix mcp23x17 precious range Tianjia Zhang (3): crypto: mediatek - Fix wrong return value in mtk_desc_ring_alloc() scsi: qla2xxx: Fix wrong return value in qla_nvme_register_hba() scsi: csiostor: Fix wrong return value in csio_hw_prep_fw() Todd Kjos (1): binder: fix UAF when releasing todo list Tom Rix (8): media: tuner-simple: fix regression in simple_set_radio_freq media: m5mols: Check function pointer in m5mols_sensor_power media: tc358743: initialize variable media: tc358743: cleanup tc358743_cec_isr brcmfmac: check ndev pointer drm/gma500: fix error check video: fbdev: sis: fix null ptr dereference mwifiex: fix double free Tong Zhang (1): tty: serial: earlycon dependency Tyrel Datwyler (1): tty: hvcs: Don't NULL tty->driver_data until hvcs_cleanup() Vadim Pasternak (1): platform/x86: mlx-platform: Remove PSU EEPROM configuration Valentin Vidic (1): net: korina: fix kfree of rx/tx descriptor array Venkateswara Naralasetty (1): ath10k: provide survey info as accumulated data Vinay Kumar Yadav (3): chelsio/chtls: fix socket lock chelsio/chtls: correct netdevice for vlan interface chelsio/chtls: correct function return and return type Wilken Gottwalt (1): net: usb: qmi_wwan: add Cellient MPL200 card Xiaoliang Pang (1): cypto: mediatek - fix leaks in mtk_desc_ring_alloc Xie He (2): net: hdlc: In hdlc_rcv, check to make sure dev is an HDLC device net: hdlc_raw_eth: Clear the IFF_TX_SKB_SHARING flag after calling ether_setup Yonghong Song (1): net: fix pos incrementment in ipv6_route_seq_next dinghao.liu(a)zju.edu.cn (1): backlight: sky81452-backlight: Fix refcount imbalance on error Łukasz Stelmach (2): spi: spi-s3c64xx: swap s3c64xx_spi_set_cs() and s3c64xx_enable_datapath() spi: spi-s3c64xx: Check return values .../admin-guide/kernel-parameters.txt | 2 +- Documentation/networking/ip-sysctl.txt | 4 +- Makefile | 2 +- arch/arc/plat-hsdk/Kconfig | 1 + arch/arm/mm/cache-l2x0.c | 16 +- arch/powerpc/include/asm/drmem.h | 18 +- arch/powerpc/include/asm/reg.h | 2 +- arch/powerpc/kernel/tau_6xx.c | 55 ++-- arch/powerpc/platforms/pseries/rng.c | 1 + arch/powerpc/sysdev/xics/icp-hv.c | 1 + arch/x86/events/amd/iommu.c | 2 +- arch/x86/kernel/fpu/init.c | 30 +- arch/x86/kernel/nmi.c | 5 +- arch/x86/kvm/mmu.c | 1 + arch/x86/kvm/svm.c | 1 + crypto/algif_aead.c | 7 +- crypto/algif_skcipher.c | 2 +- drivers/android/binder.c | 35 +-- drivers/bluetooth/hci_ldisc.c | 1 + drivers/bluetooth/hci_serdev.c | 2 + drivers/cpufreq/armada-37xx-cpufreq.c | 6 + drivers/crypto/chelsio/chtls/chtls_cm.c | 3 + drivers/crypto/chelsio/chtls/chtls_io.c | 5 +- drivers/crypto/ixp4xx_crypto.c | 2 +- drivers/crypto/mediatek/mtk-platform.c | 8 +- drivers/crypto/omap-sham.c | 3 + drivers/crypto/picoxcell_crypto.c | 9 +- drivers/edac/i5100_edac.c | 11 +- drivers/edac/ti_edac.c | 3 +- drivers/gpu/drm/gma500/cdv_intel_dp.c | 2 +- drivers/hid/hid-input.c | 4 +- drivers/hid/hid-roccat-kone.c | 23 +- drivers/hwmon/pmbus/max34440.c | 3 - drivers/infiniband/core/ucma.c | 6 +- drivers/infiniband/hw/mlx4/cm.c | 3 + drivers/infiniband/hw/mlx4/mad.c | 34 ++- drivers/infiniband/hw/mlx4/mlx4_ib.h | 2 + drivers/infiniband/hw/qedr/main.c | 2 +- drivers/infiniband/hw/qedr/verbs.c | 2 +- drivers/media/i2c/m5mols/m5mols_core.c | 3 +- drivers/media/i2c/tc358743.c | 14 +- drivers/media/platform/exynos4-is/media-dev.c | 4 +- drivers/media/platform/mx2_emmaprp.c | 7 +- drivers/media/platform/omap3isp/isp.c | 6 +- .../media/platform/qcom/camss/camss-csiphy.c | 4 +- drivers/media/platform/rcar-fcp.c | 4 +- drivers/media/platform/rcar-vin/rcar-dma.c | 4 +- drivers/media/platform/rockchip/rga/rga-buf.c | 1 + drivers/media/platform/s5p-mfc/s5p_mfc_pm.c | 4 +- drivers/media/platform/stm32/stm32-dcmi.c | 4 +- drivers/media/platform/ti-vpe/vpe.c | 2 + drivers/media/tuners/tuner-simple.c | 5 +- drivers/media/usb/uvc/uvc_ctrl.c | 6 +- drivers/media/usb/uvc/uvc_entity.c | 35 +++ drivers/mfd/sm501.c | 8 +- drivers/misc/mic/scif/scif_rma.c | 4 +- drivers/misc/vmw_vmci/vmci_queue_pair.c | 10 +- drivers/mtd/lpddr/lpddr2_nvm.c | 35 ++- drivers/mtd/mtdoops.c | 11 +- drivers/net/dsa/realtek-smi.h | 4 +- drivers/net/dsa/rtl8366.c | 280 ++++++++++-------- drivers/net/dsa/rtl8366rb.c | 2 +- drivers/net/ethernet/cisco/enic/enic.h | 1 + drivers/net/ethernet/cisco/enic/enic_api.c | 6 + drivers/net/ethernet/cisco/enic/enic_main.c | 27 +- drivers/net/ethernet/freescale/fec_main.c | 35 ++- drivers/net/ethernet/ibm/ibmveth.c | 19 +- drivers/net/ethernet/korina.c | 3 +- drivers/net/ethernet/mellanox/mlx4/en_rx.c | 3 + drivers/net/ethernet/mellanox/mlx4/en_tx.c | 2 +- .../ethernet/mellanox/mlx5/core/lib/clock.c | 5 +- drivers/net/ethernet/realtek/r8169.c | 54 ++-- .../net/ethernet/stmicro/stmmac/stmmac_main.c | 33 +-- drivers/net/usb/qmi_wwan.c | 1 + drivers/net/wan/hdlc.c | 10 +- drivers/net/wan/hdlc_raw_eth.c | 1 + drivers/net/wireless/ath/ath10k/ce.c | 2 +- drivers/net/wireless/ath/ath10k/mac.c | 2 +- drivers/net/wireless/ath/ath6kl/main.c | 3 + drivers/net/wireless/ath/ath6kl/wmi.c | 5 + drivers/net/wireless/ath/ath9k/htc_hst.c | 2 + drivers/net/wireless/ath/wcn36xx/main.c | 2 +- .../broadcom/brcm80211/brcmfmac/core.c | 2 +- .../net/wireless/intel/iwlwifi/mvm/mac80211.c | 9 +- drivers/net/wireless/marvell/mwifiex/scan.c | 2 +- drivers/net/wireless/marvell/mwifiex/sdio.c | 2 + .../net/wireless/quantenna/qtnfmac/commands.c | 2 + drivers/perf/xgene_pmu.c | 32 +- drivers/pinctrl/bcm/Kconfig | 1 + drivers/pinctrl/pinctrl-mcp23s08.c | 24 +- drivers/platform/x86/mlx-platform.c | 15 +- drivers/pwm/pwm-lpss.c | 7 +- drivers/regulator/core.c | 21 +- drivers/scsi/be2iscsi/be_main.c | 4 +- drivers/scsi/csiostor/csio_hw.c | 2 +- drivers/scsi/qla2xxx/qla_nvme.c | 2 +- drivers/scsi/qla4xxx/ql4_os.c | 2 +- drivers/slimbus/core.c | 6 +- drivers/slimbus/qcom-ngd-ctrl.c | 4 + drivers/spi/spi-s3c64xx.c | 52 +++- .../staging/rtl8192u/ieee80211/ieee80211_rx.c | 2 +- drivers/target/target_core_user.c | 2 +- drivers/tty/hvc/hvcs.c | 14 +- drivers/tty/pty.c | 2 +- drivers/tty/serial/Kconfig | 1 + drivers/usb/dwc2/gadget.c | 40 ++- drivers/usb/dwc2/params.c | 2 +- drivers/usb/gadget/function/f_ncm.c | 6 +- drivers/usb/gadget/function/u_ether.c | 2 +- drivers/video/backlight/sky81452-backlight.c | 1 + drivers/video/fbdev/aty/radeon_base.c | 2 +- drivers/video/fbdev/sis/init.c | 11 +- drivers/video/fbdev/vga16fb.c | 14 +- drivers/virt/fsl_hypervisor.c | 17 +- fs/cifs/asn1.c | 16 +- fs/cifs/smb2ops.c | 2 +- fs/proc/base.c | 3 +- fs/quota/quota_v2.c | 1 + fs/xfs/libxfs/xfs_rtbitmap.c | 11 +- fs/xfs/xfs_fsmap.c | 3 + include/linux/oom.h | 1 + include/linux/sched/coredump.h | 1 + include/net/ip.h | 6 + include/net/netfilter/nf_log.h | 1 + kernel/fork.c | 21 ++ mm/memcontrol.c | 5 +- mm/oom_kill.c | 2 + net/ipv4/icmp.c | 7 +- net/ipv4/netfilter/nf_log_arp.c | 19 +- net/ipv4/netfilter/nf_log_ipv4.c | 6 +- net/ipv4/route.c | 4 +- net/ipv4/tcp_input.c | 2 + net/ipv6/ip6_fib.c | 4 +- net/ipv6/netfilter/nf_log_ipv6.c | 8 +- net/netfilter/ipvs/ip_vs_xmit.c | 6 + net/netfilter/nf_log_common.c | 12 + net/nfc/netlink.c | 2 +- net/sched/act_tunnel_key.c | 2 +- net/smc/smc_core.c | 2 +- net/tipc/msg.c | 3 +- net/tls/tls_device.c | 11 +- net/wireless/nl80211.c | 5 +- sound/core/seq/oss/seq_oss.c | 7 +- sound/firewire/bebob/bebob_hwdep.c | 3 +- sound/pci/hda/patch_realtek.c | 42 +++ sound/soc/qcom/lpass-cpu.c | 16 - sound/soc/qcom/lpass-platform.c | 3 +- 147 files changed, 968 insertions(+), 566 deletions(-) -- 2.25.1

1 145

openEuler 21.03 5.10 kernel defconfig review
by Xie XiuQi 30 Oct '20

30 Oct '20

openEuler 21.03 5.10 kernel defconfig 基于 openEuler 20.09 适配而来，请大家评审。有问题可以在下面 issue 中讨论： https://gitee.com/openeuler/kernel/issues/I22VGE

1 1

本次 kernel sig 例会无议题，本次会议取消，欢迎积极申报议题
by Xie XiuQi 29 Oct '20

29 Oct '20

本次 kernel sig 例会无议题，本次会议取消，欢迎积极申报议题

1 0

[PATCH openEuler-1.0-LTS 0/5] kvm: arm64: emulate ID registers
by Yang Yingliang 29 Oct '20

29 Oct '20

In AArch64, guest will read the same values of the ID regsiters with host. Both of them read the values from arm64_ftr_regs. This patch series add support to emulate and configure ID registers so that we can control the value of ID registers that guest read. Peng Liang (5): arm64: add a helper function to traverse arm64_ftr_regs kvm: arm64: emulate the ID registers kvm: arm64: make ID registers configurable kvm: arm64: add KVM_CAP_ARM_CPU_FEATURE extension kvm: fix compile error when including linux/kvm.h arch/arm64/include/asm/cpufeature.h | 2 + arch/arm64/include/asm/kvm_host.h | 2 + arch/arm64/kernel/cpufeature.c | 13 ++++++ arch/arm64/kvm/sys_regs.c | 71 ++++++++++++++++++++++------- include/uapi/linux/kvm.h | 13 ++++++ virt/kvm/arm/arm.c | 23 ++++++++++ 6 files changed, 107 insertions(+), 17 deletions(-) -- 2.25.1

1 5

[PATCH 1/5] khugepaged: do not stop collapse if less than half PTEs are referenced
by Yang Yingliang 29 Oct '20

29 Oct '20

From: "Kirill A. Shutemov" <kirill.shutemov(a)linux.intel.com> mainline inclusion from mainline-v5.8-rc1 commit ffe945e633b527d5a4577b42cbadec3c7cbcf096 category: bugfix bugzilla: 36230 CVE: NA ------------------------------------------------- __collapse_huge_page_swapin() checks the number of referenced PTE to decide if the memory range is hot enough to justify swapin. We have few problems with the approach: - It is way too late: we can do the check much earlier and safe time. khugepaged_scan_pmd() already knows if we have any pages to swap in and number of referenced page. - It stops collapse altogether if there's not enough referenced pages, not only swappingin. Fix it by making the right check early. We also can avoid additional page table scanning if khugepaged_scan_pmd() haven't found any swap entries. Fixes: 0db501f7a34c ("mm, thp: convert from optimistic swapin collapsing to conservative") Signed-off-by: Kirill A. Shutemov <kirill.shutemov(a)linux.intel.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Tested-by: Zi Yan <ziy(a)nvidia.com> Reviewed-by: William Kucharski <william.kucharski(a)oracle.com> Reviewed-by: Zi Yan <ziy(a)nvidia.com> Acked-by: Yang Shi <yang.shi(a)linux.alibaba.com> Cc: Andrea Arcangeli <aarcange(a)redhat.com> Cc: John Hubbard <jhubbard(a)nvidia.com> Cc: Mike Kravetz <mike.kravetz(a)oracle.com> Cc: Ralph Campbell <rcampbell(a)nvidia.com> Link: http://lkml.kernel.org/r/20200416160026.16538-3-kirill.shutemov@linux.intel… Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Liu Shixin <liushixin2(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- mm/khugepaged.c | 27 +++++++++++---------------- 1 file changed, 11 insertions(+), 16 deletions(-) diff --git a/mm/khugepaged.c b/mm/khugepaged.c index 5883fd75d6fc..0ad9f2b2b33e 100644 --- a/mm/khugepaged.c +++ b/mm/khugepaged.c @@ -905,11 +905,6 @@ static bool __collapse_huge_page_swapin(struct mm_struct *mm, .pgoff = linear_page_index(vma, address), }; - /* we only decide to swapin, if there is enough young ptes */ - if (referenced < HPAGE_PMD_NR/2) { - trace_mm_collapse_huge_page_swapin(mm, swapped_in, referenced, 0); - return false; - } vmf.pte = pte_offset_map(pmd, address); for (; vmf.address < address + HPAGE_PMD_NR*PAGE_SIZE; vmf.pte++, vmf.address += PAGE_SIZE) { @@ -949,7 +944,7 @@ static bool __collapse_huge_page_swapin(struct mm_struct *mm, static void collapse_huge_page(struct mm_struct *mm, unsigned long address, struct page **hpage, - int node, int referenced) + int node, int referenced, int unmapped) { pmd_t *pmd, _pmd; pte_t *pte; @@ -1007,7 +1002,8 @@ static void collapse_huge_page(struct mm_struct *mm, * If it fails, we release mmap_sem and jump out_nolock. * Continuing to collapse causes inconsistency. */ - if (!__collapse_huge_page_swapin(mm, vma, address, pmd, referenced)) { + if (unmapped && !__collapse_huge_page_swapin(mm, vma, address, + pmd, referenced)) { mem_cgroup_cancel_charge(new_page, memcg, true); up_read(&mm->mmap_sem); goto out_nolock; @@ -1214,22 +1210,21 @@ static int khugepaged_scan_pmd(struct mm_struct *mm, mmu_notifier_test_young(vma->vm_mm, address)) referenced++; } - if (writable) { - if (referenced) { - result = SCAN_SUCCEED; - ret = 1; - } else { - result = SCAN_LACK_REFERENCED_PAGE; - } - } else { + if (!writable) { result = SCAN_PAGE_RO; + } else if (!referenced || (unmapped && referenced < HPAGE_PMD_NR/2)) { + result = SCAN_LACK_REFERENCED_PAGE; + } else { + result = SCAN_SUCCEED; + ret = 1; } out_unmap: pte_unmap_unlock(pte, ptl); if (ret) { node = khugepaged_find_target_node(); /* collapse_huge_page will return with the mmap_sem released */ - collapse_huge_page(mm, address, hpage, node, referenced); + collapse_huge_page(mm, address, hpage, node, + referenced, unmapped); } out: trace_mm_khugepaged_scan_pmd(mm, page, writable, referenced, -- 2.25.1

1 4

[PATCH 0/4] fixes CVE-2020-27673
by Yang Yingliang 27 Oct '20

27 Oct '20

Juergen Gross (4): xen/events: add a new "late EOI" evtchn framework xen/events: switch user event channels to lateeoi model xen/events: use a common cpu hotplug hook for event channels xen/events: defer eoi in case of excessive number of events .../admin-guide/kernel-parameters.txt | 8 + drivers/xen/events/events_2l.c | 7 +- drivers/xen/events/events_base.c | 363 +++++++++++++++++- drivers/xen/events/events_fifo.c | 70 ++-- drivers/xen/events/events_internal.h | 17 +- drivers/xen/evtchn.c | 7 +- include/xen/events.h | 21 + 7 files changed, 420 insertions(+), 73 deletions(-) -- 2.25.1

1 4

openEuler 选择 5.10 作为下一个长期维护的内核版本 (公示)
by Xie XiuQi 27 Oct '20

27 Oct '20

openEuler 选择 5.10 作为下一个长期维护的内核版本 (公示) - /本文//链接 <https://gitee.com/openeuler/community/blob/master/sig/Kernel/kernel-upgrade…>/ Linux 5.10 预计是 Linux 社区今年的 LTS (long-term support) 版本。openEuler 社区经过多轮沟通，选择 5.10 作为 openEuler 内核的下一个长期维护版本（openEuler Long-term support Kernel, OLK）。openEuler 社区提供不少于4年的维护时间和不少于2年的扩展维护时间。按照规划 openEuler 21.03、21.09 以及 openEuler 22.03 LTS 都将选择该内核版本。（具体时间以openEuler社区官网公布为准）升级安排： * v5.10-rc1 版本发布后，openEuler master 分支切换成 5.10。 * 由于主线 rc 版本尚在更新，openeuler_defconfig 暂时在 src-openeuler/kernel 仓库中提供，待 5.10 正式版本发布之后，再提交到源码仓库。 * openeuler_defconfig 基于 openEuler 20.09 的 config 修改适配，后续根据需求进行调整。 * 版本号格式，切换成 5.10.0-<devel_release>.<maintainence_release>。 * 首先支持 arm64 和 x86_64 架构，risc-v 的支持需要 risc-v sig 做适配。 * 上游社区 5.10 正式发布后（预计12月下旬）， openeuler/kernel 正式建立 OLK-5.10 分支（OLK: openEuler Long-term support Kernel），作为 5.10 的长期维护分支，接受补丁。 * 上游 5.10 正式发布前，如果有补丁需要发送 5.10，可以在 kernel(a)openeuler.org <mailto:kernel@openeuler.org> 中发 RFC 补丁，提前 Review 和讨论。 * 我们也可能提前建立 OLK-5.10 分支，以提前合入部分补丁做验证，但是在上游社区 5.10 正式发布之前，该分支可能会经常做 rebase 和 force push。对 OLK-5.10 或 openEuler 21.03 kernel 的需求 * 需求可以在 https://gitee.com/openeuler/kernel/issues 中提出。 * 如对 defconfig 有诉求也请在上述链接中提 issue。 * openEuler 21.03 计划使用该内核版本，预计2021年1月份之后，将从 OLK-5.10 拉出 openEuler-21.03 维护分支，同时补丁将受限合入 openEuler 21.03。注意事项和说明 * OLK 是 openEuler Long-term support Kernel 的缩写。openEuler LTS 版本和部分创新版本的内核基于 OLK 拉出分支进行维护。 * openEuler 5.10 内核不是基于 openEuler 4.19 内核的演进，而是基于上游社区内核的重新选型，因此如果您之前有合入 openEuler 4.19 的补丁，且这些补丁没有进入上游 5.10 内核，则需要您重新适配后推送到 openEuler 5.10 内核。 * openEuler 5.10 和 openEuler 4.19 两个版本 kabi 不兼容，您在 4.19 编译的 ko 不能直接在 5.10 上安装使用，需要重新适配和编译。 * openEuler 4.19 内核仍然处于维护周期内，如果您正在使用 4.19 内核，也不必紧张，您仍然能收到 4.19 的更新和增强。问题反馈 * kernel 升级期间，如果遇到兼容性、功能等问题，可以在 https://gitee.com/openeuler/kernel/issues 提交 issue。有疑问也可以在 PR 评论中，或者提交 issue 讨论。 * 你也可以发邮件到 kernel(a)openeuler.org <mailto:kernel@openeuler.org> 报议题在 kernel sig 例会中讨论。 * 如果 kernel sig 范围内协调或解决不了的问题，你也可以在 tc 报议题讨论。相关讨论纪要 kernel-sig 切换 5.10 的会议纪要： https://mailweb.openeuler.org/hyperkitty/list/kernel@openeuler.org/thread/6… openEuler kernel 版本号在 TC 的议题及会议纪要： * 议题及讨论记录： https://mailweb.openeuler.org/hyperkitty/list/tc@openeuler.org/thread/KOHJN… https://mailweb.openeuler.org/hyperkitty/list/tc@openeuler.org/thread/VNJQ6… * 纪要： https://mailweb.openeuler.org/hyperkitty/list/tc@openeuler.org/thread/ZLAWQ…

1 0

[PATCH] xen/events: avoid removing an event channel while handling it
by Yang Yingliang 27 Oct '20

27 Oct '20

From: Juergen Gross <jgross(a)suse.com> mainline inclusion from mainline-v5.10 commit 073d0552ead5bfc7a3a9c01de590e924f11b5dd2 category: bugfix bugzilla: NA CVE: CVE-2020-27675 -------------------------------- Today it can happen that an event channel is being removed from the system while the event handling loop is active. This can lead to a race resulting in crashes or WARN() splats when trying to access the irq_info structure related to the event channel. Fix this problem by using a rwlock taken as reader in the event handling loop and as writer when deallocating the irq_info structure. As the observed problem was a NULL dereference in evtchn_from_irq() make this function more robust against races by testing the irq_info pointer to be not NULL before dereferencing it. And finally make all accesses to evtchn_to_irq[row][col] atomic ones in order to avoid seeing partial updates of an array element in irq handling. Note that irq handling can be entered only for event channels which have been valid before, so any not populated row isn't a problem in this regard, as rows are only ever added and never removed. This is XSA-331. Cc: stable(a)vger.kernel.org Reported-by: Marek Marczykowski-Górecki <marmarek(a)invisiblethingslab.com> Reported-by: Jinoh Kang <luke1337(a)theori.io> Signed-off-by: Juergen Gross <jgross(a)suse.com> Reviewed-by: Stefano Stabellini <sstabellini(a)kernel.org> Reviewed-by: Wei Liu <wl(a)xen.org> Conflicts: drivers/xen/events/events_base.c [yyl: adjust context] Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/xen/events/events_base.c | 40 ++++++++++++++++++++++++++++---- 1 file changed, 35 insertions(+), 5 deletions(-) diff --git a/drivers/xen/events/events_base.c b/drivers/xen/events/events_base.c index 95e5a9300ff0..75eade7d2017 100644 --- a/drivers/xen/events/events_base.c +++ b/drivers/xen/events/events_base.c @@ -32,6 +32,7 @@ #include <linux/slab.h> #include <linux/irqnr.h> #include <linux/pci.h> +#include <linux/spinlock.h> #ifdef CONFIG_X86 #include <asm/desc.h> @@ -69,6 +70,23 @@ const struct evtchn_ops *evtchn_ops; */ static DEFINE_MUTEX(irq_mapping_update_lock); +/* + * Lock protecting event handling loop against removing event channels. + * Adding of event channels is no issue as the associated IRQ becomes active + * only after everything is setup (before request_[threaded_]irq() the handler + * can't be entered for an event, as the event channel will be unmasked only + * then). + */ +static DEFINE_RWLOCK(evtchn_rwlock); + +/* + * Lock hierarchy: + * + * irq_mapping_update_lock + * evtchn_rwlock + * IRQ-desc lock + */ + static LIST_HEAD(xen_irq_list_head); /* IRQ <-> VIRQ mapping. */ @@ -101,7 +119,7 @@ static void clear_evtchn_to_irq_row(unsigned row) unsigned col; for (col = 0; col < EVTCHN_PER_ROW; col++) - evtchn_to_irq[row][col] = -1; + WRITE_ONCE(evtchn_to_irq[row][col], -1); } static void clear_evtchn_to_irq_all(void) @@ -138,7 +156,7 @@ static int set_evtchn_to_irq(unsigned evtchn, unsigned irq) clear_evtchn_to_irq_row(row); } - evtchn_to_irq[row][col] = irq; + WRITE_ONCE(evtchn_to_irq[row][col], irq); return 0; } @@ -148,7 +166,7 @@ int get_evtchn_to_irq(unsigned evtchn) return -1; if (evtchn_to_irq[EVTCHN_ROW(evtchn)] == NULL) return -1; - return evtchn_to_irq[EVTCHN_ROW(evtchn)][EVTCHN_COL(evtchn)]; + return READ_ONCE(evtchn_to_irq[EVTCHN_ROW(evtchn)][EVTCHN_COL(evtchn)]); } /* Get info for IRQ */ @@ -246,10 +264,14 @@ static void xen_irq_info_cleanup(struct irq_info *info) */ unsigned int evtchn_from_irq(unsigned irq) { - if (unlikely(WARN(irq >= nr_irqs, "Invalid irq %d!\n", irq))) + const struct irq_info *info = NULL; + + if (likely(irq < nr_irqs)) + info = info_for_irq(irq); + if (!info) return 0; - return info_for_irq(irq)->evtchn; + return info->evtchn; } unsigned irq_from_evtchn(unsigned int evtchn) @@ -425,16 +447,21 @@ static int __must_check xen_allocate_irq_gsi(unsigned gsi) static void xen_free_irq(unsigned irq) { struct irq_info *info = irq_get_chip_data(irq); + unsigned long flags; if (WARN_ON(!info)) return; + write_lock_irqsave(&evtchn_rwlock, flags); + list_del(&info->list); irq_set_chip_data(irq, NULL); WARN_ON(info->refcnt > 0); + write_unlock_irqrestore(&evtchn_rwlock, flags); + kfree(info); /* Legacy IRQ descriptors are managed by the arch. */ @@ -1220,6 +1247,8 @@ static void __xen_evtchn_do_upcall(void) int cpu = get_cpu(); unsigned count; + read_lock(&evtchn_rwlock); + do { vcpu_info->evtchn_upcall_pending = 0; @@ -1236,6 +1265,7 @@ static void __xen_evtchn_do_upcall(void) out: + read_unlock(&evtchn_rwlock); put_cpu(); } -- 2.25.1

1 0

[PATCH 1/7] iommu/sysfs: Rename iommu_release_device()
by Yang Yingliang 27 Oct '20

27 Oct '20

From: Joerg Roedel <jroedel(a)suse.de> mainline inclusion from mainline-v4.20-rc1 commit 6954cf9bfda153f9544c63761aabf0199710aec3 category:bugfix bugzilla:NA CVE:NA ------------------- Remove the iommu_ prefix from the function and a few other static data structures so that the iommu_release_device name can be re-used in iommu core code. Signed-off-by: Joerg Roedel <jroedel(a)suse.de> Signed-off-by: Chen Jun <chenjun102(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/iommu/iommu-sysfs.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/iommu/iommu-sysfs.c b/drivers/iommu/iommu-sysfs.c index 36d1a7ce7fc4..71c2249d3260 100644 --- a/drivers/iommu/iommu-sysfs.c +++ b/drivers/iommu/iommu-sysfs.c @@ -22,25 +22,25 @@ static struct attribute *devices_attr[] = { NULL, }; -static const struct attribute_group iommu_devices_attr_group = { +static const struct attribute_group devices_attr_group = { .name = "devices", .attrs = devices_attr, }; -static const struct attribute_group *iommu_dev_groups[] = { - &iommu_devices_attr_group, +static const struct attribute_group *dev_groups[] = { + &devices_attr_group, NULL, }; -static void iommu_release_device(struct device *dev) +static void release_device(struct device *dev) { kfree(dev); } static struct class iommu_class = { .name = "iommu", - .dev_release = iommu_release_device, - .dev_groups = iommu_dev_groups, + .dev_release = release_device, + .dev_groups = dev_groups, }; static int __init iommu_dev_init(void) -- 2.25.1

1 6

[PATCH 1/8] bcache: use a separate data structure for the on-disk super block
by Yang Yingliang 26 Oct '20

26 Oct '20

From: Christoph Hellwig <hch(a)lst.de> mainline inclusion from mainline-5.6-rc1 commit a702a692cd7559053ea573f4e2c84828f0e62824 category: feature bugzilla: 43003 CVE: NA --------------------------- Split out an on-disk version struct cache_sb with the proper endianness annotations. This fixes a fair chunk of sparse warnings, but there are some left due to the way the checksum is defined. Signed-off-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Coly Li <colyli(a)suse.de> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Acked-by: Hanjun Guo <guohanjun(a)huawei.com> Reviewed-by: Yufen Yu <yuyufen(a)huawei.com> Signed-off-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/md/bcache/super.c | 6 ++--- include/uapi/linux/bcache.h | 51 +++++++++++++++++++++++++++++++++++++ 2 files changed, 54 insertions(+), 3 deletions(-) diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c index 78854837f398..8f8a0a69cc0e 100644 --- a/drivers/md/bcache/super.c +++ b/drivers/md/bcache/super.c @@ -63,14 +63,14 @@ static const char *read_super(struct cache_sb *sb, struct block_device *bdev, struct page **res) { const char *err; - struct cache_sb *s; + struct cache_sb_disk *s; struct buffer_head *bh = __bread(bdev, 1, SB_SIZE); unsigned int i; if (!bh) return "IO error"; - s = (struct cache_sb *) bh->b_data; + s = (struct cache_sb_disk *)bh->b_data; sb->offset = le64_to_cpu(s->offset); sb->version = le64_to_cpu(s->version); @@ -209,7 +209,7 @@ static void write_bdev_super_endio(struct bio *bio) static void __write_super(struct cache_sb *sb, struct bio *bio) { - struct cache_sb *out = page_address(bio_first_page_all(bio)); + struct cache_sb_disk *out = page_address(bio_first_page_all(bio)); unsigned int i; bio->bi_iter.bi_sector = SB_SECTOR; diff --git a/include/uapi/linux/bcache.h b/include/uapi/linux/bcache.h index 5d4f58e059fd..1d8b3a9fc080 100644 --- a/include/uapi/linux/bcache.h +++ b/include/uapi/linux/bcache.h @@ -156,6 +156,57 @@ static inline struct bkey *bkey_idx(const struct bkey *k, unsigned int nr_keys) #define BDEV_DATA_START_DEFAULT 16 /* sectors */ +struct cache_sb_disk { + __le64 csum; + __le64 offset; /* sector where this sb was written */ + __le64 version; + + __u8 magic[16]; + + __u8 uuid[16]; + union { + __u8 set_uuid[16]; + __le64 set_magic; + }; + __u8 label[SB_LABEL_SIZE]; + + __le64 flags; + __le64 seq; + __le64 pad[8]; + + union { + struct { + /* Cache devices */ + __le64 nbuckets; /* device size */ + + __le16 block_size; /* sectors */ + __le16 bucket_size; /* sectors */ + + __le16 nr_in_set; + __le16 nr_this_dev; + }; + struct { + /* Backing devices */ + __le64 data_offset; + + /* + * block_size from the cache device section is still used by + * backing devices, so don't add anything here until we fix + * things to not need it for backing devices anymore + */ + }; + }; + + __le32 last_mount; /* time overflow in y2106 */ + + __le16 first_bucket; + union { + __le16 njournal_buckets; + __le16 keys; + }; + __le64 d[SB_JOURNAL_BUCKETS]; /* journal buckets */ +}; + struct cache_sb { __u64 csum; __u64 offset; /* sector where this sb was written */ -- 2.25.1

1 7

[PATCH 1/2] list: introduce list_for_each_continue()
by Yang Yingliang 26 Oct '20

26 Oct '20

From: Pavel Begunkov <asml.silence(a)gmail.com> mainline inclusion from mainline-v5.6-rc1 commit 28ca0d6d39ab1d01c86762c82a585b7cedd2920c category: bugfix bugzilla: 35619 CVE: NA -------------------------------- As other *continue() helpers, this continues iteration from a given position. Signed-off-by: Pavel Begunkov <asml.silence(a)gmail.com> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Signed-off-by: Zhang Xiaoxu <zhangxiaoxu5(a)huawei.com> Reviewed-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/linux/list.h | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/include/linux/list.h b/include/linux/list.h index de04cc5ed536..0e540581d52c 100644 --- a/include/linux/list.h +++ b/include/linux/list.h @@ -455,6 +455,16 @@ static inline void list_splice_tail_init(struct list_head *list, #define list_for_each(pos, head) \ for (pos = (head)->next; pos != (head); pos = pos->next) +/** + * list_for_each_continue - continue iteration over a list + * @pos: the &struct list_head to use as a loop cursor. + * @head: the head for your list. + * + * Continue to iterate over a list, continuing after the current position. + */ +#define list_for_each_continue(pos, head) \ + for (pos = pos->next; pos != (head); pos = pos->next) + /** * list_for_each_prev - iterate over a list backwards * @pos: the &struct list_head to use as a loop cursor. -- 2.25.1

1 1

[PATCH 1/6] driver core: Introduce device_iommu_mapped() function
by Yang Yingliang 22 Oct '20

22 Oct '20

From: Joerg Roedel <jroedel(a)suse.de> mainline inclusion from mainline-v4.20-rc1 commit dbba197edf32209d110727a02d3a91de4c88520f category:feature bugzilla:NA CVE:NA ------------------- Some places in the kernel check the iommu_group pointer in 'struct device' in order to find out whether a device is mapped by an IOMMU. This is not good way to make this check, as the pointer will be moved to 'struct dev_iommu_data'. This way to make the check is also not very readable. Introduce an explicit function to perform this check. Acked-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Acked-by: Robin Murphy <robin.murphy(a)arm.com> Signed-off-by: Joerg Roedel <jroedel(a)suse.de> Signed-off-by: Chen Jun <chenjun102(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/linux/device.h | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/include/linux/device.h b/include/linux/device.h index ee4ed3af30d0..cb9df20a9c97 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -1090,6 +1090,16 @@ static inline struct device *kobj_to_dev(struct kobject *kobj) return container_of(kobj, struct device, kobj); } +/** + * device_iommu_mapped - Returns true when the device DMA is translated + * by an IOMMU + * @dev: Device to perform the check on + */ +static inline bool device_iommu_mapped(struct device *dev) +{ + return (dev->iommu_group != NULL); +} + /* Get the wakeup routines, which depend on struct device */ #include <linux/pm_wakeup.h> -- 2.25.1

1 5

[PATCH 1/6] net/hinic: Fix the driver does not report an error when setting MAC fails
by Yang Yingliang 22 Oct '20

22 Oct '20

From: Chiqijun <chiqijun(a)huawei.com> driver inclusion category: bugfix bugzilla: 4472 ----------------------------------------------------------------------- When using the ip link command to configure the MAC for the VF in the PF, the status 4 will be returned when the MAC is set on the VF; when the PF driver receives the status 4 returned by the firmwre, the MAC setting failed and an error should be reported. Signed-off-by: Chiqijun <chiqijun(a)huawei.com> Reviewed-by: Wangxiaoyun <cloud.wangxiaoyun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- .../net/ethernet/huawei/hinic/hinic_main.c | 7 +++- .../net/ethernet/huawei/hinic/hinic_nic_cfg.c | 41 ++++++++++++------- 2 files changed, 31 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/huawei/hinic/hinic_main.c b/drivers/net/ethernet/huawei/hinic/hinic_main.c index b5d0bdadf509..cf7b6ceef060 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_main.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c @@ -1305,8 +1305,11 @@ static int hinic_set_mac_addr(struct net_device *netdev, void *addr) err = hinic_update_mac(nic_dev->hwdev, netdev->dev_addr, saddr->sa_data, 0, func_id); - if (err) - return err; + if (err) { + nicif_err(nic_dev, drv, netdev, "Failed to update mac, err: %d\n", + err); + return err == HINIC_PF_SET_VF_ALREADY ? -EPERM : err; + } memcpy(netdev->dev_addr, saddr->sa_data, ETH_ALEN); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c b/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c index a42dea0ad707..2b349e17260b 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_nic_cfg.c @@ -231,6 +231,23 @@ int hinic_get_fw_support_func(void *hwdev) #define HINIC_ADD_VLAN_IN_MAC 0x8000 #define HINIC_VLAN_ID_MASK 0x7FFF +#define PF_SET_VF_MAC(hwdev, status) \ + (HINIC_FUNC_TYPE(hwdev) == TYPE_VF && \ + (status) == HINIC_PF_SET_VF_ALREADY) + +static int hinic_check_mac_status(struct hinic_hwdev *hwdev, u8 status, + u16 vlan_id) +{ + if ((status && status != HINIC_MGMT_STATUS_EXIST) || + (vlan_id & CHECK_IPSU_15BIT && status == HINIC_MGMT_STATUS_EXIST)) { + if (PF_SET_VF_MAC(hwdev, status)) + return 0; + + return -EINVAL; + } + + return 0; +} int hinic_set_mac(void *hwdev, const u8 *mac_addr, u16 vlan_id, u16 func_id) { @@ -255,17 +272,14 @@ int hinic_set_mac(void *hwdev, const u8 *mac_addr, u16 vlan_id, u16 func_id) err = l2nic_msg_to_mgmt_sync(hwdev, HINIC_PORT_CMD_SET_MAC, &mac_info, sizeof(mac_info), &mac_info, &out_size); if (err || !out_size || - (mac_info.status && mac_info.status != HINIC_MGMT_STATUS_EXIST && - mac_info.status != HINIC_PF_SET_VF_ALREADY) || - (mac_info.vlan_id & CHECK_IPSU_15BIT && - mac_info.status == HINIC_MGMT_STATUS_EXIST)) { + hinic_check_mac_status(hwdev, mac_info.status, mac_info.vlan_id)) { nic_err(nic_hwdev->dev_hdl, "Failed to update MAC, err: %d, status: 0x%x, out size: 0x%x\n", err, mac_info.status, out_size); - return -EINVAL; + return -EIO; } - if (mac_info.status == HINIC_PF_SET_VF_ALREADY) { + if (PF_SET_VF_MAC(nic_hwdev, mac_info.status)) { nic_warn(nic_hwdev->dev_hdl, "PF has already set VF mac, Ignore set operation\n"); return HINIC_PF_SET_VF_ALREADY; } @@ -302,13 +316,13 @@ int hinic_del_mac(void *hwdev, const u8 *mac_addr, u16 vlan_id, u16 func_id) err = l2nic_msg_to_mgmt_sync(hwdev, HINIC_PORT_CMD_DEL_MAC, &mac_info, sizeof(mac_info), &mac_info, &out_size); if (err || !out_size || - (mac_info.status && mac_info.status != HINIC_PF_SET_VF_ALREADY)) { + (mac_info.status && !PF_SET_VF_MAC(nic_hwdev, mac_info.status))) { nic_err(nic_hwdev->dev_hdl, "Failed to delete MAC, err: %d, status: 0x%x, out size: 0x%x\n", err, mac_info.status, out_size); - return -EINVAL; + return -EIO; } - if (mac_info.status == HINIC_PF_SET_VF_ALREADY) { + if (PF_SET_VF_MAC(nic_hwdev, mac_info.status)) { nic_warn(nic_hwdev->dev_hdl, "PF has already set VF mac, Ignore delete operation\n"); return HINIC_PF_SET_VF_ALREADY; } @@ -343,17 +357,14 @@ int hinic_update_mac(void *hwdev, u8 *old_mac, u8 *new_mac, u16 vlan_id, &mac_info, sizeof(mac_info), &mac_info, &out_size); if (err || !out_size || - (mac_info.status && mac_info.status != HINIC_MGMT_STATUS_EXIST && - mac_info.status != HINIC_PF_SET_VF_ALREADY) || - (mac_info.vlan_id & CHECK_IPSU_15BIT && - mac_info.status == HINIC_MGMT_STATUS_EXIST)) { + hinic_check_mac_status(hwdev, mac_info.status, mac_info.vlan_id)) { nic_err(nic_hwdev->dev_hdl, "Failed to update MAC, err: %d, status: 0x%x, out size: 0x%x\n", err, mac_info.status, out_size); - return -EINVAL; + return -EIO; } - if (mac_info.status == HINIC_PF_SET_VF_ALREADY) { + if (PF_SET_VF_MAC(nic_hwdev, mac_info.status)) { nic_warn(nic_hwdev->dev_hdl, "PF has already set VF MAC. Ignore update operation\n"); return HINIC_PF_SET_VF_ALREADY; } -- 2.25.1

1 5

[PATCH 01/20] nvme-rdma: remove redundant reference between ib_device and tagset
by Yang Yingliang 22 Oct '20

