- Kernel - mailweb.openeuler.org

[PATCH openEuler-21.03 v2] park: Reserve park mem before kexec reserved
by sangyan＠huawei.com 12 Mar '21

12 Mar '21

From: Sang Yan <sangyan(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: N/A ------------------------------ reserve_crashkernel or reserve_quick_kexec may find one sutiable memory region and reserves it, which address of the region is not fixed. As a result, cpu park reserves memory could be failed while specified address used by crashkernel or quickkexec. So, move reserve_park_mem before reserve_crashkernel and reserve_quick_kexec. Signed-off-by: Sang Yan <sangyan(a)huawei.com> Reviewed-by: Jing Xiangfeng <jingxiangfeng(a)huawei.com> --- arch/arm64/mm/init.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index b343744..dbcc801 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -497,16 +497,25 @@ void __init arm64_memblock_init(void) else arm64_dma32_phys_limit = PHYS_MASK + 1; + /* + * Reserve park memory before crashkernel and quick kexec. + * Because park memory must be specified by address, but + * crashkernel and quickkexec may be specified by memory length, + * then find one sutiable memory region to reserve. + * + * So reserve park memory firstly is better, but it may cause + * crashkernel or quickkexec reserving failed. + */ +#ifdef CONFIG_ARM64_CPU_PARK + reserve_park_mem(); +#endif + reserve_crashkernel(); #ifdef CONFIG_QUICK_KEXEC reserve_quick_kexec(); #endif -#ifdef CONFIG_ARM64_CPU_PARK - reserve_park_mem(); -#endif - reserve_pin_memory_res(); reserve_elfcorehdr(); -- 2.9.5

1 0

[PATCH openEuler-21.03 v4 1/2] arm64: Add memmap parameter and register pmem
by Zhuling 12 Mar '21