22 Oct '20

From: Max Gurtovoy <maxg(a)mellanox.com> mainline inclusion from mainline-v5.2-rc1 commit 87fd125344d68adf7699ec7396aa9f905ce79a80 category: bugfix bugzilla: NA CVE: NA Link: https://gitee.com/openeuler/kernel/issues/I1WGZE ------------------------------------------------- In the past, before adding f41725bb ("nvme-rdma: Use mr pool") commit, we needed a reference on the ib_device as long as the tagset was alive, as the MRs in the request structures needed a valid ib_device. Now, we allocate/deallocate MR pool per QP and consume on demand. Also remove nvme_rdma_free_tagset function and use blk_mq_free_tag_set instead, as it unneeded anymore. This commit also fixes a memory leakage and possible segmentation fault. When configuring the system with NIC teaming (aka bonding), we use 1 network interface to create an HA connection to the target side. In case one connection breaks down, nvme-rdma driver will get notification from rdma-cm layer that underlying address was change and will start error recovery process. During this process, we'll reconnect to the target via the second interface in the bond without destroying the tagset. This will cause a leakage of the initial rdma device (ndev) and miscount in the reference count of the new created rdma device (new ndev). In the final destruction (or in another error flow), we'll get a warning dump from the ib_dealloc_pd that we still have inflight MR's related to that pd. This happens becasue of the miscount of the reference tag of the rdma device and causing access violation to it's elements (some queues are not destroyed yet). Signed-off-by: Max Gurtovoy <maxg(a)mellanox.com> Signed-off-by: Israel Rukshin <israelr(a)mellanox.com> Signed-off-by: Christoph Hellwig <hch(a)lst.de> Reviewed-by: Chao Leng <lengchao(a)huawei.com> Reviewed-by: Jike Cheng <chengjike.cheng(a)huawei.com> Conflicts: drivers/nvme/host/rdma.c [lrz: adjust context] Signed-off-by: Ruozhu Li <liruozhu(a)huawei.com> Signed-off-by: Lijie <lijie34(a)huawei.com> Reviewed-by: Tao Hou <houtao1(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/nvme/host/rdma.c | 34 +++++----------------------------- 1 file changed, 5 insertions(+), 29 deletions(-) diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index 077c67816665..0bfc1875575d 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -666,15 +666,6 @@ static int nvme_rdma_alloc_io_queues(struct nvme_rdma_ctrl *ctrl) return ret; } -static void nvme_rdma_free_tagset(struct nvme_ctrl *nctrl, - struct blk_mq_tag_set *set) -{ - struct nvme_rdma_ctrl *ctrl = to_rdma_ctrl(nctrl); - - blk_mq_free_tag_set(set); - nvme_rdma_dev_put(ctrl->device); -} - static struct blk_mq_tag_set *nvme_rdma_alloc_tagset(struct nvme_ctrl *nctrl, bool admin) { @@ -712,24 +703,9 @@ static struct blk_mq_tag_set *nvme_rdma_alloc_tagset(struct nvme_ctrl *nctrl, ret = blk_mq_alloc_tag_set(set); if (ret) - goto out; - - /* - * We need a reference on the device as long as the tag_set is alive, - * as the MRs in the request structures need a valid ib_device. - */ - ret = nvme_rdma_dev_get(ctrl->device); - if (!ret) { - ret = -EINVAL; - goto out_free_tagset; - } + return ERR_PTR(ret); return set; - -out_free_tagset: - blk_mq_free_tag_set(set); -out: - return ERR_PTR(ret); } static void nvme_rdma_destroy_admin_queue(struct nvme_rdma_ctrl *ctrl, @@ -737,7 +713,7 @@ static void nvme_rdma_destroy_admin_queue(struct nvme_rdma_ctrl *ctrl, { if (remove) { blk_cleanup_queue(ctrl->ctrl.admin_q); - nvme_rdma_free_tagset(&ctrl->ctrl, ctrl->ctrl.admin_tagset); + blk_mq_free_tag_set(ctrl->ctrl.admin_tagset); } if (ctrl->async_event_sqe.data) { cancel_work_sync(&ctrl->ctrl.async_event_work); @@ -815,7 +791,7 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl, blk_cleanup_queue(ctrl->ctrl.admin_q); out_free_tagset: if (new) - nvme_rdma_free_tagset(&ctrl->ctrl, ctrl->ctrl.admin_tagset); + blk_mq_free_tag_set(ctrl->ctrl.admin_tagset); out_free_async_qe: if (ctrl->async_event_sqe.data) { nvme_rdma_free_qe(ctrl->device->dev, &ctrl->async_event_sqe, @@ -832,7 +808,7 @@ static void nvme_rdma_destroy_io_queues(struct nvme_rdma_ctrl *ctrl, { if (remove) { blk_cleanup_queue(ctrl->ctrl.connect_q); - nvme_rdma_free_tagset(&ctrl->ctrl, ctrl->ctrl.tagset); + blk_mq_free_tag_set(ctrl->ctrl.tagset); } nvme_rdma_free_io_queues(ctrl); } @@ -873,7 +849,7 @@ static int nvme_rdma_configure_io_queues(struct nvme_rdma_ctrl *ctrl, bool new) blk_cleanup_queue(ctrl->ctrl.connect_q); out_free_tag_set: if (new) - nvme_rdma_free_tagset(&ctrl->ctrl, ctrl->ctrl.tagset); + blk_mq_free_tag_set(ctrl->ctrl.tagset); out_free_io_queues: nvme_rdma_free_io_queues(ctrl); return ret; -- 2.25.1

1 19

【议题收集】【Meeting Notice】openEuler kernel sig meeting Time: 2020-10-16 10:00-12:00
by Xie XiuQi 20 Oct '20

20 Oct '20

openEuler kernel sig meeting 定于 2020-10-16 10:00-12:00 召开，欢迎回复邮件申报议题。 --- 会议信息 --- 会议链接：https://zoom.us/j/94156903933 会议 ID : 94156903933 --- 会议纪要归档 --- Meeting Record 20200918： https://gitee.com/openeuler/kernel/issues/I1WGN0

1 3

[PATCH 00/19] backport stable-4.19.152
by Yang Yingliang 19 Oct '20

19 Oct '20

Anant Thazhemadam (1): staging: comedi: check validity of wMaxPacketSize of usb endpoints found Arnaud Patard (1): drivers/net/ethernet/marvell/mvmdio.c: Fix non OF case Dmitry Golovin (1): ARM: 8939/1: kbuild: use correct nm executable Dominik Przychodni (1): crypto: qat - check cipher length for aead AES-CBC-HMAC-SHA Greg Kroah-Hartman (1): Linux 4.19.152 Herbert Xu (1): crypto: bcm - Verify GCM/CCM key length in setkey Jan Kara (2): reiserfs: Initialize inode keys properly reiserfs: Fix oops during mount Jason A. Donenfeld (1): ARM: 8867/1: vdso: pass --be8 to linker if necessary Leo Yan (1): perf cs-etm: Move definition of 'traceid_list' global variable from header file Leonid Bloch (1): USB: serial: option: Add Telit FT980-KS composition Luiz Augusto von Dentz (2): Bluetooth: Consolidate encryption handling in hci_encrypt_cfm Bluetooth: Disconnect if E0 is used for Level 4 Masahiro Yamada (1): ARM: 8858/1: vdso: use $(LD) instead of $(CC) to link VDSO Mychaela N. Falconia (1): USB: serial: ftdi_sio: add support for FreeCalypso JTAG+UART adapters Oliver Neukum (1): media: usbtv: Fix refcounting mixup Patrick Steinhardt (1): Bluetooth: Fix update of connection state in `hci_encrypt_cfm` Scott Chen (1): USB: serial: pl2303: add device-id for HP GC device Wilken Gottwalt (1): USB: serial: option: add Cellient MPL200 card Makefile | 2 +- arch/arm/boot/compressed/Makefile | 4 +- arch/arm/vdso/Makefile | 22 +++++------ drivers/crypto/bcm/cipher.c | 15 +++++++- drivers/crypto/qat/qat_common/qat_algs.c | 10 ++++- drivers/media/usb/usbtv/usbtv-core.c | 3 +- drivers/net/ethernet/marvell/mvmdio.c | 22 ++++++++--- drivers/staging/comedi/drivers/vmk80xx.c | 3 ++ drivers/usb/serial/ftdi_sio.c | 5 +++ drivers/usb/serial/ftdi_sio_ids.h | 7 ++++ drivers/usb/serial/option.c | 5 +++ drivers/usb/serial/pl2303.c | 1 + drivers/usb/serial/pl2303.h | 1 + fs/reiserfs/inode.c | 6 +-- fs/reiserfs/xattr.c | 7 ++++ include/net/bluetooth/hci_core.h | 30 ++++++++++++--- net/bluetooth/hci_conn.c | 17 +++++++++ net/bluetooth/hci_event.c | 48 ++++++------------------ tools/perf/util/cs-etm.c | 3 ++ tools/perf/util/cs-etm.h | 3 -- 20 files changed, 138 insertions(+), 76 deletions(-) -- 2.25.1

1 19

[PATCH 00/26] huge vmalloc mappings
by Yang Yingliang 19 Oct '20

19 Oct '20

Anshuman Khandual (1): mm/ioremap: probe platform for p4d huge map support Christoph Hellwig (8): mm: remove __get_vm_area mm: unexport unmap_kernel_range_noflush mm: only allow page table mappings for built-in zsmalloc mm: pass addr as unsigned long to vb_free mm: remove vmap_page_range_noflush and vunmap_page_range mm: rename vmap_page_range to map_kernel_range mm: don't return the number of pages from map_kernel_range{, _noflush} mm: remove map_vm_range Daniel Axtens (1): mm/memory.c: add apply_to_existing_page_range() helper Ingo Molnar (1): mm/vmalloc: Add empty <asm/vmalloc.h> headers and use them from <linux/vmalloc.h> Mike Rapoport (1): mm: move lib/ioremap.c to mm/ Nicholas Piggin (9): mm/vmalloc: fix vmalloc_to_page for huge vmap mappings mm: apply_to_pte_range warn and fail if a large pte is encountered mm/vmalloc: rename vmap_*_range vmap_pages_*_range mm/ioremap: rename ioremap_*_range to vmap_*_range mm: HUGE_VMAP arch support cleanup arm64: inline huge vmap supported functions mm: Move vmap_range from mm/ioremap.c to mm/vmalloc.c mm/vmalloc: add vmap_range_noflush variant mm/vmalloc: Hugepage vmalloc mappings Steven Price (2): mm: add generic p?d_leaf() macros arm64: mm: add p?d_leaf() definitions Will Deacon (3): ioremap: rework pXd_free_pYd_page() API lib/ioremap: ensure phys_addr actually corresponds to a physical address lib/ioremap: ensure break-before-make is used for huge p4d mappings Documentation/core-api/cachetlb.rst | 2 +- arch/Kconfig | 4 + arch/alpha/include/asm/vmalloc.h | 4 + arch/arc/include/asm/vmalloc.h | 4 + arch/arm/include/asm/vmalloc.h | 4 + arch/arm64/include/asm/pgtable.h | 2 + arch/arm64/include/asm/vmalloc.h | 29 ++ arch/arm64/mm/mmu.c | 16 - arch/c6x/include/asm/vmalloc.h | 4 + arch/csky/include/asm/vmalloc.h | 4 + arch/h8300/include/asm/vmalloc.h | 4 + arch/hexagon/include/asm/vmalloc.h | 4 + arch/ia64/include/asm/vmalloc.h | 4 + arch/m68k/include/asm/vmalloc.h | 4 + arch/microblaze/include/asm/vmalloc.h | 4 + arch/mips/include/asm/vmalloc.h | 4 + arch/nds32/include/asm/vmalloc.h | 4 + arch/nios2/include/asm/vmalloc.h | 4 + arch/openrisc/include/asm/vmalloc.h | 4 + arch/parisc/include/asm/vmalloc.h | 4 + arch/powerpc/include/asm/vmalloc.h | 12 + arch/powerpc/kernel/pci_64.c | 3 +- arch/riscv/include/asm/vmalloc.h | 4 + arch/s390/include/asm/vmalloc.h | 4 + arch/sh/include/asm/vmalloc.h | 4 + arch/sh/kernel/cpu/sh4/sq.c | 3 +- arch/sparc/include/asm/vmalloc.h | 4 + arch/um/include/asm/vmalloc.h | 4 + arch/unicore32/include/asm/vmalloc.h | 4 + arch/x86/include/asm/vmalloc.h | 12 + arch/x86/mm/ioremap.c | 11 +- arch/x86/mm/pgtable.c | 8 + arch/xtensa/include/asm/vmalloc.h | 4 + include/asm-generic/pgtable.h | 25 ++ include/linux/io.h | 8 - include/linux/mm.h | 3 + include/linux/vmalloc.h | 24 +- init/main.c | 1 - kernel/dma/mapping.c | 3 +- lib/Makefile | 1 - lib/ioremap.c | 183 --------- mm/Kconfig | 2 +- mm/Makefile | 2 +- mm/ioremap.c | 34 ++ mm/memory.c | 152 ++++++-- mm/vmalloc.c | 518 +++++++++++++++++++------- mm/zsmalloc.c | 4 +- 47 files changed, 752 insertions(+), 398 deletions(-) create mode 100644 arch/alpha/include/asm/vmalloc.h create mode 100644 arch/arc/include/asm/vmalloc.h create mode 100644 arch/arm/include/asm/vmalloc.h create mode 100644 arch/arm64/include/asm/vmalloc.h create mode 100644 arch/c6x/include/asm/vmalloc.h create mode 100644 arch/csky/include/asm/vmalloc.h create mode 100644 arch/h8300/include/asm/vmalloc.h create mode 100644 arch/hexagon/include/asm/vmalloc.h create mode 100644 arch/ia64/include/asm/vmalloc.h create mode 100644 arch/m68k/include/asm/vmalloc.h create mode 100644 arch/microblaze/include/asm/vmalloc.h create mode 100644 arch/mips/include/asm/vmalloc.h create mode 100644 arch/nds32/include/asm/vmalloc.h create mode 100644 arch/nios2/include/asm/vmalloc.h create mode 100644 arch/openrisc/include/asm/vmalloc.h create mode 100644 arch/parisc/include/asm/vmalloc.h create mode 100644 arch/powerpc/include/asm/vmalloc.h create mode 100644 arch/riscv/include/asm/vmalloc.h create mode 100644 arch/s390/include/asm/vmalloc.h create mode 100644 arch/sh/include/asm/vmalloc.h create mode 100644 arch/sparc/include/asm/vmalloc.h create mode 100644 arch/um/include/asm/vmalloc.h create mode 100644 arch/unicore32/include/asm/vmalloc.h create mode 100644 arch/x86/include/asm/vmalloc.h create mode 100644 arch/xtensa/include/asm/vmalloc.h delete mode 100644 lib/ioremap.c create mode 100644 mm/ioremap.c -- 2.25.1

1 26

[PATCH 0/4] fixes CVE-2020-12351 and CVE-2020-12352
by Yang Yingliang 19 Oct '20

19 Oct '20

Luiz Augusto von Dentz (4): Bluetooth: A2MP: Fix not initializing all members Bluetooth: L2CAP: Fix calling sk_filter on non-socket based channel Bluetooth: Disable High Speed by default Bluetooth: MGMT: Fix not checking if BT_HS is enabled include/net/bluetooth/l2cap.h | 2 ++ net/bluetooth/Kconfig | 1 - net/bluetooth/a2mp.c | 22 +++++++++++++++++++++- net/bluetooth/l2cap_core.c | 7 ++++--- net/bluetooth/l2cap_sock.c | 14 ++++++++++++++ net/bluetooth/mgmt.c | 7 ++++++- 6 files changed, 47 insertions(+), 6 deletions(-) -- 2.25.1

1 4

[PATCH 00/49] backport stable-4.19.151
by Yang Yingliang 19 Oct '20

19 Oct '20

Aaron Ma (1): platform/x86: thinkpad_acpi: re-initialize ACPI buffer size when reuse Anant Thazhemadam (3): net: wireless: nl80211: fix out-of-bounds access in nl80211_del_key() net: team: fix memory leak in __team_options_register net: usb: rtl8150: set random MAC address when set_ethernet_addr() fails Antony Antony (4): xfrm: clone XFRMA_SET_MARK in xfrm_do_migrate xfrm: clone XFRMA_REPLAY_ESN_VAL in xfrm_do_migrate xfrm: clone XFRMA_SEC_CTX in xfrm_do_migrate xfrm: clone whole liftime_cur structure in xfrm_do_migrate Aya Levin (2): net/mlx5e: Fix VLAN cleanup flow net/mlx5e: Fix VLAN create flow Chaitanya Kulkarni (1): nvme-core: put ctrl ref when module ref get fail Coly Li (1): mmc: core: don't set limits.discard_granularity as 0 Cristian Ciocaltea (1): i2c: owl: Clear NACK and BUS error bits David Howells (3): rxrpc: Downgrade the BUG() for unsupported token type in rxrpc_read() rxrpc: Fix some missing _bh annotations on locking conn->state_lock rxrpc: Fix server keyring leak Dinh Nguyen (1): arm64: dts: stratix10: add status to qspi dts node Dumitru Ceara (1): openvswitch: handle DNAT tuple collision Eric Dumazet (4): macsec: avoid use-after-free in macsec_handle_frame() sctp: fix sctp_auth_init_hmacs() error path team: set dev->needed_headroom in team_setup_by_port() bonding: set dev->needed_headroom in bond_setup_by_slave() Geert Uytterhoeven (1): Revert "ravb: Fixed to be able to unload modules" Greg Kroah-Hartman (1): Linux 4.19.151 Hans de Goede (2): platform/x86: intel-vbtn: Fix SW_TABLET_MODE always reporting 1 on the HP Pavilion 11 x360 platform/x86: intel-vbtn: Switch to an allow-list for SW_TABLET_MODE reporting Herbert Xu (1): xfrm: Use correct address family in xfrm_state_find Hugh Dickins (1): mm/khugepaged: fix filemap page_to_pgoff(page) != offset Jean Delvare (1): i2c: i801: Exclude device from suspend direct complete optimization Jerome Brunet (1): i2c: meson: fix clock setting overwrite Kajol Jain (1): perf: Fix task_function_call() error handling Karol Herbst (1): drm/nouveau/mem: guard against NULL pointer access in mem_del Linus Torvalds (1): usermodehelper: reset umask to default before executing user process Marc Dionne (1): rxrpc: Fix rxkad token xdr encoding Miquel Raynal (1): mtd: rawnand: sunxi: Fix the probe error path Necip Fazil Yildiran (1): platform/x86: fix kconfig dependency warning for FUJITSU_LAPTOP Nicolas Belin (1): i2c: meson: fixup rate calculation with filter delay Peilin Ye (3): fbdev, newport_con: Move FONT_EXTRA_WORDS macros into linux/font.h Fonts: Support FONT_EXTRA_WORDS macros for built-in fonts fbcon: Fix global-out-of-bounds read in fbcon_get_font() Philip Yang (1): drm/amdgpu: prevent double kfree ttm->sg Randy Dunlap (1): mdio: fix mdio-thunder.c dependency & build error Sabrina Dubroca (1): xfrmi: drop ignore_df check before updating pmtu Tetsuo Handa (1): driver core: Fix probe_count imbalance in really_probe() Tom Rix (1): platform/x86: thinkpad_acpi: initialize tp_nvram_state variable Vijay Balakrishna (1): mm: khugepaged: recalculate min_free_kbytes after memory hotplug as expected by khugepaged Vladimir Zapolskiy (1): cifs: Fix incomplete memory allocation on setxattr path Voon Weifeng (1): net: stmmac: removed enabling eee in EEE set callback Wilken Gottwalt (1): net: usb: ax88179_178a: fix missing stop entry in driver_info Makefile | 2 +- .../dts/altera/socfpga_stratix10_socdk.dts | 1 + drivers/base/dd.c | 5 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 1 + drivers/gpu/drm/nouveau/nouveau_mem.c | 2 + drivers/i2c/busses/i2c-i801.c | 1 + drivers/i2c/busses/i2c-meson.c | 42 +++++-- drivers/i2c/busses/i2c-owl.c | 6 + drivers/mmc/core/queue.c | 2 +- drivers/mtd/nand/raw/sunxi_nand.c | 2 +- drivers/net/bonding/bond_main.c | 1 + .../net/ethernet/mellanox/mlx5/core/en_fs.c | 14 ++- drivers/net/ethernet/renesas/ravb_main.c | 110 +++++++++--------- .../ethernet/stmicro/stmmac/stmmac_ethtool.c | 15 +-- drivers/net/macsec.c | 4 +- drivers/net/phy/Kconfig | 1 + drivers/net/team/team.c | 3 +- drivers/net/usb/ax88179_178a.c | 1 + drivers/net/usb/rtl8150.c | 16 ++- drivers/nvme/host/core.c | 4 +- drivers/platform/x86/Kconfig | 1 + drivers/platform/x86/intel-vbtn.c | 64 +++++++--- drivers/platform/x86/thinkpad_acpi.c | 6 +- drivers/video/console/newport_con.c | 7 +- drivers/video/fbdev/core/fbcon.c | 12 ++ drivers/video/fbdev/core/fbcon.h | 7 -- drivers/video/fbdev/core/fbcon_rotate.c | 1 + drivers/video/fbdev/core/tileblit.c | 1 + fs/cifs/smb2ops.c | 2 +- include/linux/font.h | 13 +++ include/linux/khugepaged.h | 5 + include/net/xfrm.h | 16 +-- kernel/events/core.c | 5 +- kernel/umh.c | 9 ++ lib/fonts/font_10x18.c | 9 +- lib/fonts/font_6x10.c | 9 +- lib/fonts/font_6x11.c | 9 +- lib/fonts/font_7x14.c | 9 +- lib/fonts/font_8x16.c | 9 +- lib/fonts/font_8x8.c | 9 +- lib/fonts/font_acorn_8x8.c | 9 +- lib/fonts/font_mini_4x6.c | 8 +- lib/fonts/font_pearl_8x8.c | 9 +- lib/fonts/font_sun12x22.c | 9 +- lib/fonts/font_sun8x16.c | 7 +- mm/khugepaged.c | 25 +++- mm/page_alloc.c | 3 + net/openvswitch/conntrack.c | 22 ++-- net/rxrpc/conn_event.c | 6 +- net/rxrpc/key.c | 18 ++- net/sctp/auth.c | 1 + net/wireless/nl80211.c | 3 + net/xfrm/xfrm_interface.c | 2 +- net/xfrm/xfrm_state.c | 42 ++++++- 54 files changed, 391 insertions(+), 209 deletions(-) -- 2.25.1

1 49

[PATCH] perf/core: Fix race in the perf_mmap_close() function
by Yang Yingliang 16 Oct '20

16 Oct '20

From: Jiri Olsa <jolsa(a)redhat.com> mainline inclusion from mainline-v5.10 commit f91072ed1b7283b13ca57fcfbece5a3b92726143 category: bugfix bugzilla: NA CVE: CVE-2020-14351 -------------------------------- There's a possible race in perf_mmap_close() when checking ring buffer's mmap_count refcount value. The problem is that the mmap_count check is not atomic because we call atomic_dec() and atomic_read() separately. perf_mmap_close: ... atomic_dec(&rb->mmap_count); ... if (atomic_read(&rb->mmap_count)) goto out_put; <ring buffer detach> free_uid out_put: ring_buffer_put(rb); /* could be last */ The race can happen when we have two (or more) events sharing same ring buffer and they go through atomic_dec() and then they both see 0 as refcount value later in atomic_read(). Then both will go on and execute code which is meant to be run just once. The code that detaches ring buffer is probably fine to be executed more than once, but the problem is in calling free_uid(), which will later on demonstrate in related crashes and refcount warnings, like: refcount_t: addition on 0; use-after-free. ... RIP: 0010:refcount_warn_saturate+0x6d/0xf ... Call Trace: prepare_creds+0x190/0x1e0 copy_creds+0x35/0x172 copy_process+0x471/0x1a80 _do_fork+0x83/0x3a0 __do_sys_wait4+0x83/0x90 __do_sys_clone+0x85/0xa0 do_syscall_64+0x5b/0x1e0 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Using atomic decrease and check instead of separated calls. Tested-by: Michael Petlan <mpetlan(a)redhat.com> Signed-off-by: Jiri Olsa <jolsa(a)kernel.org> Signed-off-by: Ingo Molnar <mingo(a)kernel.org> Acked-by: Peter Zijlstra <a.p.zijlstra(a)chello.nl> Acked-by: Namhyung Kim <namhyung(a)kernel.org> Acked-by: Wade Mealing <wmealing(a)redhat.com> Fixes: 9bb5d40cd93c ("perf: Fix mmap() accounting hole"); Link: https://lore.kernel.org/r/20200916115311.GE2301783@krava Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Conflicts: kernel/events/core.c [yyl: adjust context] Reviewed-by: Jian Cheng <cj.chengjian(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- kernel/events/core.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 30a81d1b7b5c..452775f68129 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -5496,11 +5496,11 @@ static void perf_pmu_output_stop(struct perf_event *event); static void perf_mmap_close(struct vm_area_struct *vma) { struct perf_event *event = vma->vm_file->private_data; - struct ring_buffer *rb = ring_buffer_get(event); struct user_struct *mmap_user = rb->mmap_user; int mmap_locked = rb->mmap_locked; unsigned long size = perf_data_size(rb); + bool detach_rest = false; if (event->pmu->event_unmapped) event->pmu->event_unmapped(event, vma->vm_mm); @@ -5531,7 +5531,8 @@ static void perf_mmap_close(struct vm_area_struct *vma) mutex_unlock(&event->mmap_mutex); } - atomic_dec(&rb->mmap_count); + if (atomic_dec_and_test(&rb->mmap_count)) + detach_rest = true; if (!atomic_dec_and_mutex_lock(&event->mmap_count, &event->mmap_mutex)) goto out_put; @@ -5540,7 +5541,7 @@ static void perf_mmap_close(struct vm_area_struct *vma) mutex_unlock(&event->mmap_mutex); /* If there's still other mmap()s of this buffer, we're done. */ - if (atomic_read(&rb->mmap_count)) + if (!detach_rest) goto out_put; /* -- 2.25.1

1 0

[PATCH] geneve: add transport ports in route lookup for geneve
by Yang Yingliang 15 Oct '20

15 Oct '20

From: Mark Gray <mark.d.gray(a)redhat.com> stable inclusion from linux-4.19.148 commit c797110d97c48054d1491251fd713900ff51615c CVE: CVE-2020-25645 -------------------------------- [ Upstream commit 34beb21594519ce64a55a498c2fe7d567bc1ca20 ] This patch adds transport ports information for route lookup so that IPsec can select Geneve tunnel traffic to do encryption. This is needed for OVS/OVN IPsec with encrypted Geneve tunnels. This can be tested by configuring a host-host VPN using an IKE daemon and specifying port numbers. For example, for an Openswan-type configuration, the following parameters should be configured on both hosts and IPsec set up as-per normal: $ cat /etc/ipsec.conf conn in ... left=$IP1 right=$IP2 ... leftprotoport=udp/6081 rightprotoport=udp ... conn out ... left=$IP1 right=$IP2 ... leftprotoport=udp rightprotoport=udp/6081 ... The tunnel can then be setup using "ip" on both hosts (but changing the relevant IP addresses): $ ip link add tun type geneve id 1000 remote $IP2 $ ip addr add 192.168.0.1/24 dev tun $ ip link set tun up This can then be tested by pinging from $IP1: $ ping 192.168.0.2 Without this patch the traffic is unencrypted on the wire. Fixes: 2d07dc79fe04 ("geneve: add initial netdev driver for GENEVE tunnels") Signed-off-by: Qiuyu Xiao <qiuyu.xiao.qyx(a)gmail.com> Signed-off-by: Mark Gray <mark.d.gray(a)redhat.com> Reviewed-by: Greg Rose <gvrose8192(a)gmail.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/net/geneve.c | 37 +++++++++++++++++++++++++++---------- 1 file changed, 27 insertions(+), 10 deletions(-) diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index 1506e913cebb..fdb251e1f471 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -721,7 +721,8 @@ static struct rtable *geneve_get_v4_rt(struct sk_buff *skb, struct net_device *dev, struct geneve_sock *gs4, struct flowi4 *fl4, - const struct ip_tunnel_info *info) + const struct ip_tunnel_info *info, + __be16 dport, __be16 sport) { bool use_cache = ip_tunnel_dst_cache_usable(skb, info); struct geneve_dev *geneve = netdev_priv(dev); @@ -737,6 +738,8 @@ static struct rtable *geneve_get_v4_rt(struct sk_buff *skb, fl4->flowi4_proto = IPPROTO_UDP; fl4->daddr = info->key.u.ipv4.dst; fl4->saddr = info->key.u.ipv4.src; + fl4->fl4_dport = dport; + fl4->fl4_sport = sport; tos = info->key.tos; if ((tos == 1) && !geneve->collect_md) { @@ -771,7 +774,8 @@ static struct dst_entry *geneve_get_v6_dst(struct sk_buff *skb, struct net_device *dev, struct geneve_sock *gs6, struct flowi6 *fl6, - const struct ip_tunnel_info *info) + const struct ip_tunnel_info *info, + __be16 dport, __be16 sport) { bool use_cache = ip_tunnel_dst_cache_usable(skb, info); struct geneve_dev *geneve = netdev_priv(dev); @@ -787,6 +791,9 @@ static struct dst_entry *geneve_get_v6_dst(struct sk_buff *skb, fl6->flowi6_proto = IPPROTO_UDP; fl6->daddr = info->key.u.ipv6.dst; fl6->saddr = info->key.u.ipv6.src; + fl6->fl6_dport = dport; + fl6->fl6_sport = sport; + prio = info->key.tos; if ((prio == 1) && !geneve->collect_md) { prio = ip_tunnel_get_dsfield(ip_hdr(skb), skb); @@ -833,14 +840,15 @@ static int geneve_xmit_skb(struct sk_buff *skb, struct net_device *dev, __be16 df; int err; - rt = geneve_get_v4_rt(skb, dev, gs4, &fl4, info); + sport = udp_flow_src_port(geneve->net, skb, 1, USHRT_MAX, true); + rt = geneve_get_v4_rt(skb, dev, gs4, &fl4, info, + geneve->info.key.tp_dst, sport); if (IS_ERR(rt)) return PTR_ERR(rt); skb_tunnel_check_pmtu(skb, &rt->dst, GENEVE_IPV4_HLEN + info->options_len); - sport = udp_flow_src_port(geneve->net, skb, 1, USHRT_MAX, true); if (geneve->collect_md) { tos = ip_tunnel_ecn_encap(key->tos, ip_hdr(skb), skb); ttl = key->ttl; @@ -875,13 +883,14 @@ static int geneve6_xmit_skb(struct sk_buff *skb, struct net_device *dev, __be16 sport; int err; - dst = geneve_get_v6_dst(skb, dev, gs6, &fl6, info); + sport = udp_flow_src_port(geneve->net, skb, 1, USHRT_MAX, true); + dst = geneve_get_v6_dst(skb, dev, gs6, &fl6, info, + geneve->info.key.tp_dst, sport); if (IS_ERR(dst)) return PTR_ERR(dst); skb_tunnel_check_pmtu(skb, dst, GENEVE_IPV6_HLEN + info->options_len); - sport = udp_flow_src_port(geneve->net, skb, 1, USHRT_MAX, true); if (geneve->collect_md) { prio = ip_tunnel_ecn_encap(key->tos, ip_hdr(skb), skb); ttl = key->ttl; @@ -957,13 +966,18 @@ static int geneve_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb) { struct ip_tunnel_info *info = skb_tunnel_info(skb); struct geneve_dev *geneve = netdev_priv(dev); + __be16 sport; if (ip_tunnel_info_af(info) == AF_INET) { struct rtable *rt; struct flowi4 fl4; + struct geneve_sock *gs4 = rcu_dereference(geneve->sock4); + sport = udp_flow_src_port(geneve->net, skb, + 1, USHRT_MAX, true); - rt = geneve_get_v4_rt(skb, dev, gs4, &fl4, info); + rt = geneve_get_v4_rt(skb, dev, gs4, &fl4, info, + geneve->info.key.tp_dst, sport); if (IS_ERR(rt)) return PTR_ERR(rt); @@ -973,9 +987,13 @@ static int geneve_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb) } else if (ip_tunnel_info_af(info) == AF_INET6) { struct dst_entry *dst; struct flowi6 fl6; + struct geneve_sock *gs6 = rcu_dereference(geneve->sock6); + sport = udp_flow_src_port(geneve->net, skb, + 1, USHRT_MAX, true); - dst = geneve_get_v6_dst(skb, dev, gs6, &fl6, info); + dst = geneve_get_v6_dst(skb, dev, gs6, &fl6, info, + geneve->info.key.tp_dst, sport); if (IS_ERR(dst)) return PTR_ERR(dst); @@ -986,8 +1004,7 @@ static int geneve_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb) return -EINVAL; } - info->key.tp_src = udp_flow_src_port(geneve->net, skb, - 1, USHRT_MAX, true); + info->key.tp_src = sport; info->key.tp_dst = geneve->info.key.tp_dst; return 0; } -- 2.25.1

1 0

[PATCH 00/78] bugfixes for openEuler-1.0-LTS
by Yang Yingliang 15 Oct '20

15 Oct '20

Al Grant (1): perf tools: Correct SNOOPX field offset Al Viro (2): do_epoll_ctl(): clean the failure exits up a bit fix regression in "epoll: Keep a reference on files added to the check list" Alex Williamson (1): vfio/type1: Add proper error unwind for vfio_iommu_replay() Alvin Šipraga (1): macvlan: validate setting of multiple remote source MAC addresses Bodo Stroesser (1): scsi: target: tcmu: Fix crash in tcmu_flush_dcache_range on ARM Brian Foster (3): xfs: acquire superblock freeze protection on eofblocks scans xfs: reset buffer write failure state on successful completion xfs: fix duplicate verification from xfs_qm_dqflush() Charan Teja Reddy (1): mm, page_alloc: fix core hung in free_pcppages_bulk() Chris Wilson (1): locking/lockdep: Fix overflow in presentation of average lock-time Chuck Lever (1): NFS: Zero-stateid SETATTR should first return delegation Daniel Borkmann (1): uaccess: Add non-pagefault user-space write function Darrick J. Wong (4): xfs: fix partially uninitialized structure in xfs_reflink_remap_extent xfs: fix inode quota reservation checks xfs: fix xfs_bmap_validate_extent_raw when checking attr fork of rt files xfs: initialize the shortform attr header padding entry Dave Chinner (1): xfs: Don't allow logging of XFS_ISTALE inodes David Milburn (2): nvme-fc: cancel async events before freeing event struct nvme-rdma: cancel async events before freeing event struct Doug Berger (1): mm: include CMA pages in lowmem_reserve at boot Eiichi Tsukata (1): xfs: Fix UBSAN null-ptr-deref in xfs_sysfs_init Eric Biggers (1): xfs: clear PF_MEMALLOC before exiting xfsaild thread George Kennedy (1): vt_ioctl: change VT_RESIZEX ioctl to check for error return from vc_resize() Gustav Wiklander (1): spi: Fix memory leak on splited transfers Heikki Krogerus (1): device property: Fix the secondary firmware node handling in set_primary_fwnode() Helge Deller (1): fs/signalfd.c: fix inconsistent return codes for signalfd4 Hou Pu (1): scsi: target: iscsi: Fix hang in iscsit_access_np() when getting tpg->np_login_sem Hugh Dickins (2): khugepaged: khugepaged_test_exit() check mmget_still_valid() khugepaged: adjust VM_BUG_ON_MM() in __khugepaged_enter() Jan Kara (5): ext4: fix checking of directory entry validity for inline directories ext4: don't BUG on inconsistent journal feature writeback: Protect inode->i_io_list with inode->i_lock writeback: Avoid skipping inode writeback writeback: Fix sync livelock due to b_dirty_time processing Jarkko Sakkinen (1): tpm: Unify the mismatching TPM space buffer sizes Jason Gunthorpe (1): include/linux/log2.h: add missing () around n in roundup_pow_of_two() Javed Hasan (1): scsi: fcoe: Memory leak fix in fcoe_sysfs_fcf_del() Jens Axboe (1): block: ensure bdi->io_pages is always initialized Jing Xiangfeng (1): scsi: iscsi: Do not put host in iscsi_set_flashnode_param() Li Heng (1): efi: add missed destroy_workqueue when efisubsys_init fails Lukas Czerner (3): jbd2: make sure jh have b_transaction set in refile/unfile_buffer ext4: handle read only external journal device ext4: handle option set by mount flags correctly Lukas Wunner (4): spi: Prevent adding devices below an unregistering controller serial: pl011: Fix oops on -EPROBE_DEFER serial: pl011: Don't leak amba_ports entry on driver register error serial: 8250: Avoid error message on reprobe Luo Meng (1): ext4: only set last error block when check system zone failed Mao Wenan (1): virtio_ring: Avoid loop when vq is broken in virtqueue_poll Marc Zyngier (1): epoll: Keep a reference on files added to the check list Masami Hiramatsu (2): perf probe: Fix memory leakage when the probe point is not found uaccess: Add non-pagefault user-space read functions Max Reitz (1): xfs: Fix tail rounding in xfs_alloc_file_space() Mikulas Patocka (2): xfs: don't update mtime on COW faults dm writecache: handle DAX to partitions on persistent memory correctly Ming Lei (1): blk-mq: order adding requests to hctx->dispatch and checking SCHED_RESTART Muchun Song (1): kprobes: fix kill kprobe which has been marked as gone Namhyung Kim (1): perf jevents: Fix suspicious code in fixregex() Peter Xu (1): mm/hugetlb: fix calculation of adjust_range_if_pmd_sharing_possible Peter Zijlstra (1): cpuidle: Fixup IRQ state Qiushi Wu (1): PCI: Fix pci_create_slot() reference count leak Rafael J. Wysocki (1): PM: sleep: core: Fix the handling of pending runtime resume requests Ralph Campbell (1): mm/thp: fix __split_huge_pmd_locked() for migration PMD Robin Murphy (1): iommu/iova: Don't BUG on invalid PFNs Selvin Xavier (1): RDMA/bnxt_re: Do not add user qps to flushlist Sergey Senozhatsky (1): serial: 8250: change lock order in serial8250_do_startup() Sunghyun Jin (1): percpu: fix first chunk size calculation for populated bitmap Tejun Heo (1): libata: implement ATA_HORKAGE_MAX_TRIM_128M and apply to Sandisks Tetsuo Handa (1): vt: defer kfree() of vc_screenbuf in vc_do_resize() Valmer Huhn (1): serial: 8250_exar: Fix number of ports for Commtech PCIe cards Varun Prakash (1): scsi: target: iscsi: Fix data digest calculation Wei Yongjun (1): kernel/relay.c: fix memleak on destroy relay channel Xianting Tian (1): fs: prevent BUG_ON in submit_bh_wbc() Xunlei Pang (1): mm: memcg: fix memcg reclaim soft lockup Yang Xu (1): KEYS: reaching the keys quotas correctly kaixuxia (1): xfs: Fix deadlock between AGI and AGF with RENAME_WHITEOUT zhangyi (F) (1): jbd2: add the missing unlock_buffer() in the error path of jbd2_write_superblock() block/blk-core.c | 2 + block/blk-mq-sched.c | 9 ++ block/blk-mq.c | 9 ++ drivers/ata/libata-core.c | 5 +- drivers/ata/libata-scsi.c | 8 +- drivers/base/core.c | 12 +- drivers/base/power/main.c | 16 ++- drivers/char/tpm/tpm-chip.c | 9 +- drivers/char/tpm/tpm.h | 6 +- drivers/char/tpm/tpm2-space.c | 26 ++-- drivers/char/tpm/tpmrm-dev.c | 2 +- drivers/cpuidle/cpuidle.c | 3 +- drivers/firmware/efi/efi.c | 2 + drivers/infiniband/hw/bnxt_re/main.c | 3 +- drivers/iommu/iova.c | 4 +- drivers/md/dm-writecache.c | 12 +- drivers/net/macvlan.c | 21 ++- drivers/nvme/host/fc.c | 1 + drivers/nvme/host/rdma.c | 1 + drivers/pci/slot.c | 6 +- drivers/scsi/fcoe/fcoe_ctlr.c | 2 +- drivers/scsi/scsi_transport_iscsi.c | 2 +- drivers/spi/Kconfig | 3 + drivers/spi/spi.c | 30 +++- drivers/target/iscsi/iscsi_target.c | 17 ++- drivers/target/iscsi/iscsi_target_login.c | 6 +- drivers/target/iscsi/iscsi_target_login.h | 3 +- drivers/target/iscsi/iscsi_target_nego.c | 3 +- drivers/target/target_core_user.c | 2 +- drivers/tty/serial/8250/8250_core.c | 11 +- drivers/tty/serial/8250/8250_exar.c | 24 +++- drivers/tty/serial/8250/8250_port.c | 9 +- drivers/tty/serial/amba-pl011.c | 16 ++- drivers/tty/vt/vt.c | 5 +- drivers/tty/vt/vt_ioctl.c | 12 +- drivers/vfio/vfio_iommu_type1.c | 71 ++++++++- drivers/virtio/virtio_ring.c | 3 + fs/buffer.c | 9 ++ fs/eventpoll.c | 23 +-- fs/ext4/block_validity.c | 3 +- fs/ext4/namei.c | 6 +- fs/ext4/super.c | 147 ++++++++++++------- fs/fs-writeback.c | 83 ++++++----- fs/jbd2/journal.c | 4 +- fs/jbd2/transaction.c | 10 ++ fs/nfs/nfs4proc.c | 4 +- fs/signalfd.c | 10 +- fs/xfs/libxfs/xfs_attr_leaf.c | 4 +- fs/xfs/libxfs/xfs_bmap.c | 2 +- fs/xfs/xfs_bmap_util.c | 4 +- fs/xfs/xfs_buf.c | 8 +- fs/xfs/xfs_dquot.c | 9 +- fs/xfs/xfs_file.c | 12 +- fs/xfs/xfs_icache.c | 13 +- fs/xfs/xfs_inode.c | 110 ++++++++------ fs/xfs/xfs_ioctl.c | 5 +- fs/xfs/xfs_reflink.c | 1 + fs/xfs/xfs_sysfs.h | 6 +- fs/xfs/xfs_trans_ail.c | 4 +- fs/xfs/xfs_trans_dquot.c | 2 +- fs/xfs/xfs_trans_inode.c | 2 + include/linux/fs.h | 8 +- include/linux/libata.h | 1 + include/linux/log2.h | 2 +- include/linux/uaccess.h | 26 ++++ include/trace/events/writeback.h | 13 +- kernel/kprobes.c | 9 +- kernel/locking/lockdep_proc.c | 2 +- kernel/relay.c | 1 + mm/huge_memory.c | 40 +++--- mm/hugetlb.c | 24 ++-- mm/khugepaged.c | 7 +- mm/maccess.c | 167 ++++++++++++++++++++-- mm/page_alloc.c | 7 +- mm/percpu.c | 2 +- mm/vmscan.c | 8 ++ security/keys/key.c | 2 +- security/keys/keyctl.c | 4 +- tools/include/uapi/linux/perf_event.h | 2 +- tools/perf/pmu-events/jevents.c | 2 +- tools/perf/util/probe-finder.c | 2 +- 81 files changed, 864 insertions(+), 322 deletions(-) -- 2.25.1

1 78

[PATCH 01/11] xfs: Fix tail rounding in xfs_alloc_file_space()
by Yang Yingliang 15 Oct '20

15 Oct '20

From: Max Reitz <mreitz(a)redhat.com> mainline inclusion from mainline-v5.4-rc3 commit e093c4be760ebf46c131ae0dd6138865a22f46fa category: bugfix bugzilla: NA CVE: NA --------------------------- To ensure that all blocks touched by the range [offset, offset + count) are allocated, we need to calculate the block count from the difference of the range end (rounded up) and the range start (rounded down). Before this patch, we just round up the byte count, which may lead to unaligned ranges not being fully allocated: $ touch test_file $ block_size=$(stat -fc '%S' test_file) $ fallocate -o $((block_size / 2)) -l $block_size test_file $ xfs_bmap test_file test_file: 0: [0..7]: 1396264..1396271 1: [8..15]: hole There should not be a hole there. Instead, the first two blocks should be fully allocated. With this patch applied, the result is something like this: $ touch test_file $ block_size=$(stat -fc '%S' test_file) $ fallocate -o $((block_size / 2)) -l $block_size test_file $ xfs_bmap test_file test_file: 0: [0..15]: 11024..11039 Signed-off-by: Max Reitz <mreitz(a)redhat.com> Reviewed-by: Carlos Maiolino <cmaiolino(a)redhat.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Reviewed-by: Darrick J. Wong <darrick.wong(a)oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong(a)oracle.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/xfs/xfs_bmap_util.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 3e1dd66bd676..e54f5aed30a9 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -869,6 +869,7 @@ xfs_alloc_file_space( xfs_filblks_t allocatesize_fsb; xfs_extlen_t extsz, temp; xfs_fileoff_t startoffset_fsb; + xfs_fileoff_t endoffset_fsb; int nimaps; int quota_flag; int rt; @@ -896,7 +897,8 @@ xfs_alloc_file_space( imapp = &imaps[0]; nimaps = 1; startoffset_fsb = XFS_B_TO_FSBT(mp, offset); - allocatesize_fsb = XFS_B_TO_FSB(mp, count); + endoffset_fsb = XFS_B_TO_FSB(mp, offset + count); + allocatesize_fsb = endoffset_fsb - startoffset_fsb; /* * Allocate file space until done or until there is an error -- 2.25.1

1 10

[PATCH kernel-5.5] RISCV-V: KVM: use __u64 define variables instead of u64 in kvm.h
by l00484210 13 Oct '20

13 Oct '20

From: MingWang Li <limingwang(a)huawei.com> euleros inclusion category: bugfix bugzilla: NA CVE: NA one error occurred while building qemu on riscv system, which is unknown type name 'u64'. Link: https://gitee.com/openeuler/kernel/issues/I1XUB0 Signed-off-by: MingWang Li <limingwang(a)huawei.com> --- arch/riscv/include/uapi/asm/kvm.h | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/arch/riscv/include/uapi/asm/kvm.h b/arch/riscv/include/uapi/asm/kvm.h index 65cd00654..c2b4ad6b1 100644 --- a/arch/riscv/include/uapi/asm/kvm.h +++ b/arch/riscv/include/uapi/asm/kvm.h @@ -75,10 +75,10 @@ struct kvm_riscv_csr { /* TIMER registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */ struct kvm_riscv_timer { - u64 frequency; - u64 time; - u64 compare; - u64 state; + __u64 frequency; + __u64 time; + __u64 compare; + __u64 state; }; /* Possible states for kvm_riscv_timer */ @@ -110,17 +110,17 @@ struct kvm_riscv_timer { /* Timer registers are mapped as type 4 */ #define KVM_REG_RISCV_TIMER (0x04 << KVM_REG_RISCV_TYPE_SHIFT) #define KVM_REG_RISCV_TIMER_REG(name) \ - (offsetof(struct kvm_riscv_timer, name) / sizeof(u64)) + (offsetof(struct kvm_riscv_timer, name) / sizeof(__u64)) /* F extension registers are mapped as type 5 */ #define KVM_REG_RISCV_FP_F (0x05 << KVM_REG_RISCV_TYPE_SHIFT) #define KVM_REG_RISCV_FP_F_REG(name) \ - (offsetof(struct __riscv_f_ext_state, name) / sizeof(u32)) + (offsetof(struct __riscv_f_ext_state, name) / sizeof(__u32)) /* D extension registers are mapped as type 6 */ #define KVM_REG_RISCV_FP_D (0x06 << KVM_REG_RISCV_TYPE_SHIFT) #define KVM_REG_RISCV_FP_D_REG(name) \ - (offsetof(struct __riscv_d_ext_state, name) / sizeof(u64)) + (offsetof(struct __riscv_d_ext_state, name) / sizeof(__u64)) #endif -- 2.19.1

2 1

[PATCH 1/2] netfilter: nf_tables: incorrect enum nft_list_attributes definition
by Yang Yingliang 13 Oct '20

13 Oct '20

From: Pablo Neira Ayuso <pablo(a)netfilter.org> stable inclusion from linux-4.19.144 commit 3f21d1dd7cafb0230dc141e64ec5da622b3b1c46 -------------------------------- [ Upstream commit da9125df854ea48a6240c66e8a67be06e2c12c03 ] This should be NFTA_LIST_UNSPEC instead of NFTA_LIST_UNPEC, all other similar attribute definitions are postfixed with _UNSPEC. Fixes: 96518518cc41 ("netfilter: add nftables") Signed-off-by: Pablo Neira Ayuso <pablo(a)netfilter.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Signed-off-by: Aichun Li <liaichun(a)huawei.com> Reviewed-by: wangxiaopeng <wangxiaopeng7(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/uapi/linux/netfilter/nf_tables.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h index 44c8ba1f3652..8586c816bea0 100644 --- a/include/uapi/linux/netfilter/nf_tables.h +++ b/include/uapi/linux/netfilter/nf_tables.h @@ -132,7 +132,7 @@ enum nf_tables_msg_types { * @NFTA_LIST_ELEM: list element (NLA_NESTED) */ enum nft_list_attributes { - NFTA_LIST_UNPEC, + NFTA_LIST_UNSPEC, NFTA_LIST_ELEM, __NFTA_LIST_MAX }; -- 2.25.1

1 1

[PATCH 1/6] blk-mq: insert passthrough request into hctx->dispatch directly
by Yang Yingliang 13 Oct '20

13 Oct '20

From: Ming Lei <ming.lei(a)redhat.com> mainline inclusion from mainline-5.6-rc4 commit 01e99aeca3979600302913cef3f89076786f32c8 category: bugfix bugzilla: 42777 CVE: NA --------------------------- For some reason, device may be in one situation which can't handle FS request, so STS_RESOURCE is always returned and the FS request will be added to hctx->dispatch. However passthrough request may be required at that time for fixing the problem. If passthrough request is added to scheduler queue, there isn't any chance for blk-mq to dispatch it given we prioritize requests in hctx->dispatch. Then the FS IO request may never be completed, and IO hang is caused. So passthrough request has to be added to hctx->dispatch directly for fixing the IO hang. Fix this issue by inserting passthrough request into hctx->dispatch directly together withing adding FS request to the tail of hctx->dispatch in blk_mq_dispatch_rq_list(). Actually we add FS request to tail of hctx->dispatch at default, see blk_mq_request_bypass_insert(). Then it becomes consistent with original legacy IO request path, in which passthrough request is always added to q->queue_head. Cc: Dongli Zhang <dongli.zhang(a)oracle.com> Cc: Christoph Hellwig <hch(a)infradead.org> Cc: Ewan D. Milne <emilne(a)redhat.com> Signed-off-by: Ming Lei <ming.lei(a)redhat.com> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Conflicts: block/blk-flush.c block/blk-mq.c block/blk-mq-sched.c Signed-off-by: Yu Kuai <yukuai3(a)huawei.com> Reviewed-by: Yufen Yu <yuyufen(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- block/blk-flush.c | 2 +- block/blk-mq-sched.c | 22 +++++++++++++++------- block/blk-mq.c | 18 +++++++++++------- block/blk-mq.h | 3 ++- 4 files changed, 29 insertions(+), 16 deletions(-) diff --git a/block/blk-flush.c b/block/blk-flush.c index 256fa1ccc2bd..2a8369eb6c1c 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -495,7 +495,7 @@ void blk_insert_flush(struct request *rq) if ((policy & REQ_FSEQ_DATA) && !(policy & (REQ_FSEQ_PREFLUSH | REQ_FSEQ_POSTFLUSH))) { if (q->mq_ops) - blk_mq_request_bypass_insert(rq, false); + blk_mq_request_bypass_insert(rq, false, false); else list_add_tail(&rq->queuelist, &q->queue_head); return; diff --git a/block/blk-mq-sched.c b/block/blk-mq-sched.c index e98f51b86342..623c258d0b34 100644 --- a/block/blk-mq-sched.c +++ b/block/blk-mq-sched.c @@ -357,13 +357,19 @@ static bool blk_mq_sched_bypass_insert(struct blk_mq_hw_ctx *hctx, bool has_sched, struct request *rq) { - /* dispatch flush rq directly */ - if (rq->rq_flags & RQF_FLUSH_SEQ) { - spin_lock(&hctx->lock); - list_add(&rq->queuelist, &hctx->dispatch); - spin_unlock(&hctx->lock); + /* + * dispatch flush and passthrough rq directly + * + * passthrough request has to be added to hctx->dispatch directly. + * For some reason, device may be in one situation which can't + * handle FS request, so STS_RESOURCE is always returned and the + * FS request will be added to hctx->dispatch. However passthrough + * request may be required at that time for fixing the problem. If + * passthrough request is added to scheduler queue, there isn't any + * chance to dispatch it given we prioritize requests in hctx->dispatch. + */ + if ((rq->rq_flags & RQF_FLUSH_SEQ) || blk_rq_is_passthrough(rq)) return true; - } if (has_sched) rq->rq_flags |= RQF_SORTED; @@ -387,8 +393,10 @@ void blk_mq_sched_insert_request(struct request *rq, bool at_head, WARN_ON(e && (rq->tag != -1)); - if (blk_mq_sched_bypass_insert(hctx, !!e, rq)) + if (blk_mq_sched_bypass_insert(hctx, !!e, rq)) { + blk_mq_request_bypass_insert(rq, at_head, false); goto run; + } if (e && e->type->ops.mq.insert_requests) { LIST_HEAD(list); diff --git a/block/blk-mq.c b/block/blk-mq.c index 59c92961cc20..a9138235b870 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -721,7 +721,7 @@ static void blk_mq_requeue_work(struct work_struct *work) * merge. */ if (rq->rq_flags & RQF_DONTPREP) - blk_mq_request_bypass_insert(rq, false); + blk_mq_request_bypass_insert(rq, false, false); else blk_mq_sched_insert_request(rq, true, false, false); } @@ -1227,7 +1227,7 @@ bool blk_mq_dispatch_rq_list(struct request_queue *q, struct list_head *list, bool needs_restart; spin_lock(&hctx->lock); - list_splice_init(list, &hctx->dispatch); + list_splice_tail_init(list, &hctx->dispatch); spin_unlock(&hctx->lock); /* @@ -1590,13 +1590,17 @@ void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, * Should only be used carefully, when the caller knows we want to * bypass a potential IO scheduler on the target device. */ -void blk_mq_request_bypass_insert(struct request *rq, bool run_queue) +void blk_mq_request_bypass_insert(struct request *rq, bool at_head, + bool run_queue) { struct blk_mq_ctx *ctx = rq->mq_ctx; struct blk_mq_hw_ctx *hctx = blk_mq_map_queue(rq->q, ctx->cpu); spin_lock(&hctx->lock); - list_add_tail(&rq->queuelist, &hctx->dispatch); + if (at_head) + list_add(&rq->queuelist, &hctx->dispatch); + else + list_add_tail(&rq->queuelist, &hctx->dispatch); spin_unlock(&hctx->lock); if (run_queue) @@ -1776,7 +1780,7 @@ static blk_status_t __blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx, if (bypass_insert) return BLK_STS_RESOURCE; - blk_mq_request_bypass_insert(rq, run_queue); + blk_mq_request_bypass_insert(rq, false, run_queue); return BLK_STS_OK; } @@ -1792,7 +1796,7 @@ static void blk_mq_try_issue_directly(struct blk_mq_hw_ctx *hctx, ret = __blk_mq_try_issue_directly(hctx, rq, cookie, false); if (ret == BLK_STS_RESOURCE || ret == BLK_STS_DEV_RESOURCE) - blk_mq_request_bypass_insert(rq, true); + blk_mq_request_bypass_insert(rq, false, true); else if (ret != BLK_STS_OK) blk_mq_end_request(rq, ret); @@ -1827,7 +1831,7 @@ void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx, if (ret != BLK_STS_OK) { if (ret == BLK_STS_RESOURCE || ret == BLK_STS_DEV_RESOURCE) { - blk_mq_request_bypass_insert(rq, + blk_mq_request_bypass_insert(rq, false, list_empty(list)); break; } diff --git a/block/blk-mq.h b/block/blk-mq.h index a6094c27b827..debc646e1bed 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -64,7 +64,8 @@ int blk_mq_alloc_rqs(struct blk_mq_tag_set *set, struct blk_mq_tags *tags, */ void __blk_mq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, bool at_head); -void blk_mq_request_bypass_insert(struct request *rq, bool run_queue); +void blk_mq_request_bypass_insert(struct request *rq, bool at_head, + bool run_queue); void blk_mq_insert_requests(struct blk_mq_hw_ctx *hctx, struct blk_mq_ctx *ctx, struct list_head *list); -- 2.25.1

1 5

[PATCH] arm64/ascend: Fix register_persistent_clock definition
by Yang Yingliang 13 Oct '20

13 Oct '20

From: Fang Lijun <fanglijun3(a)huawei.com> ascend inclusion category: bugfix bugzilla: NA CVE: NA ------------------------------------------------- The register_persistent_clock will be called after kernel init, so it can not be defined as __init. Fixes: 76ab899d73d6 ("arm64/ascend: Implement the read_persistend_clock64 for aarch64") Signed-off-by: Fang Lijun <fanglijun3(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Reviewed-by: Ding Tianhong <dingtianhong(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/kernel/time.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kernel/time.c b/arch/arm64/kernel/time.c index c06c3feb6772..902d0f0b4f7b 100644 --- a/arch/arm64/kernel/time.c +++ b/arch/arm64/kernel/time.c @@ -77,7 +77,7 @@ void read_persistent_clock64(struct timespec64 *ts) __read_persistent_clock(ts); } -int __init register_persistent_clock(clock_access_fn read_persistent) +int register_persistent_clock(clock_access_fn read_persistent) { /* Only allow the clockaccess functions to be registered once */ if (__read_persistent_clock == dummy_clock_access) { -- 2.25.1

1 0

[PATCH 00/38] backport stable-4.19.150
by Yang Yingliang 13 Oct '20

13 Oct '20

Al Viro (4): epoll: do not insert into poll queues until all sanity checks are done epoll: replace ->visited/visited_list with generation count epoll: EPOLL_CTL_ADD: close the race in decision to take fast path ep_create_wakeup_source(): dentry name can change under you... Bartosz Golaszewski (1): gpio: mockup: fix resource leak in error path Bryan O'Donoghue (1): USB: gadget: f_ncm: Fix NDP16 datagram validation Chaitanya Kulkarni (1): nvme-core: get/put ctrl and transport module in nvme_dev_open/release() Chris Packham (2): spi: fsl-espi: Only process interrupts for expected events pinctrl: mvebu: Fix i2c sda definition for 98DX3236 Dinh Nguyen (1): clk: socfpga: stratix10: fix the divider for the emac_ptp_free_clk Felix Fietkau (1): mac80211: do not allow bigger VHT MPDUs than the hardware supports Greg Kroah-Hartman (1): Linux 4.19.150 Hans de Goede (1): mmc: sdhci: Workaround broken command queuing on Intel GLK based IRBIS models James Smart (1): nvme-fc: fail new connections to a deleted host or remote port Jean Delvare (1): drm/amdgpu: restore proper ref count in amdgpu_display_crtc_set_config Jeffrey Mitchell (1): nfs: Fix security label length not being reset Jiri Kosina (1): Input: i8042 - add nopnp quirk for Acer Aspire 5 A515 Laurent Dufour (2): mm: replace memmap_context by meminit_context mm: don't rely on system state to detect hot-plug operations Lucy Yan (1): net: dec: de2104x: Increase receive ring size for Tulip Marek Szyprowski (1): clk: samsung: exynos4: mark 'chipid' clock as CLK_IGNORE_UNUSED Martin Cerveny (1): drm/sun4i: mixer: Extend regmap max_register Nicolas VINCENT (1): i2c: cpm: Fix i2c_ram structure Olympia Giannou (1): rndis_host: increase sleep time in the query-response loop Sebastien Boeuf (1): net: virtio_vsock: Enhance connection semantics Stefano Garzarella (3): vsock/virtio: use RCU to avoid use-after-free on the_virtio_vsock vsock/virtio: stop workers during the .remove() vsock/virtio: add transport parameter to the virtio_transport_reset_no_sock() Steven Rostedt (VMware) (1): ftrace: Move RCU is watching check after recursion check Taiping Lai (1): gpio: sprd: Clear interrupt when setting the type as edge Thibaut Sautereau (1): random32: Restore __latent_entropy attribute on net_rand_state Vincent Huang (1): Input: trackpoint - enable Synaptics trackpoints Will McVicker (1): netfilter: ctnetlink: add a range check for l3/l4 protonum Xie He (3): drivers/net/wan/hdlc_fr: Add needed_headroom for PVC devices drivers/net/wan/lapbether: Make skb->protocol consistent with the header drivers/net/wan/hdlc: Set skb->protocol before transmitting Yu Kuai (1): iommu/exynos: add missing put_device() call in exynos_iommu_of_xlate() dillon min (1): gpio: tc35894: fix up tc35894 interrupt configuration Makefile | 2 +- arch/ia64/mm/init.c | 6 +- drivers/base/node.c | 84 ++++--- drivers/clk/samsung/clk-exynos4.c | 4 +- drivers/clk/socfpga/clk-s10.c | 2 +- drivers/gpio/gpio-mockup.c | 2 + drivers/gpio/gpio-sprd.c | 3 + drivers/gpio/gpio-tc3589x.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 2 +- drivers/gpu/drm/sun4i/sun8i_mixer.c | 2 +- drivers/i2c/busses/i2c-cpm.c | 3 + drivers/input/mouse/trackpoint.c | 2 + drivers/input/serio/i8042-x86ia64io.h | 7 + drivers/iommu/exynos-iommu.c | 8 +- drivers/mmc/host/sdhci-pci-core.c | 3 +- drivers/net/ethernet/dec/tulip/de2104x.c | 2 +- drivers/net/usb/rndis_host.c | 2 +- drivers/net/wan/hdlc_cisco.c | 1 + drivers/net/wan/hdlc_fr.c | 6 +- drivers/net/wan/hdlc_ppp.c | 1 + drivers/net/wan/lapbether.c | 4 +- drivers/nvme/host/core.c | 15 ++ drivers/nvme/host/fc.c | 6 +- drivers/pinctrl/mvebu/pinctrl-armada-xp.c | 2 +- drivers/spi/spi-fsl-espi.c | 5 +- drivers/usb/gadget/function/f_ncm.c | 30 +-- drivers/vhost/vsock.c | 94 +++---- fs/eventpoll.c | 71 +++--- fs/nfs/dir.c | 3 + include/linux/mm.h | 2 +- include/linux/mmzone.h | 11 +- include/linux/node.h | 11 +- include/linux/virtio_vsock.h | 3 +- kernel/trace/ftrace.c | 6 +- lib/random32.c | 2 +- mm/memory_hotplug.c | 5 +- mm/page_alloc.c | 11 +- net/mac80211/vht.c | 8 +- net/netfilter/nf_conntrack_netlink.c | 2 + net/vmw_vsock/virtio_transport.c | 265 +++++++++++++------- net/vmw_vsock/virtio_transport_common.c | 13 +- 41 files changed, 416 insertions(+), 297 deletions(-) -- 2.25.1

1 38

openEuler 20.09 kernel release
by Xie XiuQi 10 Oct '20

10 Oct '20

很高兴 openEuler 20.09 kernel 如期而至，该版本包含众多特性，欢迎通过 openEuler 版本来体验。 openEuler-20.09 kernel tree 位于: git@gitee.com:openeuler/kernel.git openEuler-20.09 openEuler kernel 项目位于: https://gitee.com/openeuler/kernel openEuler 20.09 kernel 依然基于社区的 Linux 4.19 LTS 内核，在 openEuler 1.0 LTS 基础上往前演进(跟踪到 4.19.140)，在这大概6个月的时间内，合入并使能了多个特性, 下面从架构支持，驱动支持，以及内核特性支持等几个方面来进行介绍。 1 架构支持 1.1 支持海光CPU-Hygon Dhyana 海光的Peng Wu向openEuler贡献了Hygon CPU的支持补丁集，通过这个补丁集，新增了 Hygon CPU 的ACPI, cpufreq，perf，KVM，NTB以及Hygon CPU的识别等支持，通过这个补丁集，openEuler完整支持海光x86 CPU。 1.2 支持昇腾芯片包括CPU和AI CPU共享虚拟地址的SVM（shared virtual memory）特性，以及芯片的硬件支持。 1.3 支持鲲鹏PC平台鲲鹏PC CPU基于鲲鹏920,8核。20.09 版本内核加入了对PC的休眠支持（架构特性），以及I2C控制器等支持。 1.4 支持 ARM v8.x 特性 - ARMv8.0 CRC32 instructions - ARMv8.0-SB Control of speculation - Armv8.2 Cache clean to Point of Deep Persistence - ARMv8.2-TTCNP, Translation table Common not private translations - ARMv8.4-TLBI, TLB maintenance and TLB range instructions - Armv8.5-RNG, Random number generator - Armv8.5-BTI, Branch target identification - Armv8.5-E0PD, Preventing EL0 access to halves of address map - ARMv8.6 Data gathering hint - ARMv8.6 Matrix multiply extension 1.5 支持 risc-v 支持 risc-v kvm. (openEuler 20.09 通过 kernel 5.5 来支持 risc-v) kernel tree 位于： git@gitee.com:openeuler/kernel.git kernel-5.5 2. 新介质以及新驱动支持 2.1 ARM64支持 persistence memory ARMv8.2引入新的指令支持 persistent memory，鲲鹏920处理器完整支持该特性，通过打开内核的 ACPI NFIT以及libnvdimm的配置文件，就能在鲲鹏上完美支持NVDIMM介质内存。 2.2 1822 网卡驱动增强 - 1822网卡支持x86架构 - 1822网卡支持128队列 2.3 3408/3416 RAID卡规避 3408iMR/3416iMRraid卡在ARM64平台上（ARM64生态问题）使用存在问题，开启SMMU后，会存在不兼容，通过SMMU 设备bypass特性，规避不兼容问题。 2.4 Huawei iBMA 驱动 iBMA 驱动是一组 BMC 与带内系统通信的驱动，系统宕机时，能辅助保存宕机时的运行状态信息，帮助定位故障原因。 2.5 hns3: update hns3 version to 1.9.38.8 2.6 TIPC: enable TIPC module by default 3. 存储/IO特性支持 3.1 bcache稳定性提升 bcache Maintainer Coly回合了大量bcache的bugfix以及特性，大幅提升bcache稳定性和可用性 4. 性能优化特性 4.1 sched Task Steal 回合社区的Task Steal特性，通过提升CPU利用率，替身MySQL数据库mix场景的性能10%+； 4.2 REFCOUNT_FULL以及CMPXCHG_LOOP()优化通过atomic_fetch_*操作来降低cmpxchg()带来的性能开销，从而提升refcount机制性能，带来文件读写等场景的性能提升，在ARM64服务器上尤其明显，空文件读写的benchmark 20+提升。 4.3 Linux's vmalloc Seeing "Large Performance Benefits" 优化 vmalloc 性能，内核针对vmalloc区的查找做了优化，从O(N)到~O(log(N) 。 See: https://www.phoronix.com/scan.php?page=news_item&px=Linux-5.2-vmalloc-Perfo… 4.4 MPAM特性增强支持通过ACPI获取系统MPAM资源； 4.5 percpu refcount for pagecache pagecache的生命周期管理是通过refcount来管理的，在多核并发压力下，读取文件cache 的atomic操作成为瓶颈，因此引入percpu refcount mechanism来提升性能，针对nginx http测试，性能有1倍+的提升 4.6 ARM64 clear_page()性能优化通过‘stnp’代替'dc zva'指令，提升clear_page()的性能，在鲲鹏920上，针对hugetlb 测试，有53%的性能提升。 4.7 调度，优化关键进程的抢占时延，提升响应速度对 SCHED_IDLE 策略的优化，之前选核的时候只选择 idle 的CPU，现在是如果那个CPU 上运行的是 SCHED_IDLE 的进程，也可以选择到，由于运行SCHED_IDLE进程的CPU并不是 idle，没有唤醒时延，能提升响应速度。 5. 安全特性 5.1 支持AppArmor AppArmor相比Selinux使用起来简单，但之前版本未开启AppArmor支持。20.09版本支持 AppArmor，但默认的安全策略依然是Selinux，需要在启动参数传入security=apparmor使能 5.2 IMA摘要列表增强相比内核社区原生 IMA 机制，IMA 摘要列表扩展从安全性、性能、易用性三个方面进行了改良，助力完整性保护机制在生产场景下落地，并具有三大优势：极致安全原生 IMA 机制要求在现网环境下预先生成并标记文件扩展属性，访问文件时将文件扩展属性作为参考值，信任链不完整。摘要列表扩展将文件参考摘要值保存在内核空间中，构建阶段通过摘要列表的形式携带在发布的 rpm 包中，安装 rpm 包的同时导入摘要列表并执行验签，确保了参考值来自于软件发行商，实现了完整的信任链。惊艳的性能表现 IMA 度量场景下，摘要列表扩展在确保安全性的前提下，减少了不必要的 PCR 扩展操作，开启度量时性能损失小于 5%，相比原生 IMA 度量性能提升高达 50%。IMA 评估场景下，摘要列表扩展将签名验证统一移动到启动阶段进行，避免每次访问文件时都执行验签，相比原生 IMA 评估场景提升运行阶段文件访问的性能约 20%。快速部署，平滑升级原生 IMA 机制在初次部署或每次更新软件包时，需要切换到 ﬁx 模式手动标记文件扩展属性后再重启进入 enforce 模式，才能正常访问安装的程序。摘要列表扩展可实现安装完成后开箱即用，且允许直接在enforce 模式下安装或升级 rpm 包，无需重启和手动标记即可使用，实现了用户感知最小化，适合现网环境下快速批量部署。 6. 功耗特性 6.1 支持 CPU idle TEO governor TEO 是新引入的CPU调频算法，更适合 tickless 系统，可以为CPU选择更合适的节能状态。 7 调测特性 7.1 support livepatch without ftrace 8. 虚拟化特性 8.1 支持双层调度双层调度就是让Hypervisor的调度器感知到VM的VCPU上跑什么应用。让VM的调度器感知到 Hypervisor 层物理CPU压力。两层调度感知，整机达到最好的业务性能。 8.2 支持 PMU nmi watchdog 8.3 支持 SmartPolling （guest idle poll) ARM 平台上 cpu idle 支持 smart polling，用来提升性能。 8.4 支持 vmtop, 用于虚拟机性能指标监测 8.5 支持ARM架构PV qspinlock功能，提升vcpu的自旋锁等锁性能 --- openEuler Kernel SIG

1 0

[PATCH 01/43] mpls: fix warning with multi-label encap
by Yang Yingliang 10 Oct '20

10 Oct '20

From: George Wilkie <gwilkie(a)vyatta.att-mail.com> [ Upstream commit 2f3f7d1fa0d1039b24a55d127ed190f196fc3e79 ] If you configure a route with multiple labels, e.g. ip route add 10.10.3.0/24 encap mpls 16/100 via 10.10.2.2 dev ens4 A warning is logged: kernel: [ 130.561819] netlink: 'ip': attribute type 1 has an invalid length. This happens because mpls_iptunnel_policy has set the type of MPLS_IPTUNNEL_DST to fixed size NLA_U32. Change it to a minimum size. nla_get_labels() does the remaining validation. Fixes: e3e4712ec096 ("mpls: ip tunnel support") Signed-off-by: George Wilkie <gwilkie(a)vyatta.att-mail.com> Reviewed-by: David Ahern <dsahern(a)gmail.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Signed-off-by: Aichun Li <liaichun(a)huawei.com> Reviewed-by: wangxiaopeng <wangxiaopeng7(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- net/mpls/mpls_iptunnel.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/mpls/mpls_iptunnel.c b/net/mpls/mpls_iptunnel.c index 8141eb10752f..b556acebdc86 100644 --- a/net/mpls/mpls_iptunnel.c +++ b/net/mpls/mpls_iptunnel.c @@ -28,7 +28,7 @@ #include "internal.h" static const struct nla_policy mpls_iptunnel_policy[MPLS_IPTUNNEL_MAX + 1] = { - [MPLS_IPTUNNEL_DST] = { .type = NLA_U32 }, + [MPLS_IPTUNNEL_DST] = { .len = sizeof(u32) }, [MPLS_IPTUNNEL_TTL] = { .type = NLA_U8 }, }; -- 2.25.1

1 42

[PATCH] hdlc_ppp: add range checks in ppp_cp_parse_cr()
by Yang Yingliang 10 Oct '20

10 Oct '20

From: Dan Carpenter <dan.carpenter(a)oracle.com> stable inclusion from linux-4.19.148 commit 45676c0bc28eff8f46455b28e2db80a77676488b CVE: CVE-2020-25643 -------------------------------- [ Upstream commit 66d42ed8b25b64eb63111a2b8582c5afc8bf1105 ] There are a couple bugs here: 1) If opt[1] is zero then this results in a forever loop. If the value is less than 2 then it is invalid. 2) It assumes that "len" is more than sizeof(valid_accm) or 6 which can result in memory corruption. In the case of LCP_OPTION_ACCM, then we should check "opt[1]" instead of "len" because, if "opt[1]" is less than sizeof(valid_accm) then "nak_len" gets out of sync and it can lead to memory corruption in the next iterations through the loop. In case of LCP_OPTION_MAGIC, the only valid value for opt[1] is 6, but the code is trying to log invalid data so we should only discard the data when "len" is less than 6 because that leads to a read overflow. Reported-by: ChenNan Of Chaitin Security Research Lab <whutchennan(a)gmail.com> Fixes: e022c2f07ae5 ("WAN: new synchronous PPP implementation for generic HDLC.") Signed-off-by: Dan Carpenter <dan.carpenter(a)oracle.com> Reviewed-by: Eric Dumazet <edumazet(a)google.com> Reviewed-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/net/wan/hdlc_ppp.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/drivers/net/wan/hdlc_ppp.c b/drivers/net/wan/hdlc_ppp.c index ab8b3cbbb205..85844f26547d 100644 --- a/drivers/net/wan/hdlc_ppp.c +++ b/drivers/net/wan/hdlc_ppp.c @@ -386,11 +386,8 @@ static void ppp_cp_parse_cr(struct net_device *dev, u16 pid, u8 id, } for (opt = data; len; len -= opt[1], opt += opt[1]) { - if (len < 2 || len < opt[1]) { - dev->stats.rx_errors++; - kfree(out); - return; /* bad packet, drop silently */ - } + if (len < 2 || opt[1] < 2 || len < opt[1]) + goto err_out; if (pid == PID_LCP) switch (opt[0]) { @@ -398,6 +395,8 @@ static void ppp_cp_parse_cr(struct net_device *dev, u16 pid, u8 id, continue; /* MRU always OK and > 1500 bytes? */ case LCP_OPTION_ACCM: /* async control character map */ + if (opt[1] < sizeof(valid_accm)) + goto err_out; if (!memcmp(opt, valid_accm, sizeof(valid_accm))) continue; @@ -409,6 +408,8 @@ static void ppp_cp_parse_cr(struct net_device *dev, u16 pid, u8 id, } break; case LCP_OPTION_MAGIC: + if (len < 6) + goto err_out; if (opt[1] != 6 || (!opt[2] && !opt[3] && !opt[4] && !opt[5])) break; /* reject invalid magic number */ @@ -427,6 +428,11 @@ static void ppp_cp_parse_cr(struct net_device *dev, u16 pid, u8 id, ppp_cp_event(dev, pid, RCR_GOOD, CP_CONF_ACK, id, req_len, data); kfree(out); + return; + +err_out: + dev->stats.rx_errors++; + kfree(out); } static int ppp_rx(struct sk_buff *skb) -- 2.25.1