12 Mar '21

From: ZhuLing <zhuling8(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: NA Register pmem in arm64: Use memmap(memmap=nn[KMG]!ss[KMG]) reserve memory and e820(driver/nvdimm/e820.c) function to register persistent memory in arm64. when the kernel restart or update, the data in PMEM will not be lost and can be loaded faster. this is a general features. driver/nvdimm/e820.c: The function of this file is scan "iomem_resource" and take advantage of nvdimm resource discovery mechanism by registering a resource named "Persistent Memory (legacy)", this function doesn't depend on architecture. We will push the feature to linux kernel community and discuss to modify the file name. because people have a mistaken notion that the e820.c is depend on x86. If you want use this features, you need do as follows: 1.Reserve memory: add memmap to reserve memory in grub.cfg memmap=nn[KMG]!ss[KMG] exp:memmap=100K!0x1a0000000. 2.Insmod nd_e820.ko: modprobe nd_e820. 3.Check pmem device in /dev exp: /dev/pmem0 Signed-off-by: ZhuLing <zhuling8(a)huawei.com> Signed-off-by: Sang Yan <sangyan(a)huawei.com> Acked-by: Hanjun Guo <guohanjun(a)huawei.com> --- arch/arm64/Kconfig | 21 +++++++++ arch/arm64/kernel/Makefile | 1 + arch/arm64/kernel/pmem.c | 35 ++++++++++++++ arch/arm64/kernel/setup.c | 10 ++++ arch/arm64/mm/init.c | 97 ++++++++++++++++++++++++++++++++++++++ drivers/nvdimm/Kconfig | 5 ++ drivers/nvdimm/Makefile | 2 +- 7 files changed, 170 insertions(+), 1 deletion(-) create mode 100644 arch/arm64/kernel/pmem.c diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index c451137ab..326f26d40 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -360,6 +360,27 @@ config ARM64_CPU_PARK config ARCH_HAS_CPU_RELAX def_bool y +config ARM64_PMEM_RESERVE + bool "Reserve memory for persistent storage" + default n + help + Use memmap=nn[KMG]!ss[KMG](memmap=100K!0x1a0000000) reserve + memory for persistent storage. + + Say y here to enable this feature. + +config ARM64_PMEM_LEGACY_DEVICE + bool "Create persistent storage" + depends on BLK_DEV + depends on LIBNVDIMM + select ARM64_PMEM_RESERVE + help + Use reserved memory for persistent storage when the kernel + restart or update. the data in PMEM will not be lost and + can be loaded faster. + + Say y if unsure. + source "arch/arm64/Kconfig.platforms" menu "Kernel Features" diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile index 967cb3c6d..be996f3c1 100644 --- a/arch/arm64/kernel/Makefile +++ b/arch/arm64/kernel/Makefile @@ -67,6 +67,7 @@ obj-$(CONFIG_ARM64_PTR_AUTH) += pointer_auth.o obj-$(CONFIG_SHADOW_CALL_STACK) += scs.o obj-$(CONFIG_ARM64_MTE) += mte.o obj-$(CONFIG_MPAM) += mpam/ +obj-$(CONFIG_ARM64_PMEM_LEGACY_DEVICE) += pmem.o obj-y += vdso/ probes/ obj-$(CONFIG_COMPAT_VDSO) += vdso32/ diff --git a/arch/arm64/kernel/pmem.c b/arch/arm64/kernel/pmem.c new file mode 100644 index 000000000..16eaf706f --- /dev/null +++ b/arch/arm64/kernel/pmem.c @@ -0,0 +1,35 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright(c) 2021 Huawei Technologies Co., Ltd + * + * Derived from x86 and arm64 implement PMEM. + */ +#include <linux/platform_device.h> +#include <linux/init.h> +#include <linux/ioport.h> +#include <linux/module.h> + +static int found(struct resource *res, void *data) +{ + return 1; +} + +static int __init register_e820_pmem(void) +{ + struct platform_device *pdev; + int rc; + + rc = walk_iomem_res_desc(IORES_DESC_PERSISTENT_MEMORY_LEGACY, + IORESOURCE_MEM, 0, -1, NULL, found); + if (rc <= 0) + return 0; + + /* + * See drivers/nvdimm/e820.c for the implementation, this is + * simply here to trigger the module to load on demand. + */ + pdev = platform_device_alloc("e820_pmem", -1); + + return platform_device_add(pdev); +} +device_initcall(register_e820_pmem); diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index 5e282d31a..84c71c88d 100644 --- a/arch/arm64/kernel/setup.c +++ b/arch/arm64/kernel/setup.c @@ -57,6 +57,10 @@ static int num_standard_resources; static struct resource *standard_resources; +#ifdef CONFIG_ARM64_PMEM_RESERVE +extern struct resource pmem_res; +#endif + phys_addr_t __fdt_pointer __initdata; /* @@ -270,6 +274,12 @@ static void __init request_standard_resources(void) request_resource(res, &pin_memory_resource); #endif } + +#ifdef CONFIG_ARM64_PMEM_RESERVE + if (pmem_res.end && pmem_res.start) + request_resource(&iomem_resource, &pmem_res); +#endif + } static int __init reserve_memblock_reserved_regions(void) diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index b3437440d..f22faea1a 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -66,6 +66,18 @@ EXPORT_SYMBOL(memstart_addr); phys_addr_t arm64_dma_phys_limit __ro_after_init; phys_addr_t arm64_dma32_phys_limit __ro_after_init; +static unsigned long long pmem_size, pmem_start; + +#ifdef CONFIG_ARM64_PMEM_RESERVE +struct resource pmem_res = { + .name = "Persistent Memory (legacy)", + .start = 0, + .end = 0, + .flags = IORESOURCE_MEM, + .desc = IORES_DESC_PERSISTENT_MEMORY_LEGACY +}; +#endif + #ifndef CONFIG_KEXEC_CORE static void __init reserve_crashkernel(void) { @@ -378,6 +390,87 @@ static int __init reserve_park_mem(void) } #endif +static int __init is_mem_valid(unsigned long long mem_size, unsigned long long mem_start) +{ + if (!memblock_is_region_memory(mem_start, mem_size)) { + pr_warn("cannot reserve mem: region is not memory!\n"); + return -EINVAL; + } + + if (memblock_is_region_reserved(mem_start, mem_size)) { + pr_warn("cannot reserve mem: region overlaps reserved memory!\n"); + return -EINVAL; + } + + if (!IS_ALIGNED(mem_start, SZ_2M)) { + pr_warn("cannot reserve mem: base address is not 2MB aligned!\n"); + return -EINVAL; + } + + return 0; +} + +static int __init parse_memmap_one(char *p) +{ + char *oldp; + phys_addr_t start_at, mem_size; + + if (!p) + return -EINVAL; + + oldp = p; + mem_size = memparse(p, &p); + if (p == oldp) + return -EINVAL; + + if (!mem_size) + return -EINVAL; + + mem_size = PAGE_ALIGN(mem_size); + + if (*p == '!') { + start_at = memparse(p+1, &p); + + if (is_mem_valid(mem_size, start_at) != 0) + return -EINVAL; + + pr_info("pmem reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + start_at, start_at + mem_size, mem_size >> 20); + pmem_start = start_at; + pmem_size = mem_size; + } else + pr_info("Unrecognized memmap option, please check the parameter.\n"); + + return *p == '\0' ? 0 : -EINVAL; +} + +static int __init parse_memmap_opt(char *str) +{ + while (str) { + char *k = strchr(str, ','); + + if (k) + *k++ = 0; + + parse_memmap_one(str); + str = k; + } + + return 0; +} +early_param("memmap", parse_memmap_opt); + +#ifdef CONFIG_ARM64_PMEM_RESERVE +static void __init reserve_pmem(void) +{ + memblock_remove(pmem_start, pmem_size); + pr_info("pmem reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + pmem_start, pmem_start + pmem_size, pmem_size >> 20); + pmem_res.start = pmem_start; + pmem_res.end = pmem_start + pmem_size - 1; +} +#endif + void __init arm64_memblock_init(void) { const s64 linear_region_size = BIT(vabits_actual - 1); @@ -511,6 +604,10 @@ void __init arm64_memblock_init(void) reserve_elfcorehdr(); +#ifdef CONFIG_ARM64_PMEM_RESERVE + reserve_pmem(); +#endif + high_memory = __va(memblock_end_of_DRAM() - 1) + 1; dma_contiguous_reserve(arm64_dma32_phys_limit); diff --git a/drivers/nvdimm/Kconfig b/drivers/nvdimm/Kconfig index b7d1eb38b..ce4de7526 100644 --- a/drivers/nvdimm/Kconfig +++ b/drivers/nvdimm/Kconfig @@ -132,3 +132,8 @@ config NVDIMM_TEST_BUILD infrastructure. endif + +config PMEM_LEGACY + tristate "Pmem_legacy" + select X86_PMEM_LEGACY if X86 + select ARM64_PMEM_LEGACY_DEVICE if ARM64 diff --git a/drivers/nvdimm/Makefile b/drivers/nvdimm/Makefile index 29203f3d3..6f8dc9242 100644 --- a/drivers/nvdimm/Makefile +++ b/drivers/nvdimm/Makefile @@ -3,7 +3,7 @@ obj-$(CONFIG_LIBNVDIMM) += libnvdimm.o obj-$(CONFIG_BLK_DEV_PMEM) += nd_pmem.o obj-$(CONFIG_ND_BTT) += nd_btt.o obj-$(CONFIG_ND_BLK) += nd_blk.o -obj-$(CONFIG_X86_PMEM_LEGACY) += nd_e820.o +obj-$(CONFIG_PMEM_LEGACY) += nd_e820.o obj-$(CONFIG_OF_PMEM) += of_pmem.o obj-$(CONFIG_VIRTIO_PMEM) += virtio_pmem.o nd_virtio.o -- 2.19.1