1 0

[PATCH 000/237] backport stable-4.19.149
by Yang Yingliang 10 Oct '20

10 Oct '20

Adrian Hunter (1): perf kcore_copy: Fix module map when there are no modules loaded Al Viro (1): fix dget_parent() fastpath race Alain Michaud (1): Bluetooth: guard against controllers sending zero'd events Alex Deucher (2): drm/amdgpu/powerplay: fix AVFS handling with custom powerplay table drm/amdgpu/powerplay/smu7: fix AVFS handling with custom powerplay table Alex Williamson (1): vfio/pci: Clear error and request eventfd ctx after releasing Alexander Duyck (1): e1000: Do not perform reset in reset_task if we are already down Alexandre Belloni (2): rtc: sa1100: fix possible race condition rtc: ds1374: fix possible race condition Amelie Delaunay (2): dmaengine: stm32-mdma: use vchan_terminate_vdesc() in .terminate_all dmaengine: stm32-dma: use vchan_terminate_vdesc() in .terminate_all Andreas Steinmetz (1): ALSA: usb-audio: Fix case when USB MIDI interface has more than one extra endpoint descriptor Andy Lutomirski (1): selftests/x86/syscall_nt: Clear weird flags after each test Anshuman Khandual (1): arm64/cpufeature: Drop TraceFilt feature exposure from ID_DFR0 register Anthony Iliopoulos (1): nvme: explicitly update mpath disk capacity on revalidation Aric Cyr (1): drm/amd/display: dal_ddc_i2c_payloads_create can fail causing panic Ayush Sawal (1): crypto: chelsio - This fixes the kernel panic which occurs during a libkcapi test Balsundar P (1): scsi: aacraid: fix illegal IO beyond last LBA Bart Van Assche (3): scsi: ufs: Make ufshcd_add_command_trace() easier to read scsi: ufs: Fix a race condition in the tracing code RDMA/rxe: Fix configuration of atomic queue pair attributes Bob Peterson (1): gfs2: clean up iopen glock mess in gfs2_create_inode Boris Brezillon (1): mtd: parser: cmdline: Support MTD names containing one or more colons Bradley Bolen (1): mmc: core: Fix size overflow for mmc partitions Brian Foster (1): xfs: fix attr leaf header freemap.size underflow Chris Wilson (1): dma-fence: Serialise signal enabling (dma_fence_enable_sw_signaling) Christian Borntraeger (1): s390/zcrypt: Fix ZCRYPT_PERDEV_REQCNT ioctl Christophe JAILLET (4): RDMA/iw_cgxb4: Fix an error handling path in 'c4iw_connect()' perf cpumap: Fix snprintf overflow check SUNRPC: Fix a potential buffer overflow in 'svc_print_xprts()' scsi: aacraid: Fix error handling paths in aac_probe_one() Chuck Lever (1): svcrdma: Fix leak of transport addresses Colin Ian King (2): media: tda10071: fix unsigned sign extension overflow USB: EHCI: ehci-mv: fix less than zero comparison of an unsigned int Cong Wang (1): atm: fix a memory leak of vcc->user_back Dan Carpenter (1): media: staging/imx: Missing assignment in imx_media_capture_device_register() Daniel Borkmann (1): bpf: Fix clobbering of r2 in bpf_gen_ld_abs Darrick J. Wong (3): xfs: fix log reservation overflows when allocating large rt extents xfs: don't ever return a stale pointer from __xfs_dir3_free_read xfs: mark dir corrupt when lookup-by-hash fails Dave Hansen (1): x86/pkeys: Add check for pkey "overflow" David Sterba (1): btrfs: don't force read-only after error in drop snapshot Dennis Li (1): drm/amdkfd: fix a memory leak issue Dinghao Liu (8): drm/nouveau/debugfs: fix runtime pm imbalance on error drm/nouveau: fix runtime pm imbalance on error drm/nouveau/dispnv50: fix runtime pm imbalance on error ASoC: img-i2s-out: Fix runtime PM imbalance on error wlcore: fix runtime pm imbalance in wl1271_tx_work wlcore: fix runtime pm imbalance in wlcore_regdomain_config mtd: rawnand: omap_elm: Fix runtime PM imbalance on error PCI: tegra: Fix runtime PM imbalance on error Dinh Nguyen (1): clk: stratix10: use do_div() for 64-bit calculation Divya Indi (1): tracing: Adding NULL checks for trace_array descriptor pointer Dmitry Baryshkov (1): regmap: fix page selection for noinc reads Dmitry Bogdanov (1): net: qed: RDMA personality shouldn't fail VF load Dmitry Osipenko (2): PM / devfreq: tegra30: Fix integer overflow on CPU's freq max out dmaengine: tegra-apb: Prevent race conditions on channel's freeing Don Brace (1): scsi: hpsa: correct race condition in offload enabled Doug Smythies (1): tools/power/x86/intel_pstate_tracer: changes for python 3 compatibility Douglas Anderson (1): bdev: Reduce time holding bd_mutex in sync in blkdev_close() Eric Dumazet (2): net: silence data-races on sk_backlog.tail mac802154: tx: fix use-after-free Felix Fietkau (1): mt76: clear skb pointers from rx aggregation reorder buffer during cleanup Fuqian Huang (1): m68k: q40: Fix info-leak in rtc_ioctl Gabriel Ravier (1): tools: gpio-hammer: Avoid potential overflow in main Gao Xiang (1): mm, THP, swap: fix allocating cluster for swapfile by mistake Greg Kroah-Hartman (1): Linux 4.19.149 Gustavo Romero (1): KVM: PPC: Book3S HV: Treat TM-related invalid form instructions on P9 like the valid ones Hans de Goede (2): ASoC: Intel: bytcr_rt5640: Add quirk for MPMAN Converter9 2-in-1 i2c: core: Call i2c_acpi_install_space_handler() before i2c_acpi_register_devices() Hillf Danton (1): Bluetooth: prefetch channel before killing sock Hou Tao (2): mtd: cfi_cmdset_0002: don't free cfi->cfiq in error path of cfi_amdstd_setup() ubi: fastmap: Free unused fastmap anchor peb during detach Howard Chung (1): Bluetooth: L2CAP: handle l2cap config request during open state Hui Wang (1): ALSA: hda/realtek - Couldn't detect Mic if booting with headset plugged Ian Rogers (5): perf parse-events: Fix 3 use after frees found with clang ASAN perf mem2node: Avoid double free related to realloc perf evsel: Fix 2 memory leaks perf trace: Fix the selection for architectures to generate the errno name tables perf metricgroup: Free metric_events on error Ilya Leoshkevich (1): s390/init: add missing __init annotations Israel Rukshin (2): nvme: Fix controller creation races with teardown flow nvmet-rdma: fix double free of rdma queue Ivan Lazeev (1): tpm_crb: fix fTPM on AMD Zen+ CPUs Ivan Safonov (1): staging:r8188eu: avoid skb_clone for amsdu to msdu conversion Jaewon Kim (1): mm/mmap.c: initialize align_offset explicitly for vm_unmapped_area James Morse (1): firmware: arm_sdei: Use cpus_read_lock() to avoid races with cpuhp James Smart (3): scsi: lpfc: Fix kernel crash at lpfc_nvme_info_show during remote port bounce scsi: lpfc: Fix RQ buffer leakage when no IOCBs available scsi: lpfc: Fix coverity errors in fmdi attribute handling Jan Höppner (1): s390/dasd: Fix zero write for FBA devices Jason Gunthorpe (1): RDMA/cm: Remove a race freeing timewait_info Javed Hasan (2): scsi: libfc: Handling of extra kref scsi: libfc: Skip additional kref updating work event Jeff Layton (2): ceph: ensure we have a new cap before continuing in fill_inode ceph: fix potential race in ceph_check_caps Jia He (1): mm: fix double page fault on arm64 if PTE_AF is cleared Jin Yao (1): perf parse-events: Use strcmp() to compare the PMU name Jing Xiangfeng (1): atm: eni: fix the missed pci_disable_device() for eni_init_one() Jiri Olsa (1): perf stat: Fix duration_time value for higher intervals Jiri Slaby (3): ata: define AC_ERR_OK ata: make qc_prep return ata_completion_errors ata: sata_mv, avoid trigerrable BUG_ON Joakim Tjernlund (1): ALSA: usb-audio: Add delay quirk for H570e USB headsets Joe Perches (1): kernel/sys.c: avoid copying possible padding bytes in copy_to_user John Clements (1): drm/amdgpu: increase atombios cmd timeout John Garry (2): bus: hisi_lpc: Fixup IO ports addresses to avoid use-after-free in host removal perf jevents: Fix leak of mapfile memory John Meneghini (1): nvme-multipath: do not reset on unknown status Jonathan Bakker (3): power: supply: max17040: Correct voltage reading phy: samsung: s5pv210-usb2: Add delay after reset tty: serial: samsung: Correct clock selection logic Jordan Crouse (1): drm/msm/a5xx: Always set an OPP supported hardware value Josef Bacik (1): tracing: Set kernel_stack's caller size properly Josh Poimboeuf (1): objtool: Fix noreturn detection for ignored functions Kai-Heng Feng (1): ALSA: hda/realtek: Enable front panel headset LED on Lenovo ThinkStation P520 Kangjie Lu (1): gma/gma500: fix a memory disclosure bug due to uninitialized bytes Kevin Kou (1): sctp: move trace_sctp_probe_path into sctp_outq_sack Kirill A. Shutemov (1): mm: avoid data corruption on CoW fault into PFN-mapped VMA Krzysztof Kozlowski (1): dt-bindings: sound: wm8994: Correct required supplies based on actual implementaion Kusanagi Kouichi (1): debugfs: Fix !DEBUG_FS debugfs_create_automount Lee Jones (1): mfd: mfd-core: Protect against NULL call-back function pointer Linus Lüssing (4): batman-adv: bla: fix type misuse for backbone_gw hash indexing batman-adv: mcast/TT: fix wrongly dropped or rerouted packets batman-adv: mcast: fix duplicate mcast packets in BLA backbone from mesh batman-adv: mcast: fix duplicate mcast packets from BLA backbone to mesh Liu Jian (1): ieee802154: fix one possible memleak in ca8210_dev_com_init Liu Song (1): ubifs: Fix out-of-bounds memory access caused by abnormal value of node_len Madhuparna Bhowmik (2): drivers: char: tlclk.c: Avoid data race between init and interrupt handler rapidio: avoid data race between file operation callbacks and mport_cdev_add(). Manish Mandlik (1): Bluetooth: Fix refcount use-after-free issue Marc Zyngier (1): KVM: arm64: Assume write fault on S1PTW permission fault on instruction fetch Marco Elver (1): seqlock: Require WRITE_ONCE surrounding raw_seqcount_barrier Marek Szyprowski (1): drm/vc4/vc4_hdmi: fill ASoC card owner Martin Cerveny (1): drm/sun4i: sun8i-csc: Secondary CSC register correction Masami Hiramatsu (1): kprobes: Fix to check probe enabled before disarm_kprobe_ftrace() Matthias Fend (1): dmaengine: zynqmp_dma: fix burst length configuration Maxim Mikityanskiy (1): Bluetooth: btrtl: Use kvmalloc for FW allocations Maximilian Luz (1): mwifiex: Increase AES key storage size to 256 bits Mert Dirik (1): ar5523: Add USB ID of SMCWUSBT-G2 wireless adapter Miaohe Lin (1): KVM: arm/arm64: vgic: Fix potential double free dist->spis in __kvm_vgic_destroy() Miaoqing Pan (2): ath10k: fix array out-of-bounds access ath10k: fix memory leak for tpc_stats_final Mikel Rychliski (1): PCI: Use ioremap(), not phys_to_virt() for platform ROM Miklos Szeredi (1): fuse: don't check refcount after stealing page Mikulas Patocka (1): arch/x86/lib/usercopy_64.c: fix __copy_user_flushcache() cache writeback Mohan Kumar (1): ALSA: hda: Clear RIRB status before reading WP Nathan Chancellor (2): tracing: Use address-of operator on section symbols mm/kmemleak.c: use address-of operator on section symbols Nicholas Piggin (1): powerpc/traps: Make unrecoverable NMIs die instead of panic Nick Desaulniers (1): lib/string.c: implement stpcpy Nikhil Devshatwar (1): media: ti-vpe: cal: Restrict DMA to avoid memory corruption Niklas Söderlund (1): thermal: rcar_thermal: Handle probe error gracefully Nilesh Javali (1): scsi: qedi: Fix termination timeouts in session logout Oleh Kravchenko (1): leds: mlxreg: Fix possible buffer overflow Oliver O'Halloran (1): powerpc/eeh: Only dump stack once if an MMIO loop is detected Palmer Dabbelt (1): RISC-V: Take text_mutex in ftrace_init_nop() Pan Bian (3): scsi: fnic: fix use after free RDMA/qedr: Fix potential use after free RDMA/i40iw: Fix potential use after free Paolo Bonzini (1): KVM: x86: fix incorrect comparison in trace event Pavel Machek (1): drm/msm: fix leaks if initialization fails Pavel Shilovsky (1): CIFS: Properly process SMB3 lease breaks Peter Ujfalusi (1): serial: 8250_omap: Fix sleeping function called from invalid context during probe Pratik Rajesh Sampat (1): cpufreq: powernv: Fix frame-size-overflow in powernv_cpufreq_work_fn Qian Cai (4): skbuff: fix a data race in skb_queue_len() random: fix data races at timer_rand_state mm/vmscan.c: fix data races using kswapd_classzone_idx vfio/pci: fix memory leaks of eventfd ctx Qu Wenruo (1): btrfs: qgroup: fix data leak caused by race between writeback and truncate Rafael J. Wysocki (1): ACPI: EC: Reference count query handlers under lock Raviteja Narayanam (1): serial: uartps: Wait for tx_empty in console setup Rodrigo Siqueira (1): drm/amd/display: Stop if retimer is not available Russell King (1): ASoC: kirkwood: fix IRQ error handling Sagar Biradar (1): scsi: aacraid: Disabling TM path and only processing IOP reset Sagi Grimberg (1): nvme: fix possible deadlock when I/O is blocked Sakari Ailus (1): media: smiapp: Fix error handling at NVM reading Sascha Hauer (1): ubi: Fix producing anchor PEBs Satendra Singh Thakur (1): dmaengine: mediatek: hsdma_probe: fixed a memory leak when devm_request_irq fails Sean Christopherson (1): KVM: x86: Reset MMU context if guest toggles CR4.SMAP or CR4.PKE Shreyas Joshi (1): printk: handle blank console arguments passed in. Sonny Sasaka (1): Bluetooth: Handle Inquiry Cancel error after Inquiry Complete Stefan Berger (1): tpm: ibmvtpm: Wait for buffer to be set before proceeding Stephen Kitt (1): clk/ti/adpll: allocate room for terminating null Steve Grubb (1): audit: CONFIG_CHANGE don't log internal bookkeeping as an event Steve Rutherford (1): KVM: Remove CREATE_IRQCHIP/SET_PIT2 race Stuart Hayes (1): PCI: pciehp: Fix MSI interrupt race Sven Eckelmann (1): batman-adv: Add missing include for in_interrupt() Sven Schnelle (2): selftests/ftrace: fix glob selftest lockdep: fix order in trace_hardirqs_off_caller() Sylwester Nawrocki (2): ASoC: wm8994: Skip setting of the WM8994_MICBIAS register for WM1811 ASoC: wm8994: Ensure the device is resumed in wm89xx_mic_detect functions Takashi Iwai (3): ALSA: usb-audio: Don't create a mixer element with bogus volume range media: go7007: Fix URB type for interrupt handling ALSA: hda: Fix potential race in unsol event handler Tang Bin (1): USB: EHCI: ehci-mv: fix error handling in mv_ehci_probe() Thomas Gleixner (3): x86/ioapic: Unbreak check_timer() bpf: Remove recursion prevention from rcu free callback x86/speculation/mds: Mark mds_user_clear_cpu_buffers() __always_inline Thomas Richter (2): s390/cpum_sf: Use kzalloc and minor changes perf test: Fix test trace+probe_vfs_getname.sh on s390 Tianjia Zhang (1): clocksource/drivers/h8300_timer8: Fix wrong return value in h8300_8timer_init() Tom Lendacky (1): KVM: SVM: Add a dedicated INVD intercept routine Tom Rix (3): ieee802154/adf7242: check status of adf7242_read_reg ALSA: asihpi: fix iounmap in error handler tracing: fix double free Tonghao Zhang (2): net: openvswitch: use u64 for meter bucket net: openvswitch: use div_u64() for 64-by-32 divisions Trond Myklebust (2): nfsd: Don't add locks to closed or closing open stateids NFS: Fix races nfs_page_group_destroy() vs nfs_destroy_unlinked_subrequests() Tuong Lien (1): tipc: fix memory leak in service subscripting Tzung-Bi Shih (1): ASoC: max98090: remove msleep in PLL unlocked workaround Vasily Averin (5): neigh_stat_seq_next() should increase position index rt_cpu_seq_next should increase position index ipv6_route_seq_next should increase position index mm/swapfile.c: swap_next should increase position index selinux: sel_avc_get_stat_idx should increase position index Vignesh Raghavendra (2): serial: 8250_port: Don't service RX FIFO if throttled serial: 8250: 8250_omap: Terminate DMA before pushing data on RX timeout Vincent Whitchurch (1): ARM: 8948/1: Prevent OOB access in stacktrace Wei Li (1): MIPS: Add the missing 'CPU_1074K' into __get_cpu_type() Wei Yongjun (2): sparc64: vcc: Fix error return code in vcc_probe() scsi: cxlflash: Fix error return code in cxlflash_probe() Wen Gong (1): ath10k: use kzalloc to read for ath10k_sdio_hif_diag_read Wen Yang (1): drm/omap: fix possible object reference leak Will Deacon (1): arm64: cpufeature: Relax checks for AArch32 support at EL[0-2] Xianting Tian (1): mm/filemap.c: clear page error before actual read Xie XiuQi (1): perf util: Fix memory leak of prefix_if_not_in Yonghong Song (1): bpf: Fix a rcu warning for bpffs map pretty-print Yu Chen (1): usb: dwc3: Increase timeout for CmdAct cleared by device controller Zeng Tao (1): vfio/pci: fix racy on error and request eventfd ctx Zenghui Yu (1): KVM: arm64: vgic-its: Fix memory leak on the error path of vgic_add_lpi() Zhang Xiaoxu (1): cifs: Fix double add page to memcg when cifs_readpages Zhu Yanjun (1): RDMA/rxe: Set sys_image_guid to be aligned with HW IB devices Zhuang Yanying (1): KVM: fix overflow of zero page refcount with ksm running peter chang (1): scsi: pm80xx: Cleanup command when a reset times out zhengbin (1): media: mc-device.c: fix memleak in media_device_register_entity .../devicetree/bindings/sound/wm8994.txt | 18 ++- Documentation/driver-api/libata.rst | 2 +- Makefile | 2 +- arch/arm/include/asm/kvm_emulate.h | 11 +- arch/arm/kernel/stacktrace.c | 2 + arch/arm/kernel/traps.c | 6 +- arch/arm64/include/asm/kvm_emulate.h | 9 +- arch/arm64/kernel/cpufeature.c | 12 +- arch/arm64/kvm/hyp/switch.c | 2 +- arch/m68k/q40/config.c | 1 + arch/mips/include/asm/cpu-type.h | 1 + arch/powerpc/include/asm/kvm_asm.h | 3 + arch/powerpc/kernel/eeh.c | 2 +- arch/powerpc/kernel/traps.c | 6 +- arch/powerpc/kvm/book3s_hv_tm.c | 28 +++- arch/powerpc/kvm/book3s_hv_tm_builtin.c | 16 +- arch/riscv/include/asm/ftrace.h | 7 + arch/riscv/kernel/ftrace.c | 19 +++ arch/s390/kernel/perf_cpum_sf.c | 9 +- arch/s390/kernel/setup.c | 6 +- arch/x86/include/asm/nospec-branch.h | 4 +- arch/x86/include/asm/pkeys.h | 5 + arch/x86/kernel/apic/io_apic.c | 1 + arch/x86/kernel/fpu/xstate.c | 9 +- arch/x86/kvm/mmutrace.h | 2 +- arch/x86/kvm/svm.c | 8 +- arch/x86/kvm/x86.c | 13 +- arch/x86/lib/usercopy_64.c | 2 +- drivers/acpi/ec.c | 16 +- drivers/ata/acard-ahci.c | 6 +- drivers/ata/libahci.c | 6 +- drivers/ata/libata-core.c | 9 +- drivers/ata/libata-sff.c | 12 +- drivers/ata/pata_macio.c | 6 +- drivers/ata/pata_pxa.c | 8 +- drivers/ata/pdc_adma.c | 7 +- drivers/ata/sata_fsl.c | 4 +- drivers/ata/sata_inic162x.c | 4 +- drivers/ata/sata_mv.c | 34 ++-- drivers/ata/sata_nv.c | 18 ++- drivers/ata/sata_promise.c | 6 +- drivers/ata/sata_qstor.c | 8 +- drivers/ata/sata_rcar.c | 6 +- drivers/ata/sata_sil.c | 8 +- drivers/ata/sata_sil24.c | 6 +- drivers/ata/sata_sx4.c | 6 +- drivers/atm/eni.c | 2 +- drivers/base/regmap/regmap.c | 12 +- drivers/bluetooth/btrtl.c | 20 +-- drivers/bus/hisi_lpc.c | 27 +++- drivers/char/random.c | 12 +- drivers/char/tlclk.c | 17 +- drivers/char/tpm/tpm_crb.c | 123 +++++++++++---- drivers/char/tpm/tpm_ibmvtpm.c | 9 ++ drivers/char/tpm/tpm_ibmvtpm.h | 1 + drivers/clk/socfpga/clk-pll-s10.c | 4 +- drivers/clk/ti/adpll.c | 11 +- drivers/clocksource/h8300_timer8.c | 2 +- drivers/cpufreq/powernv-cpufreq.c | 13 +- drivers/crypto/chelsio/chcr_algo.c | 5 +- drivers/crypto/chelsio/chtls/chtls_io.c | 10 +- drivers/devfreq/tegra-devfreq.c | 4 +- drivers/dma-buf/dma-fence.c | 78 +++++----- drivers/dma/mediatek/mtk-hsdma.c | 4 +- drivers/dma/stm32-dma.c | 9 +- drivers/dma/stm32-mdma.c | 9 +- drivers/dma/tegra20-apb-dma.c | 3 +- drivers/dma/xilinx/zynqmp_dma.c | 24 +-- drivers/firmware/arm_sdei.c | 26 ++-- drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c | 31 ++-- drivers/gpu/drm/amd/amdgpu/atom.c | 4 +- .../drm/amd/amdkfd/kfd_device_queue_manager.c | 2 + drivers/gpu/drm/amd/display/dc/core/dc_link.c | 67 ++++---- .../gpu/drm/amd/display/dc/core/dc_link_ddc.c | 52 +++---- .../gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c | 7 + .../drm/amd/powerplay/hwmgr/vega10_hwmgr.c | 7 + drivers/gpu/drm/gma500/cdv_intel_display.c | 2 + drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 27 +++- drivers/gpu/drm/msm/msm_drv.c | 6 +- drivers/gpu/drm/nouveau/dispnv50/disp.c | 4 +- drivers/gpu/drm/nouveau/nouveau_debugfs.c | 5 +- drivers/gpu/drm/nouveau/nouveau_gem.c | 4 +- .../drm/nouveau/nvkm/subdev/bios/shadowpci.c | 17 +- .../gpu/drm/omapdrm/dss/omapdss-boot-init.c | 4 +- drivers/gpu/drm/radeon/radeon_bios.c | 30 ++-- drivers/gpu/drm/sun4i/sun8i_csc.h | 2 +- drivers/gpu/drm/vc4/vc4_hdmi.c | 1 + drivers/i2c/i2c-core-base.c | 2 +- drivers/infiniband/core/cm.c | 25 +-- drivers/infiniband/hw/cxgb4/cm.c | 4 +- drivers/infiniband/hw/i40iw/i40iw_cm.c | 2 +- drivers/infiniband/hw/qedr/qedr_iw_cm.c | 2 +- drivers/infiniband/sw/rxe/rxe.c | 2 + drivers/infiniband/sw/rxe/rxe_qp.c | 7 +- drivers/leds/leds-mlxreg.c | 4 +- drivers/media/dvb-frontends/tda10071.c | 9 +- drivers/media/i2c/smiapp/smiapp-core.c | 3 +- drivers/media/media-device.c | 65 ++++---- drivers/media/platform/ti-vpe/cal.c | 6 +- drivers/media/usb/go7007/go7007-usb.c | 4 +- drivers/mfd/mfd-core.c | 10 ++ drivers/mmc/core/mmc.c | 9 +- drivers/mtd/chips/cfi_cmdset_0002.c | 1 - drivers/mtd/cmdlinepart.c | 23 ++- drivers/mtd/nand/raw/omap_elm.c | 1 + drivers/mtd/ubi/fastmap-wl.c | 46 ++++-- drivers/mtd/ubi/fastmap.c | 14 +- drivers/mtd/ubi/ubi.h | 6 +- drivers/mtd/ubi/wl.c | 32 ++-- drivers/mtd/ubi/wl.h | 1 - drivers/net/ethernet/intel/e1000/e1000_main.c | 18 ++- drivers/net/ethernet/qlogic/qed/qed_sriov.c | 1 + drivers/net/ieee802154/adf7242.c | 4 +- drivers/net/ieee802154/ca8210.c | 1 + drivers/net/wireless/ath/ar5523/ar5523.c | 2 + drivers/net/wireless/ath/ath10k/debug.c | 3 +- drivers/net/wireless/ath/ath10k/sdio.c | 18 ++- drivers/net/wireless/ath/ath10k/wmi.c | 49 +++--- drivers/net/wireless/marvell/mwifiex/fw.h | 2 +- .../wireless/marvell/mwifiex/sta_cmdresp.c | 4 +- drivers/net/wireless/mediatek/mt76/agg-rx.c | 1 + drivers/net/wireless/ti/wlcore/main.c | 4 +- drivers/net/wireless/ti/wlcore/tx.c | 1 + drivers/nvme/host/core.c | 12 +- drivers/nvme/host/multipath.c | 21 ++- drivers/nvme/host/nvme.h | 19 ++- drivers/nvme/target/rdma.c | 30 ++-- drivers/pci/controller/pci-tegra.c | 3 +- drivers/pci/hotplug/pciehp_hpc.c | 26 +++- drivers/pci/rom.c | 17 -- drivers/phy/samsung/phy-s5pv210-usb2.c | 4 + drivers/power/supply/max17040_battery.c | 2 +- drivers/rapidio/devices/rio_mport_cdev.c | 14 +- drivers/rtc/rtc-ds1374.c | 15 +- drivers/rtc/rtc-sa1100.c | 18 ++- drivers/s390/block/dasd_fba.c | 9 +- drivers/s390/crypto/zcrypt_api.c | 3 +- drivers/scsi/aacraid/aachba.c | 8 +- drivers/scsi/aacraid/commsup.c | 2 +- drivers/scsi/aacraid/linit.c | 46 ++++-- drivers/scsi/cxlflash/main.c | 1 + drivers/scsi/fnic/fnic_scsi.c | 3 +- drivers/scsi/hpsa.c | 80 +++++++--- drivers/scsi/libfc/fc_rport.c | 13 +- drivers/scsi/lpfc/lpfc_attr.c | 35 +++-- drivers/scsi/lpfc/lpfc_ct.c | 137 +++++++++-------- drivers/scsi/lpfc/lpfc_hw.h | 36 ++--- drivers/scsi/lpfc/lpfc_sli.c | 4 + drivers/scsi/pm8001/pm8001_sas.c | 50 ++++-- drivers/scsi/qedi/qedi_iscsi.c | 3 + drivers/scsi/ufs/ufshcd.c | 14 +- drivers/staging/media/imx/imx-media-capture.c | 2 +- drivers/staging/rtl8188eu/core/rtw_recv.c | 19 +-- drivers/thermal/rcar_thermal.c | 6 +- drivers/tty/serial/8250/8250_omap.c | 8 +- drivers/tty/serial/8250/8250_port.c | 16 +- drivers/tty/serial/samsung.c | 8 +- drivers/tty/serial/xilinx_uartps.c | 8 + drivers/tty/vcc.c | 1 + drivers/usb/dwc3/gadget.c | 2 +- drivers/usb/host/ehci-mv.c | 8 +- drivers/vfio/pci/vfio_pci.c | 13 ++ fs/block_dev.c | 10 ++ fs/btrfs/extent-tree.c | 2 - fs/btrfs/inode.c | 23 ++- fs/ceph/caps.c | 14 +- fs/ceph/inode.c | 5 +- fs/cifs/cifsglob.h | 9 +- fs/cifs/file.c | 21 ++- fs/cifs/misc.c | 17 +- fs/cifs/smb1ops.c | 8 +- fs/cifs/smb2misc.c | 32 +--- fs/cifs/smb2ops.c | 44 ++++-- fs/cifs/smb2pdu.h | 2 +- fs/dcache.c | 4 +- fs/fuse/dev.c | 1 - fs/gfs2/inode.c | 13 +- fs/nfs/pagelist.c | 67 +++++--- fs/nfs/write.c | 10 +- fs/nfsd/nfs4state.c | 73 +++++---- fs/ubifs/io.c | 16 +- fs/xfs/libxfs/xfs_attr_leaf.c | 4 +- fs/xfs/libxfs/xfs_dir2_node.c | 1 + fs/xfs/libxfs/xfs_trans_resv.c | 96 +++++++++--- fs/xfs/scrub/dir.c | 3 + include/linux/debugfs.h | 5 +- include/linux/libata.h | 13 +- include/linux/mmc/card.h | 2 +- include/linux/nfs_page.h | 2 + include/linux/pci.h | 1 - include/linux/seqlock.h | 11 +- include/linux/skbuff.h | 14 +- include/net/sock.h | 4 +- include/trace/events/sctp.h | 9 -- kernel/audit_watch.c | 2 - kernel/bpf/hashtab.c | 8 - kernel/bpf/inode.c | 4 +- kernel/kprobes.c | 5 +- kernel/printk/printk.c | 3 + kernel/sys.c | 4 +- kernel/trace/trace.c | 5 +- kernel/trace/trace_entries.h | 2 +- kernel/trace/trace_events.c | 2 + kernel/trace/trace_events_hist.c | 1 - kernel/trace/trace_preemptirq.c | 4 +- lib/string.c | 24 +++ mm/filemap.c | 8 + mm/kmemleak.c | 2 +- mm/memory.c | 121 +++++++++++++-- mm/mmap.c | 2 + mm/swapfile.c | 4 +- mm/vmscan.c | 45 +++--- net/atm/lec.c | 6 + net/batman-adv/bridge_loop_avoidance.c | 145 ++++++++++++++---- net/batman-adv/bridge_loop_avoidance.h | 4 +- net/batman-adv/routing.c | 4 + net/batman-adv/soft-interface.c | 6 +- net/bluetooth/hci_event.c | 25 ++- net/bluetooth/l2cap_core.c | 29 ++-- net/bluetooth/l2cap_sock.c | 18 ++- net/core/filter.c | 4 +- net/core/neighbour.c | 1 + net/ipv4/route.c | 1 + net/ipv4/tcp.c | 2 +- net/ipv6/ip6_fib.c | 7 +- net/llc/af_llc.c | 2 +- net/mac802154/tx.c | 8 +- net/openvswitch/meter.c | 4 +- net/openvswitch/meter.h | 2 +- net/sctp/outqueue.c | 6 + net/sunrpc/svc_xprt.c | 19 ++- net/sunrpc/xprtrdma/svc_rdma_backchannel.c | 1 + net/tipc/topsrv.c | 4 +- net/unix/af_unix.c | 11 +- security/selinux/selinuxfs.c | 1 + sound/hda/hdac_bus.c | 4 + sound/pci/asihpi/hpioctl.c | 4 +- sound/pci/hda/hda_controller.c | 11 +- sound/pci/hda/patch_realtek.c | 13 +- sound/soc/codecs/max98090.c | 8 +- sound/soc/codecs/wm8994.c | 10 ++ sound/soc/codecs/wm_hubs.c | 3 + sound/soc/codecs/wm_hubs.h | 1 + sound/soc/img/img-i2s-out.c | 8 +- sound/soc/intel/boards/bytcr_rt5640.c | 10 ++ sound/soc/kirkwood/kirkwood-dma.c | 2 +- sound/usb/midi.c | 29 +++- sound/usb/mixer.c | 10 ++ sound/usb/quirks.c | 7 +- tools/gpio/gpio-hammer.c | 17 +- tools/objtool/check.c | 2 +- tools/perf/builtin-stat.c | 2 +- tools/perf/pmu-events/jevents.c | 15 +- .../perf/tests/shell/lib/probe_vfs_getname.sh | 2 +- tools/perf/trace/beauty/arch_errno_names.sh | 2 +- tools/perf/util/cpumap.c | 10 +- tools/perf/util/evsel.c | 3 + tools/perf/util/mem2node.c | 3 +- tools/perf/util/metricgroup.c | 3 + tools/perf/util/parse-events.c | 9 +- tools/perf/util/sort.c | 2 +- tools/perf/util/symbol-elf.c | 7 + .../intel_pstate_tracer.py | 22 +-- .../ftrace/test.d/ftrace/func-filter-glob.tc | 2 +- tools/testing/selftests/x86/syscall_nt.c | 1 + virt/kvm/arm/mmio.c | 2 +- virt/kvm/arm/mmu.c | 5 +- virt/kvm/arm/vgic/vgic-init.c | 1 + virt/kvm/arm/vgic/vgic-its.c | 11 +- virt/kvm/kvm_main.c | 1 + 270 files changed, 2313 insertions(+), 1196 deletions(-) -- 2.25.1

1 237

[PATCH 01/65] mm: memcg: make memory.oom.group tolerable to task migration
by Yang Yingliang 10 Oct '20

10 Oct '20

From: Roman Gushchin <guro(a)fb.com> mainline inclusion from mainline-5.7-rc1 commit 48fe267c503ec22014ba4e83d002b07caad034d0 category: bugfix bugzilla: 33351 CVE: NA ------------------------------------------------- If a task is getting moved out of the OOMing cgroup, it might result in unexpected OOM killings if memory.oom.group is used anywhere in the cgroup tree. Imagine the following example: A (oom.group = 1) / \ (OOM) B C Let's say B's memory.max is exceeded and it's OOMing. The OOM killer selects a task in B as a victim, but someone asynchronously moves the task into C. mem_cgroup_get_oom_group() will iterate over all ancestors of C up to the root cgroup. In theory it had to stop at the oom_domain level - the memory cgroup which is OOMing. But because B is not an ancestor of C, it's not happening. Instead it chooses A (because it's oom.group is set), and kills all tasks in A. This behavior is wrong because the OOM happened in B, so there is no reason to kill anything outside. Fix this by checking it the memory cgroup to which the task belongs is a descendant of the oom_domain. If not, memory.oom.group should be ignored, and the OOM killer should kill only the victim task. Reported-by: Dan Schatzberg <dschatzberg(a)fb.com> Signed-off-by: Roman Gushchin <guro(a)fb.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Acked-by: Michal Hocko <mhocko(a)suse.com> Acked-by: Johannes Weiner <hannes(a)cmpxchg.org> Link: http://lkml.kernel.org/r/20200316223510.3176148-1-guro@fb.com Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> (cherry picked from commit 48fe267c503ec22014ba4e83d002b07caad034d0) Signed-off-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Liu Shixin <liushixin2(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- mm/memcontrol.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 1342b9540476..8611f3301686 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1856,6 +1856,14 @@ struct mem_cgroup *mem_cgroup_get_oom_group(struct task_struct *victim, if (memcg == root_mem_cgroup) goto out; + /* + * If the victim task has been asynchronously moved to a different + * memory cgroup, we might end up killing tasks outside oom_domain. + * In this case it's better to ignore memory.group.oom. + */ + if (unlikely(!mem_cgroup_is_descendant(memcg, oom_domain))) + goto out; + /* * Traverse the memory cgroup hierarchy from the victim task's * cgroup up to the OOMing cgroup (or root) to find the -- 2.25.1

1 64

[PATCH] mm/swapfile: fix and annotate various data races
by Yang Yingliang 30 Sep '20

30 Sep '20

From: Qian Cai <cai(a)lca.pw> mainline inclusion from mainline-v5.9-rc1 commit a449bf58e45abf919b27ba62a9dcbae3ad36e725 category: bugfix bugzilla: NA CVE: NA ------------------------------------------------- swap_info_struct si.highest_bit, si.swap_map[offset] and si.flags could be accessed concurrently separately as noticed by KCSAN, === si.highest_bit === write to 0xffff8d5abccdc4d4 of 4 bytes by task 5353 on cpu 24: swap_range_alloc+0x81/0x130 swap_range_alloc at mm/swapfile.c:681 scan_swap_map_slots+0x371/0xb90 get_swap_pages+0x39d/0x5c0 get_swap_page+0xf2/0x524 add_to_swap+0xe4/0x1c0 shrink_page_list+0x1795/0x2870 shrink_inactive_list+0x316/0x880 shrink_lruvec+0x8dc/0x1380 shrink_node+0x317/0xd80 do_try_to_free_pages+0x1f7/0xa10 try_to_free_pages+0x26c/0x5e0 __alloc_pages_slowpath+0x458/0x1290 read to 0xffff8d5abccdc4d4 of 4 bytes by task 6672 on cpu 70: scan_swap_map_slots+0x4a6/0xb90 scan_swap_map_slots at mm/swapfile.c:892 get_swap_pages+0x39d/0x5c0 get_swap_page+0xf2/0x524 add_to_swap+0xe4/0x1c0 shrink_page_list+0x1795/0x2870 shrink_inactive_list+0x316/0x880 shrink_lruvec+0x8dc/0x1380 shrink_node+0x317/0xd80 do_try_to_free_pages+0x1f7/0xa10 try_to_free_pages+0x26c/0x5e0 __alloc_pages_slowpath+0x458/0x1290 Reported by Kernel Concurrency Sanitizer on: CPU: 70 PID: 6672 Comm: oom01 Tainted: G W L 5.5.0-next-20200205+ #3 Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385 Gen10, BIOS A40 07/10/2019 === si.swap_map[offset] === write to 0xffffbc370c29a64c of 1 bytes by task 6856 on cpu 86: __swap_entry_free_locked+0x8c/0x100 __swap_entry_free_locked at mm/swapfile.c:1209 (discriminator 4) __swap_entry_free.constprop.20+0x69/0xb0 free_swap_and_cache+0x53/0xa0 unmap_page_range+0x7f8/0x1d70 unmap_single_vma+0xcd/0x170 unmap_vmas+0x18b/0x220 exit_mmap+0xee/0x220 mmput+0x10e/0x270 do_exit+0x59b/0xf40 do_group_exit+0x8b/0x180 read to 0xffffbc370c29a64c of 1 bytes by task 6855 on cpu 20: _swap_info_get+0x81/0xa0 _swap_info_get at mm/swapfile.c:1140 free_swap_and_cache+0x40/0xa0 unmap_page_range+0x7f8/0x1d70 unmap_single_vma+0xcd/0x170 unmap_vmas+0x18b/0x220 exit_mmap+0xee/0x220 mmput+0x10e/0x270 do_exit+0x59b/0xf40 do_group_exit+0x8b/0x180 === si.flags === write to 0xffff956c8fc6c400 of 8 bytes by task 6087 on cpu 23: scan_swap_map_slots+0x6fe/0xb50 scan_swap_map_slots at mm/swapfile.c:887 get_swap_pages+0x39d/0x5c0 get_swap_page+0x377/0x524 add_to_swap+0xe4/0x1c0 shrink_page_list+0x1795/0x2870 shrink_inactive_list+0x316/0x880 shrink_lruvec+0x8dc/0x1380 shrink_node+0x317/0xd80 do_try_to_free_pages+0x1f7/0xa10 try_to_free_pages+0x26c/0x5e0 __alloc_pages_slowpath+0x458/0x1290 read to 0xffff956c8fc6c400 of 8 bytes by task 6207 on cpu 63: _swap_info_get+0x41/0xa0 __swap_info_get at mm/swapfile.c:1114 put_swap_page+0x84/0x490 __remove_mapping+0x384/0x5f0 shrink_page_list+0xff1/0x2870 shrink_inactive_list+0x316/0x880 shrink_lruvec+0x8dc/0x1380 shrink_node+0x317/0xd80 do_try_to_free_pages+0x1f7/0xa10 try_to_free_pages+0x26c/0x5e0 __alloc_pages_slowpath+0x458/0x1290 The writes are under si->lock but the reads are not. For si.highest_bit and si.swap_map[offset], data race could trigger logic bugs, so fix them by having WRITE_ONCE() for the writes and READ_ONCE() for the reads except those isolated reads where they compare against zero which a data race would cause no harm. Thus, annotate them as intentional data races using the data_race() macro. For si.flags, the readers are only interested in a single bit where a data race there would cause no issue there. [cai(a)lca.pw: add a missing annotation for si->flags in memory.c] Link: http://lkml.kernel.org/r/1581612647-5958-1-git-send-email-cai@lca.pw Signed-off-by: Qian Cai <cai(a)lca.pw> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Cc: Marco Elver <elver(a)google.com> Cc: Hugh Dickins <hughd(a)google.com> Link: http://lkml.kernel.org/r/1581095163-12198-1-git-send-email-cai@lca.pw Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> [yyl: remove data_race(), this macro is used when KCSAN is enabled] Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- mm/swapfile.c | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) diff --git a/mm/swapfile.c b/mm/swapfile.c index 60ccb3072318..41400ae4e7c1 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -639,7 +639,7 @@ static void swap_range_alloc(struct swap_info_struct *si, unsigned long offset, if (offset == si->lowest_bit) si->lowest_bit += nr_entries; if (end == si->highest_bit) - si->highest_bit -= nr_entries; + WRITE_ONCE(si->highest_bit, si->highest_bit - nr_entries); si->inuse_pages += nr_entries; if (si->inuse_pages == si->pages) { si->lowest_bit = si->max; @@ -671,7 +671,7 @@ static void swap_range_free(struct swap_info_struct *si, unsigned long offset, if (end > si->highest_bit) { bool was_full = !si->highest_bit; - si->highest_bit = end; + WRITE_ONCE(si->highest_bit, end); if (was_full && (si->flags & SWP_WRITEOK)) add_to_avail_list(si); } @@ -804,7 +804,7 @@ static int scan_swap_map_slots(struct swap_info_struct *si, else goto done; } - si->swap_map[offset] = usage; + WRITE_ONCE(si->swap_map[offset], usage); inc_cluster_info_page(si, si->cluster_info, offset); unlock_cluster(ci); @@ -850,12 +850,13 @@ static int scan_swap_map_slots(struct swap_info_struct *si, scan: spin_unlock(&si->lock); - while (++offset <= si->highest_bit) { + while (++offset <= READ_ONCE(si->highest_bit)) { if (!si->swap_map[offset]) { spin_lock(&si->lock); goto checks; } - if (vm_swap_full() && si->swap_map[offset] == SWAP_HAS_CACHE) { + if (vm_swap_full() && + READ_ONCE(si->swap_map[offset]) == SWAP_HAS_CACHE) { spin_lock(&si->lock); goto checks; } @@ -870,7 +871,8 @@ static int scan_swap_map_slots(struct swap_info_struct *si, spin_lock(&si->lock); goto checks; } - if (vm_swap_full() && si->swap_map[offset] == SWAP_HAS_CACHE) { + if (vm_swap_full() && + READ_ONCE(si->swap_map[offset]) == SWAP_HAS_CACHE) { spin_lock(&si->lock); goto checks; } @@ -1166,7 +1168,10 @@ static unsigned char __swap_entry_free_locked(struct swap_info_struct *p, } usage = count | has_cache; - p->swap_map[offset] = usage ? : SWAP_HAS_CACHE; + if (usage) + WRITE_ONCE(p->swap_map[offset], usage); + else + WRITE_ONCE(p->swap_map[offset], SWAP_HAS_CACHE); return usage; } @@ -3522,7 +3527,7 @@ static int __swap_duplicate(swp_entry_t entry, unsigned char usage) } else err = -ENOENT; /* unused swap entry */ - p->swap_map[offset] = count | has_cache; + WRITE_ONCE(p->swap_map[offset], count | has_cache); unlock_out: unlock_cluster_or_swap_info(p, ci); -- 2.25.1

1 0

[PATCH] iommu: fix a mistake for iommu_unregister_device_fault_handler
by Yang Yingliang 30 Sep '20

30 Sep '20

From: Bixuan Cui <cuibixuan(a)huawei.com> ascend inclusion category: bugfix bugzilla: NA CVE: NA ------------------------------------------------- Fix a mistake check for commit bd6c06e0917d ("iommu: introduce device fault report API") Fixes: bd6c06e0917d ("iommu: introduce device fault report API") Signed-off-by: Bixuan Cui <cuibixuan(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/iommu/iommu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 831c5065f7f8..bee6b8fcfe0e 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -982,7 +982,7 @@ int iommu_unregister_device_fault_handler(struct device *dev) mutex_lock(&param->lock); /* we cannot unregister handler if there are pending faults */ - if (list_empty(&param->fault_param->faults)) { + if (!list_empty(&param->fault_param->faults)) { ret = -EBUSY; goto unlock; } -- 2.25.1

1 0

[PATCH] arm64: Kconfig: change fix compile error if gcc don't support armv8.4-a
by Yang Yingliang 30 Sep '20

30 Sep '20

hulk inclusion category: feature CVE: NA ----------------------- After 4d0831e8a029 ("kconfig: unify cc-option and as-option") in mainline, we can use cc-option. But in linux-4.19, this patch is not merged, so we will get the follow error: CC scripts/mod/empty.o CC scripts/mod/devicetable-offsets.s Assembler messages: Error: unknown architecture `armv8.4-a' Error: unrecognized option -march=armv8.4-a Fix this by changing cc-option to as-option. Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 1ef7862d1b75..e379fca3c8ad 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -1307,7 +1307,7 @@ endmenu menu "ARMv8.4 architectural features" config AS_HAS_ARMV8_4 - def_bool $(cc-option,-Wa$(comma)-march=armv8.4-a) + def_bool $(as-option,-Wa$(comma)-march=armv8.4-a) config ARM64_TLB_RANGE bool "Enable support for tlbi range feature" -- 2.25.1

1 0

[PATCH 1/2] block: loop: set discard granularity and alignment for block device backed loop
by Yang Yingliang 30 Sep '20

30 Sep '20

From: Ming Lei <ming.lei(a)redhat.com> mainline inclusion from mainline-v5.9-rc3 commit bcb21c8cc9947286211327d663ace69f07d37a76 category: bugfix bugzilla: NA CVE: NA ------------------------------------------------- In case of block device backend, if the backend supports write zeros, the loop device will set queue flag of QUEUE_FLAG_DISCARD. However, limits.discard_granularity isn't setup, and this way is wrong, see the following description in Documentation/ABI/testing/sysfs-block: A discard_granularity of 0 means that the device does not support discard functionality. Especially 9b15d109a6b2 ("block: improve discard bio alignment in __blkdev_issue_discard()") starts to take q->limits.discard_granularity for computing max discard sectors. And zero discard granularity may cause kernel oops, or fail discard request even though the loop queue claims discard support via QUEUE_FLAG_DISCARD. Fix the issue by setup discard granularity and alignment. Fixes: c52abf563049 ("loop: Better discard support for block devices") Signed-off-by: Ming Lei <ming.lei(a)redhat.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Acked-by: Coly Li <colyli(a)suse.de> Cc: Hannes Reinecke <hare(a)suse.com> Cc: Xiao Ni <xni(a)redhat.com> Cc: Martin K. Petersen <martin.petersen(a)oracle.com> Cc: Evan Green <evgreen(a)chromium.org> Cc: Gwendal Grignou <gwendal(a)chromium.org> Cc: Chaitanya Kulkarni <chaitanya.kulkarni(a)wdc.com> Cc: Andrzej Pietrasiewicz <andrzej.p(a)collabora.com> Cc: Christoph Hellwig <hch(a)lst.de> Cc: <stable(a)vger.kernel.org> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Yufen Yu <yuyufen(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/block/loop.c | 33 ++++++++++++++++++--------------- 1 file changed, 18 insertions(+), 15 deletions(-) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index da68c42aed68..19042b42a8ba 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -864,6 +864,7 @@ static void loop_config_discard(struct loop_device *lo) struct file *file = lo->lo_backing_file; struct inode *inode = file->f_mapping->host; struct request_queue *q = lo->lo_queue; + u32 granularity, max_discard_sectors; /* * If the backing device is a block device, mirror its zeroing @@ -876,11 +877,10 @@ static void loop_config_discard(struct loop_device *lo) struct request_queue *backingq; backingq = bdev_get_queue(inode->i_bdev); - blk_queue_max_discard_sectors(q, - backingq->limits.max_write_zeroes_sectors); - blk_queue_max_write_zeroes_sectors(q, - backingq->limits.max_write_zeroes_sectors); + max_discard_sectors = backingq->limits.max_write_zeroes_sectors; + granularity = backingq->limits.discard_granularity ?: + queue_physical_block_size(backingq); /* * We use punch hole to reclaim the free space used by the @@ -889,23 +889,26 @@ static void loop_config_discard(struct loop_device *lo) * useful information. */ } else if (!file->f_op->fallocate || lo->lo_encrypt_key_size) { - q->limits.discard_granularity = 0; - q->limits.discard_alignment = 0; - blk_queue_max_discard_sectors(q, 0); - blk_queue_max_write_zeroes_sectors(q, 0); + max_discard_sectors = 0; + granularity = 0; } else { - q->limits.discard_granularity = inode->i_sb->s_blocksize; - q->limits.discard_alignment = 0; - - blk_queue_max_discard_sectors(q, UINT_MAX >> 9); - blk_queue_max_write_zeroes_sectors(q, UINT_MAX >> 9); + max_discard_sectors = UINT_MAX >> 9; + granularity = inode->i_sb->s_blocksize; } - if (q->limits.max_write_zeroes_sectors) + if (max_discard_sectors) { + q->limits.discard_granularity = granularity; + blk_queue_max_discard_sectors(q, max_discard_sectors); + blk_queue_max_write_zeroes_sectors(q, max_discard_sectors); blk_queue_flag_set(QUEUE_FLAG_DISCARD, q); - else + } else { + q->limits.discard_granularity = 0; + blk_queue_max_discard_sectors(q, 0); + blk_queue_max_write_zeroes_sectors(q, 0); blk_queue_flag_clear(QUEUE_FLAG_DISCARD, q); + } + q->limits.discard_alignment = 0; } static void loop_unprepare_queue(struct loop_device *lo) -- 2.25.1

1 1

[PATCH 1/2] Revert "KVM: arm64: Add kvm_extable for vaxorcism code"
by Yang Yingliang 30 Sep '20

30 Sep '20

This reverts commit 8cc25436c41592b2236b4d2911a3031131a62872. --- arch/arm64/include/asm/kvm_asm.h | 15 ---------- arch/arm64/kernel/vmlinux.lds.S | 8 ----- arch/arm64/kvm/hyp/entry.S | 16 ++++------ arch/arm64/kvm/hyp/hyp-entry.S | 51 +++++++++++++------------------- arch/arm64/kvm/hyp/switch.c | 31 ------------------- 5 files changed, 26 insertions(+), 95 deletions(-) diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h index 5df55a4dab42..400cb2af5ba6 100644 --- a/arch/arm64/include/asm/kvm_asm.h +++ b/arch/arm64/include/asm/kvm_asm.h @@ -146,21 +146,6 @@ extern u32 __kvm_get_mdcr_el2(void); kern_hyp_va \vcpu .endm -/* - * KVM extable for unexpected exceptions. - * In the same format _asm_extable, but output to a different section so that - * it can be mapped to EL2. The KVM version is not sorted. The caller must - * ensure: - * x18 has the hypervisor value to allow any Shadow-Call-Stack instrumented - * code to write to it, and that SPSR_EL2 and ELR_EL2 are restored by the fixup. - */ -.macro _kvm_extable, from, to - .pushsection __kvm_ex_table, "a" - .align 3 - .long (\from - .), (\to - .) - .popsection -.endm - #endif #endif /* __ARM_KVM_ASM_H__ */ diff --git a/arch/arm64/kernel/vmlinux.lds.S b/arch/arm64/kernel/vmlinux.lds.S index 69e7c8d4a00f..d6050c6e65bc 100644 --- a/arch/arm64/kernel/vmlinux.lds.S +++ b/arch/arm64/kernel/vmlinux.lds.S @@ -24,13 +24,6 @@ ENTRY(_text) jiffies = jiffies_64; - -#define HYPERVISOR_EXTABLE \ - . = ALIGN(SZ_8); \ - __start___kvm_ex_table = .; \ - *(__kvm_ex_table) \ - __stop___kvm_ex_table = .; - #define HYPERVISOR_TEXT \ /* \ * Align to 4 KB so that \ @@ -46,7 +39,6 @@ jiffies = jiffies_64; __hyp_idmap_text_end = .; \ __hyp_text_start = .; \ *(.hyp.text) \ - HYPERVISOR_EXTABLE \ __hyp_text_end = .; #define IDMAP_TEXT \ diff --git a/arch/arm64/kvm/hyp/entry.S b/arch/arm64/kvm/hyp/entry.S index 90e012fa3ca5..a0552b5177c5 100644 --- a/arch/arm64/kvm/hyp/entry.S +++ b/arch/arm64/kvm/hyp/entry.S @@ -164,22 +164,18 @@ alternative_endif // This is our single instruction exception window. A pending // SError is guaranteed to occur at the earliest when we unmask // it, and at the latest just after the ISB. + .global abort_guest_exit_start abort_guest_exit_start: isb + .global abort_guest_exit_end abort_guest_exit_end: - msr daifset, #4 // Mask aborts - ret - - _kvm_extable abort_guest_exit_start, 9997f - _kvm_extable abort_guest_exit_end, 9997f -9997: - msr daifset, #4 // Mask aborts - mov x0, #(1 << ARM_EXIT_WITH_SERROR_BIT) - // restore the EL1 exception context so that we can report some - // information. Merge the exception code with the SError pending bit. + // If the exception took place, restore the EL1 exception + // context so that we can report some information. + // Merge the exception code with the SError pending bit. + tbz x0, #ARM_EXIT_WITH_SERROR_BIT, 1f msr elr_el2, x2 msr esr_el2, x3 msr spsr_el2, x4 diff --git a/arch/arm64/kvm/hyp/hyp-entry.S b/arch/arm64/kvm/hyp/hyp-entry.S index c3e4ae84f3a4..71591a6ee63e 100644 --- a/arch/arm64/kvm/hyp/hyp-entry.S +++ b/arch/arm64/kvm/hyp/hyp-entry.S @@ -26,30 +26,6 @@ #include <asm/kvm_mmu.h> #include <asm/mmu.h> -.macro save_caller_saved_regs_vect - /* x0 and x1 were saved in the vector entry */ - stp x2, x3, [sp, #-16]! - stp x4, x5, [sp, #-16]! - stp x6, x7, [sp, #-16]! - stp x8, x9, [sp, #-16]! - stp x10, x11, [sp, #-16]! - stp x12, x13, [sp, #-16]! - stp x14, x15, [sp, #-16]! - stp x16, x17, [sp, #-16]! -.endm - -.macro restore_caller_saved_regs_vect - ldp x16, x17, [sp], #16 - ldp x14, x15, [sp], #16 - ldp x12, x13, [sp], #16 - ldp x10, x11, [sp], #16 - ldp x8, x9, [sp], #16 - ldp x6, x7, [sp], #16 - ldp x4, x5, [sp], #16 - ldp x2, x3, [sp], #16 - ldp x0, x1, [sp], #16 -.endm - .text .pushsection .hyp.text, "ax" @@ -209,14 +185,27 @@ el2_sync: el2_error: - save_caller_saved_regs_vect - stp x29, x30, [sp, #-16]! - - bl kvm_unexpected_el2_exception - - ldp x29, x30, [sp], #16 - restore_caller_saved_regs_vect + ldp x0, x1, [sp], #16 + /* + * Only two possibilities: + * 1) Either we come from the exit path, having just unmasked + * PSTATE.A: change the return code to an EL2 fault, and + * carry on, as we're already in a sane state to handle it. + * 2) Or we come from anywhere else, and that's a bug: we panic. + * + * For (1), x0 contains the original return code and x1 doesn't + * contain anything meaningful at that stage. We can reuse them + * as temp registers. + * For (2), who cares? + */ + mrs x0, elr_el2 + adr x1, abort_guest_exit_start + cmp x0, x1 + adr x1, abort_guest_exit_end + ccmp x0, x1, #4, ne + b.ne __hyp_panic + mov x0, #(1 << ARM_EXIT_WITH_SERROR_BIT) eret sb diff --git a/arch/arm64/kvm/hyp/switch.c b/arch/arm64/kvm/hyp/switch.c index e9ea7cf3e98f..acd2d84b190c 100644 --- a/arch/arm64/kvm/hyp/switch.c +++ b/arch/arm64/kvm/hyp/switch.c @@ -25,7 +25,6 @@ #include <asm/barrier.h> #include <asm/cpufeature.h> -#include <asm/extable.h> #include <asm/kprobes.h> #include <asm/kvm_asm.h> #include <asm/kvm_emulate.h> @@ -37,9 +36,6 @@ #include <asm/processor.h> #include <asm/thread_info.h> -extern struct exception_table_entry __start___kvm_ex_table; -extern struct exception_table_entry __stop___kvm_ex_table; - /* Check whether the FP regs were dirtied while in the host-side run loop: */ static bool __hyp_text update_fp_enabled(struct kvm_vcpu *vcpu) { @@ -730,30 +726,3 @@ void __hyp_text __noreturn hyp_panic(struct kvm_cpu_context *host_ctxt) unreachable(); } - -asmlinkage void __hyp_text kvm_unexpected_el2_exception(void) -{ - unsigned long addr, fixup; - struct kvm_cpu_context *host_ctxt; - struct exception_table_entry *entry, *end; - unsigned long elr_el2 = read_sysreg(elr_el2); - - entry = hyp_symbol_addr(__start___kvm_ex_table); - end = hyp_symbol_addr(__stop___kvm_ex_table); - host_ctxt = __hyp_this_cpu_ptr(kvm_host_cpu_state); - - while (entry < end) { - addr = (unsigned long)&entry->insn + entry->insn; - fixup = (unsigned long)&entry->fixup + entry->fixup; - - if (addr != elr_el2) { - entry++; - continue; - } - - write_sysreg(fixup, elr_el2); - return; - } - - hyp_panic(host_ctxt); -} -- 2.25.1

2 3

[PATCH 1/3] net: hns3: modify the sensitive words
by Yang Yingliang 30 Sep '20

30 Sep '20

From: Shengzui You <youshengzui(a)huawei.com> driver inclusion category: bugfix bugzilla: NA CVE: NA ----------------------------------- Reviewed-by: Weiwei Deng <dengweiwei(a)huawei.com> Reviewed-by: Zhaohui Zhong <zhongzhaohui(a)huawei.com> Reviewed-by: Junxin Chen <chenjunxin1(a)huawei.com> Signed-off-by: Shengzui You <youshengzui(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- .../hisilicon/hns3/hns-customer/hns3pf/hclge_main_it.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns-customer/hns3pf/hclge_main_it.h b/drivers/net/ethernet/hisilicon/hns3/hns-customer/hns3pf/hclge_main_it.h index cd1636e6dd65..343e036412d4 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns-customer/hns3pf/hclge_main_it.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns-customer/hns3pf/hclge_main_it.h @@ -36,14 +36,14 @@ typedef void (*nic_event_fn_t) (struct net_device *netdev, enum hnae3_event_type_custom); /** - * nic_register_event - register for nic event listening + * nic_register_event - register for nic event handling * @event_call: nic event handler * return 0 - success , negative - fail */ int nic_register_event(nic_event_fn_t event_call); /** - * nic_unregister_event - quit nic event listening + * nic_unregister_event - quit nic event handling * return 0 - success , negative - fail */ int nic_unregister_event(void); -- 2.25.1

1 2

[PATCH kernel-5.5] riscv/kvm: Add Linux-syscall-note at kvm.h
by Yipeng Yin 29 Sep '20

29 Sep '20

euleros inclusion category: bugfix bugzilla: NA CVE: NA It will raise 'missing "WITH Linux-syscall-note" for SPDX-License-Identifier' when make headers_install. this patch fix it. Link: https://gitee.com/openeuler/kernel/issues/I1WZL5 Signed-off-by: Yipeng Yin <yinyipeng1(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> --- arch/riscv/include/uapi/asm/kvm.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/riscv/include/uapi/asm/kvm.h b/arch/riscv/include/uapi/asm/kvm.h index f4274c2e5..65cd00654 100644 --- a/arch/riscv/include/uapi/asm/kvm.h +++ b/arch/riscv/include/uapi/asm/kvm.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: GPL-2.0 */ +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ /* * Copyright (C) 2019 Western Digital Corporation or its affiliates. * -- 2.19.1

2 1

[PATCH] block: allow for_each_bvec to support zero len bvec
by Yang Yingliang 29 Sep '20

29 Sep '20

From: Ming Lei <ming.lei(a)redhat.com> stable inclusion from linux-4.19.144 commit b48bcb664b657ae94b19c0728978c88e012f7a37 CVE: CVE-2020-25641 -------------------------------- commit 7e24969022cbd61ddc586f14824fc205661bb124 upstream. Block layer usually doesn't support or allow zero-length bvec. Since commit 1bdc76aea115 ("iov_iter: use bvec iterator to implement iterate_bvec()"), iterate_bvec() switches to bvec iterator. However, Al mentioned that 'Zero-length segments are not disallowed' in iov_iter. Fixes for_each_bvec() so that it can move on after seeing one zero length bvec. Fixes: 1bdc76aea115 ("iov_iter: use bvec iterator to implement iterate_bvec()") Reported-by: syzbot <syzbot+61acc40a49a3e46e25ea(a)syzkaller.appspotmail.com> Signed-off-by: Ming Lei <ming.lei(a)redhat.com> Tested-by: Tetsuo Handa <penguin-kernel(a)i-love.sakura.ne.jp> Cc: Al Viro <viro(a)zeniv.linux.org.uk> Cc: Matthew Wilcox <willy(a)infradead.org> Cc: <stable(a)vger.kernel.org> Link: https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg2262077.html Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/linux/bvec.h | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/include/linux/bvec.h b/include/linux/bvec.h index fe7a22dd133b..bc1f16e9f3f4 100644 --- a/include/linux/bvec.h +++ b/include/linux/bvec.h @@ -119,11 +119,18 @@ static inline bool bvec_iter_rewind(const struct bio_vec *bv, return true; } +static inline void bvec_iter_skip_zero_bvec(struct bvec_iter *iter) +{ + iter->bi_bvec_done = 0; + iter->bi_idx++; +} + #define for_each_bvec(bvl, bio_vec, iter, start) \ for (iter = (start); \ (iter).bi_size && \ ((bvl = bvec_iter_bvec((bio_vec), (iter))), 1); \ - bvec_iter_advance((bio_vec), &(iter), (bvl).bv_len)) + (bvl).bv_len ? (void)bvec_iter_advance((bio_vec), &(iter), \ + (bvl).bv_len) : bvec_iter_skip_zero_bvec(&(iter))) /* for iterating one bio from start to end */ #define BVEC_ITER_ALL_INIT (struct bvec_iter) \ -- 2.25.1

1 0

[PATCH 1/6] loop: Report EOPNOTSUPP properly
by Yang Yingliang 28 Sep '20

28 Sep '20

From: Evan Green <evgreen(a)chromium.org> mainline inclusion from mainline-v5.7-rc1 commit 8cd55087dc45b2e1a73ed2a197cbf405f32deb08 category: bugfix bugzilla: 38877 CVE: NA Properly plumb out EOPNOTSUPP from loop driver operations, which may get returned when for instance a discard operation is attempted but not supported by the underlying block device. Before this change, everything was reported in the log as an I/O error, which is scary and not helpful in debugging. Signed-off-by: Evan Green <evgreen(a)chromium.org> Reviewed-by: Gwendal Grignou <gwendal(a)chromium.org> Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Signed-off-by: Andrzej Pietrasiewicz <andrzej.p(a)collabora.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Signed-off-by: Luo Meng <luomeng12(a)huawei.com> Reviewed-by: Hou Tao <houtao1(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/block/loop.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index da68c42aed68..19b64ca8c4e3 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -461,7 +461,7 @@ static void lo_complete_rq(struct request *rq) if (!cmd->use_aio || cmd->ret < 0 || cmd->ret == blk_rq_bytes(rq) || req_op(rq) != REQ_OP_READ) { if (cmd->ret < 0) - ret = BLK_STS_IOERR; + ret = errno_to_blk_status(cmd->ret); goto end_io; } @@ -1924,7 +1924,10 @@ static void loop_handle_cmd(struct loop_cmd *cmd) failed: /* complete non-aio request */ if (!cmd->use_aio || ret) { - cmd->ret = ret ? -EIO : 0; + if (ret == -EOPNOTSUPP) + cmd->ret = ret; + else + cmd->ret = ret ? -EIO : 0; blk_mq_complete_request(rq); } } -- 2.25.1

1 5

[PATCH 00/38] backport stable-4.19.148
by Yang Yingliang 28 Sep '20

28 Sep '20

Dan Carpenter (1): hdlc_ppp: add range checks in ppp_cp_parse_cr() David Ahern (1): ipv4: Update exception handling for multipath routes via same device Dmitry Golovin (1): x86/boot: kbuild: allow readelf executable to be specified Edwin Peer (1): bnxt_en: return proper error codes in bnxt_show_temp Eric Dumazet (3): ipv6: avoid lockdep issue in fib6_del() net: qrtr: check skb_put_padto() return value net: add __must_check to skb_put_padto() Fangrui Song (1): Documentation/llvm: fix the name of llvm-size Florian Fainelli (1): net: phy: Avoid NPD upon phy_detach() when driver is unbound Ganji Aravind (1): cxgb4: Fix offset when clearing filter byte counters Greg Kroah-Hartman (1): Linux 4.19.148 Jakub Kicinski (1): nfp: use correct define to return NONE fec Linus Walleij (1): net: dsa: rtl8366: Properly clear member config Lukas Wunner (1): serial: 8250: Avoid error message on reprobe Mark Gray (1): geneve: add transport ports in route lookup for geneve Mark Salyzyn (1): af_key: pfkey_dump needs parameter validation Masahiro Yamada (5): net: wan: wanxl: use allow to pass CROSS_COMPILE_M68k for rebuilding firmware net: wan: wanxl: use $(M68KCC) instead of $(M68KAS) for rebuilding firmware kbuild: remove AS variable kbuild: replace AS=clang with LLVM_IAS=1 kbuild: support LLVM=1 to switch the default tools to Clang/LLVM Michael Chan (1): bnxt_en: Protect bnxt_set_eee() and bnxt_set_pauseparam() with mutex. Muchun Song (1): kprobes: fix kill kprobe which has been marked as gone Necip Fazil Yildiran (1): net: ipv6: fix kconfig dependency warning for IPV6_SEG6_HMAC Nick Desaulniers (2): MAINTAINERS: add CLANG/LLVM BUILD SUPPORT info Documentation/llvm: add documentation on building w/ Clang/LLVM Peilin Ye (1): tipc: Fix memory leak in tipc_group_create_member() Petr Machata (1): net: DCB: Validate DCB_ATTR_DCB_BUFFER argument Priyaranjan Jha (2): tcp_bbr: refactor bbr_target_cwnd() for general inflight provisioning tcp_bbr: adapt cwnd based on ack aggregation estimation Ralph Campbell (1): mm/thp: fix __split_huge_pmd_locked() for migration PMD Rustam Kovhaev (1): KVM: fix memory leak in kvm_io_bus_unregister_dev() Tetsuo Handa (1): tipc: fix shutdown() of connection oriented socket Vasily Gorbik (1): kbuild: add OBJSIZE variable for the size tool Wei Wang (1): ip: fix tos reflection in ack and reset packets Xin Long (1): tipc: use skb_unshare() instead in tipc_buf_append() Xunlei Pang (1): mm: memcg: fix memcg reclaim soft lockup Yunsheng Lin (1): net: sch_generic: aviod concurrent reset and enqueue op for lockless qdisc Documentation/kbuild/llvm.rst | 87 +++++++++ MAINTAINERS | 9 + Makefile | 38 +++- arch/x86/boot/compressed/Makefile | 2 +- drivers/net/dsa/rtl8366.c | 20 +- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 19 +- .../net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 31 +-- .../net/ethernet/chelsio/cxgb4/cxgb4_filter.c | 9 +- .../ethernet/netronome/nfp/nfp_net_ethtool.c | 4 +- drivers/net/geneve.c | 37 +++- drivers/net/phy/phy_device.c | 3 +- drivers/net/wan/Kconfig | 2 +- drivers/net/wan/Makefile | 12 +- drivers/net/wan/hdlc_ppp.c | 16 +- drivers/tty/serial/8250/8250_core.c | 11 +- include/linux/skbuff.h | 7 +- include/net/inet_connection_sock.h | 4 +- kernel/kprobes.c | 9 +- mm/huge_memory.c | 40 ++-- mm/vmscan.c | 8 + net/dcb/dcbnl.c | 8 + net/ipv4/ip_output.c | 3 +- net/ipv4/route.c | 13 +- net/ipv4/tcp_bbr.c | 180 ++++++++++++++++-- net/ipv6/Kconfig | 1 + net/ipv6/ip6_fib.c | 13 +- net/key/af_key.c | 7 + net/qrtr/qrtr.c | 20 +- net/sched/sch_generic.c | 49 +++-- net/tipc/group.c | 14 +- net/tipc/msg.c | 3 +- net/tipc/socket.c | 5 +- tools/objtool/Makefile | 6 + virt/kvm/kvm_main.c | 21 +- 34 files changed, 550 insertions(+), 161 deletions(-) create mode 100644 Documentation/kbuild/llvm.rst -- 2.25.1

1 38

【议题收集】【Meeting Notice】openEuler kernel sig meeting Time: 2020-10-16 10:00-12:00
by Xie XiuQi 28 Sep '20

28 Sep '20

openEuler kernel sig meeting 定于 2020-10-16 10:00-12:00 召开，欢迎回复邮件申报议题。 (受十一假期影响，中间跳过一次例会，如有问题需要交流，可以先发邮件列表） --- 会议信息 --- 会议链接：https://zoom.us/j/94156903933 会议 ID : 94156903933 --- 会议纪要归档 --- Meeting Record 20200918： https://gitee.com/openeuler/kernel/issues/I1WGN0

1 0

[PATCH kernel-5.5] riscv/kvm: Add Linux-syscall-note at kvm.h
by Yipeng Yin 28 Sep '20

28 Sep '20

euleros inclusion category: bugfix bugzilla: NA CVE: NA It will raise 'missing "WITH Linux-syscall-note" for SPDX-License-Identifier' when make headers_install. this patch fix it. Link: https://gitee.com/openeuler/kernel/issues/I1WZL5 Signed-off-by: Yipeng Yin <yinyipeng1(a)huawei.com> --- arch/riscv/include/uapi/asm/kvm.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/riscv/include/uapi/asm/kvm.h b/arch/riscv/include/uapi/asm/kvm.h index f4274c2e5..65cd00654 100644 --- a/arch/riscv/include/uapi/asm/kvm.h +++ b/arch/riscv/include/uapi/asm/kvm.h @@ -1,4 +1,4 @@ -/* SPDX-License-Identifier: GPL-2.0 */ +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ /* * Copyright (C) 2019 Western Digital Corporation or its affiliates. * -- 2.19.1

2 1

[PATCH openEuler-1.0-LTS 00/13] steal tasks to improve CPU utilization
by Wei Li 27 Sep '20

27 Sep '20

Steal tasks to improve CPU utilization, backported from branch 'kernel-4.19'. Cheng Jian (3): disable stealing by default sched/fair: introduce SCHED_STEAL config: enable CONFIG_SCHED_STEAL by default Steve Sistare (10): sched: Provide sparsemask, a reduced contention bitmap sched/topology: Provide hooks to allocate data shared per LLC sched/topology: Provide cfs_overload_cpus bitmap sched/fair: Dynamically update cfs_overload_cpus sched/fair: Hoist idle_stamp up from idle_balance sched/fair: Generalize the detach_task interface sched/fair: Provide can_migrate_task_llc sched/fair: Steal work from an overloaded CPU when CPU goes idle sched/fair: disable stealing if too many NUMA nodes sched/fair: Provide idle search schedstats arch/arm64/configs/euleros_defconfig | 1 + arch/arm64/configs/hulk_defconfig | 1 + arch/arm64/configs/openeuler_defconfig | 1 + arch/arm64/configs/storage_ci_defconfig | 1 + arch/arm64/configs/syzkaller_defconfig | 1 + arch/x86/configs/hulk_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + arch/x86/configs/storage_ci_defconfig | 1 + include/linux/sched/topology.h | 3 + init/Kconfig | 15 + kernel/sched/core.c | 35 ++- kernel/sched/fair.c | 367 ++++++++++++++++++++++-- kernel/sched/features.h | 8 + kernel/sched/sched.h | 20 ++ kernel/sched/sparsemask.h | 210 ++++++++++++++ kernel/sched/stats.c | 15 + kernel/sched/stats.h | 20 ++ kernel/sched/topology.c | 141 ++++++++- 18 files changed, 810 insertions(+), 32 deletions(-) create mode 100644 kernel/sched/sparsemask.h -- 2.17.1

1 13

[PATCH 00/50] backport stable-4.19.147
by Yang Yingliang 24 Sep '20

24 Sep '20

Adam Borowski (1): x86/defconfig: Enable CONFIG_USB_XHCI_HCD=y Alexey Kardashevskiy (1): powerpc/dma: Fix dma_map_ops::get_required_mask Arvind Sankar (1): x86/boot/compressed: Disable relocation relaxation Bob Peterson (1): gfs2: initialize transaction tr_ailX_lists earlier Christophe JAILLET (1): clk: davinci: Use the correct size when allocating memory Chuck Lever (1): NFS: Zero-stateid SETATTR should first return delegation Daniel Mack (1): dsa: Allow forwarding of redirected IGMP traffic David Milburn (2): nvme-fc: cancel async events before freeing event struct nvme-rdma: cancel async events before freeing event struct Dinghao Liu (1): scsi: pm8001: Fix memleak in pm8001_exec_internal_task_abort Evan Nimmo (1): i2c: algo: pca: Reapply i2c bus settings after reset Gabriel Krisman Bertazi (1): f2fs: Return EOF on unaligned end of file DIO read Greg Kroah-Hartman (2): Revert "ALSA: hda - Fix silent audio output and corrupted input on MSI X570-A PRO" Linux 4.19.147 Gustav Wiklander (1): spi: Fix memory leak on splited transfers Haiyang Zhang (1): hv_netvsc: Remove "unlikely" from netvsc_select_queue Hans de Goede (1): Input: i8042 - add Entroware Proteus EL07R4 to nomux and reset lists Huacai Chen (1): KVM: MIPS: Change the definition of kvm type J. Bruce Fields (1): SUNRPC: stop printk reading past end of string James Smart (1): scsi: lpfc: Fix FLOGI/PLOGI receive race condition in pt2pt discovery Javed Hasan (1): scsi: libfc: Fix for double free() Jiri Olsa (1): perf test: Fix the "signal" test inline assembly Laurent Pinchart (1): rapidio: Replace 'select' DMAENGINES 'with depends on' Miaohe Lin (1): net: handle the return value of pskb_carve_frag_list() correctly Michael Kelley (1): Drivers: hv: vmbus: Add timeout to vmbus_wait_for_unload Namhyung Kim (1): perf test: Free formats for perf pmu parse test Naresh Kumar PBS (1): RDMA/bnxt_re: Restrict the max_gids to 256 Nathan Chancellor (1): clk: rockchip: Fix initialization of mux_pll_src_4plls_p Olga Kornievskaia (1): NFSv4.1 handle ERR_DELAY error reclaiming locking state on delegation recall Oliver Neukum (2): USB: UAS: fix disconnect by unplugging a hub usblp: fix race between disconnect() and read() Penghao (1): USB: quirks: Add USB_QUIRK_IGNORE_REMOTE_WAKEUP quirk for BYD zhaoxin notebook Quentin Perret (1): ehci-hcd: Move include to keep CRC stable Quinn Tran (3): scsi: qla2xxx: Update rscn_rcvd field to more meaningful scan_needed scsi: qla2xxx: Move rport registration out of internal work_list scsi: qla2xxx: Reduce holding sess_lock to prevent CPU lock-up Sahitya Tummala (1): f2fs: fix indefinite loop scanning for free nid Stafford Horne (1): openrisc: Fix cache API compile issue when not inlining Stephan Gerhold (1): ASoC: qcom: Set card->owner to avoid warnings Sunghyun Jin (1): percpu: fix first chunk size calculation for populated bitmap Tetsuo Handa (1): fbcon: Fix user font detection test at fbcon_resize(). Thomas Bogendoerfer (2): MIPS: SNI: Fix MIPS_L1_CACHE_SHIFT MIPS: SNI: Fix spurious interrupts Tobias Diedrich (1): serial: 8250_pci: Add Realtek 816a and 816b Vincent Huang (1): Input: trackpoint - add new trackpoint variant IDs Vincent Whitchurch (2): regulator: pwm: Fix machine constraints application spi: spi-loopback-test: Fix out-of-bounds read Volker Rümelin (1): i2c: i801: Fix resume bug Yu Kuai (2): drm/mediatek: Add exception handing in mtk_drm_probe() if component init fail drm/mediatek: Add missing put_device() call in mtk_hdmi_dt_parse_pdata() Makefile | 2 +- arch/mips/Kconfig | 1 + arch/mips/kvm/mips.c | 2 + arch/mips/sni/a20r.c | 9 +- arch/openrisc/mm/cache.c | 2 +- arch/powerpc/kernel/dma-iommu.c | 3 +- arch/x86/boot/compressed/Makefile | 2 + arch/x86/configs/i386_defconfig | 1 + arch/x86/configs/x86_64_defconfig | 1 + drivers/clk/davinci/pll.c | 2 +- drivers/clk/rockchip/clk-rk3228.c | 2 +- drivers/gpu/drm/mediatek/mtk_drm_drv.c | 7 +- drivers/gpu/drm/mediatek/mtk_hdmi.c | 26 ++++-- drivers/hv/channel_mgmt.c | 7 +- drivers/i2c/algos/i2c-algo-pca.c | 35 +++++--- drivers/i2c/busses/i2c-i801.c | 21 +++-- drivers/infiniband/hw/bnxt_re/qplib_sp.c | 2 +- drivers/infiniband/hw/bnxt_re/qplib_sp.h | 1 + drivers/input/mouse/trackpoint.c | 10 ++- drivers/input/mouse/trackpoint.h | 10 ++- drivers/input/serio/i8042-x86ia64io.h | 16 ++++ drivers/net/hyperv/netvsc_drv.c | 2 +- drivers/nvme/host/fc.c | 1 + drivers/nvme/host/rdma.c | 1 + drivers/rapidio/Kconfig | 2 +- drivers/regulator/pwm-regulator.c | 2 +- drivers/scsi/libfc/fc_disc.c | 2 - drivers/scsi/lpfc/lpfc_els.c | 4 +- drivers/scsi/pm8001/pm8001_sas.c | 2 +- drivers/scsi/qla2xxx/qla_def.h | 10 ++- drivers/scsi/qla2xxx/qla_gbl.h | 5 +- drivers/scsi/qla2xxx/qla_gs.c | 30 ++++--- drivers/scsi/qla2xxx/qla_init.c | 101 +++++++++++++++++------ drivers/scsi/qla2xxx/qla_os.c | 29 ++++--- drivers/scsi/qla2xxx/qla_target.c | 85 +++++++++++++++---- drivers/spi/spi-loopback-test.c | 2 +- drivers/spi/spi.c | 9 +- drivers/tty/serial/8250/8250_pci.c | 11 +++ drivers/usb/class/usblp.c | 5 ++ drivers/usb/core/quirks.c | 4 + drivers/usb/host/ehci-hcd.c | 1 + drivers/usb/host/ehci-hub.c | 1 - drivers/usb/storage/uas.c | 14 +++- drivers/video/fbdev/core/fbcon.c | 2 +- fs/f2fs/data.c | 3 + fs/f2fs/node.c | 3 + fs/gfs2/glops.c | 2 + fs/gfs2/log.c | 2 - fs/gfs2/trans.c | 2 + fs/nfs/nfs4proc.c | 11 ++- include/linux/i2c-algo-pca.h | 15 ++++ include/uapi/linux/kvm.h | 5 +- mm/percpu.c | 2 +- net/core/skbuff.c | 10 ++- net/dsa/tag_edsa.c | 37 ++++++++- net/sunrpc/rpcb_clnt.c | 4 +- sound/pci/hda/patch_realtek.c | 1 - sound/soc/qcom/apq8016_sbc.c | 1 + sound/soc/qcom/apq8096.c | 1 + sound/soc/qcom/sdm845.c | 1 + sound/soc/qcom/storm.c | 1 + tools/perf/tests/bp_signal.c | 5 +- tools/perf/tests/pmu.c | 1 + tools/perf/util/pmu.c | 11 +++ tools/perf/util/pmu.h | 1 + 65 files changed, 457 insertions(+), 149 deletions(-) -- 2.25.1

1 50

[PATCH 1/5] loop: Report EOPNOTSUPP properly
by Yang Yingliang 24 Sep '20

24 Sep '20

From: Evan Green <evgreen(a)chromium.org> mainline inclusion from mainline-v5.7-rc1 commit 8cd55087dc45b2e1a73ed2a197cbf405f32deb08 category: bugfix bugzilla: 38877 CVE: NA Properly plumb out EOPNOTSUPP from loop driver operations, which may get returned when for instance a discard operation is attempted but not supported by the underlying block device. Before this change, everything was reported in the log as an I/O error, which is scary and not helpful in debugging. Signed-off-by: Evan Green <evgreen(a)chromium.org> Reviewed-by: Gwendal Grignou <gwendal(a)chromium.org> Reviewed-by: Bart Van Assche <bvanassche(a)acm.org> Signed-off-by: Andrzej Pietrasiewicz <andrzej.p(a)collabora.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Signed-off-by: Luo Meng <luomeng12(a)huawei.com> Reviewed-by: Hou Tao <houtao1(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/block/loop.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 19042b42a8ba..d0687e90e5d4 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -461,7 +461,7 @@ static void lo_complete_rq(struct request *rq) if (!cmd->use_aio || cmd->ret < 0 || cmd->ret == blk_rq_bytes(rq) || req_op(rq) != REQ_OP_READ) { if (cmd->ret < 0) - ret = BLK_STS_IOERR; + ret = errno_to_blk_status(cmd->ret); goto end_io; } @@ -1927,7 +1927,10 @@ static void loop_handle_cmd(struct loop_cmd *cmd) failed: /* complete non-aio request */ if (!cmd->use_aio || ret) { - cmd->ret = ret ? -EIO : 0; + if (ret == -EOPNOTSUPP) + cmd->ret = ret; + else + cmd->ret = ret ? -EIO : 0; blk_mq_complete_request(rq); } } -- 2.25.1

1 4

[PATCH 00/75] backport stable-4.19.146
by Yang Yingliang 23 Sep '20

23 Sep '20

Adam Ford (2): ARM: dts: logicpd-torpedo-baseboard: Fix broken audio ARM: dts: logicpd-som-lv-baseboard: Fix broken audio Aleksander Morgado (1): USB: serial: option: add support for SIM7070/SIM7080/SIM7090 modules Angelo Compagnucci (2): iio: adc: mcp3422: fix locking scope iio: adc: mcp3422: fix locking on error path Bjørn Mork (1): USB: serial: option: support dynamic Quectel USB compositions Chris Healy (1): ARM: dts: vfxxx: Add syscon compatible with OCOTP Darrick J. Wong (1): xfs: initialize the shortform attr header padding entry Dinghao Liu (4): RDMA/rxe: Fix memleak in rxe_mem_init_user NFC: st95hf: Fix memleak in st95hf_in_send_cmd firestream: Fix memleak in fs_open HID: elan: Fix memleak in elan_input_configured Dinh Nguyen (1): ARM: dts: socfpga: fix register entry for timer3 on Arria10 Douglas Anderson (1): mmc: sdhci-msm: Add retries when all tuning phases are found valid Evgeniy Didin (1): ARC: [plat-hsdk]: Switch ethernet phy-mode to rgmii-id Filipe Manana (1): btrfs: fix wrong address when faulting in pages in the search ioctl Florian Fainelli (4): ARM: dts: bcm: HR2: Fixed QSPI compatible string ARM: dts: NSP: Fixed QSPI compatible string ARM: dts: BCM5301X: Fixed QSPI compatible string arm64: dts: ns2: Fixed QSPI compatible string Florian Westphal (1): netfilter: conntrack: allow sctp hearbeat after connection re-use Francisco Jerez (1): cpufreq: intel_pstate: Fix intel_pstate_get_hwp_max() for turbo disabled Greg Kroah-Hartman (1): Linux 4.19.146 Hanjun Guo (1): dmaengine: acpi: Put the CSRT table after using it Heikki Krogerus (1): usb: typec: ucsi: acpi: Check the _DEP dependencies Hou Pu (1): scsi: target: iscsi: Fix hang in iscsit_access_np() when getting tpg->np_login_sem Joerg Roedel (1): iommu/amd: Do not use IOMMUv2 functionality when SME is active Jonathan Cameron (12): iio:light:ltr501 Fix timestamp alignment issue. iio:accel:bmc150-accel: Fix timestamp alignment and prevent data leak. iio:adc:ti-adc084s021 Fix alignment and data leak issues. iio:adc:ina2xx Fix timestamp alignment issue. iio:adc:max1118 Fix alignment of timestamp and data leak issues iio:adc:ti-adc081c Fix alignment and data leak issues iio:magnetometer:ak8975 Fix alignment and data leak issues. iio:light:max44000 Fix timestamp alignment and prevent data leak. iio:chemical:ccs811: Fix timestamp alignment and prevent data leak. iio: accel: kxsd9: Fix alignment of local buffer. iio:accel:mma7455: Fix timestamp alignment and prevent data leak. iio:accel:mma8452: Fix timestamp alignment and prevent data leak. Jordan Crouse (1): drm/msm: Disable preemption on all 5xx targets Josef Bacik (1): btrfs: fix lockdep splat in add_missing_dev Kamal Heib (2): RDMA/rxe: Drop pointless checks in rxe_init_ports RDMA/core: Fix reported speed and width Leon Romanovsky (1): gcov: Disable gcov build with GCC 10 Linus Torvalds (1): vgacon: remove software scrollback support Linus Walleij (1): drm/tve200: Stabilize enable/disable Mathias Nyman (1): usb: Fix out of sync data toggle if a configured device is reconfigured Matthias Schiffer (1): ARM: dts: ls1021a: fix QuadSPI-memory reg range Maxim Kochetkov (1): iio: adc: ti-ads1015: fix conversion when CONFIG_PM is not set Michał Mirosław (1): regulator: push allocation in set_consumer_device_supply() out of lock Mohan Kumar (1): ALSA: hda: Fix 2 channel swapping for Tegra Nirenjan Krishnan (1): HID: quirks: Set INCREMENT_USAGE_ON_DUPLICATE for all Saitek X52 devices Ondrej Jirman (1): drm/sun4i: Fix dsi dcs long write function Patrick Riphagen (1): USB: serial: ftdi_sio: add IDs for Xsens Mti USB converter Peter Oberparleiter (1): gcov: add support for GCC 10.1 Qu Wenruo (1): btrfs: require only sector size alignment for parent eb bytenr Rafael J. Wysocki (1): cpufreq: intel_pstate: Refuse to turn off with HWP enabled Rander Wang (1): ALSA: hda: fix a runtime pm issue in SOF when integrated GPU is disabled Rustam Kovhaev (1): staging: wlan-ng: fix out of bounds read in prism2sta_probe_usb() Sagi Grimberg (2): nvme-fabrics: don't check state NVME_CTRL_NEW for request acceptance nvme-rdma: serialize controller teardown sequences Sandeep Raghuraman (1): drm/amdgpu: Fix bug in reporting voltage for CIK Selvin Xavier (1): RDMA/bnxt_re: Do not report transparent vlan from QP1 Sivaprakash Murugesan (1): phy: qcom-qmp: Use correct values for ipq8074 PCIe Gen2 PHY init Tetsuo Handa (1): video: fbdev: fix OOB read in vga_8planes_imageblit() Vaibhav Agarwal (1): staging: greybus: audio: fix uninitialized value issue Varun Prakash (1): scsi: target: iscsi: Fix data digest calculation Vineet Gupta (2): ARC: HSDK: wireup perf irq irqchip/eznps: Fix build error for !ARC700 builds Wanpeng Li (1): KVM: VMX: Don't freeze guest when event delivery causes an APIC-access exit Xie He (3): drivers/net/wan/lapbether: Added needed_tailroom drivers/net/wan/lapbether: Set network_header before transmitting drivers/net/wan/hdlc_cisco: Add hard_header_len Yi Zhang (1): RDMA/rxe: Fix the parent sysfs read when the interface has 15 chars Zeng Tao (1): usb: core: fix slab-out-of-bounds Read in read_descriptors Makefile | 2 +- arch/arc/boot/dts/hsdk.dts | 6 +- arch/arc/plat-eznps/include/plat/ctop.h | 1 - arch/arm/boot/dts/bcm-hr2.dtsi | 2 +- arch/arm/boot/dts/bcm-nsp.dtsi | 2 +- arch/arm/boot/dts/bcm5301x.dtsi | 2 +- .../boot/dts/logicpd-som-lv-baseboard.dtsi | 2 + .../boot/dts/logicpd-torpedo-baseboard.dtsi | 2 + arch/arm/boot/dts/ls1021a.dtsi | 2 +- arch/arm/boot/dts/socfpga_arria10.dtsi | 2 +- arch/arm/boot/dts/vfxxx.dtsi | 2 +- .../boot/dts/broadcom/northstar2/ns2.dtsi | 2 +- arch/powerpc/configs/pasemi_defconfig | 1 - arch/powerpc/configs/ppc6xx_defconfig | 1 - arch/x86/configs/i386_defconfig | 1 - arch/x86/configs/x86_64_defconfig | 1 - arch/x86/kvm/vmx.c | 1 + drivers/atm/firestream.c | 1 + drivers/cpufreq/intel_pstate.c | 14 +- drivers/dma/acpi-dma.c | 4 +- .../gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c | 3 +- drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 3 +- drivers/gpu/drm/sun4i/sun6i_mipi_dsi.c | 4 +- drivers/gpu/drm/tve200/tve200_display.c | 22 +- drivers/hid/hid-elan.c | 2 + drivers/hid/hid-ids.h | 2 + drivers/hid/hid-quirks.c | 2 + drivers/iio/accel/bmc150-accel-core.c | 15 +- drivers/iio/accel/kxsd9.c | 16 +- drivers/iio/accel/mma7455_core.c | 16 +- drivers/iio/accel/mma8452.c | 11 +- drivers/iio/adc/ina2xx-adc.c | 11 +- drivers/iio/adc/max1118.c | 10 +- drivers/iio/adc/mcp3422.c | 16 +- drivers/iio/adc/ti-adc081c.c | 11 +- drivers/iio/adc/ti-adc084s021.c | 10 +- drivers/iio/adc/ti-ads1015.c | 10 + drivers/iio/chemical/ccs811.c | 13 +- drivers/iio/light/ltr501.c | 15 +- drivers/iio/light/max44000.c | 12 +- drivers/iio/magnetometer/ak8975.c | 16 +- drivers/infiniband/core/verbs.c | 2 +- drivers/infiniband/hw/bnxt_re/ib_verbs.c | 21 +- drivers/infiniband/sw/rxe/rxe.c | 3 - drivers/infiniband/sw/rxe/rxe_mr.c | 1 + drivers/infiniband/sw/rxe/rxe_verbs.c | 2 +- drivers/iommu/amd_iommu_v2.c | 7 + drivers/mmc/host/sdhci-msm.c | 18 +- drivers/net/wan/hdlc_cisco.c | 1 + drivers/net/wan/lapbether.c | 3 + drivers/nfc/st95hf/core.c | 2 +- drivers/nvme/host/fabrics.c | 1 - drivers/nvme/host/rdma.c | 6 + drivers/phy/qualcomm/phy-qcom-qmp.c | 16 +- drivers/phy/qualcomm/phy-qcom-qmp.h | 2 + drivers/regulator/core.c | 46 ++-- drivers/staging/greybus/audio_topology.c | 29 +-- drivers/staging/wlan-ng/hfa384x_usb.c | 5 - drivers/staging/wlan-ng/prism2usb.c | 19 +- drivers/target/iscsi/iscsi_target.c | 17 +- drivers/target/iscsi/iscsi_target_login.c | 6 +- drivers/target/iscsi/iscsi_target_login.h | 3 +- drivers/target/iscsi/iscsi_target_nego.c | 3 +- drivers/usb/core/message.c | 91 ++++---- drivers/usb/core/sysfs.c | 5 + drivers/usb/serial/ftdi_sio.c | 1 + drivers/usb/serial/ftdi_sio_ids.h | 1 + drivers/usb/serial/option.c | 22 +- drivers/usb/typec/ucsi/ucsi_acpi.c | 4 + drivers/video/console/Kconfig | 46 ---- drivers/video/console/vgacon.c | 221 +----------------- drivers/video/fbdev/vga16fb.c | 2 +- fs/btrfs/extent-tree.c | 19 +- fs/btrfs/ioctl.c | 3 +- fs/btrfs/print-tree.c | 12 +- fs/btrfs/volumes.c | 10 + fs/xfs/libxfs/xfs_attr_leaf.c | 4 +- include/linux/netfilter/nf_conntrack_sctp.h | 2 + include/soc/nps/common.h | 6 + kernel/gcov/gcc_4_7.c | 4 +- net/netfilter/nf_conntrack_proto_sctp.c | 39 +++- sound/hda/hdac_device.c | 2 + sound/pci/hda/patch_hdmi.c | 5 + 83 files changed, 479 insertions(+), 504 deletions(-) -- 2.25.1

1 75

[PATCH] config: add digest list options for arm64
by Yang Yingliang 22 Sep '20

22 Sep '20

From: Roberto Sassu <roberto.sassu(a)huawei.com> hulk inclusion category: feature feature: digest-lists --------------------------- Enable digest lists and PGP keys preload. Signed-off-by: Roberto Sassu <roberto.sassu(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 73 +++++++++++++++++++------- 1 file changed, 54 insertions(+), 19 deletions(-) diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 809350e03968..b2abedf899b5 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -3048,20 +3048,23 @@ CONFIG_HW_RANDOM_CAVIUM=y # CONFIG_RAW_DRIVER=y CONFIG_MAX_RAW_DEVS=8192 -CONFIG_TCG_TPM=m +CONFIG_TCG_TPM=y CONFIG_HW_RANDOM_TPM=y -CONFIG_TCG_TIS_CORE=m -CONFIG_TCG_TIS=m -CONFIG_TCG_TIS_SPI=m -# CONFIG_TCG_TIS_I2C_ATMEL is not set -# CONFIG_TCG_TIS_I2C_INFINEON is not set -# CONFIG_TCG_TIS_I2C_NUVOTON is not set -CONFIG_TCG_ATMEL=m -# CONFIG_TCG_INFINEON is not set -CONFIG_TCG_CRB=m +CONFIG_TCG_TIS_CORE=y +CONFIG_TCG_TIS=y +CONFIG_TCG_TIS_SPI=y +CONFIG_TCG_TIS_I2C_ATMEL=y +CONFIG_TCG_TIS_I2C_INFINEON=y +CONFIG_TCG_TIS_I2C_NUVOTON=y +CONFIG_TCG_NSC=y +CONFIG_TCG_ATMEL=y +CONFIG_TCG_INFINEON=y +# CONFIG_TCG_XEN is not set +CONFIG_TCG_CRB=y # CONFIG_TCG_VTPM_PROXY is not set -# CONFIG_TCG_TIS_ST33ZP24_I2C is not set -# CONFIG_TCG_TIS_ST33ZP24_SPI is not set +CONFIG_TCG_TIS_ST33ZP24=y +CONFIG_TCG_TIS_ST33ZP24_I2C=y +CONFIG_TCG_TIS_ST33ZP24_SPI=y # CONFIG_DEVPORT is not set # CONFIG_XILLYBUS is not set CONFIG_HISI_SVM=y @@ -5425,8 +5428,8 @@ CONFIG_KEYS=y CONFIG_KEYS_COMPAT=y CONFIG_PERSISTENT_KEYRINGS=y CONFIG_BIG_KEYS=y -CONFIG_TRUSTED_KEYS=m -CONFIG_ENCRYPTED_KEYS=m +CONFIG_TRUSTED_KEYS=y +CONFIG_ENCRYPTED_KEYS=y # CONFIG_KEY_DH_OPERATIONS is not set # CONFIG_SECURITY_DMESG_RESTRICT is not set CONFIG_SECURITY=y @@ -5459,7 +5462,39 @@ CONFIG_SECURITY_APPARMOR_HASH_DEFAULT=y # CONFIG_SECURITY_APPARMOR_DEBUG is not set # CONFIG_SECURITY_LOADPIN is not set CONFIG_SECURITY_YAMA=y -# CONFIG_INTEGRITY is not set +CONFIG_INTEGRITY=y +CONFIG_INTEGRITY_SIGNATURE=y +CONFIG_INTEGRITY_ASYMMETRIC_KEYS=y +CONFIG_INTEGRITY_TRUSTED_KEYRING=y +CONFIG_INTEGRITY_AUDIT=y +CONFIG_IMA=y +CONFIG_IMA_MEASURE_PCR_IDX=10 +CONFIG_IMA_LSM_RULES=y +# CONFIG_IMA_TEMPLATE is not set +CONFIG_IMA_NG_TEMPLATE=y +# CONFIG_IMA_SIG_TEMPLATE is not set +CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng" +# CONFIG_IMA_DEFAULT_HASH_SHA1 is not set +CONFIG_IMA_DEFAULT_HASH_SHA256=y +CONFIG_IMA_DEFAULT_HASH="sha256" +# CONFIG_IMA_WRITE_POLICY is not set +CONFIG_IMA_READ_POLICY=y +CONFIG_IMA_APPRAISE=y +# CONFIG_IMA_APPRAISE_BUILD_POLICY is not set +CONFIG_IMA_APPRAISE_BOOTPARAM=y +CONFIG_IMA_TRUSTED_KEYRING=y +# CONFIG_IMA_BLACKLIST_KEYRING is not set +CONFIG_IMA_LOAD_X509=y +CONFIG_IMA_X509_PATH="/etc/keys/x509_ima.der" +# CONFIG_IMA_APPRAISE_SIGNED_INIT is not set +CONFIG_IMA_DIGEST_LIST=y +CONFIG_IMA_DIGEST_LISTS_DIR="/etc/ima/digest_lists" +CONFIG_IMA_PARSER_BINARY_PATH="/usr/bin/upload_digest_lists" +CONFIG_EVM=y +CONFIG_EVM_ATTR_FSUUID=y +# CONFIG_EVM_ADD_XATTRS is not set +CONFIG_EVM_LOAD_X509=y +CONFIG_EVM_X509_PATH="/etc/keys/x509_evm.der" CONFIG_DEFAULT_SECURITY_SELINUX=y # CONFIG_DEFAULT_SECURITY_APPARMOR is not set # CONFIG_DEFAULT_SECURITY_DAC is not set @@ -5646,9 +5681,9 @@ CONFIG_X509_CERTIFICATE_PARSER=y CONFIG_PKCS7_MESSAGE_PARSER=y # CONFIG_PKCS7_TEST_KEY is not set CONFIG_SIGNED_PE_FILE_VERIFICATION=y -# CONFIG_PGP_LIBRARY is not set -# CONFIG_PGP_KEY_PARSER is not set -# CONFIG_PGP_PRELOAD is not set +CONFIG_PGP_LIBRARY=y +CONFIG_PGP_KEY_PARSER=y +CONFIG_PGP_PRELOAD=y # # Certificates for signature checking @@ -5659,7 +5694,7 @@ CONFIG_SYSTEM_TRUSTED_KEYS="" # CONFIG_SYSTEM_EXTRA_CERTIFICATE is not set # CONFIG_SECONDARY_TRUSTED_KEYRING is not set # CONFIG_SYSTEM_BLACKLIST_KEYRING is not set -# CONFIG_PGP_PRELOAD_PUBLIC_KEYS is not set +CONFIG_PGP_PRELOAD_PUBLIC_KEYS=y CONFIG_BINARY_PRINTF=y # -- 2.25.1