1 1

[PATCH v2] arm: keep the original function for non-RPi
by Zheng Zengkai 11 Mar '21

11 Mar '21

raspberrypi inclusion category: feature bugzilla: 50432 ------------------------------ This patch adjusts following arch arm related patches for raspberry pi on non-Raspberry Pi platforms, using specific config CONFIG_OPENEULER_RASPBERRYPI to distinguish them: d5c13edbd8 Improve __copy_to_user and __copy_from_user performance 97145d2a6a Update vfpmodule.c bffc462cbd Main bcm2708/bcm2709 linux port 588cfce788 cache: export clean and invalidate 41cd350cca ARM: Activate FIQs to avoid __irq_startup warnings 90607c7aaf ARM: proc-v7: Force misalignment of early stmia ee46d0fadf reboot: Use power off rather than busy spinning when halt is requested Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> --- arch/arm/include/asm/string.h | 2 ++ arch/arm/include/asm/uaccess.h | 2 ++ arch/arm/kernel/fiq.c | 4 ++++ arch/arm/kernel/reboot.c | 6 ++++++ arch/arm/lib/Makefile | 15 +++++++++++++++ arch/arm/lib/copy_from_user.S | 6 ++++++ arch/arm/lib/uaccess_with_memcpy.c | 22 +++++++++++++++++++++- arch/arm/mm/cache-v6.S | 8 ++++++++ arch/arm/mm/cache-v7.S | 8 ++++++++ arch/arm/mm/proc-v6.S | 8 ++++++++ arch/arm/mm/proc-v7.S | 8 ++++++++ arch/arm/vfp/vfpmodule.c | 26 ++++++++++++++++++++++++++ 12 files changed, 114 insertions(+), 1 deletion(-) diff --git a/arch/arm/include/asm/string.h b/arch/arm/include/asm/string.h index c22d5869e7b6..3c4ae6b3c3a6 100644 --- a/arch/arm/include/asm/string.h +++ b/arch/arm/include/asm/string.h @@ -45,10 +45,12 @@ static inline void *memset64(uint64_t *p, uint64_t v, __kernel_size_t n) return __memset64(p, v, n * 8, v >> 32); } +#ifdef CONFIG_OPENEULER_RASPBERRYPI #ifdef CONFIG_BCM2835_FAST_MEMCPY #define __HAVE_ARCH_MEMCMP extern int memcmp(const void *, const void *, size_t); #endif +#endif /* * For files that are not instrumented (e.g. mm/slub.c) we diff --git a/arch/arm/include/asm/uaccess.h b/arch/arm/include/asm/uaccess.h index f24d3fabccd6..0d8cba7a9dde 100644 --- a/arch/arm/include/asm/uaccess.h +++ b/arch/arm/include/asm/uaccess.h @@ -516,8 +516,10 @@ do { \ extern unsigned long __must_check arm_copy_from_user(void *to, const void __user *from, unsigned long n); +#ifdef CONFIG_OPENEULER_RASPBERRYPI extern unsigned long __must_check __copy_from_user_std(void *to, const void __user *from, unsigned long n); +#endif static inline unsigned long __must_check raw_copy_from_user(void *to, const void __user *from, unsigned long n) diff --git a/arch/arm/kernel/fiq.c b/arch/arm/kernel/fiq.c index c3fe7d3cf482..8116f1e52e1b 100644 --- a/arch/arm/kernel/fiq.c +++ b/arch/arm/kernel/fiq.c @@ -56,7 +56,9 @@ static unsigned long dfl_fiq_insn; static struct pt_regs dfl_fiq_regs; +#ifdef CONFIG_OPENEULER_RASPBERRYPI extern int irq_activate(struct irq_desc *desc); +#endif /* Default reacquire function * - we always relinquish FIQ control @@ -142,8 +144,10 @@ static int fiq_start; void enable_fiq(int fiq) { +#ifdef CONFIG_OPENEULER_RASPBERRYPI struct irq_desc *desc = irq_to_desc(fiq + fiq_start); irq_activate(desc); +#endif enable_irq(fiq + fiq_start); } diff --git a/arch/arm/kernel/reboot.c b/arch/arm/kernel/reboot.c index 63373adab475..ffb170568dcc 