1 0

[PATCH] acpi/arm64: check the returned logical CPU number of 'acpi_map_cpuid()'
by Yang Yingliang 22 Sep '20

22 Sep '20

From: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> hulk inclusion category: bugfix bugzilla: NA CVE: NA --------------------------- When we set 'nr_cpus=1' in kernel parameter, we get the following error. It's because 'acpi_map_cpuid()' return -ENODEV in 'acpi_map_cpu()' when there are not enough logical CPU numbers. So we need to check the returned logical CPU number and return error if it is negative. 'acpi_map_cpu' when there are not enough logical CPU [ 0.025955] Unable to handle kernel paging request at virtual address ffff00002915b828 [ 0.025958] Mem abort info: [ 0.025959] ESR = 0x96000006 [ 0.025961] Exception class = DABT (current EL), IL = 32 bits [ 0.025963] SET = 0, FnV = 0 [ 0.025965] EA = 0, S1PTW = 0 [ 0.025966] Data abort info: [ 0.025968] ISV = 0, ISS = 0x00000006 [ 0.025970] CM = 0, WnR = 0 [ 0.025972] swapper pgtable: 4k pages, 48-bit VAs, pgdp = (____ptrval____) [ 0.025974] [ffff00002915b828] pgd=000000013fffe003, pud=000000013fffd003, pmd=0000000000000000 [ 0.025979] Internal error: Oops: 96000006 [#1] SMP [ 0.025981] Modules linked in: [ 0.025983] Process swapper/0 (pid: 1, stack limit = 0x(____ptrval____)) [ 0.025986] CPU: 0 PID: 1 Comm: swapper/0 Tainted: G W 4.19.141+ #37 [ 0.025988] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015 [ 0.025991] pstate: a0c00005 (NzCv daif +PAN +UAO) [ 0.025993] pc : acpi_map_cpu+0xe0/0x170 [ 0.025996] lr : acpi_map_cpu+0xb8/0x170 [ 0.025997] sp : ffff8000fef1fa50 [ 0.025999] x29: ffff8000fef1fa50 x28: ffff000008e22058 [ 0.026001] x27: ffff0000092a6000 x26: ffff000008d60778 [ 0.026004] x25: ffff0000094c3000 x24: 0000000000000001 [ 0.026006] x23: ffff8000fe802c18 x22: 00000000ffffffff [ 0.026008] x21: ffff000008a16000 x20: ffff000008a16a20 [ 0.026011] x19: 00000000ffffffed x18: ffffffffffffffff [ 0.026013] x17: 0000000087411dcf x16: 00000000b93a5600 [ 0.026015] x15: ffff000009159708 x14: 0720072007200720 [ 0.026018] x13: 0720072007200720 x12: 0720072007200720 [ 0.026020] x11: 072007200720073d x10: 073d073d073d073d [ 0.026022] x9 : 073d073d073d0764 x8 : 0765076607660766 [ 0.026024] x7 : 0766076607660778 x6 : 0000000000000130 [ 0.026027] x5 : ffff0000085955c8 x4 : 0000000000000000 [ 0.026029] x3 : 0000000000000000 x2 : ffff00000915b830 [ 0.026031] x1 : ffff00002915b828 x0 : 0000200000000000 [ 0.026033] Call trace: [ 0.026035] acpi_map_cpu+0xe0/0x170 [ 0.026038] acpi_processor_add+0x44c/0x640 [ 0.026040] acpi_bus_attach+0x174/0x218 [ 0.026043] acpi_bus_attach+0xa8/0x218 [ 0.026045] acpi_bus_attach+0xa8/0x218 [ 0.026047] acpi_bus_attach+0xa8/0x218 [ 0.026049] acpi_bus_scan+0x58/0xb8 [ 0.026052] acpi_scan_init+0xf4/0x234 [ 0.026054] acpi_init+0x318/0x384 [ 0.026056] do_one_initcall+0x54/0x250 [ 0.026059] kernel_init_freeable+0x2d4/0x3c0 [ 0.026061] kernel_init+0x18/0x118 [ 0.026063] ret_from_fork+0x10/0x18 [ 0.026066] Code: d2800020 9120c042 8b010c41 9ad32000 (f820303f) Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Reviewed-by: Keqian Zhu <zhukeqian1(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/kernel/acpi.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/arm64/kernel/acpi.c b/arch/arm64/kernel/acpi.c index 73eb04f997a7..729733536cb4 100644 --- a/arch/arm64/kernel/acpi.c +++ b/arch/arm64/kernel/acpi.c @@ -270,6 +270,10 @@ int acpi_map_cpu(acpi_handle handle, phys_cpuid_t physid, u32 acpi_id, int cpu, nid; cpu = acpi_map_cpuid(physid, acpi_id); + if (cpu < 0) { + pr_info("Unable to map GICC to logical cpu number\n"); + return cpu; + } nid = acpi_get_node(handle); if (nid != NUMA_NO_NODE) { set_cpu_numa_node(cpu, nid); -- 2.25.1

1 0

[PATCH 0/9] backport stable-4.19.145
by Yang Yingliang 22 Sep '20

22 Sep '20

Greg Kroah-Hartman (1): Linux 4.19.145 Jakub Kicinski (1): net: disable netpoll on fresh napis Jens Axboe (1): block: ensure bdi->io_pages is always initialized Kamil Lorenc (1): net: usb: dm9601: Add USB ID of Keenetic Plus DSL Paul Moore (1): netlabel: fix problems with mapping removal Roi Dayan (1): net/mlx5e: Don't support phys switch id if not in switchdev mode Takashi Sakamoto (1): ALSA; firewire-tascam: exclude Tascam FE-8 from detection Tetsuo Handa (1): tipc: fix shutdown() of connectionless socket Xin Long (1): sctp: not disable bh in the whole sctp_get_port_local() Makefile | 2 +- block/blk-core.c | 2 + .../net/ethernet/mellanox/mlx5/core/en_rep.c | 2 +- drivers/net/usb/dm9601.c | 4 ++ net/core/dev.c | 3 +- net/core/netpoll.c | 2 +- net/netlabel/netlabel_domainhash.c | 59 ++++++++++--------- net/sctp/socket.c | 16 ++--- net/tipc/socket.c | 9 ++- sound/firewire/tascam/tascam.c | 30 +++++++++- 10 files changed, 82 insertions(+), 47 deletions(-) -- 2.25.1

1 9

[PATCH 00/79] backport stable-4.19.144
by Yang Yingliang 22 Sep '20

22 Sep '20

Al Grant (1): perf tools: Correct SNOOPX field offset Al Viro (1): fix regression in "epoll: Keep a reference on files added to the check list" Amit Engel (1): nvmet: Disable keep-alive timer when kato is cleared to 0h Bodo Stroesser (2): scsi: target: tcmu: Fix size in calls to tcmu_flush_dcache_range scsi: target: tcmu: Optimize use of flush_dcache_page Christophe JAILLET (1): nvmet-fc: Fix a missed _irqsave version of spin_lock in 'nvmet_fc_fod_op_done()' Dan Carpenter (1): net: gemini: Fix another missing clk_disable_unprepare() in probe Dan Crawford (1): ALSA: hda - Fix silent audio output and corrupted input on MSI X570-A PRO Daniel Borkmann (1): uaccess: Add non-pagefault user-space write function Daniele Palmas (1): net: usb: qmi_wwan: add Telit 0x1050 composition Darrick J. Wong (1): xfs: fix xfs_bmap_validate_extent_raw when checking attr fork of rt files Dinghao Liu (3): net: hns: Fix memleak in hns_nic_dev_probe net: systemport: Fix memleak in bcm_sysport_probe net: arc_emac: Fix memleak in arc_mdio_probe Dmitry Baryshkov (1): drm/msm/a6xx: fix gmu start on newer firmware Edwin Peer (1): bnxt_en: fix HWRM error when querying VF temperature Florian Fainelli (2): MIPS: mm: BMIPS5000 has inclusive physical caches MIPS: BMIPS: Also call bmips_cpu_setup() for secondary cores Florian Westphal (1): netfilter: nf_tables: fix destination register zeroing Greg Kroah-Hartman (1): Linux 4.19.144 Himadri Pandya (1): net: usb: Fix uninit-was-stored issue in asix_read_phy_addr() Huang Ying (1): x86, fakenuma: Fix invalid starting node ID Jakub Kicinski (1): bnxt: don't enable NAPI until rings are ready James Morse (4): KVM: arm64: Add kvm_extable for vaxorcism code KVM: arm64: Defer guest entry when an asynchronous exception is pending KVM: arm64: Survive synchronous exceptions caused by AT instructions KVM: arm64: Set HCR_EL2.PTW to prevent AT taking synchronous exception Jason Gunthorpe (1): include/linux/log2.h: add missing () around n in roundup_pow_of_two() Jeff Layton (1): ceph: don't allow setlease on cephfs Jesper Dangaard Brouer (1): selftests/bpf: Fix massive output from test_maps Johannes Berg (1): cfg80211: regulatory: reject invalid hints John Stultz (1): tty: serial: qcom_geni_serial: Drop __init from qcom_geni_console_setup Josef Bacik (3): btrfs: drop path before adding new uuid tree entry btrfs: set the lockdep class for log tree extent buffers btrfs: fix potential deadlock in the search ioctl Jussi Kivilinna (1): batman-adv: bla: use netif_rx_ni when not in interrupt context Kai Vehmanen (1): ALSA: hda/hdmi: always check pin power status in i915 pin fixup Kim Phillips (1): perf record/stat: Explicitly call out event modifiers in the documentation Krishna Manikandan (1): drm/msm: add shutdown support for display platform_driver Linus Lüssing (1): batman-adv: Fix own OGM check in aggregated OGMs Lu Baolu (1): iommu/vt-d: Serialize IOMMU GCMD register modifications Marc Zyngier (2): HID: core: Correctly handle ReportSize being zero HID: core: Sanitize event code and type when mapping input Marek Szyprowski (1): dmaengine: pl330: Fix burst length if burst size is smaller than bus width Masami Hiramatsu (1): uaccess: Add non-pagefault user-space read functions Max Staudt (1): affs: fix basic permission bits to actually work Michael Chan (1): tg3: Fix soft lockup when tg3_reset_task() fails. Mikulas Patocka (3): ext2: don't update mtime on COW faults xfs: don't update mtime on COW faults dm writecache: handle DAX to partitions on persistent memory correctly Ming Lei (1): block: allow for_each_bvec to support zero len bvec Mrinal Pandey (1): checkpatch: fix the usage of capture group ( ... ) Namhyung Kim (1): perf jevents: Fix suspicious code in fixregex() Nicolas Dichtel (1): gtp: add GTPA_LINK info to msg sent to userspace Nikolay Borisov (2): btrfs: Remove redundant extent_buffer_get in get_old_root btrfs: Remove extraneous extent_buffer_get from tree_mod_log_rewind Pablo Neira Ayuso (3): netfilter: nf_tables: add NFTA_SET_USERDATA if not null netfilter: nf_tables: incorrect enum nft_list_attributes definition netfilter: nfnetlink: nfnetlink_unicast() reports EAGAIN instead of ENOBUFS Pavan Chebbi (1): bnxt_en: Don't query FW when netif_running() is false. Peter Ujfalusi (1): dmaengine: of-dma: Fix of_dma_router_xlate's of_dma_xlate handling Peter Zijlstra (1): cpuidle: Fixup IRQ state Rogan Dawes (1): usb: qmi_wwan: add D-Link DWM-222 A2 device ID Sean Young (2): media: rc: do not access device via sysfs after rc_unregister_device() media: rc: uevent sysfs file races with rc_unregister_device() Shung-Hsi Yu (1): net: ethernet: mlx4: Fix memory allocation in mlx4_buddy_init() Simon Leiner (1): xen/xenbus: Fix granting of vmalloc'd memory Sven Eckelmann (1): batman-adv: Avoid uninitialized chaddr when handling DHCP Sven Schnelle (1): s390: don't trace preemption in percpu macros Takashi Iwai (1): ALSA: pcm: oss: Remove superfluous WARN_ON() for mulaw sanity check Takashi Sakamoto (1): ALSA: firewire-digi00x: exclude Avid Adrenaline from detection Tejun Heo (1): libata: implement ATA_HORKAGE_MAX_TRIM_128M and apply to Sandisks Tom Rix (1): hwmon: (applesmc) check status earlier. Tong Zhang (1): ALSA: ca0106: fix error code handling Tony Lindgren (1): thermal: ti-soc-thermal: Fix bogus thermal shutdowns for omap4430 Vasundhara Volam (2): bnxt_en: Check for zero dir entries in NVRAM. bnxt_en: Fix PCI AER error recovery flow Yu Kuai (1): dmaengine: at_hdmac: check return value of of_find_device_by_node() in at_dma_xlate() Yuusuke Ashizuka (1): ravb: Fixed to be able to unload modules Documentation/filesystems/affs.txt | 16 +- Makefile | 2 +- arch/arm64/include/asm/kvm_arm.h | 3 +- arch/arm64/include/asm/kvm_asm.h | 43 +++++ arch/arm64/kernel/vmlinux.lds.S | 8 + arch/arm64/kvm/hyp/entry.S | 31 +++- arch/arm64/kvm/hyp/hyp-entry.S | 66 ++++--- arch/arm64/kvm/hyp/switch.c | 39 +++- arch/mips/kernel/smp-bmips.c | 2 + arch/mips/mm/c-r4k.c | 4 + arch/s390/include/asm/percpu.h | 28 +-- arch/x86/mm/numa_emulation.c | 2 +- drivers/ata/libata-core.c | 5 +- drivers/ata/libata-scsi.c | 8 +- drivers/cpuidle/cpuidle.c | 3 +- drivers/dma/at_hdmac.c | 2 + drivers/dma/of-dma.c | 8 +- drivers/dma/pl330.c | 2 +- drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 12 +- drivers/gpu/drm/msm/msm_drv.c | 8 + drivers/hid/hid-core.c | 15 +- drivers/hid/hid-input.c | 4 + drivers/hid/hid-multitouch.c | 2 + drivers/hwmon/applesmc.c | 31 ++-- drivers/iommu/intel_irq_remapping.c | 10 +- drivers/md/dm-writecache.c | 12 +- drivers/media/rc/rc-main.c | 44 +++-- drivers/net/ethernet/arc/emac_mdio.c | 1 + drivers/net/ethernet/broadcom/bcmsysport.c | 6 +- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 26 +-- .../net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 5 +- drivers/net/ethernet/broadcom/tg3.c | 17 +- drivers/net/ethernet/cortina/gemini.c | 34 ++-- drivers/net/ethernet/hisilicon/hns/hns_enet.c | 9 +- drivers/net/ethernet/mellanox/mlx4/mr.c | 2 +- drivers/net/ethernet/renesas/ravb_main.c | 110 ++++++------ drivers/net/gtp.c | 1 + drivers/net/usb/asix_common.c | 2 +- drivers/net/usb/qmi_wwan.c | 2 + drivers/nvme/target/core.c | 6 + drivers/nvme/target/fc.c | 4 +- drivers/target/target_core_user.c | 15 +- .../ti-soc-thermal/omap4-thermal-data.c | 23 +-- .../thermal/ti-soc-thermal/omap4xxx-bandgap.h | 10 +- drivers/tty/serial/qcom_geni_serial.c | 2 +- drivers/xen/xenbus/xenbus_client.c | 10 +- fs/affs/amigaffs.c | 27 +++ fs/affs/file.c | 26 ++- fs/btrfs/ctree.c | 8 +- fs/btrfs/extent_io.c | 8 +- fs/btrfs/extent_io.h | 6 +- fs/btrfs/ioctl.c | 27 ++- fs/btrfs/volumes.c | 3 +- fs/ceph/file.c | 1 + fs/eventpoll.c | 6 +- fs/ext2/file.c | 6 +- fs/xfs/libxfs/xfs_bmap.c | 2 +- fs/xfs/xfs_file.c | 12 +- include/linux/bvec.h | 9 +- include/linux/hid.h | 42 +++-- include/linux/libata.h | 1 + include/linux/log2.h | 2 +- include/linux/netfilter/nfnetlink.h | 3 +- include/linux/uaccess.h | 26 +++ include/net/netfilter/nf_tables.h | 2 + include/uapi/linux/netfilter/nf_tables.h | 2 +- mm/maccess.c | 167 ++++++++++++++++-- net/batman-adv/bat_v_ogm.c | 11 +- net/batman-adv/bridge_loop_avoidance.c | 5 +- net/batman-adv/gateway_client.c | 6 +- net/netfilter/nf_tables_api.c | 64 ++++--- net/netfilter/nfnetlink.c | 11 +- net/netfilter/nfnetlink_log.c | 3 +- net/netfilter/nfnetlink_queue.c | 2 +- net/netfilter/nft_payload.c | 4 +- net/wireless/reg.c | 3 + scripts/checkpatch.pl | 4 +- sound/core/oss/mulaw.c | 4 +- sound/firewire/digi00x/digi00x.c | 5 + sound/pci/ca0106/ca0106_main.c | 3 +- sound/pci/hda/patch_hdmi.c | 1 + sound/pci/hda/patch_realtek.c | 1 + tools/include/uapi/linux/perf_event.h | 2 +- tools/perf/Documentation/perf-record.txt | 4 + tools/perf/Documentation/perf-stat.txt | 4 + tools/perf/pmu-events/jevents.c | 2 +- tools/testing/selftests/bpf/test_maps.c | 2 + 87 files changed, 885 insertions(+), 337 deletions(-) -- 2.25.1

1 79

[PATCH 000/120] backport stable-4.19.143
by Yang Yingliang 22 Sep '20

22 Sep '20

Aditya Pakki (5): drm/radeon: fix multiple reference count leak omapfb: fix multiple reference count leaks due to pm_runtime_get_sync drm/nouveau/drm/noveau: fix reference count leak in nouveau_fbcon_open drm/nouveau: fix reference count leak in nv50_disp_atomic_commit drm/nouveau: Fix reference count leak in nouveau_connector_detect Adrian Hunter (1): scsi: ufs: Improve interrupt handling for shared interrupts Alan Stern (3): USB: yurex: Fix bad gfp argument USB: quirks: Ignore duplicate endpoint on Sound Devices MixPre-D usb: storage: Add unusual_uas entry for Sony PSZ drives Alex Deucher (1): drm/amdgpu: Fix buffer overflow in INFO ioctl Alexey Kardashevskiy (1): powerpc/xive: Ignore kmemleak false positives Alvin Šipraga (1): macvlan: validate setting of multiple remote source MAC addresses Amelie Delaunay (1): spi: stm32: fix stm32_spi_prepare_mbr in case of odd clk_rate Andrey Konovalov (1): efi: provide empty efi_enter_virtual_mode implementation Andy Shevchenko (2): mfd: intel-lpss: Add Intel Emmitsburg PCH PCI IDs USB: gadget: u_f: Unbreak offset calculation in VLAs Arnd Bergmann (1): powerpc/spufs: add CONFIG_COREDUMP dependency Athira Rajeev (1): powerpc/perf: Fix soft lockups due to missed interrupt accounting Bodo Stroesser (1): scsi: target: tcmu: Fix crash on ARM during cmd completion Brooke Basile (2): USB: gadget: u_f: add overflow checks to VLA macros USB: gadget: f_ncm: add bounds checks to ncm_unwrap_ntb() Changming Liu (1): USB: sisusbvga: Fix a potential UB casued by left shifting a negative value Chao Yu (1): f2fs: fix error path in do_recover_data() Chris Wilson (1): locking/lockdep: Fix overflow in presentation of average lock-time Christophe JAILLET (1): usb: gadget: f_tcm: Fix some resource leaks in some error paths Cong Wang (1): tipc: fix uninit skb->data in tipc_nl_compat_dumpit() Cyril Roelandt (1): USB: Ignore UAS for JMicron JMS567 ATA/ATAPI Bridge Dave Chinner (1): xfs: Don't allow logging of XFS_ISTALE inodes David Brazdil (1): KVM: arm64: Fix symbol dependency in __hyp_call_panic_nvhe Desnes A. Nunes do Rosario (1): selftests/powerpc: Purge extra count_pmc() calls of ebb selftests Dick Kennedy (1): scsi: lpfc: Fix shost refcount mismatch when deleting vport Ding Hui (1): xhci: Always restore EP_SOFT_CLEAR_TOGGLE even if ep reset failed Evan Quan (2): drm/amd/pm: correct Vega10 swctf limit setting drm/amd/pm: correct Vega12 swctf limit setting Evgeny Novikov (1): USB: lvtest: return proper error code in probe Filipe Manana (1): btrfs: fix space cache memory leak after transaction abort George Kennedy (2): fbcon: prevent user font height or width change from causing potential out-of-bounds access vt_ioctl: change VT_RESIZEX ioctl to check for error return from vc_resize() Greg Kroah-Hartman (1): Linux 4.19.143 Hans Verkuil (1): cec-api: prevent leaking memory through hole in structure Hans de Goede (1): HID: i2c-hid: Always sleep 60ms after I2C_HID_PWR_ON commands Hector Martin (1): ALSA: usb-audio: Update documentation comment for MS2109 quirk Heikki Krogerus (1): device property: Fix the secondary firmware node handling in set_primary_fwnode() Hou Pu (1): null_blk: fix passing of REQ_FUA flag in null_handle_rq Ikjoon Jang (1): HID: quirks: add NOGET quirk for Logitech GROUP Jan Kara (4): ext4: don't BUG on inconsistent journal feature writeback: Protect inode->i_io_list with inode->i_lock writeback: Avoid skipping inode writeback writeback: Fix sync livelock due to b_dirty_time processing Jarkko Sakkinen (1): tpm: Unify the mismatching TPM space buffer sizes Jason Baron (1): EDAC/ie31200: Fallback if host bridge device is already initialized Javed Hasan (1): scsi: fcoe: Memory leak fix in fcoe_sysfs_fcf_del() Jia-Ju Bai (1): media: pci: ttpci: av7110: fix possible buffer overflow caused by bad DMA value in debiirq() Jing Xiangfeng (1): scsi: iscsi: Do not put host in iscsi_set_flashnode_param() Josef Bacik (1): btrfs: check the right error variable in btrfs_del_dir_entries_in_log Kai-Heng Feng (2): xhci: Do warm-reset when both CAS and XDEV_RESUME are set USB: quirks: Add no-lpm quirk for another Raydium touchscreen Li Guifu (1): f2fs: fix use-after-free issue Li Jun (1): usb: host: xhci: fix ep context print mismatch in debugfs Lukas Czerner (3): jbd2: make sure jh have b_transaction set in refile/unfile_buffer ext4: handle read only external journal device ext4: handle option set by mount flags correctly Lukas Wunner (2): serial: pl011: Fix oops on -EPROBE_DEFER serial: pl011: Don't leak amba_ports entry on driver register error Mahesh Bandewar (1): ipvlan: fix device features Marcos Paulo de Souza (1): btrfs: reset compression level for lzo on remount Mark Tomlinson (1): gre6: Fix reception with IP6_TNL_F_RCV_DSCP_COPY Miaohe Lin (1): net: Fix potential wrong skb->protocol in skb_vlan_untag() Michael Ellerman (1): powerpc/64s: Don't init FSCR_DSCR in __init_FSCR() Mike Christie (1): scsi: fcoe: Fix I/O path allocation Ming Lei (1): blk-mq: order adding requests to hctx->dispatch and checking SCHED_RESTART Navid Emamdoost (4): drm/amdgpu: fix ref count leak in amdgpu_driver_open_kms drm/amd/display: fix ref count leak in amdgpu_drm_ioctl drm/amdgpu: fix ref count leak in amdgpu_display_crtc_set_config drm/amdgpu/display: fix ref count leak when pm_runtime_get_sync fails Necip Fazil Yildiran (1): net: qrtr: fix usage of idr in port assignment to socket Peilin Ye (2): net/smc: Prevent kernel-infoleak in __smc_diag_dump() HID: hiddev: Fix slab-out-of-bounds write in hiddev_ioctl_usage() Peng Fan (1): mips/vdso: Fix resource leaks in genvdso.c Qiushi Wu (5): ASoC: img: Fix a reference count leak in img_i2s_in_set_fmt ASoC: img-parallel-out: Fix a reference count leak ASoC: tegra: Fix reference count leaks. drm/amdkfd: Fix reference count leaks. PCI: Fix pci_create_slot() reference count leak Qu Wenruo (1): btrfs: file: reserve qgroup space after the hole punch range is locked Quinn Tran (1): scsi: qla2xxx: Fix null pointer access during disconnect from subsystem Rafael J. Wysocki (1): PM: sleep: core: Fix the handling of pending runtime resume requests Randy Dunlap (1): ALSA: pci: delete repeated words in comments Reto Schneider (1): rtlwifi: rtl8192cu: Prevent leaking urb Rob Clark (1): drm/msm/adreno: fix updating ring fence Robin Murphy (1): iommu/iova: Don't BUG on invalid PFNs Saurav Kashyap (2): scsi: qla2xxx: Check if FW supports MQ before enabling Revert "scsi: qla2xxx: Fix crash on qla2x00_mailbox_command" Sean Young (1): media: gpio-ir-tx: improve precision of transmitted signal due to scheduling Sergey Senozhatsky (1): serial: 8250: change lock order in serial8250_do_startup() Shay Agroskin (1): net: ena: Make missed_tx stat incremental Stanley Chu (2): scsi: ufs: Fix possible infinite loop in ufshcd_hold scsi: ufs: Clean up completed request without interrupt notification Stephan Gerhold (1): arm64: dts: qcom: msm8916: Pull down PDM GPIOs during sleep Sumera Priyadarsini (1): net: gianfar: Add of_node_put() before goto statement Sylwester Nawrocki (1): ASoC: wm8994: Avoid attempts to read unreadable registers Tamseel Shams (1): serial: samsung: Removes the IRQ not found warning Tang Bin (1): usb: host: ohci-exynos: Fix error handling in exynos_ohci_probe() Tetsuo Handa (1): vt: defer kfree() of vc_screenbuf in vc_do_resize() Thinh Nguyen (4): usb: uas: Add quirk for PNY Pro Elite usb: dwc3: gadget: Don't setup more than requested usb: dwc3: gadget: Fix handling ZLP usb: dwc3: gadget: Handle ZLP for sg requests Thomas Gleixner (2): XEN uses irqdesc::irq_data_common::handler_data to store a per interrupt XEN data pointer which contains XEN specific information. genirq/matrix: Deal with the sillyness of for_each_cpu() on UP Tianjia Zhang (1): nvme-fc: Fix wrong return value in __nvme_fc_init_request() Tom Rix (1): USB: cdc-acm: rework notification_buffer resizing Valmer Huhn (1): serial: 8250_exar: Fix number of ports for Commtech PCIe cards Vineeth Vijayan (1): s390/cio: add cond_resched() in the slow_eval_known_fn() loop Wolfram Sang (1): i2c: rcar: in slave mode, clear NACK earlier Xianting Tian (1): fs: prevent BUG_ON in submit_bh_wbc() Xiubo Li (1): ceph: fix potential mdsc use-after-free crash Yangbo Lu (1): ARM: dts: ls1021a: output PPS signal on FIPER2 Zhi Chen (1): Revert "ath10k: fix DMA related firmware crashes on multiple devices" qiuguorui1 (1): irqchip/stm32-exti: Avoid losing interrupts due to clearing pending bits by mistake Makefile | 2 +- arch/arm/boot/dts/ls1021a.dtsi | 2 +- arch/arm64/boot/dts/qcom/msm8916-pins.dtsi | 2 +- arch/arm64/kvm/hyp/switch.c | 2 +- arch/mips/vdso/genvdso.c | 10 ++ arch/powerpc/kernel/cpu_setup_power.S | 2 +- arch/powerpc/perf/core-book3s.c | 4 + arch/powerpc/platforms/cell/Kconfig | 1 + arch/powerpc/sysdev/xive/native.c | 2 + block/blk-mq-sched.c | 9 ++ block/blk-mq.c | 9 ++ drivers/base/core.c | 12 +- drivers/base/power/main.c | 16 +- drivers/block/null_blk_main.c | 2 +- drivers/char/tpm/tpm-chip.c | 9 +- drivers/char/tpm/tpm.h | 6 +- drivers/char/tpm/tpm2-space.c | 26 ++-- drivers/char/tpm/tpmrm-dev.c | 2 +- drivers/edac/ie31200_edac.c | 50 +++++- .../gpu/drm/amd/amdgpu/amdgpu_connectors.c | 16 +- drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 5 +- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 +- drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 7 +- drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 20 ++- .../drm/amd/powerplay/hwmgr/vega10_thermal.c | 7 +- .../drm/amd/powerplay/hwmgr/vega12_thermal.c | 6 +- drivers/gpu/drm/msm/adreno/adreno_gpu.c | 2 +- drivers/gpu/drm/nouveau/dispnv50/disp.c | 4 +- drivers/gpu/drm/nouveau/nouveau_connector.c | 4 +- drivers/gpu/drm/nouveau/nouveau_fbcon.c | 4 +- drivers/gpu/drm/radeon/radeon_connectors.c | 20 ++- drivers/hid/hid-ids.h | 1 + drivers/hid/hid-quirks.c | 1 + drivers/hid/i2c-hid/i2c-hid-core.c | 22 +-- drivers/hid/usbhid/hiddev.c | 4 + drivers/i2c/busses/i2c-rcar.c | 1 + drivers/iommu/iova.c | 4 +- drivers/irqchip/irq-stm32-exti.c | 14 +- drivers/media/cec/cec-api.c | 8 +- drivers/media/pci/ttpci/av7110.c | 5 +- drivers/media/rc/gpio-ir-tx.c | 7 +- drivers/mfd/intel-lpss-pci.c | 3 + drivers/net/ethernet/amazon/ena/ena_netdev.c | 5 +- drivers/net/ethernet/freescale/gianfar.c | 4 +- drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c | 2 +- drivers/net/ipvlan/ipvlan_main.c | 27 +++- drivers/net/macvlan.c | 21 ++- drivers/net/wireless/ath/ath10k/hw.h | 2 +- drivers/net/wireless/realtek/rtlwifi/usb.c | 5 +- drivers/nvme/host/fc.c | 4 +- drivers/pci/slot.c | 6 +- drivers/s390/cio/css.c | 5 + drivers/scsi/fcoe/fcoe_ctlr.c | 2 +- drivers/scsi/lpfc/lpfc_vport.c | 26 +--- drivers/scsi/qla2xxx/qla_mbx.c | 8 - drivers/scsi/qla2xxx/qla_nvme.c | 5 + drivers/scsi/qla2xxx/qla_os.c | 5 + drivers/scsi/scsi_transport_iscsi.c | 2 +- drivers/scsi/ufs/ufshcd.c | 14 +- drivers/spi/spi-stm32.c | 3 +- drivers/target/target_core_user.c | 9 +- drivers/tty/serial/8250/8250_exar.c | 24 ++- drivers/tty/serial/8250/8250_port.c | 9 +- drivers/tty/serial/amba-pl011.c | 16 +- drivers/tty/serial/samsung.c | 8 +- drivers/tty/vt/vt.c | 5 +- drivers/tty/vt/vt_ioctl.c | 12 +- drivers/usb/class/cdc-acm.c | 22 ++- drivers/usb/core/quirks.c | 7 + drivers/usb/dwc3/gadget.c | 104 ++++++++++--- drivers/usb/gadget/function/f_ncm.c | 81 ++++++++-- drivers/usb/gadget/function/f_tcm.c | 7 +- drivers/usb/gadget/u_f.h | 38 +++-- drivers/usb/host/ohci-exynos.c | 5 +- drivers/usb/host/xhci-debugfs.c | 8 +- drivers/usb/host/xhci-hub.c | 19 +-- drivers/usb/host/xhci.c | 3 +- drivers/usb/misc/lvstest.c | 2 +- drivers/usb/misc/sisusbvga/sisusb.c | 2 +- drivers/usb/misc/yurex.c | 2 +- drivers/usb/storage/unusual_devs.h | 2 +- drivers/usb/storage/unusual_uas.h | 14 ++ drivers/video/fbdev/core/fbcon.c | 25 ++- drivers/video/fbdev/omap2/omapfb/dss/dispc.c | 7 +- drivers/video/fbdev/omap2/omapfb/dss/dsi.c | 7 +- drivers/video/fbdev/omap2/omapfb/dss/dss.c | 7 +- drivers/video/fbdev/omap2/omapfb/dss/hdmi4.c | 5 +- drivers/video/fbdev/omap2/omapfb/dss/hdmi5.c | 5 +- drivers/video/fbdev/omap2/omapfb/dss/venc.c | 7 +- drivers/xen/events/events_base.c | 16 +- fs/btrfs/disk-io.c | 1 + fs/btrfs/file.c | 8 +- fs/btrfs/free-space-cache.c | 2 +- fs/btrfs/super.c | 1 + fs/btrfs/tree-log.c | 10 +- fs/buffer.c | 9 ++ fs/ceph/mds_client.c | 14 +- fs/ext4/super.c | 147 ++++++++++++------ fs/f2fs/f2fs.h | 4 +- fs/f2fs/inline.c | 19 ++- fs/f2fs/node.c | 6 +- fs/f2fs/recovery.c | 10 +- fs/f2fs/super.c | 5 +- fs/fs-writeback.c | 83 +++++----- fs/jbd2/transaction.c | 10 ++ fs/xfs/xfs_icache.c | 3 +- fs/xfs/xfs_inode.c | 25 ++- fs/xfs/xfs_trans_inode.c | 2 + include/linux/efi.h | 4 + include/linux/fs.h | 8 +- include/trace/events/writeback.h | 13 +- kernel/irq/matrix.c | 7 + kernel/locking/lockdep_proc.c | 2 +- net/core/skbuff.c | 4 +- net/ipv6/ip6_tunnel.c | 10 +- net/qrtr/qrtr.c | 20 +-- net/smc/smc_diag.c | 16 +- net/tipc/netlink_compat.c | 12 +- sound/pci/cs46xx/cs46xx_lib.c | 2 +- sound/pci/cs46xx/dsp_spos_scb_lib.c | 2 +- sound/pci/hda/hda_codec.c | 2 +- sound/pci/hda/hda_generic.c | 2 +- sound/pci/hda/patch_sigmatel.c | 2 +- sound/pci/ice1712/prodigy192.c | 2 +- sound/pci/oxygen/xonar_dg.c | 2 +- sound/soc/codecs/wm8958-dsp2.c | 4 + sound/soc/img/img-i2s-in.c | 4 +- sound/soc/img/img-parallel-out.c | 4 +- sound/soc/tegra/tegra30_ahub.c | 4 +- sound/soc/tegra/tegra30_i2s.c | 4 +- sound/usb/quirks-table.h | 4 +- .../powerpc/pmu/ebb/back_to_back_ebbs_test.c | 2 - .../selftests/powerpc/pmu/ebb/cycles_test.c | 2 - .../powerpc/pmu/ebb/cycles_with_freeze_test.c | 2 - .../powerpc/pmu/ebb/cycles_with_mmcr2_test.c | 2 - tools/testing/selftests/powerpc/pmu/ebb/ebb.c | 2 - .../pmu/ebb/ebb_on_willing_child_test.c | 2 - .../powerpc/pmu/ebb/lost_exception_test.c | 1 - .../powerpc/pmu/ebb/multi_counter_test.c | 7 - .../powerpc/pmu/ebb/multi_ebb_procs_test.c | 2 - .../powerpc/pmu/ebb/pmae_handling_test.c | 2 - .../powerpc/pmu/ebb/pmc56_overflow_test.c | 2 - 142 files changed, 1037 insertions(+), 442 deletions(-) -- 2.25.1

1 120

[PATCH 01/37] arm64/ascend: enable ascend features for Ascend910 platform
by Yang Yingliang 22 Sep '20

22 Sep '20

From: Ding Tianhong <dingtianhong(a)huawei.com> ascend inclusion category: feature bugzilla: NA CVE: NA ------------------------------------------------- The Ascend910 platform only use the acpi mode to boot system, the oem message is recorded in the oem_table_id of IORT table, so use the oem message to enable the ascend features. Signed-off-by: Ding Tianhong <dingtianhong(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/mm/init.c | 7 ++++++- drivers/acpi/arm64/iort.c | 24 ++++++++++++++++++++++++ include/linux/init.h | 7 +++++++ 3 files changed, 37 insertions(+), 1 deletion(-) diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index 883350f9cc42..f23da539a476 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -769,7 +769,7 @@ __setup("keepinitrd", keepinitrd_setup); #endif #ifdef CONFIG_ASCEND_FEATURES -static int __init ascend_enable_setup(char *__unused) +void ascend_enable_all_features(void) { if (IS_ENABLED(CONFIG_ASCEND_DVPP_MMAP)) enable_mmap_dvpp = 1; @@ -782,6 +782,11 @@ static int __init ascend_enable_setup(char *__unused) if (IS_ENABLED(CONFIG_SUSPEND)) mem_sleep_current = PM_SUSPEND_ON; +} + +static int __init ascend_enable_setup(char *__unused) +{ + ascend_enable_all_features(); return 1; } diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c index 5a724bdb4a4d..7408ffab7303 100644 --- a/drivers/acpi/arm64/iort.c +++ b/drivers/acpi/arm64/iort.c @@ -25,6 +25,7 @@ #include <linux/pci.h> #include <linux/platform_device.h> #include <linux/slab.h> +#include <linux/init.h> #define IORT_TYPE_MASK(type) (1 << (type)) #define IORT_MSI_TYPE (1 << ACPI_IORT_NODE_ITS_GROUP) @@ -1639,6 +1640,26 @@ static void __init iort_init_platform_devices(void) } } +/* + * This function detects the ascend platform by oem table id. + */ +static bool ascend_platform_detected(struct acpi_table_header *h) +{ + if (!memcmp(h->oem_table_id, "HI19801P", ACPI_OEM_TABLE_ID_SIZE)) + return true; + + if (!memcmp(h->oem_table_id, "HI19802P", ACPI_OEM_TABLE_ID_SIZE)) + return true; + + if (!memcmp(h->oem_table_id, "HI19804P", ACPI_OEM_TABLE_ID_SIZE)) + return true; + + if (!memcmp(h->oem_table_id, "HI1980\0\0", ACPI_OEM_TABLE_ID_SIZE)) + return true; + + return false; +} + void __init acpi_iort_init(void) { acpi_status status; @@ -1654,5 +1675,8 @@ void __init acpi_iort_init(void) return; } + if (ascend_platform_detected(iort_table)) + ascend_enable_all_features(); + iort_init_platform_devices(); } diff --git a/include/linux/init.h b/include/linux/init.h index 2538d176dd1f..e6f970ac7bf4 100644 --- a/include/linux/init.h +++ b/include/linux/init.h @@ -306,4 +306,11 @@ void __init parse_early_options(char *cmdline); #define __exit_p(x) NULL #endif +#ifndef __ASSEMBLY__ +#ifdef CONFIG_ASCEND_FEATURES +extern void ascend_enable_all_features(void); +#else +static inline void ascend_enable_all_features(void) { } +#endif +#endif #endif /* _LINUX_INIT_H */ -- 2.25.1

1 36

[PATCH] export walk_page_range and tlb flush function
by Miaohe Lin 18 Sep '20

18 Sep '20

From: yanxiaodan <yanxiaodan(a)huawei.com> Export walk_page_range,__kvm_tlb_flush_vmid and flush_tlb_mm_range, so that memory-scan module can use these functions to scan the memory pages and flush TLB. memory-scan link: https://gitee.com/openeuler/memory-scan Signed-off-by: yanxiaodan <yanxiaodan(a)huawei.com> --- arch/arm64/kvm/hyp/tlb.c | 1 + arch/x86/mm/tlb.c | 2 +- mm/pagewalk.c | 1 + 3 files changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/arm64/kvm/hyp/tlb.c b/arch/arm64/kvm/hyp/tlb.c index d063a57..7158622 100644 --- a/arch/arm64/kvm/hyp/tlb.c +++ b/arch/arm64/kvm/hyp/tlb.c @@ -205,6 +205,7 @@ void __hyp_text __kvm_tlb_flush_vmid(struct kvm *kvm) __tlb_switch_to_host(kvm, &cxt); } +EXPORT_SYMBOL(__kvm_tlb_flush_vmid); void __hyp_text __kvm_tlb_flush_local_vmid(struct kvm_vcpu *vcpu) { diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index 1a3569b..be19bf5 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -945,7 +945,7 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, put_flush_tlb_info(); put_cpu(); } - +EXPORT_SYMBOL(flush_tlb_mm_range); static void do_flush_tlb_all(void *info) { diff --git a/mm/pagewalk.c b/mm/pagewalk.c index e81640d..44a9ab1 100644 --- a/mm/pagewalk.c +++ b/mm/pagewalk.c @@ -430,6 +430,7 @@ int walk_page_range(struct mm_struct *mm, unsigned long start, } while (start = next, start < end); return err; } +EXPORT_SYMBOL(walk_page_range); /* * Similar to walk_page_range() but can walk any page tables even if they are -- 1.8.3.1

1 0

【Meeting Notice】openEuler kernel sig meeting Time: 2020-09-18 10:00-12:00
by Meeting Book 17 Sep '20

17 Sep '20

5 7

[PATCH v2] delete python2 dependency
by Zhipeng Xie 15 Sep '20

15 Sep '20

hulk inclusion category: tools bugzilla: NA CVE: NA Link: https://gitee.com/src-openeuler/kernel/issues/I1SMDG ------------------------------------------------- Python2 is no longer supported by the upstream community. The dependency on python2 should be removed from the kernel code. Signed-off-by: Zhipeng Xie <xiezhipeng1(a)huawei.com> --- tools/perf/scripts/python/call-graph-from-sql.py | 2 +- tools/perf/scripts/python/export-to-postgresql.py | 2 +- tools/power/pm-graph/bootgraph.py | 2 +- tools/power/pm-graph/sleepgraph.py | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/tools/perf/scripts/python/call-graph-from-sql.py b/tools/perf/scripts/python/call-graph-from-sql.py index b494a67a1c67..099b472df4a2 100644 --- a/tools/perf/scripts/python/call-graph-from-sql.py +++ b/tools/perf/scripts/python/call-graph-from-sql.py @@ -1,4 +1,4 @@ -#!/usr/bin/python2 +#!/usr/bin/python # call-graph-from-sql.py: create call-graph from sql database # Copyright (c) 2014-2017, Intel Corporation. # diff --git a/tools/perf/scripts/python/export-to-postgresql.py b/tools/perf/scripts/python/export-to-postgresql.py index e46f51b17513..e97a03697aa2 100644 --- a/tools/perf/scripts/python/export-to-postgresql.py +++ b/tools/perf/scripts/python/export-to-postgresql.py @@ -171,7 +171,7 @@ import datetime # SELECT * FROM samples_view WHERE event = 'transactions' AND branch_type_name = 'transaction abort'; # # To print a call stack requires walking the call_paths table. For example this python script: -# #!/usr/bin/python2 +# #!/usr/bin/python # # import sys # from PySide.QtSql import * diff --git a/tools/power/pm-graph/bootgraph.py b/tools/power/pm-graph/bootgraph.py index 8ee626c0f6a5..abb4c38f029b 100755 --- a/tools/power/pm-graph/bootgraph.py +++ b/tools/power/pm-graph/bootgraph.py @@ -1,4 +1,4 @@ -#!/usr/bin/python2 +#!/usr/bin/python # # Tool for analyzing boot timing # Copyright (c) 2013, Intel Corporation. diff --git a/tools/power/pm-graph/sleepgraph.py b/tools/power/pm-graph/sleepgraph.py index 0c760478f7d7..420102e2cd08 100755 --- a/tools/power/pm-graph/sleepgraph.py +++ b/tools/power/pm-graph/sleepgraph.py @@ -1,4 +1,4 @@ -#!/usr/bin/python2 +#!/usr/bin/python # # Tool for analyzing suspend/resume timing # Copyright (c) 2013, Intel Corporation. -- 2.18.1

3 2

[PATCH] arm64: configs: enable CONFIG_NUMA_AWARE_SPINLOCKS
by Yang Yingliang 15 Sep '20

15 Sep '20

hulk inclusion category: feature CVE: NA ----------------------- enable CONFIG_NUMA_AWARE_SPINLOCKS. Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/configs/euleros_defconfig | 3 ++- arch/arm64/configs/hulk_defconfig | 2 +- arch/arm64/configs/openeuler_defconfig | 2 +- 3 files changed, 4 insertions(+), 3 deletions(-) diff --git a/arch/arm64/configs/euleros_defconfig b/arch/arm64/configs/euleros_defconfig index 09c604738074..2762ee8afe05 100644 --- a/arch/arm64/configs/euleros_defconfig +++ b/arch/arm64/configs/euleros_defconfig @@ -418,6 +418,7 @@ CONFIG_ARM64_ERR_RECOV=y CONFIG_MPAM=y CONFIG_NUMA=y CONFIG_NODES_SHIFT=3 +CONFIG_NUMA_AWARE_SPINLOCKS=y CONFIG_USE_PERCPU_NUMA_NODE_ID=y CONFIG_HAVE_SETUP_PER_CPU_AREA=y CONFIG_NEED_PER_CPU_EMBED_FIRST_CHUNK=y @@ -440,7 +441,7 @@ CONFIG_ARCH_WANT_HUGE_PMD_SHARE=y CONFIG_ARCH_HAS_CACHE_LINE_SIZE=y CONFIG_SECCOMP=y CONFIG_PARAVIRT=y -# CONFIG_PARAVIRT_SPINLOCKS is not set +CONFIG_PARAVIRT_SPINLOCKS=y CONFIG_PARAVIRT_TIME_ACCOUNTING=y CONFIG_KEXEC=y CONFIG_CRASH_DUMP=y diff --git a/arch/arm64/configs/hulk_defconfig b/arch/arm64/configs/hulk_defconfig index 9952d0b89744..4979492bb484 100644 --- a/arch/arm64/configs/hulk_defconfig +++ b/arch/arm64/configs/hulk_defconfig @@ -440,7 +440,7 @@ CONFIG_ARCH_WANT_HUGE_PMD_SHARE=y CONFIG_ARCH_HAS_CACHE_LINE_SIZE=y CONFIG_SECCOMP=y CONFIG_PARAVIRT=y -# CONFIG_PARAVIRT_SPINLOCKS is not set +CONFIG_PARAVIRT_SPINLOCKS=y CONFIG_PARAVIRT_TIME_ACCOUNTING=y CONFIG_KEXEC=y CONFIG_CRASH_DUMP=y diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index beb292fa9b47..809350e03968 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -421,7 +421,7 @@ CONFIG_ARM64_ERR_RECOV=y CONFIG_MPAM=y CONFIG_NUMA=y CONFIG_NODES_SHIFT=4 -# CONFIG_NUMA_AWARE_SPINLOCKS is not set +CONFIG_NUMA_AWARE_SPINLOCKS=y CONFIG_USE_PERCPU_NUMA_NODE_ID=y CONFIG_HAVE_SETUP_PER_CPU_AREA=y CONFIG_NEED_PER_CPU_EMBED_FIRST_CHUNK=y -- 2.25.1

1 0

[PATCH 1/2] random: only read from /dev/random after its pool has received 128 bits
by Yang Yingliang 15 Sep '20

15 Sep '20

From: Theodore Ts'o <tytso(a)mit.edu> mainline inclusion from mainline-v5.2-rc1 commit eb9d1bf079bb438d1a066d72337092935fc770f6 category: bugfix bugzilla: NA CVE: NA https://gitee.com/openeuler/kernel/issues/I1CVQ4?from=project-issue ------------------------------------------------- Immediately after boot, we allow reads from /dev/random before its entropy pool has been fully initialized. Fix this so that we don't allow this until the blocking pool has received 128 bits. We do this by repurposing the initialized flag in the entropy pool struct, and use the initialized flag in the blocking pool to indicate whether it is safe to pull from the blocking pool. To do this, we needed to rework when we decide to push entropy from the input pool to the blocking pool, since the initialized flag for the input pool was used for this purpose. To simplify things, we no longer use the initialized flag for that purpose, nor do we use the entropy_total field any more. Signed-off-by: Theodore Ts'o <tytso(a)mit.edu> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/char/random.c | 44 +++++++++++++++++------------------ include/trace/events/random.h | 13 ++++------- 2 files changed, 27 insertions(+), 30 deletions(-) diff --git a/drivers/char/random.c b/drivers/char/random.c index 6a5d4dfafc47..750d2ab2ac54 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -471,7 +471,6 @@ struct entropy_store { unsigned short add_ptr; unsigned short input_rotate; int entropy_count; - int entropy_total; unsigned int initialized:1; unsigned int last_data_init:1; __u8 last_data[EXTRACT_SIZE]; @@ -644,7 +643,7 @@ static void process_random_ready_list(void) */ static void credit_entropy_bits(struct entropy_store *r, int nbits) { - int entropy_count, orig; + int entropy_count, orig, has_initialized = 0; const int pool_size = r->poolinfo->poolfracbits; int nfrac = nbits << ENTROPY_SHIFT; @@ -699,23 +698,25 @@ static void credit_entropy_bits(struct entropy_store *r, int nbits) entropy_count = 0; } else if (entropy_count > pool_size) entropy_count = pool_size; + if ((r == &blocking_pool) && !r->initialized && + (entropy_count >> ENTROPY_SHIFT) > 128) + has_initialized = 1; if (cmpxchg(&r->entropy_count, orig, entropy_count) != orig) goto retry; - r->entropy_total += nbits; - if (!r->initialized && r->entropy_total > 128) { + if (has_initialized) r->initialized = 1; - r->entropy_total = 0; - } trace_credit_entropy_bits(r->name, nbits, - entropy_count >> ENTROPY_SHIFT, - r->entropy_total, _RET_IP_); + entropy_count >> ENTROPY_SHIFT, _RET_IP_); if (r == &input_pool) { int entropy_bits = entropy_count >> ENTROPY_SHIFT; + struct entropy_store *other = &blocking_pool; - if (crng_init < 2 && entropy_bits >= 128) { + if (crng_init < 2) { + if (entropy_bits < 128) + return; crng_reseed(&primary_crng, r); entropy_bits = r->entropy_count >> ENTROPY_SHIFT; } @@ -726,20 +727,14 @@ static void credit_entropy_bits(struct entropy_store *r, int nbits) wake_up_interruptible(&random_read_wait); kill_fasync(&fasync, SIGIO, POLL_IN); } - /* If the input pool is getting full, send some - * entropy to the blocking pool until it is 75% full. + /* If the input pool is getting full, and the blocking + * pool has room, send some entropy to the blocking + * pool. */ - if (entropy_bits > random_write_wakeup_bits && - r->initialized && - r->entropy_total >= 2*random_read_wakeup_bits) { - struct entropy_store *other = &blocking_pool; - - if (other->entropy_count <= - 3 * other->poolinfo->poolfracbits / 4) { - schedule_work(&other->push_work); - r->entropy_total = 0; - } - } + if (!work_pending(&other->push_work) && + (ENTROPY_BITS(r) > 6 * r->poolinfo->poolbytes) && + (ENTROPY_BITS(other) <= 6 * other->poolinfo->poolbytes)) + schedule_work(&other->push_work); } } @@ -1558,6 +1553,11 @@ static ssize_t extract_entropy_user(struct entropy_store *r, void __user *buf, int large_request = (nbytes > 256); trace_extract_entropy_user(r->name, nbytes, ENTROPY_BITS(r), _RET_IP_); + if (!r->initialized && r->pull) { + xfer_secondary_pool(r, ENTROPY_BITS(r->pull)/8); + if (!r->initialized) + return 0; + } xfer_secondary_pool(r, nbytes); nbytes = account(r, nbytes, 0, 0); diff --git a/include/trace/events/random.h b/include/trace/events/random.h index 0560dfc33f1c..32c10a515e2d 100644 --- a/include/trace/events/random.h +++ b/include/trace/events/random.h @@ -62,15 +62,14 @@ DEFINE_EVENT(random__mix_pool_bytes, mix_pool_bytes_nolock, TRACE_EVENT(credit_entropy_bits, TP_PROTO(const char *pool_name, int bits, int entropy_count, - int entropy_total, unsigned long IP), + unsigned long IP), - TP_ARGS(pool_name, bits, entropy_count, entropy_total, IP), + TP_ARGS(pool_name, bits, entropy_count, IP), TP_STRUCT__entry( __field( const char *, pool_name ) __field( int, bits ) __field( int, entropy_count ) - __field( int, entropy_total ) __field(unsigned long, IP ) ), @@ -78,14 +77,12 @@ TRACE_EVENT(credit_entropy_bits, __entry->pool_name = pool_name; __entry->bits = bits; __entry->entropy_count = entropy_count; - __entry->entropy_total = entropy_total; __entry->IP = IP; ), - TP_printk("%s pool: bits %d entropy_count %d entropy_total %d " - "caller %pS", __entry->pool_name, __entry->bits, - __entry->entropy_count, __entry->entropy_total, - (void *)__entry->IP) + TP_printk("%s pool: bits %d entropy_count %d caller %pS", + __entry->pool_name, __entry->bits, + __entry->entropy_count, (void *)__entry->IP) ); TRACE_EVENT(push_to_pool, -- 2.25.1

1 1

[PATCH 0/6] Support E0PD
by Yang Yingliang 14 Sep '20

14 Sep '20

Chen Jun (1): config: set default value of CONFIG_ARM64_E0PD Mark Brown (4): arm64: Add initial support for E0PD arm64: Factor out checks for KASLR in KPTI code into separate function arm64: Don't use KPTI where we have E0PD arm64: Use a variable to store non-global mappings decision Will Deacon (1): arm64: kpti: Fix "kpti=off" when KASLR is enabled arch/arm64/Kconfig | 16 +++++++ arch/arm64/configs/euleros_defconfig | 1 + arch/arm64/configs/hulk_defconfig | 1 + arch/arm64/configs/openeuler_defconfig | 2 + arch/arm64/include/asm/cpucaps.h | 3 +- arch/arm64/include/asm/mmu.h | 4 +- arch/arm64/include/asm/pgtable-hwdef.h | 2 + arch/arm64/include/asm/pgtable-prot.h | 6 ++- arch/arm64/include/asm/sysreg.h | 1 + arch/arm64/kernel/cpufeature.c | 61 ++++++++++++++++++++++++-- arch/arm64/kernel/setup.c | 7 +++ 11 files changed, 95 insertions(+), 9 deletions(-) -- 2.25.1

1 6

[PATCH 0/7] Add auto-tuning for huagepage
by Yang Yingliang 14 Sep '20

14 Sep '20

Weilong Chen (7): arm64/ascend: Add new CONFIG for auto-tuning hugepage arm64/ascend: Add mmap hook when alloc hugepage arm64/ascend: Add set hugepage number helper function arm64/ascend: Add hugepage flags change interface arm64/ascend: Notifier will return a freed val to indecate print logs arm64/ascend: Enable CONFIG_ASCEND_AUTO_TUNING_HUGEPAGE for hulk_defconfig arm64/ascend: Add auto tuning hugepage module arch/arm64/Kconfig | 6 + arch/arm64/configs/hulk_defconfig | 1 + mm/Makefile | 1 + mm/hugepage_tuning.c | 693 ++++++++++++++++++++++++++++++ mm/hugepage_tuning.h | 70 +++ mm/hugetlb.c | 22 + mm/mmap.c | 30 ++ mm/oom_kill.c | 8 +- 8 files changed, 829 insertions(+), 2 deletions(-) create mode 100644 mm/hugepage_tuning.c create mode 100644 mm/hugepage_tuning.h -- 2.25.1

1 7

[PATCH 00/10] Support v8.5-RNG
by Yang Yingliang 11 Sep '20

11 Sep '20

Guenter Roeck (1): arm64: kaslr: Use standard early random function Linus Torvalds (1): random: random.h should include archrandom.h, not the other way around Mark Brown (3): arm64: kaslr: Announce KASLR status on boot arm64: kaslr: Check command line before looking for a seed arm64: Use v8.5-RNG entropy for KASLR seed Mark Rutland (2): arm64: add credited/trusted RNG support random: add arch_get_random_*long_early() Richard Henderson (1): arm64: Implement archrandom.h for ARMv8.5-RNG Robin Murphy (1): arm64: Fix CONFIG_ARCH_RANDOM=n build Yang Yingliang (1): config: set default value of CONFIG_ARCH_RANDOM Documentation/arm64/cpu-feature-registers.txt | 2 + Documentation/arm64/elf_hwcaps.txt | 5 ++ arch/arm64/Kconfig | 12 +++ arch/arm64/configs/euleros_defconfig | 5 ++ arch/arm64/configs/hulk_defconfig | 5 ++ arch/arm64/configs/openeuler_defconfig | 5 ++ arch/arm64/include/asm/archrandom.h | 88 +++++++++++++++++++ arch/arm64/include/asm/cpucaps.h | 3 +- arch/arm64/include/asm/hwcap.h | 1 + arch/arm64/include/asm/sysreg.h | 4 + arch/arm64/include/uapi/asm/hwcap.h | 1 + arch/arm64/kernel/cpufeature.c | 14 +++ arch/arm64/kernel/cpuinfo.c | 1 + arch/arm64/kernel/kaslr.c | 54 +++++++++++- include/linux/random.h | 22 +++++ 15 files changed, 217 insertions(+), 5 deletions(-) create mode 100644 arch/arm64/include/asm/archrandom.h -- 2.25.1

1 10

[PATCH 0/2] Add support for SDEI_WATCHDOG/PMU_WATCHDOG selection
by Yang Yingliang 11 Sep '20

11 Sep '20

Jingyi Wang (2): arm64: watchdog: add switch to select sdei_watchdog/pmu_watchdog arm64: update hulk_defconfig and openeuler_defconfig arch/arm64/configs/hulk_defconfig | 2 +- arch/arm64/configs/openeuler_defconfig | 3 ++- arch/arm64/kernel/watchdog_sdei.c | 23 ++++++++++++++------ include/linux/nmi.h | 11 ++++++++++ kernel/watchdog.c | 29 +++++++++++++++++++------- lib/Kconfig.debug | 7 ++----- 6 files changed, 55 insertions(+), 20 deletions(-) -- 2.25.1

1 2

[PATCH 0/3] Fix use-after-free in dm_bm_set_read_only
by Yang Yingliang 11 Sep '20

11 Sep '20

Ye Bin (3): dm cache metadata: Avoid returning cmd->bm wild pointer on error dm thin metadata: Avoid returning cmd->bm wild pointer on error dm thin metadata: Fix use-after-free in dm_bm_set_read_only drivers/md/dm-cache-metadata.c | 8 ++++++-- drivers/md/dm-thin-metadata.c | 10 +++++++--- drivers/md/persistent-data/dm-block-manager.c | 14 ++++++++------ 3 files changed, 21 insertions(+), 11 deletions(-) -- 2.25.1

1 3

[PATCH] mm: slub: fix conversion of freelist_corrupted()
by Yang Yingliang 11 Sep '20

11 Sep '20

From: Eugeniu Rosca <erosca(a)de.adit-jv.com> mainline inclusion from mainline-v5.9-rc4 commit dc07a728d49cf025f5da2c31add438d839d076c0 category: bugfix bugzilla: NA CVE: NA ------------------------------------------------- Commit 52f23478081ae0 ("mm/slub.c: fix corrupted freechain in deactivate_slab()") suffered an update when picked up from LKML [1]. Specifically, relocating 'freelist = NULL' into 'freelist_corrupted()' created a no-op statement. Fix it by sticking to the behavior intended in the original patch [1]. In addition, make freelist_corrupted() immune to passing NULL instead of &freelist. The issue has been spotted via static analysis and code review. [1] https://lore.kernel.org/linux-mm/20200331031450.12182-1-dongli.zhang@oracle… Fixes: 52f23478081ae0 ("mm/slub.c: fix corrupted freechain in deactivate_slab()") Signed-off-by: Eugeniu Rosca <erosca(a)de.adit-jv.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Cc: Dongli Zhang <dongli.zhang(a)oracle.com> Cc: Joe Jin <joe.jin(a)oracle.com> Cc: Christoph Lameter <cl(a)linux.com> Cc: Pekka Enberg <penberg(a)kernel.org> Cc: David Rientjes <rientjes(a)google.com> Cc: Joonsoo Kim <iamjoonsoo.kim(a)lge.com> Cc: <stable(a)vger.kernel.org> Link: https://lkml.kernel.org/r/20200824130643.10291-1-erosca@de.adit-jv.com Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- mm/slub.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/mm/slub.c b/mm/slub.c index 0415902a2749..887e3d67a4d7 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -646,12 +646,12 @@ static void slab_fix(struct kmem_cache *s, char *fmt, ...) } static bool freelist_corrupted(struct kmem_cache *s, struct page *page, - void *freelist, void *nextfree) + void **freelist, void *nextfree) { if ((s->flags & SLAB_CONSISTENCY_CHECKS) && - !check_valid_pointer(s, page, nextfree)) { - object_err(s, page, freelist, "Freechain corrupt"); - freelist = NULL; + !check_valid_pointer(s, page, nextfree) && freelist) { + object_err(s, page, *freelist, "Freechain corrupt"); + *freelist = NULL; slab_fix(s, "Isolate corrupted freechain"); return true; } @@ -1342,7 +1342,7 @@ static inline void dec_slabs_node(struct kmem_cache *s, int node, int objects) {} static bool freelist_corrupted(struct kmem_cache *s, struct page *page, - void *freelist, void *nextfree) + void **freelist, void *nextfree) { return false; } @@ -2036,7 +2036,7 @@ static void deactivate_slab(struct kmem_cache *s, struct page *page, * 'freelist' is already corrupted. So isolate all objects * starting at 'freelist'. */ - if (freelist_corrupted(s, page, freelist, nextfree)) + if (freelist_corrupted(s, page, &freelist, nextfree)) break; do { -- 2.25.1

1 0

[PATCH 0/7] Support some features of ARMv8.x
by Yang Yingliang 10 Sep '20

10 Sep '20

Support - SVE2 - ARMv8.5-CondM, Condition flag manipulation - ARMv8.5-FRINT, Armv8.5 Floating-point to integer - ARMv8.6 Matrix multiply extension Dave Martin (2): arm64: Expose SVE2 features for userspace arm64: cpufeature: Fix missing ZFR0 in __read_sysreg_by_encoding() Julien Grall (2): arm64: cpufeature: Treat ID_AA64ZFR0_EL1 as RAZ when SVE is not enabled arm64: cpufeature: Effectively expose FRINT capability to userspace Mark Brown (2): arm64: Expose ARMv8.5 CondM capability to userspace arm64: Expose FRINT capabilities to userspace Steven Price (1): arm64: cpufeature: Export matrix and other features to userspace Documentation/arm64/cpu-feature-registers.txt | 30 ++++++++++ Documentation/arm64/elf_hwcaps.txt | 59 +++++++++++++++++++ Documentation/arm64/sve.txt | 17 ++++++ arch/arm64/Kconfig | 3 + arch/arm64/include/asm/hwcap.h | 15 +++++ arch/arm64/include/asm/sysreg.h | 27 +++++++++ arch/arm64/include/uapi/asm/hwcap.h | 15 +++++ arch/arm64/kernel/cpufeature.c | 45 +++++++++++++- arch/arm64/kernel/cpuinfo.c | 15 +++++ 9 files changed, 225 insertions(+), 1 deletion(-) -- 2.25.1

1 7

[PATCH 0/6] Support TLB maintenance and TLB range instructions
by Yang Yingliang 10 Sep '20

10 Sep '20

Anshuman Khandual (1): arm64/cpufeature: Add remaining feature bits in ID_AA64ISAR0 register Sami Tolvanen (1): arm64: use a common .arch preamble for inline assembly Yuan Can (1): config: arm64: update defconfig Zhenyu Ye (3): arm64: tlb: Detect the ARMv8.4 TLBI RANGE feature arm64: enable tlbi range instructions arm64: tlb: Use the TLBI RANGE feature in arm64 arch/arm64/Kconfig | 18 +++ arch/arm64/Makefile | 15 ++- arch/arm64/configs/euleros_defconfig | 5 + arch/arm64/configs/hulk_defconfig | 5 + arch/arm64/configs/openeuler_defconfig | 5 + arch/arm64/include/asm/compiler.h | 6 + arch/arm64/include/asm/cpucaps.h | 3 +- arch/arm64/include/asm/cpufeature.h | 6 + arch/arm64/include/asm/sysreg.h | 4 + arch/arm64/include/asm/tlbflush.h | 160 ++++++++++++++++++++----- arch/arm64/kernel/cpufeature.c | 11 ++ 11 files changed, 205 insertions(+), 33 deletions(-) -- 2.25.1

1 6

[PATCH 0/2] Add interface to configure timeout
by Yang Yingliang 10 Sep '20

10 Sep '20

Wang Wensheng (2): watchdog: Add interface to config timeout and pretimeout in sysfs watchdog: Enable CONFIG_ASCEND_WATCHDOG_SYSFS_CONFIGURE in hulk_defconfig arch/arm64/Kconfig | 10 +++++ arch/arm64/configs/hulk_defconfig | 1 + drivers/watchdog/watchdog_dev.c | 63 +++++++++++++++++++++++++++++++ 3 files changed, 74 insertions(+) -- 2.25.1

1 2

[PATCH 0/6] arm64: Initial support for CVADP
by Yang Yingliang 09 Sep '20

09 Sep '20

ARMv8.5 introduces a DC CVADP instruction which cleans the data cache to the point of deep persistence. This series makes the instruction available to userspace and advertises the presence of this CPU feature. At present when CONFIG_ARM64_PMEM is enabled and the CVAP feature is present (ARMv8.2) the CVAP instruction is used (from memcpy_flushcache and arch_wb_cache_pmem). No changes have been made to use CVADP in these functions or similar. As we have moved beyond 32 capabilities we now begin using AT_HWCAP2 for userspace. Tested as follows: $ dmesg | grep "Deep" [ 0.166496] CPU features: detected: Data cache clean to Point of Deep Persistence $ LD_SHOW_AUXV=1 sleep 2>&1 | grep AT_HWCAP AT_HWCAP: ef91ff87 AT_HWCAP2: 0x1 Andrew Murray (6): arm64: HWCAP: add support for AT_HWCAP2 arm64: HWCAP: encapsulate elf_hwcap arm64: Handle trapped DC CVADP arm64: Expose DC CVADP to userspace arm64: add CVADP support to the cache maintenance helper arm64: Advertise ARM64_HAS_DCPODP cpu feature Documentation/arm64/elf_hwcaps.txt | 17 +++- arch/arm64/crypto/aes-ce-ccm-glue.c | 2 +- arch/arm64/crypto/aes-neonbs-glue.c | 2 +- arch/arm64/crypto/chacha20-neon-glue.c | 2 +- arch/arm64/crypto/ghash-ce-glue.c | 6 +- arch/arm64/crypto/sha256-glue.c | 4 +- arch/arm64/include/asm/assembler.h | 4 + arch/arm64/include/asm/cpucaps.h | 3 +- arch/arm64/include/asm/cpufeature.h | 21 ++--- arch/arm64/include/asm/esr.h | 3 +- arch/arm64/include/asm/hwcap.h | 54 +++++++++++- arch/arm64/include/uapi/asm/hwcap.h | 7 +- arch/arm64/kernel/cpufeature.c | 110 +++++++++++++++++-------- arch/arm64/kernel/cpuinfo.c | 3 +- arch/arm64/kernel/fpsimd.c | 4 +- arch/arm64/kernel/traps.c | 3 + drivers/clocksource/arm_arch_timer.c | 8 ++ 17 files changed, 186 insertions(+), 67 deletions(-) -- 2.25.1

1 6

[PATCH 1/3] mm/swap_state: fix a data race in swapin_nr_pages
by Yang Yingliang 09 Sep '20

09 Sep '20

From: Qian Cai <cai(a)lca.pw> mainline inclusion from mainline-v5.8-rc1 commit d6c1f098f2a7ba62627c9bc17cda28f534ef9e4a category: bugfix bugzilla: 35806 CVE: NA ------------------------------------------------- "prev_offset" is a static variable in swapin_nr_pages() that can be accessed concurrently with only mmap_sem held in read mode as noticed by KCSAN, BUG: KCSAN: data-race in swap_cluster_readahead / swap_cluster_readahead write to 0xffffffff92763830 of 8 bytes by task 14795 on cpu 17: swap_cluster_readahead+0x2a6/0x5e0 swapin_readahead+0x92/0x8dc do_swap_page+0x49b/0xf20 __handle_mm_fault+0xcfb/0xd70 handle_mm_fault+0xfc/0x2f0 do_page_fault+0x263/0x715 page_fault+0x34/0x40 1 lock held by (dnf)/14795: #0: ffff897bd2e98858 (&mm->mmap_sem#2){++++}-{3:3}, at: do_page_fault+0x143/0x715 do_user_addr_fault at arch/x86/mm/fault.c:1405 (inlined by) do_page_fault at arch/x86/mm/fault.c:1535 irq event stamp: 83493 count_memcg_event_mm+0x1a6/0x270 count_memcg_event_mm+0x119/0x270 __do_softirq+0x365/0x589 irq_exit+0xa2/0xc0 read to 0xffffffff92763830 of 8 bytes by task 1 on cpu 22: swap_cluster_readahead+0xfd/0x5e0 swapin_readahead+0x92/0x8dc do_swap_page+0x49b/0xf20 __handle_mm_fault+0xcfb/0xd70 handle_mm_fault+0xfc/0x2f0 do_page_fault+0x263/0x715 page_fault+0x34/0x40 1 lock held by systemd/1: #0: ffff897c38f14858 (&mm->mmap_sem#2){++++}-{3:3}, at: do_page_fault+0x143/0x715 irq event stamp: 43530289 count_memcg_event_mm+0x1a6/0x270 count_memcg_event_mm+0x119/0x270 __do_softirq+0x365/0x589 irq_exit+0xa2/0xc0 Signed-off-by: Qian Cai <cai(a)lca.pw> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Cc: Marco Elver <elver(a)google.com> Cc: Hugh Dickins <hughd(a)google.com> Link: http://lkml.kernel.org/r/20200402213748.2237-1-cai@lca.pw Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> Signed-off-by: Liu Shixin <liushixin2(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- mm/swap_state.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/mm/swap_state.c b/mm/swap_state.c index b776b472974c..2137e2d57196 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -548,10 +548,11 @@ static unsigned long swapin_nr_pages(unsigned long offset) return 1; hits = atomic_xchg(&swapin_readahead_hits, 0); - pages = __swapin_nr_pages(prev_offset, offset, hits, max_pages, + pages = __swapin_nr_pages(READ_ONCE(prev_offset), offset, hits, + max_pages, atomic_read(&last_readahead_pages)); if (!hits) - prev_offset = offset; + WRITE_ONCE(prev_offset, offset); atomic_set(&last_readahead_pages, pages); return pages; -- 2.25.1

1 2

[PATCH 0/4] arm64: perf: Add support for ARMv8.5-PMU
by Yang Yingliang 09 Sep '20

09 Sep '20

Backport perf support patches for ARMv8.5-PMU. Andrew Murray (3): arm64: cpufeature: Extract capped perfmon fields KVM: arm64: limit PMU version to PMUv3 for ARMv8.1 arm64: perf: Add support for ARMv8.5-PMU 64-bit counters Robin Murphy (1): arm64: perf: Clean up enable/disable calls arch/arm64/include/asm/cpufeature.h | 23 ++++ arch/arm64/include/asm/perf_event.h | 3 +- arch/arm64/include/asm/sysreg.h | 10 ++ arch/arm64/kernel/perf_event.c | 170 +++++++++++++++++----------- arch/arm64/kvm/sys_regs.c | 10 ++ include/linux/perf/arm_pmu.h | 1 + 6 files changed, 150 insertions(+), 67 deletions(-) -- 2.25.1

1 4

[PATCH 0/8] ARMv8.4 TTL feature
by Yang Yingliang 09 Sep '20

09 Sep '20

ARMv8.4 TTL feature. Catalin Marinas (1): arm64: Shift the __tlbi_level() indentation left Marc Zyngier (2): arm64: Detect the ARMv8.4 TTL feature arm64: Add level-hinted TLB invalidation helper Peter Zijlstra (Intel) (1): tlb: mmu_gather: add tlb_flush_*_range APIs Zhenyu Ye (4): arm64: Add tlbi_user_level TLB invalidation helper arm64: tlb: Set the TTL field in flush_tlb_range arm64: tlb: Set the TTL field in flush_*_tlb_range arm64: tlb: don't set the ttl value in flush_tlb_page_nosync arch/arm64/include/asm/cpucaps.h | 3 +- arch/arm64/include/asm/pgtable.h | 10 ++++ arch/arm64/include/asm/stage2_pgtable.h | 9 ++++ arch/arm64/include/asm/sysreg.h | 1 + arch/arm64/include/asm/tlb.h | 29 +++++++++++- arch/arm64/include/asm/tlbflush.h | 63 ++++++++++++++++++++++--- arch/arm64/kernel/cpufeature.c | 11 +++++ include/asm-generic/tlb.h | 55 +++++++++++++++------ 8 files changed, 158 insertions(+), 23 deletions(-) -- 2.25.1

1 8

BPF 新特性合入openEuler内核策略讨论
by 陆志浩 09 Sep '20

09 Sep '20

Hi： BPF 特性在上游内核社区非常活跃，该特性在网络、内核维护、容器、安全等方面都有使用价值，当前openEuler 4.19内核的BPF特性相较最新版本比较陈旧，不利于下游场景的一些价值场景的业务开发。我希望在openEuler 4.19内核版本逐步合入BPF最新特性，并且长期更新BPF最新特性及bugfix。 PS: 会上提供BPF新特性列表以及相关的文件列表范围，初步给出对周边模块的影响性。 by luzhihao

1 0

[PATCH] net: hns3: fix ping failed after S3 since removing HNAE3_RESTORE_CLIENT
by Yonglong Liu 09 Sep '20

09 Sep '20

commit: 412c29318f9b ("net: hns3: refactor the MAC address configure") removed the operation of HNAE3_RESTORE_CLIENT, so after S3 or S4, the MAC address is missing, then cause ping failed. According to the commit, need to assign a reset type to restore the MAC address. Signed-off-by: Yonglong Liu <liuyonglong(a)huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 823f161..ac1789e 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -3850,6 +3850,11 @@ static int hclge_resume(struct hnae3_ae_dev *ae_dev) rtnl_lock(); + memset(hdev->vlan_table, 0, sizeof(hdev->vlan_table)); + memset(hdev->vf_vlan_full, 0, sizeof(hdev->vf_vlan_full)); + bitmap_set(hdev->vport_config_block, 0, hdev->num_alloc_vport); + hclge_reset_umv_space(hdev); + ret = hclge_reset_ae_dev(hdev->ae_dev); if (ret) goto err_reset_lock; -- 2.8.1

1 0

[PATCH] net: hns3: fix ping failed after S3 since removing HNAE3_RESTORE_CLIENT
by Yonglong Liu 08 Sep '20

08 Sep '20

commit: 412c29318f9b ("net: hns3: refactor the MAC address configure") removed the operation of HNAE3_RESTORE_CLIENT, so after S3 or S4, the MAC address is missing, then cause ping failed. According to the commit, need to assign a reset type to restore the MAC address. Signed-off-by: Yonglong Liu <liuyonglong(a)huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 823f161..25aee93 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -3850,6 +3850,8 @@ static int hclge_resume(struct hnae3_ae_dev *ae_dev) rtnl_lock(); + hdev->reset_type = HNAE3_GLOBAL_RESET; + ret = hclge_reset_ae_dev(hdev->ae_dev); if (ret) goto err_reset_lock; -- 2.8.1

1 1

[PATCH 00/72] backport stable-4.19.142
by Yang Yingliang 08 Sep '20

08 Sep '20

Al Viro (1): do_epoll_ctl(): clean the failure exits up a bit Alex Williamson (1): vfio/type1: Add proper error unwind for vfio_iommu_replay() Bodo Stroesser (1): scsi: target: tcmu: Fix crash in tcmu_flush_dcache_range on ARM Charan Teja Reddy (1): mm, page_alloc: fix core hung in free_pcppages_bulk() Chris Wilson (1): drm/vgem: Replace opencoded version of drm_gem_dumb_map_offset() Chuck Lever (1): svcrdma: Fix another Receive buffer leak Chuhong Yuan (1): media: budget-core: Improve exception handling in budget_register() Cong Wang (1): bonding: fix a potential double-unregister Darrick J. Wong (1): xfs: fix inode quota reservation checks David Howells (1): afs: Fix NULL deref in afs_dynroot_depopulate() Dinghao Liu (1): ASoC: intel: Fix memleak in sst_media_open Doug Berger (1): mm: include CMA pages in lowmem_reserve at boot Eiichi Tsukata (1): xfs: Fix UBSAN null-ptr-deref in xfs_sysfs_init Eric Sandeen (1): ext4: fix potential negative array index in do_split() Evgeny Novikov (1): media: vpss: clean up resources in init Fugang Duan (1): net: fec: correct the error path for regulator disable in probe Gaurav Singh (1): tools/testing/selftests/cgroup/cgroup_util.c: cg_read_strcmp: fix null pointer dereference Greg Kroah-Hartman (1): Linux 4.19.142 Greg Ungerer (1): m68knommu: fix overwriting of bits in ColdFire V3 cache control Grzegorz Szczurek (1): i40e: Fix crash during removing i40e driver Haiyang Zhang (1): hv_netvsc: Fix the queue_mapping in netvsc_vf_xmit() Heiko Carstens (2): s390/runtime_instrumentation: fix storage key handling s390/ptrace: fix storage key handling Helge Deller (1): fs/signalfd.c: fix inconsistent return codes for signalfd4 Huacai Chen (1): rtc: goldfish: Enable interrupt in set_alarm() when necessary Hugh Dickins (2): khugepaged: khugepaged_test_exit() check mmget_still_valid() khugepaged: adjust VM_BUG_ON_MM() in __khugepaged_enter() Jan Kara (1): ext4: fix checking of directory entry validity for inline directories Jann Horn (1): romfs: fix uninitialized memory leak in romfs_dev_read() Jarod Wilson (1): bonding: show saner speed for broadcast mode Javed Hasan (1): scsi: libfc: Free skb in fc_disc_gpn_id_resp() for valid cases Jim Mattson (2): kvm: x86: Toggling CR4.SMAP does not load PDPTEs in PAE mode kvm: x86: Toggling CR4.PKE does not load PDPTEs in PAE mode Jiri Wiesner (1): bonding: fix active-backup failover for current ARP slave Josef Bacik (2): btrfs: don't show full path of bind mounts in subvol= btrfs: sysfs: use NOFS for device creation Juergen Gross (2): efi: avoid error message when booting under Xen xen: don't reschedule in preemption off sections Krunoslav Kovac (1): drm/amd/display: fix pow() crashing when given base 0 Li Heng (1): efi: add missed destroy_workqueue when efisubsys_init fails Liang Chen (1): kthread: Do not preempt current task if it is going to call schedule() Luc Van Oostenryck (1): alpha: fix annotation of io{read,write}{16,32}be() Lukas Wunner (1): spi: Prevent adding devices below an unregistering controller Mao Wenan (1): virtio_ring: Avoid loop when vq is broken in virtqueue_poll Marc Zyngier (1): epoll: Keep a reference on files added to the check list Marcos Paulo de Souza (1): btrfs: export helpers for subvolume name/id resolution Masahiro Yamada (2): kconfig: qconf: do not limit the pop-up menu to the first row kconfig: qconf: fix signal connection to invalid slots Masami Hiramatsu (1): perf probe: Fix memory leakage when the probe point is not found Nikolay Borisov (1): btrfs: Move free_pages_out label in inline extent handling branch in compress_file_range Peter Xu (1): mm/hugetlb: fix calculation of adjust_range_if_pmd_sharing_possible Przemyslaw Patynowski (1): i40e: Set RX_ONLY mode for unicast promiscuous on VLAN Qu Wenruo (1): btrfs: inode: fix NULL pointer dereference if inode doesn't need compression Randy Dunlap (1): Fix build error when CONFIG_ACPI is not set/enabled: Selvin Xavier (1): RDMA/bnxt_re: Do not add user qps to flushlist Shay Agroskin (1): net: ena: Prevent reset after device destruction Srinivas Kandagatla (2): ASoC: q6routing: add dummy register read/write function ASoC: msm8916-wcd-analog: fix register Interrupt offset Srinivas Pandruvada (1): cpufreq: intel_pstate: Fix cpuinfo_max_freq when MSR_TURBO_RATIO_LIMIT is 0 Stanley Chu (1): scsi: ufs: Add DELAY_BEFORE_LPM quirk for Micron devices Steffen Maier (1): scsi: zfcp: Fix use-after-free in request timeout handlers Stephen Boyd (1): clk: Evict unregistered clks from parent caches Tom Rix (1): net: dsa: b53: check for timeout Vasant Hegde (1): powerpc/pseries: Do not initiate shutdown when system is running on UPS Wang Hai (1): net: gemini: Fix missing free_netdev() in error path of gemini_ethernet_port_probe() Wei Yongjun (1): kernel/relay.c: fix memleak on destroy relay channel Will Deacon (2): KVM: Pass MMU notifier range flags to kvm_unmap_hva_range() KVM: arm64: Only reschedule if MMU_NOTIFIER_RANGE_BLOCKABLE is not set Xiongfeng Wang (1): Input: psmouse - add a newline when printing 'proto' by sysfs Xiubo Li (1): ceph: fix use-after-free for fsc->mdsc Zhe Li (1): jffs2: fix UAF problem zhangyi (F) (1): jbd2: add the missing unlock_buffer() in the error path of jbd2_write_superblock() Makefile | 2 +- arch/alpha/include/asm/io.h | 8 +-- arch/arm/include/asm/kvm_host.h | 2 +- arch/arm64/include/asm/kvm_host.h | 2 +- arch/m68k/include/asm/m53xxacr.h | 6 +- arch/mips/include/asm/kvm_host.h | 2 +- arch/mips/kvm/mmu.c | 3 +- arch/powerpc/include/asm/kvm_host.h | 3 +- arch/powerpc/kvm/book3s.c | 3 +- arch/powerpc/kvm/e500_mmu_host.c | 3 +- arch/powerpc/platforms/pseries/ras.c | 1 - arch/s390/kernel/ptrace.c | 7 +- arch/s390/kernel/runtime_instr.c | 2 +- arch/x86/include/asm/kvm_host.h | 3 +- arch/x86/kvm/mmu.c | 3 +- arch/x86/kvm/x86.c | 2 +- arch/x86/pci/xen.c | 1 + drivers/clk/clk.c | 52 +++++++++++--- drivers/cpufreq/intel_pstate.c | 1 + drivers/firmware/efi/efi.c | 2 + .../gpu/drm/amd/display/include/fixed31_32.h | 3 + drivers/gpu/drm/vgem/vgem_drv.c | 27 ------- drivers/infiniband/hw/bnxt_re/main.c | 3 +- drivers/input/mouse/psmouse-base.c | 2 +- drivers/media/pci/ttpci/budget-core.c | 11 ++- drivers/media/platform/davinci/vpss.c | 20 ++++-- drivers/net/bonding/bond_main.c | 42 +++++++++-- drivers/net/dsa/b53/b53_common.c | 2 + drivers/net/ethernet/amazon/ena/ena_netdev.c | 19 ++--- drivers/net/ethernet/cortina/gemini.c | 4 +- drivers/net/ethernet/freescale/fec_main.c | 4 +- .../net/ethernet/intel/i40e/i40e_adminq_cmd.h | 2 +- drivers/net/ethernet/intel/i40e/i40e_common.c | 35 ++++++--- drivers/net/ethernet/intel/i40e/i40e_main.c | 3 + drivers/net/hyperv/netvsc_drv.c | 2 +- drivers/rtc/rtc-goldfish.c | 1 + drivers/s390/scsi/zfcp_fsf.c | 4 +- drivers/scsi/libfc/fc_disc.c | 12 +++- drivers/scsi/ufs/ufs_quirks.h | 1 + drivers/scsi/ufs/ufshcd.c | 2 + drivers/spi/Kconfig | 3 + drivers/spi/spi.c | 21 +++++- drivers/target/target_core_user.c | 2 +- drivers/vfio/vfio_iommu_type1.c | 71 +++++++++++++++++-- drivers/video/fbdev/efifb.c | 2 +- drivers/virtio/virtio_ring.c | 3 + drivers/xen/preempt.c | 2 +- fs/afs/dynroot.c | 20 +++--- fs/btrfs/ctree.h | 2 + fs/btrfs/export.c | 8 +-- fs/btrfs/export.h | 5 ++ fs/btrfs/inode.c | 23 +++--- fs/btrfs/super.c | 18 +++-- fs/btrfs/sysfs.c | 4 ++ fs/ceph/mds_client.c | 3 +- fs/eventpoll.c | 19 ++--- fs/ext4/namei.c | 22 ++++-- fs/jbd2/journal.c | 4 +- fs/jffs2/dir.c | 6 +- fs/romfs/storage.c | 4 +- fs/signalfd.c | 10 +-- fs/xfs/xfs_sysfs.h | 6 +- fs/xfs/xfs_trans_dquot.c | 2 +- kernel/kthread.c | 17 ++++- kernel/relay.c | 1 + mm/hugetlb.c | 24 +++---- mm/khugepaged.c | 7 +- mm/page_alloc.c | 7 +- net/sunrpc/xprtrdma/svc_rdma_recvfrom.c | 2 + scripts/kconfig/qconf.cc | 70 +++++++++--------- sound/soc/codecs/msm8916-wcd-analog.c | 4 +- sound/soc/intel/atom/sst-mfld-platform-pcm.c | 5 +- sound/soc/qcom/qdsp6/q6routing.c | 16 +++++ tools/perf/util/probe-finder.c | 2 +- tools/testing/selftests/cgroup/cgroup_util.c | 2 +- virt/kvm/arm/mmu.c | 18 +++-- virt/kvm/kvm_main.c | 2 +- 77 files changed, 508 insertions(+), 236 deletions(-) -- 2.25.1

1 72

[PATCH 1/6] can: j1939: fix kernel-infoleak in j1939_sk_sock2sockaddr_can()
by Yang Yingliang 08 Sep '20

08 Sep '20

From: Eric Dumazet <edumazet(a)google.com> mainline inclusion from mainline-v5.9-rc2 commit 38ba8b9241f5848a49b80fddac9ab5f4692e434e category: bugfix bugzilla: 38684 CVE: NA --------------------------- syzbot found that at least 2 bytes of kernel information were leaked during getsockname() on AF_CAN CAN_J1939 socket. Since struct sockaddr_can has in fact two holes, simply clear the whole area before filling it with useful data. BUG: KMSAN: kernel-infoleak in kmsan_copy_to_user+0x81/0x90 mm/kmsan/kmsan_hooks.c:253 CPU: 0 PID: 8466 Comm: syz-executor511 Not tainted 5.8.0-rc5-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x21c/0x280 lib/dump_stack.c:118 kmsan_report+0xf7/0x1e0 mm/kmsan/kmsan_report.c:121 kmsan_internal_check_memory+0x238/0x3d0 mm/kmsan/kmsan.c:423 kmsan_copy_to_user+0x81/0x90 mm/kmsan/kmsan_hooks.c:253 instrument_copy_to_user include/linux/instrumented.h:91 [inline] _copy_to_user+0x18e/0x260 lib/usercopy.c:39 copy_to_user include/linux/uaccess.h:186 [inline] move_addr_to_user+0x3de/0x670 net/socket.c:237 __sys_getsockname+0x407/0x5e0 net/socket.c:1909 __do_sys_getsockname net/socket.c:1920 [inline] __se_sys_getsockname+0x91/0xb0 net/socket.c:1917 __x64_sys_getsockname+0x4a/0x70 net/socket.c:1917 do_syscall_64+0xad/0x160 arch/x86/entry/common.c:386 entry_SYSCALL_64_after_hwframe+0x44/0xa9 RIP: 0033:0x440219 Code: Bad RIP value. RSP: 002b:00007ffe5ee150c8 EFLAGS: 00000246 ORIG_RAX: 0000000000000033 RAX: ffffffffffffffda RBX: 00000000004002c8 RCX: 0000000000440219 RDX: 0000000020000240 RSI: 0000000020000100 RDI: 0000000000000003 RBP: 00000000006ca018 R08: 0000000000000000 R09: 00000000004002c8 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000401a20 R13: 0000000000401ab0 R14: 0000000000000000 R15: 0000000000000000 Local variable ----address@__sys_getsockname created at: __sys_getsockname+0x91/0x5e0 net/socket.c:1894 __sys_getsockname+0x91/0x5e0 net/socket.c:1894 Bytes 2-3 of 24 are uninitialized Memory access of size 24 starts at ffff8880ba2c7de8 Data copied to user address 0000000020000100 Fixes: 9d71dd0c7009 ("can: add support of SAE J1939 protocol") Signed-off-by: Eric Dumazet <edumazet(a)google.com> Reported-by: syzbot <syzkaller(a)googlegroups.com> Cc: Robin van der Gracht <robin(a)protonic.nl> Cc: Oleksij Rempel <o.rempel(a)pengutronix.de> Cc: Pengutronix Kernel Team <kernel(a)pengutronix.de> Cc: linux-can(a)vger.kernel.org Acked-by: Oleksij Rempel <o.rempel(a)pengutronix.de> Link: https://lore.kernel.org/r/20200813161834.4021638-1-edumazet@google.com Signed-off-by: Marc Kleine-Budde <mkl(a)pengutronix.de> Signed-off-by: Zhang Changzhong <zhangchangzhong(a)huawei.com> Reviewed-by: Yue Haibing <yuehaibing(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- net/can/j1939/socket.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/can/j1939/socket.c b/net/can/j1939/socket.c index f7587428febdd..2f4e2142ef268 100644 --- a/net/can/j1939/socket.c +++ b/net/can/j1939/socket.c @@ -553,6 +553,11 @@ static int j1939_sk_connect(struct socket *sock, struct sockaddr *uaddr, static void j1939_sk_sock2sockaddr_can(struct sockaddr_can *addr, const struct j1939_sock *jsk, int peer) { + /* There are two holes (2 bytes and 3 bytes) to clear to avoid + * leaking kernel information to user space. + */ + memset(addr, 0, J1939_MIN_NAMELEN); + addr->can_family = AF_CAN; addr->can_ifindex = jsk->ifindex; addr->can_addr.j1939.pgn = jsk->addr.pgn; -- 2.25.1

1 5

[PATCH 0/4] Support Common Not Private
by Yang Yingliang 08 Sep '20

08 Sep '20

Cheng Jian (1): config: arm64: set defalut value of config CNP Vladimir Murzin (3): arm64: mm: Support Common Not Private translations arm64: KVM: Enable Common Not Private translations arm64: Introduce command line parameter to disable CNP .../admin-guide/kernel-parameters.txt | 4 ++ arch/arm/include/asm/kvm_arm.h | 1 + arch/arm/include/asm/kvm_mmu.h | 5 +++ arch/arm64/Kconfig | 14 ++++++ arch/arm64/configs/euleros_defconfig | 1 + arch/arm64/configs/hulk_defconfig | 1 + arch/arm64/configs/openeuler_defconfig | 1 + arch/arm64/configs/storage_ci_defconfig | 1 + arch/arm64/configs/syzkaller_defconfig | 1 + arch/arm64/include/asm/cpucaps.h | 3 +- arch/arm64/include/asm/cpufeature.h | 6 +++ arch/arm64/include/asm/kvm_arm.h | 1 + arch/arm64/include/asm/kvm_mmu.h | 5 +++ arch/arm64/include/asm/mmu_context.h | 17 +++++++- arch/arm64/include/asm/pgtable-hwdef.h | 2 + arch/arm64/kernel/cpufeature.c | 43 +++++++++++++++++++ arch/arm64/kernel/suspend.c | 4 ++ arch/arm64/kvm/hyp-init.S | 3 ++ arch/arm64/mm/context.c | 3 ++ arch/arm64/mm/proc.S | 11 +++-- virt/kvm/arm/arm.c | 4 +- 21 files changed, 123 insertions(+), 8 deletions(-) -- 2.25.1

1 4

[PATCH 1/8] posix-cpu-timers: Stop disabling timers on mt-exec
by Yang Yingliang 08 Sep '20

08 Sep '20

From: "Eric W. Biederman" <ebiederm(a)xmission.com> mainline inclusion from mainline-v5.6 commit b95e31c07c5eb4f5c0769f12b38b0343d7961040 category: bugfix bugzilla: 32426 CVE: NA ------------------------ The reasons why the extra posix_cpu_timers_exit_group() invocation has been added are not entirely clear from the commit message. Today all that posix_cpu_timers_exit_group() does is stop timers that are tracking the task from firing. Every other operation on those timers is still allowed. The practical implication of this is posix_cpu_timer_del() which could not get the siglock after the thread group leader has exited (because sighand == NULL) would be able to run successfully because the timer was already dequeued. With that locking issue fixed there is no point in disabling all of the timers. So remove this ``tempoary'' hack. Fixes: e0a70217107e ("posix-cpu-timers: workaround to suppress the problems with mt exec") Signed-off-by: "Eric W. Biederman" <ebiederm(a)xmission.com> Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Link: https://lkml.kernel.org/r/87o8tityzs.fsf@x220.int.ebiederm.org Reviewed-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- kernel/exit.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) diff --git a/kernel/exit.c b/kernel/exit.c index 35adc3b1e9be9..891d65e3ffd51 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -102,17 +102,8 @@ static void __exit_signal(struct task_struct *tsk) #ifdef CONFIG_POSIX_TIMERS posix_cpu_timers_exit(tsk); - if (group_dead) { + if (group_dead) posix_cpu_timers_exit_group(tsk); - } else { - /* - * This can only happen if the caller is de_thread(). - * FIXME: this is the temporary hack, we should teach - * posix-cpu-timers to handle this case correctly. - */ - if (unlikely(has_group_leader_pid(tsk))) - posix_cpu_timers_exit_group(tsk); - } #endif if (group_dead) { -- 2.25.1

1 7

[PATCH] defconfig: update the defconfigs to support NVDIMM
by Yang Yingliang 08 Sep '20

08 Sep '20

From: Hanjun Guo <guohanjun(a)huawei.com> hulk inclusion category: feature bugzilla: NA CVE: NA --------------------------- openEuler wants NVDIMM support, update the defconfig. Signed-off-by: Hanjun Guo <guohanjun(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/configs/hulk_defconfig | 2 +- arch/arm64/configs/openeuler_defconfig | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/configs/hulk_defconfig b/arch/arm64/configs/hulk_defconfig index 809f808363616..a696b55daaae0 100644 --- a/arch/arm64/configs/hulk_defconfig +++ b/arch/arm64/configs/hulk_defconfig @@ -644,7 +644,7 @@ CONFIG_ACPI_HED=y # CONFIG_ACPI_CUSTOM_METHOD is not set # CONFIG_ACPI_BGRT is not set CONFIG_ACPI_REDUCED_HARDWARE_ONLY=y -# CONFIG_ACPI_NFIT is not set +CONFIG_ACPI_NFIT=m CONFIG_HAVE_ACPI_APEI=y CONFIG_ACPI_APEI=y CONFIG_ACPI_APEI_GHES=y diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index d3144ca393452..91bef6fcdbf88 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -640,7 +640,7 @@ CONFIG_ACPI_HED=y # CONFIG_ACPI_CUSTOM_METHOD is not set # CONFIG_ACPI_BGRT is not set CONFIG_ACPI_REDUCED_HARDWARE_ONLY=y -# CONFIG_ACPI_NFIT is not set +CONFIG_ACPI_NFIT=m CONFIG_HAVE_ACPI_APEI=y CONFIG_ACPI_APEI=y CONFIG_ACPI_APEI_GHES=y -- 2.25.1

1 0

[PATCH 1/5] net: hns3: check cmdq message parameters sent from VF
by Yang Yingliang 08 Sep '20

08 Sep '20

From: Yufeng Mo <moyufeng(a)huawei.com> driver inclusion category: bugfix bugzilla: NA CVE: NA ----------------------------------- The parameters sent from VF may be unreliable. If these parameters are used directly, memory overwriting may occur. Therefore, we need to check parameters before using. Signed-off-by: Yufeng Mo <moyufeng(a)huawei.com> Reviewed-by: Peng Li <lipeng321(a)huawei.com> Reviewed-by: Weiwei Deng <dengweiwei(a)huawei.com> Reviewed-by: Zhaohui Zhong <zhongzhaohui(a)huawei.com> Reviewed-by: Junxin Chen <chenjunxin1(a)huawei.com> Signed-off-by: Shengzui You <youshengzui(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- .../hisilicon/hns3/hns3pf/hclge_main.c | 7 +++++ .../hisilicon/hns3/hns3pf/hclge_mbx.c | 26 ++++++++++--------- 2 files changed, 21 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 24b86d704e038..9c3ee09a67d7c 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -9542,12 +9542,19 @@ int hclge_reset_tqp(struct hnae3_handle *handle, u16 queue_id) void hclge_reset_vf_queue(struct hclge_vport *vport, u16 queue_id) { + struct hnae3_handle *handle = &vport->nic; struct hclge_dev *hdev = vport->back; int reset_try_times = 0; int reset_status; u16 queue_gid; int ret; + if (queue_id >= handle->kinfo.num_tqps) { + dev_warn(&hdev->pdev->dev, "Invalid vf queue id(%u)\n", + queue_id); + return; + } + queue_gid = hclge_covert_handle_qid_global(&vport->nic, queue_id); ret = hclge_send_reset_tqp_cmd(hdev, queue_gid, true); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c index 72fd997576016..33ca01408cfcb 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c @@ -167,21 +167,31 @@ static int hclge_get_ring_chain_from_mbx( struct hclge_vport *vport) { struct hnae3_ring_chain_node *cur_chain, *new_chain; + struct hclge_dev *hdev = vport->back; int ring_num; - int i = 0; + int i; ring_num = req->msg.ring_num; if (ring_num > HCLGE_MBX_MAX_RING_CHAIN_PARAM_NUM) return -ENOMEM; + for (i = 0; i < ring_num; i++) { + if (req->msg.param[i].tqp_index >= vport->nic.kinfo.rss_size) { + dev_err(&hdev->pdev->dev, "tqp index(%u) is out of range(0-%u)\n", + req->msg.param[i].tqp_index, + vport->nic.kinfo.rss_size - 1); + return -EINVAL; + } + } + hnae3_set_bit(ring_chain->flag, HNAE3_RING_TYPE_B, - req->msg.param[i].ring_type); + req->msg.param[0].ring_type); ring_chain->tqp_index = hclge_get_queue_id(vport->nic.kinfo.tqp - [req->msg.param[i].tqp_index]); + [req->msg.param[0].tqp_index]); hnae3_set_field(ring_chain->int_gl_idx, HNAE3_RING_GL_IDX_M, - HNAE3_RING_GL_IDX_S, req->msg.param[i].int_gl_index); + HNAE3_RING_GL_IDX_S, req->msg.param[0].int_gl_index); cur_chain = ring_chain; @@ -270,7 +280,6 @@ static int hclge_get_vf_ring_vector_map(struct hclge_vport *vport, #define HCLGE_VECTOR_ID_OFFSET 3 #define HCLGE_RING_VECTOR_MAP_INFO_LEN 4 struct hnae3_ring_chain_node ring_chain; - struct hclge_dev *hdev = vport->back; struct hclge_desc desc; struct hclge_ctrl_vector_chain_cmd *data = (struct hclge_ctrl_vector_chain_cmd *)desc.data; @@ -278,13 +287,6 @@ static int hclge_get_vf_ring_vector_map(struct hclge_vport *vport, u8 int_gl_index; int ret; - if (req->msg.param[0].tqp_index >= vport->nic.kinfo.rss_size) { - dev_err(&hdev->pdev->dev, "tqp index(%u) is out of range(0-%u)\n", - req->msg.param[0].tqp_index, - vport->nic.kinfo.rss_size - 1); - return -EINVAL; - } - req->msg.ring_num = HCLGE_LIMIT_RING_NUM; memset(&ring_chain, 0, sizeof(ring_chain)); -- 2.25.1

1 4

[PATCH 1/2] config: add certs dir to CONFIG_MODULE_SIG_KEY
by Yang Yingliang 08 Sep '20

08 Sep '20

hulk inclusion category: bugfix bugzilla: NA CVE: NA --------------------------- make[1]: *** No rule to make target 'signing_key.pem', needed by 'certs/signing_key.x509'. Stop. Fix the compile error by add certs dir to CONFIG_MODULE_SIG_KEY. Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/configs/euleros_defconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/configs/euleros_defconfig b/arch/arm64/configs/euleros_defconfig index c0ec409e48ac7..6b8a71fd4f846 100644 --- a/arch/arm64/configs/euleros_defconfig +++ b/arch/arm64/configs/euleros_defconfig @@ -5317,7 +5317,7 @@ CONFIG_SIGNED_PE_FILE_VERIFICATION=y # # Certificates for signature checking # -CONFIG_MODULE_SIG_KEY="./signing_key.pem" +CONFIG_MODULE_SIG_KEY="certs/signing_key.pem" CONFIG_SYSTEM_TRUSTED_KEYRING=y CONFIG_SYSTEM_TRUSTED_KEYS="" # CONFIG_SYSTEM_EXTRA_CERTIFICATE is not set -- 2.25.1

1 1

[PATCH] Revert "net: hns3: fix for ping failed when resume from s3 or s4"
by Yonglong Liu 08 Sep '20

08 Sep '20

This reverts commit 4c6d4e21ed9b2309f15ef9074e80c7fc1c09ccfc. Because commit: 412c29318f9b ("net: hns3: refactor the MAC address configure") removed the operation of HNAE3_RESTORE_CLIENT. --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 6ee6b4e..823f161 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -3858,10 +3858,6 @@ static int hclge_resume(struct hnae3_ae_dev *ae_dev) if (ret) goto err_reset_lock; - ret = hclge_notify_client(hdev, HNAE3_RESTORE_CLIENT); - if (ret) - goto err_reset_lock; - rtnl_unlock(); ret = hclge_notify_roce_client(hdev, HNAE3_INIT_CLIENT); -- 2.8.1

1 0

[PATCH 1/3] config: set config BLK_CGROUP_IOCOST default value
by Yang Yingliang 02 Sep '20

02 Sep '20

hulk inclusion category: feature bugzilla: 38688 CVE: NA --------------------------- set config BLK_CGROUP_IOCOST default value. Reviewed-by: zhangyi (F) <yi.zhang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/configs/euleros_defconfig | 1 + arch/arm64/configs/hulk_defconfig | 1 + arch/arm64/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + 4 files changed, 4 insertions(+) diff --git a/arch/arm64/configs/euleros_defconfig b/arch/arm64/configs/euleros_defconfig index 673dcb491a5ac..c0ec409e48ac7 100644 --- a/arch/arm64/configs/euleros_defconfig +++ b/arch/arm64/configs/euleros_defconfig @@ -795,6 +795,7 @@ CONFIG_BLK_DEV_THROTTLING=y CONFIG_BLK_WBT=y # CONFIG_BLK_CGROUP_IOLATENCY is not set # CONFIG_BLK_WBT_SQ is not set +# CONFIG_BLK_CGROUP_IOCOST is not set CONFIG_BLK_WBT_MQ=y CONFIG_BLK_DEBUG_FS=y CONFIG_BLK_DEBUG_FS_ZONED=y diff --git a/arch/arm64/configs/hulk_defconfig b/arch/arm64/configs/hulk_defconfig index 02090f911f17a..809f808363616 100644 --- a/arch/arm64/configs/hulk_defconfig +++ b/arch/arm64/configs/hulk_defconfig @@ -809,6 +809,7 @@ CONFIG_BLK_DEV_THROTTLING=y CONFIG_BLK_WBT=y # CONFIG_BLK_CGROUP_IOLATENCY is not set # CONFIG_BLK_WBT_SQ is not set +CONFIG_BLK_CGROUP_IOCOST=y CONFIG_BLK_WBT_MQ=y CONFIG_BLK_DEBUG_FS=y CONFIG_BLK_DEBUG_FS_ZONED=y diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 49fd2e5c20185..1f18ab6957849 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -803,6 +803,7 @@ CONFIG_BLK_DEV_THROTTLING=y CONFIG_BLK_WBT=y # CONFIG_BLK_CGROUP_IOLATENCY is not set # CONFIG_BLK_WBT_SQ is not set +# CONFIG_BLK_CGROUP_IOCOST is not set CONFIG_BLK_WBT_MQ=y CONFIG_BLK_DEBUG_FS=y CONFIG_BLK_DEBUG_FS_ZONED=y diff --git a/arch/x86/configs/openeuler_defconfig b/arch/x86/configs/openeuler_defconfig index 747c6bd5967f2..912933ebd9817 100644 --- a/arch/x86/configs/openeuler_defconfig +++ b/arch/x86/configs/openeuler_defconfig @@ -883,6 +883,7 @@ CONFIG_BLK_DEV_THROTTLING=y CONFIG_BLK_WBT=y # CONFIG_BLK_CGROUP_IOLATENCY is not set # CONFIG_BLK_WBT_SQ is not set +# CONFIG_BLK_CGROUP_IOCOST is not set CONFIG_BLK_WBT_MQ=y CONFIG_BLK_DEBUG_FS=y # CONFIG_BLK_SED_OPAL is not set -- 2.25.1

1 2

[PATCH 00/28] ARMv8.3 pointer authentication
by Yang Yingliang 02 Sep '20

02 Sep '20

Firstly, this series improves function return address protection for the arm64 kernel, by compiling the kernel with ARMv8.3 Pointer Authentication instructions (referred ptrauth hereafter). This should help protect the kernel against attacks using return-oriented programming. Secondly, Suzuki's patches optimize checking and enabling cpu capabilities. Thirdly, three patches on kernel config enable ptrauth compiling and fix compile warnings. Finally, two patches of Mark Rutland and one patch of Marc Zyngier are necessary simplifications and cleanups. Amit Daniel Kachhap (9): arm64: cpufeature: Fix meta-capability cpufeature check arm64: ptrauth: Add bootup/runtime flags for __cpu_setup arm64: cpufeature: Move cpu capability helpers inside C file arm64: initialize ptrauth keys for kernel booting task arm64: mask PAC bits of __builtin_return_address arm64: __show_regs: strip PAC from lr in printk arm64: suspend: restore the kernel ptrauth keys lkdtm: arm64: test kernel pointer authentication arm64: Kconfig: ptrauth: Add binutils version check to fix mismatch Catalin Marinas (1): kbuild: Add support for 'as-instr' to be used in Kconfig files Kristina Martsenko (7): arm64: cpufeature: add pointer auth meta-capabilities arm64: rename ptrauth key structures to be user-specific arm64: install user ptrauth keys at kernel exit time arm64: cpufeature: handle conflicts based on capability arm64: enable ptrauth earlier arm64: initialize and switch ptrauth kernel keys arm64: compile the kernel with ptrauth return address signing Marc Zyngier (1): arm64: Drop unnecessary include from asm/smp.h Mark Rutland (3): arm64: unwind: strip PAC from kernel addresses arm64: remove ptrauth_keys_install_kernel sync arg arm64: simplify ptrauth initialization Nick Desaulniers (1): arm64: Kconfig: verify binutils support for ARM64_PTR_AUTH Suzuki K Poulose (4): arm64: capabilities: Speed up capability lookup arm64: capabilities: Optimize this_cpu_has_cap arm64: capabilities: Use linear array for detection and verification arm64: capabilities: Batch cpu_enable callbacks Vincenzo Frascino (1): kconfig: Add support for 'as-option' Zheng Zengkai (1): config: enable ARM64 pointer authentication configs by default arch/arm64/Kconfig | 38 +++- arch/arm64/Makefile | 11 + arch/arm64/configs/hulk_defconfig | 7 + arch/arm64/configs/openeuler_defconfig | 7 + arch/arm64/include/asm/asm_pointer_auth.h | 98 ++++++++ arch/arm64/include/asm/compiler.h | 19 ++ arch/arm64/include/asm/cpucaps.h | 4 +- arch/arm64/include/asm/cpufeature.h | 42 ++-- arch/arm64/include/asm/pointer_auth.h | 50 ++--- arch/arm64/include/asm/processor.h | 3 +- arch/arm64/include/asm/stackprotector.h | 5 + arch/arm64/kernel/asm-offsets.c | 13 ++ arch/arm64/kernel/cpufeature.c | 258 ++++++++++++++-------- arch/arm64/kernel/entry.S | 6 + arch/arm64/kernel/head.S | 10 + arch/arm64/kernel/pointer_auth.c | 7 +- arch/arm64/kernel/process.c | 5 +- arch/arm64/kernel/ptrace.c | 16 +- arch/arm64/kernel/sleep.S | 1 + arch/arm64/kernel/stacktrace.c | 5 +- arch/arm64/mm/proc.S | 27 ++- drivers/misc/lkdtm/bugs.c | 36 +++ drivers/misc/lkdtm/core.c | 1 + drivers/misc/lkdtm/lkdtm.h | 1 + include/linux/stackprotector.h | 2 +- scripts/Kconfig.include | 10 + 26 files changed, 513 insertions(+), 169 deletions(-) create mode 100644 arch/arm64/include/asm/asm_pointer_auth.h -- 2.25.1

1 28

[PATCH 00/90] backport stable-4.19.141
by Yang Yingliang 02 Sep '20

02 Sep '20

Adrian Hunter (1): perf intel-pt: Fix FUP packet state Ahmad Fatoum (3): watchdog: f71808e_wdt: indicate WDIOF_CARDRESET support in watchdog_info.options watchdog: f71808e_wdt: remove use of wrong watchdog_info option watchdog: f71808e_wdt: clear watchdog timeout occurred flag Alexandru Ardelean (1): iio: dac: ad5592r: fix unbalanced mutex unlocks in ad5592r_read_raw() Anand Jain (1): btrfs: don't traverse into the seed devices in show_devname Andy Shevchenko (1): mfd: dln2: Run event handler loop under spinlock Aneesh Kumar K.V (3): selftests/powerpc: ptrace-pkey: Rename variables to make it easier to follow code selftests/powerpc: ptrace-pkey: Update the test to mark an invalid pkey correctly selftests/powerpc: ptrace-pkey: Don't update expected UAMOR value Ansuel Smith (2): PCI: qcom: Define some PARF params needed for ipq8064 SoC PCI: qcom: Add support for tx term offset for rev 2.1.0 Anton Blanchard (1): pseries: Fix 64 bit logical memory block panic ChangSyun Peng (1): md/raid5: Fix Force reconstruct-write io stuck in degraded raid5 Charles Keepax (1): mfd: arizona: Ensure 32k clock is put on driver unbind and error Chengming Zhou (1): ftrace: Setup correct FTRACE_FL_REGS flags for module Christian Eggers (1): dt-bindings: iio: io-channel-mux: Fix compatible string in example code Colin Ian King (3): iommu/omap: Check for failure of a call to omap_iommu_dump_ctx Input: sentelic - fix error return when fsp_reg_write fails fs/ufs: avoid potential u32 multiplication overflow Coly Li (2): bcache: allocate meta data pages as compound pages bcache: fix overflow in offset_to_stripe() Dan Carpenter (2): drm/vmwgfx: Use correct vmw_legacy_display_unit pointer drm/vmwgfx: Fix two list_for_each loop exit tests Daniel Díaz (1): tools build feature: Quote CC and CXX for their arguments David Sterba (1): btrfs: fix messages after changing compression level by remount Dinghao Liu (1): ALSA: echoaudio: Fix potential Oops in snd_echo_resume() Eric Biggers (3): fs/minix: set s_maxbytes correctly fs/minix: fix block limit check for V1 filesystems fs/minix: remove expected error message in block_to_path() Eugeniu Rosca (1): media: vsp1: dl: Fix NULL pointer dereference on unbind Ewan D. Milne (1): scsi: lpfc: nvmet: Avoid hang / use-after-free again when destroying targetport Filipe Manana (1): btrfs: fix memory leaks after failure to lookup checksums during inode logging Geert Uytterhoeven (1): sh: landisk: Add missing initialization of sh_io_port_base Greg Kroah-Hartman (1): Linux 4.19.141 Huacai Chen (1): MIPS: CPU#0 is not hotpluggable Hugh Dickins (1): khugepaged: retract_page_tables() remember to test exit Jason Gunthorpe (1): RDMA/ipoib: Fix ABBA deadlock with ipoib_reap_ah() Jeffrey Mitchell (1): nfs: Fix getxattr kernel panic and memory overflow Johan Hovold (2): USB: serial: ftdi_sio: make process-packet buffer unsigned USB: serial: ftdi_sio: clean up receive processing Johannes Berg (1): mac80211: fix misplaced while instead of if Jonathan McDowell (2): net: ethernet: stmmac: Disable hardware multicast filter net: stmmac: dwmac1000: provide multicast filter fallback Josef Bacik (2): btrfs: open device without device_list_mutex btrfs: only search for left_info if there is no right_info in try_merge_free_space Junxiao Bi (1): ocfs2: change slot number type s16 to u16 Kai-Heng Feng (1): PCI: Mark AMD Navi10 GPU rev 0x00 ATS as broken Kamal Heib (1): RDMA/ipoib: Return void from ipoib_ib_dev_stop() Kees Cook (2): net/compat: Add missing sock updates for SCM_RIGHTS module: Correctly truncate sysfs sections output Kevin Hao (1): tracing/hwlat: Honor the tracing_cpumask Krzysztof Sobota (1): watchdog: initialize device before misc_register Liu Yi L (1): iommu/vt-d: Enforce PASID devTLB field mask Liu Ying (1): drm/imx: imx-ldb: Disable both channels for split mode in enc->disable() Lukas Wunner (1): driver core: Avoid binding drivers to dead devices Marius Iacob (1): drm: Added orientation quirk for ASUS tablet model T103HAF Max Filippov (1): xtensa: fix xtensa_pmu_setup prototype Michael Ellerman (2): powerpc: Allow 4224 bytes of stack expansion for the signal frame powerpc: Fix circular dependency between percpu.h and mmu.h Michal Koutný (1): mm/page_counter.c: fix protection usage propagation Mikulas Patocka (1): ext2: fix missing percpu_counter_inc Ming Lei (1): dm rq: don't call blk_mq_queue_stopped() in dm_stop_queue() Muchun Song (1): kprobes: Fix NULL pointer dereference at kprobe_ftrace_handler Paul Aurich (1): cifs: Fix leak when handling lease break for cached root fid Paul Kocialkowski (2): media: rockchip: rga: Introduce color fmt macros and refactor CSC mode logic media: rockchip: rga: Only set output CSC mode for RGB input Pavel Machek (1): btrfs: fix return value mixup in btrfs_get_extent Qu Wenruo (2): btrfs: free anon block device right after subvolume deletion btrfs: don't allocate anonymous block device for user invisible roots Rafael J. Wysocki (1): PCI: hotplug: ACPI: Fix context refcounting in acpiphp_grab_context() Rajat Jain (1): PCI: Add device even if driver attach failed Rayagonda Kokatanur (1): pwm: bcm-iproc: handle clk_get_rate() return Sandeep Raghuraman (1): drm/amdgpu: Fix bug where DPM is not enabled after hibernate and resume Sibi Sankar (1): remoteproc: qcom: q6v5: Update running state before requesting stop Stafford Horne (1): openrisc: Fix oops caused when dumping stack Steve French (1): smb3: warn on confusing error scenario with sec=krb5 Steve Longerbeam (1): gpu: ipu-v3: image-convert: Combine rotate/no-rotate irq handlers Steven Rostedt (VMware) (1): tracing: Use trace_sched_process_free() instead of exit() for pid tracing Thomas Gleixner (1): genirq/affinity: Make affinity setting if activated opt-in Thomas Hebb (1): tools build feature: Use CC and CXX from parent Tiezhu Yang (1): test_kmod: avoid potential double free in trigger_config_run_type() Tom Rix (1): btrfs: ref-verify: fix memory leak in add_block_entry Tomasz Maciej Nowak (1): arm64: dts: marvell: espressobin: add ethernet alias Vincent Whitchurch (1): perf bench mem: Always memset source before memcpy Wang Hai (1): net: qcom/emac: add missed clk_disable_unprepare in error path of emac_clks_phase1_init Wolfram Sang (2): i2c: rcar: slave: only send STOP event when we have been addressed i2c: rcar: avoid race when unregistering slave Xu Wang (1): clk: clk-atlas6: fix return value check in atlas6_clk_init() Yoshihiro Shimoda (1): mmc: renesas_sdhi_internal_dmac: clean up the code for dma complete .../iio/multiplexer/io-channel-mux.txt | 2 +- Makefile | 2 +- .../dts/marvell/armada-3720-espressobin.dts | 6 ++ arch/mips/kernel/topology.c | 2 +- arch/openrisc/kernel/stacktrace.c | 18 ++++- arch/powerpc/include/asm/percpu.h | 4 +- arch/powerpc/mm/fault.c | 7 +- .../platforms/pseries/hotplug-memory.c | 2 +- arch/sh/boards/mach-landisk/setup.c | 3 + arch/x86/kernel/apic/vector.c | 4 + arch/xtensa/kernel/perf_event.c | 2 +- drivers/base/dd.c | 4 +- drivers/clk/sirf/clk-atlas6.c | 2 +- .../gpu/drm/amd/powerplay/smumgr/ci_smumgr.c | 5 +- .../gpu/drm/drm_panel_orientation_quirks.c | 6 ++ drivers/gpu/drm/imx/imx-ldb.c | 7 +- drivers/gpu/drm/vmwgfx/vmwgfx_kms.c | 8 +- drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c | 5 +- drivers/gpu/ipu-v3/ipu-image-convert.c | 58 +++++---------- drivers/i2c/busses/i2c-rcar.c | 15 ++-- drivers/iio/dac/ad5592r-base.c | 4 +- drivers/infiniband/ulp/ipoib/ipoib.h | 2 +- drivers/infiniband/ulp/ipoib/ipoib_ib.c | 67 ++++++++--------- drivers/infiniband/ulp/ipoib/ipoib_main.c | 2 + drivers/input/mouse/sentelic.c | 2 +- drivers/iommu/omap-iommu-debug.c | 3 + drivers/irqchip/irq-gic-v3-its.c | 5 +- drivers/md/bcache/bcache.h | 2 +- drivers/md/bcache/bset.c | 2 +- drivers/md/bcache/btree.c | 2 +- drivers/md/bcache/journal.c | 4 +- drivers/md/bcache/super.c | 2 +- drivers/md/bcache/writeback.c | 14 ++-- drivers/md/bcache/writeback.h | 19 ++++- drivers/md/dm-rq.c | 3 - drivers/md/raid5.c | 3 +- drivers/media/platform/rockchip/rga/rga-hw.c | 29 ++++---- drivers/media/platform/rockchip/rga/rga-hw.h | 5 ++ drivers/media/platform/vsp1/vsp1_dl.c | 2 + drivers/mfd/arizona-core.c | 18 +++++ drivers/mfd/dln2.c | 4 + drivers/mmc/host/renesas_sdhi_internal_dmac.c | 18 +++-- drivers/net/ethernet/qualcomm/emac/emac.c | 17 ++++- .../ethernet/stmicro/stmmac/dwmac-ipq806x.c | 1 + .../ethernet/stmicro/stmmac/dwmac1000_core.c | 3 + drivers/pci/bus.c | 6 +- drivers/pci/controller/dwc/pcie-qcom.c | 41 ++++++++++- drivers/pci/hotplug/acpiphp_glue.c | 14 +++- drivers/pci/quirks.c | 5 +- drivers/pwm/pwm-bcm-iproc.c | 9 ++- drivers/remoteproc/qcom_q6v5.c | 2 + drivers/scsi/lpfc/lpfc_nvmet.c | 2 +- drivers/usb/serial/ftdi_sio.c | 37 +++++----- drivers/watchdog/f71808e_wdt.c | 13 +++- drivers/watchdog/watchdog_dev.c | 18 ++--- fs/btrfs/disk-io.c | 13 +++- fs/btrfs/free-space-cache.c | 4 +- fs/btrfs/inode.c | 4 +- fs/btrfs/ref-verify.c | 2 + fs/btrfs/super.c | 35 ++++----- fs/btrfs/tree-log.c | 8 +- fs/btrfs/volumes.c | 21 +++++- fs/cifs/smb2misc.c | 73 +++++++++++++------ fs/cifs/smb2pdu.c | 2 + fs/ext2/ialloc.c | 3 +- fs/minix/inode.c | 12 +-- fs/minix/itree_v1.c | 12 +-- fs/minix/itree_v2.c | 13 ++-- fs/minix/minix.h | 1 - fs/nfs/nfs4proc.c | 2 - fs/nfs/nfs4xdr.c | 6 +- fs/ocfs2/ocfs2.h | 4 +- fs/ocfs2/suballoc.c | 4 +- fs/ocfs2/super.c | 4 +- fs/ufs/super.c | 2 +- include/linux/intel-iommu.h | 4 +- include/linux/irq.h | 13 ++++ include/net/sock.h | 4 + kernel/irq/manage.c | 6 +- kernel/kprobes.c | 7 ++ kernel/module.c | 22 +++++- kernel/trace/ftrace.c | 15 ++-- kernel/trace/trace_events.c | 4 +- kernel/trace/trace_hwlat.c | 5 +- lib/test_kmod.c | 2 +- mm/khugepaged.c | 22 +++--- mm/page_counter.c | 6 +- net/compat.c | 1 + net/core/sock.c | 21 ++++++ net/mac80211/sta_info.c | 2 +- sound/pci/echoaudio/echoaudio.c | 2 - tools/build/Makefile.feature | 2 +- tools/build/feature/Makefile | 2 - tools/perf/bench/mem-functions.c | 21 +++--- .../util/intel-pt-decoder/intel-pt-decoder.c | 21 ++---- .../selftests/powerpc/ptrace/ptrace-pkey.c | 55 +++++++------- 96 files changed, 639 insertions(+), 365 deletions(-) -- 2.25.1

1 90

[PATCH] delete python2 dependency
by Zhipeng Xie 01 Sep '20

01 Sep '20

Python2 is no longer supported by the upstream community. The dependency on python2 should be removed from the kernel code. Signed-off-by: Zhipeng Xie <xiezhipeng1(a)huawei.com> --- tools/perf/scripts/python/call-graph-from-sql.py | 2 +- tools/perf/scripts/python/export-to-postgresql.py | 2 +- tools/power/pm-graph/bootgraph.py | 2 +- tools/power/pm-graph/sleepgraph.py | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/tools/perf/scripts/python/call-graph-from-sql.py b/tools/perf/scripts/python/call-graph-from-sql.py index b494a67a1c67..099b472df4a2 100644 --- a/tools/perf/scripts/python/call-graph-from-sql.py +++ b/tools/perf/scripts/python/call-graph-from-sql.py @@ -1,4 +1,4 @@ -#!/usr/bin/python2 +#!/usr/bin/python # call-graph-from-sql.py: create call-graph from sql database # Copyright (c) 2014-2017, Intel Corporation. # diff --git a/tools/perf/scripts/python/export-to-postgresql.py b/tools/perf/scripts/python/export-to-postgresql.py index e46f51b17513..e97a03697aa2 100644 --- a/tools/perf/scripts/python/export-to-postgresql.py +++ b/tools/perf/scripts/python/export-to-postgresql.py @@ -171,7 +171,7 @@ import datetime # SELECT * FROM samples_view WHERE event = 'transactions' AND branch_type_name = 'transaction abort'; # # To print a call stack requires walking the call_paths table. For example this python script: -# #!/usr/bin/python2 +# #!/usr/bin/python # # import sys # from PySide.QtSql import * diff --git a/tools/power/pm-graph/bootgraph.py b/tools/power/pm-graph/bootgraph.py index 8ee626c0f6a5..abb4c38f029b 100755 --- a/tools/power/pm-graph/bootgraph.py +++ b/tools/power/pm-graph/bootgraph.py @@ -1,4 +1,4 @@ -#!/usr/bin/python2 +#!/usr/bin/python # # Tool for analyzing boot timing # Copyright (c) 2013, Intel Corporation. diff --git a/tools/power/pm-graph/sleepgraph.py b/tools/power/pm-graph/sleepgraph.py index 0c760478f7d7..420102e2cd08 100755 --- a/tools/power/pm-graph/sleepgraph.py +++ b/tools/power/pm-graph/sleepgraph.py @@ -1,4 +1,4 @@ -#!/usr/bin/python2 +#!/usr/bin/python # # Tool for analyzing suspend/resume timing # Copyright (c) 2013, Intel Corporation. -- 2.18.1

2 1

[PATCH] delete python2 dependency
by Zhipeng Xie 01 Sep '20

01 Sep '20

Python2 is no longer supported by the upstream community. The dependency on python2 should be removed from the kernel code. Signed-off-by: Zhipeng Xie <xiezhipeng1(a)huawei.com> --- tools/perf/scripts/python/call-graph-from-sql.py | 2 +- tools/perf/scripts/python/export-to-postgresql.py | 2 +- tools/power/pm-graph/bootgraph.py | 2 +- tools/power/pm-graph/sleepgraph.py | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/tools/perf/scripts/python/call-graph-from-sql.py b/tools/perf/scripts/python/call-graph-from-sql.py index b494a67a1c67..099b472df4a2 100644 --- a/tools/perf/scripts/python/call-graph-from-sql.py +++ b/tools/perf/scripts/python/call-graph-from-sql.py @@ -1,4 +1,4 @@ -#!/usr/bin/python2 +#!/usr/bin/python # call-graph-from-sql.py: create call-graph from sql database # Copyright (c) 2014-2017, Intel Corporation. # diff --git a/tools/perf/scripts/python/export-to-postgresql.py b/tools/perf/scripts/python/export-to-postgresql.py index e46f51b17513..e97a03697aa2 100644 --- a/tools/perf/scripts/python/export-to-postgresql.py +++ b/tools/perf/scripts/python/export-to-postgresql.py @@ -171,7 +171,7 @@ import datetime # SELECT * FROM samples_view WHERE event = 'transactions' AND branch_type_name = 'transaction abort'; # # To print a call stack requires walking the call_paths table. For example this python script: -# #!/usr/bin/python2 +# #!/usr/bin/python # # import sys # from PySide.QtSql import * diff --git a/tools/power/pm-graph/bootgraph.py b/tools/power/pm-graph/bootgraph.py index 8ee626c0f6a5..abb4c38f029b 100755 --- a/tools/power/pm-graph/bootgraph.py +++ b/tools/power/pm-graph/bootgraph.py @@ -1,4 +1,4 @@ -#!/usr/bin/python2 +#!/usr/bin/python # # Tool for analyzing boot timing # Copyright (c) 2013, Intel Corporation. diff --git a/tools/power/pm-graph/sleepgraph.py b/tools/power/pm-graph/sleepgraph.py index 0c760478f7d7..420102e2cd08 100755 --- a/tools/power/pm-graph/sleepgraph.py +++ b/tools/power/pm-graph/sleepgraph.py @@ -1,4 +1,4 @@ -#!/usr/bin/python2 +#!/usr/bin/python # # Tool for analyzing suspend/resume timing # Copyright (c) 2013, Intel Corporation. -- 2.18.1

1 0

2025

2024

2023

2022

2021

2020

2019

Kernel