100644 --- a/arch/arm/kernel/reboot.c +++ b/arch/arm/kernel/reboot.c @@ -102,7 +102,13 @@ void machine_shutdown(void) */ void machine_halt(void) { +#ifdef CONFIG_OPENEULER_RASPBERRYPI machine_power_off(); +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ + local_irq_disable(); + smp_send_stop(); + while (1); +#endif } /* diff --git a/arch/arm/lib/Makefile b/arch/arm/lib/Makefile index 8271cde92dec..63820a487d4b 100644 --- a/arch/arm/lib/Makefile +++ b/arch/arm/lib/Makefile @@ -5,6 +5,7 @@ # Copyright (C) 1995-2000 Russell King # +ifeq ($(CONFIG_OPENEULER_RASPBERRYPI),y) lib-y := changebit.o csumipv6.o csumpartial.o \ csumpartialcopy.o csumpartialcopyuser.o clearbit.o \ delay.o delay-loop.o findbit.o memchr.o \ @@ -15,6 +16,18 @@ lib-y := changebit.o csumipv6.o csumpartial.o \ ucmpdi2.o lib1funcs.o div64.o \ io-readsb.o io-writesb.o io-readsl.o io-writesl.o \ call_with_stack.o bswapsdi2.o +else +lib-y := changebit.o csumipv6.o csumpartial.o \ + csumpartialcopy.o csumpartialcopyuser.o clearbit.o \ + delay.o delay-loop.o findbit.o memchr.o memcpy.o \ + memmove.o memset.o setbit.o \ + strchr.o strrchr.o \ + testchangebit.o testclearbit.o testsetbit.o \ + ashldi3.o ashrdi3.o lshrdi3.o muldi3.o \ + ucmpdi2.o lib1funcs.o div64.o \ + io-readsb.o io-writesb.o io-readsl.o io-writesl.o \ + call_with_stack.o bswapsdi2.o +endif mmu-y := clear_user.o copy_page.o getuser.o putuser.o \ copy_from_user.o copy_to_user.o @@ -25,6 +38,7 @@ else lib-y += backtrace.o endif +ifeq ($(CONFIG_OPENEULER_RASPBERRYPI),y) # Choose optimised implementations for Raspberry Pi ifeq ($(CONFIG_BCM2835_FAST_MEMCPY),y) CFLAGS_uaccess_with_memcpy.o += -DCOPY_FROM_USER_THRESHOLD=1600 @@ -34,6 +48,7 @@ ifeq ($(CONFIG_BCM2835_FAST_MEMCPY),y) else lib-y += memcpy.o memmove.o memset.o endif +endif # using lib_ here won't override already available weak symbols obj-$(CONFIG_UACCESS_WITH_MEMCPY) += uaccess_with_memcpy.o diff --git a/arch/arm/lib/copy_from_user.S b/arch/arm/lib/copy_from_user.S index ab7bf28dbec0..3f83d8b18b0d 100644 --- a/arch/arm/lib/copy_from_user.S +++ b/arch/arm/lib/copy_from_user.S @@ -107,8 +107,12 @@ .text +#ifdef CONFIG_OPENEULER_RASPBERRYPI ENTRY(__copy_from_user_std) WEAK(arm_copy_from_user) +#else +ENTRY(arm_copy_from_user) +#endif #ifdef CONFIG_CPU_SPECTRE get_thread_info r3 ldr r3, [r3, #TI_ADDR_LIMIT] @@ -118,7 +122,9 @@ WEAK(arm_copy_from_user) #include "copy_template.S" ENDPROC(arm_copy_from_user) +#ifdef CONFIG_OPENEULER_RASPBERRYPI ENDPROC(__copy_from_user_std) +#endif .pushsection .text.fixup,"ax" .align 0 diff --git a/arch/arm/lib/uaccess_with_memcpy.c b/arch/arm/lib/uaccess_with_memcpy.c index b483e5713039..ab15ed7f599a 100644 --- a/arch/arm/lib/uaccess_with_memcpy.c +++ b/arch/arm/lib/uaccess_with_memcpy.c @@ -19,6 +19,7 @@ #include <asm/current.h> #include <asm/page.h> +#ifdef CONFIG_OPENEULER_RASPBERRYPI #ifndef COPY_FROM_USER_THRESHOLD #define COPY_FROM_USER_THRESHOLD 64 #endif @@ -26,6 +27,7 @@ #ifndef COPY_TO_USER_THRESHOLD #define COPY_TO_USER_THRESHOLD 64 #endif +#endif static int pin_page_for_write(const void __user *_addr, pte_t **ptep, spinlock_t **ptlp) @@ -51,7 +53,11 @@ pin_page_for_write(const void __user *_addr, pte_t **ptep, spinlock_t **ptlp) return 0; pmd = pmd_offset(pud, addr); +#ifdef CONFIG_OPENEULER_RASPBERRYPI if (unlikely(pmd_none(*pmd) || pmd_bad(*pmd))) +#else + if (unlikely(pmd_none(*pmd))) +#endif return 0; /* @@ -94,6 +100,7 @@ pin_page_for_write(const void __user *_addr, pte_t **ptep, spinlock_t **ptlp) return 1; } +#ifdef CONFIG_OPENEULER_RASPBERRYPI static int pin_page_for_read(const void __user *_addr, pte_t **ptep, spinlock_t **ptlp) { @@ -132,8 +139,13 @@ pin_page_for_read(const void __user *_addr, pte_t **ptep, spinlock_t **ptlp) return 1; } +#endif +#ifdef CONFIG_OPENEULER_RASPBERRYPI unsigned long noinline +#else +static unsigned long noinline +#endif __copy_to_user_memcpy(void __user *to, const void *from, unsigned long n) { unsigned long ua_flags; @@ -186,6 +198,7 @@ __copy_to_user_memcpy(void __user *to, const void *from, unsigned long n) return n; } +#ifdef CONFIG_OPENEULER_RASPBERRYPI unsigned long noinline __copy_from_user_memcpy(void *to, const void __user *from, unsigned long n) { @@ -236,6 +249,7 @@ __copy_from_user_memcpy(void *to, const void __user *from, unsigned long n) out: return n; } +#endif unsigned long arm_copy_to_user(void __user *to, const void *from, unsigned long n) @@ -247,7 +261,11 @@ arm_copy_to_user(void __user *to, const void *from, unsigned long n) * With frame pointer disabled, tail call optimization kicks in * as well making this test almost invisible. */ +#ifdef CONFIG_OPENEULER_RASPBERRYPI if (n < COPY_TO_USER_THRESHOLD) { +#else + if (n < 64) { +#endif unsigned long ua_flags = uaccess_save_and_enable(); n = __copy_to_user_std(to, from, n); uaccess_restore(ua_flags); @@ -258,6 +276,7 @@ arm_copy_to_user(void __user *to, const void *from, unsigned long n) return n; } +#ifdef CONFIG_OPENEULER_RASPBERRYPI unsigned long __must_check arm_copy_from_user(void *to, const void __user *from, unsigned long n) { @@ -283,7 +302,8 @@ arm_copy_from_user(void *to, const void __user *from, unsigned long n) #endif return n; } - +#endif + static unsigned long noinline __clear_user_memset(void __user *addr, unsigned long n) { diff --git a/arch/arm/mm/cache-v6.S b/arch/arm/mm/cache-v6.S index 868011801521..614d4ff2a760 100644 --- a/arch/arm/mm/cache-v6.S +++ b/arch/arm/mm/cache-v6.S @@ -198,7 +198,11 @@ ENTRY(v6_flush_kern_dcache_area) * - start - virtual start address of region * - end - virtual end address of region */ +#ifdef CONFIG_OPENEULER_RASPBERRYPI ENTRY(v6_dma_inv_range) +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ +v6_dma_inv_range: +#endif #ifdef CONFIG_DMA_CACHE_RWFO ldrb r2, [r0] @ read for ownership strb r2, [r0] @ write for ownership @@ -243,7 +247,11 @@ ENTRY(v6_dma_inv_range) * - start - virtual start address of region * - end - virtual end address of region */ +#ifdef CONFIG_OPENEULER_RASPBERRYPI ENTRY(v6_dma_clean_range) +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ +v6_dma_clean_range: +#endif bic r0, r0, #D_CACHE_LINE_SIZE - 1 1: #ifdef CONFIG_DMA_CACHE_RWFO diff --git a/arch/arm/mm/cache-v7.S b/arch/arm/mm/cache-v7.S index 536df5db66e4..58df59734e40 100644 --- a/arch/arm/mm/cache-v7.S +++ b/arch/arm/mm/cache-v7.S @@ -363,8 +363,12 @@ ENDPROC(v7_flush_kern_dcache_area) * - start - virtual start address of region * - end - virtual end address of region */ +#ifdef CONFIG_OPENEULER_RASPBERRYPI ENTRY(b15_dma_inv_range) ENTRY(v7_dma_inv_range) +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ +v7_dma_inv_range: +#endif dcache_line_size r2, r3 sub r3, r2, #1 tst r0, r3 @@ -394,8 +398,12 @@ ENDPROC(v7_dma_inv_range) * - start - virtual start address of region * - end - virtual end address of region */ +#ifdef CONFIG_OPENEULER_RASPBERRYPI ENTRY(b15_dma_clean_range) ENTRY(v7_dma_clean_range) +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ +v7_dma_clean_range: +#endif dcache_line_size r2, r3 sub r3, r2, #1 bic r0, r0, r3 diff --git a/arch/arm/mm/proc-v6.S b/arch/arm/mm/proc-v6.S index b3a2fce22eac..b651eaa1ee40 100644 --- a/arch/arm/mm/proc-v6.S +++ b/arch/arm/mm/proc-v6.S @@ -71,6 +71,7 @@ ENDPROC(cpu_v6_reset) * IRQs are already disabled. */ +#ifdef CONFIG_OPENEULER_RASPBERRYPI /* See jira SW-5991 for details of this workaround */ ENTRY(cpu_v6_do_idle) .align 5 @@ -84,6 +85,13 @@ ENTRY(cpu_v6_do_idle) nop bne 1b ret lr +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ +ENTRY(cpu_v6_do_idle) + mov r1, #0 + mcr p15, 0, r1, c7, c10, 4 @ DWB - WFI may enter a low-power mode + mcr p15, 0, r1, c7, c0, 4 @ wait for interrupt + ret lr +#endif ENTRY(cpu_v6_dcache_clean_area) 1: mcr p15, 0, r0, c7, c10, 1 @ clean D entry diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S index 3e77e8982df3..ef0e00249515 100644 --- a/arch/arm/mm/proc-v7.S +++ b/arch/arm/mm/proc-v7.S @@ -287,8 +287,10 @@ __v7_ca17mp_setup: mov r10, #0 1: adr r0, __v7_setup_stack_ptr ldr r12, [r0] +#ifdef CONFIG_OPENEULER_RASPBERRYPI tst r12, #0x1f addeq r12, r12, #4 +#endif add r12, r12, r0 @ the local stack stmia r12, {r1-r6, lr} @ v7_invalidate_l1 touches r0-r6 bl v7_invalidate_l1 @@ -476,8 +478,10 @@ __v7_setup: adr r0, __v7_setup_stack_ptr ldr r12, [r0] add r12, r12, r0 @ the local stack +#ifdef CONFIG_OPENEULER_RASPBERRYPI tst r12, #0x1f addeq r12, r12, #4 +#endif stmia r12, {r1-r6, lr} @ v7_invalidate_l1 touches r0-r6 bl v7_invalidate_l1 ldmia r12, {r1-r6, lr} @@ -561,7 +565,11 @@ ENDPROC(__v7_setup) .bss .align 2 __v7_setup_stack: +#ifdef CONFIG_OPENEULER_RASPBERRYPI .space 4 * 8 @ 7 registers + 1 spare +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ + .space 4 * 7 @ 7 registers +#endif __INITDATA diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c index 1e2dcf81aefa..f7238e00db7c 100644 --- a/arch/arm/vfp/vfpmodule.c +++ b/arch/arm/vfp/vfpmodule.c @@ -176,11 +176,16 @@ static int vfp_notifier(struct notifier_block *self, unsigned long cmd, void *v) * case the thread migrates to a different CPU. The * restoring is done lazily. */ +#ifdef CONFIG_OPENEULER_RASPBERRYPI if ((fpexc & FPEXC_EN) && vfp_current_hw_state[cpu]) { /* vfp_save_state oopses on VFP11 if EX bit set */ fmxr(FPEXC, fpexc & ~FPEXC_EX); vfp_save_state(vfp_current_hw_state[cpu], fpexc); } +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ + if ((fpexc & FPEXC_EN) && vfp_current_hw_state[cpu]) + vfp_save_state(vfp_current_hw_state[cpu], fpexc); +#endif #endif /* @@ -457,16 +462,22 @@ static int vfp_pm_suspend(void) /* if vfp is on, then save state for resumption */ if (fpexc & FPEXC_EN) { pr_debug("%s: saving vfp state\n", __func__); +#ifdef CONFIG_OPENEULER_RASPBERRYPI /* vfp_save_state oopses on VFP11 if EX bit set */ fmxr(FPEXC, fpexc & ~FPEXC_EX); +#endif vfp_save_state(&ti->vfpstate, fpexc); /* disable, just in case */ fmxr(FPEXC, fmrx(FPEXC) & ~FPEXC_EN); } else if (vfp_current_hw_state[ti->cpu]) { #ifndef CONFIG_SMP +#ifdef CONFIG_OPENEULER_RASPBERRYPI /* vfp_save_state oopses on VFP11 if EX bit set */ fmxr(FPEXC, (fpexc & ~FPEXC_EX) | FPEXC_EN); +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ + fmxr(FPEXC, fpexc | FPEXC_EN); +#endif vfp_save_state(vfp_current_hw_state[ti->cpu], fpexc); fmxr(FPEXC, fpexc); #endif @@ -529,8 +540,12 @@ void vfp_sync_hwstate(struct thread_info *thread) /* * Save the last VFP state on this CPU. */ +#ifdef CONFIG_OPENEULER_RASPBERRYPI /* vfp_save_state oopses on VFP11 if EX bit set */ fmxr(FPEXC, (fpexc & ~FPEXC_EX) | FPEXC_EN); +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ + fmxr(FPEXC, fpexc | FPEXC_EN); +#endif vfp_save_state(&thread->vfpstate, fpexc | FPEXC_EN); fmxr(FPEXC, fpexc); } @@ -596,7 +611,9 @@ int vfp_restore_user_hwstate(struct user_vfp *ufp, struct user_vfp_exc *ufp_exc) struct thread_info *thread = current_thread_info(); struct vfp_hard_struct *hwstate = &thread->vfpstate.hard; unsigned long fpexc; +#ifdef CONFIG_OPENEULER_RASPBERRYPI u32 fpsid = fmrx(FPSID); +#endif /* Disable VFP to avoid corrupting the new thread state. */ vfp_flush_hwstate(thread); @@ -619,11 +636,16 @@ int vfp_restore_user_hwstate(struct user_vfp *ufp, struct user_vfp_exc *ufp_exc) /* Ensure the VFP is enabled. */ fpexc |= FPEXC_EN; +#ifdef CONFIG_OPENEULER_RASPBERRYPI /* Mask FPXEC_EX and FPEXC_FP2V if not required by VFP arch */ if ((fpsid & FPSID_ARCH_MASK) != (1 << FPSID_ARCH_BIT)) { /* Ensure FPINST2 is invalid and the exception flag is cleared. */ fpexc &= ~(FPEXC_EX | FPEXC_FP2V); } +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ + /* Ensure FPINST2 is invalid and the exception flag is cleared. */ + fpexc &= ~(FPEXC_EX | FPEXC_FP2V); +#endif hwstate->fpexc = fpexc; @@ -738,8 +760,12 @@ void kernel_neon_begin(void) cpu = get_cpu(); fpexc = fmrx(FPEXC) | FPEXC_EN; +#ifdef CONFIG_OPENEULER_RASPBERRYPI /* vfp_save_state oopses on VFP11 if EX bit set */ fmxr(FPEXC, fpexc & ~FPEXC_EX); +#else /* !CONFIG_OPENEULER_RASPBERRYPI */ + fmxr(FPEXC, fpexc); +#endif /* * Save the userland NEON/VFP state. Under UP, -- 2.20.1

2 1

[PATCH openEuler-1.0-LTS] sysctl: control if check validity of freelist and connector debug by sysctl
by Cheng Jian 11 Mar '21

11 Mar '21

From: Yang Yingliang <yangyingliang(a)huawei.com> hulk inclusion category: bugfix bugzilla: 47452 CVE: NA ------------------------------------------------- Change control methods of check validity of freelist and connector debug to sysctl. 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> Signed-off-by: Cheng Jian <cj.chengjian(a)huawei.com> --- drivers/connector/connector.c | 1 + include/linux/slab.h | 4 --- kernel/sysctl.c | 20 +++++++++++ mm/slub.c | 63 ++++++++++++++--------------------- 4 files changed, 46 insertions(+), 42 deletions(-) diff --git a/drivers/connector/connector.c b/drivers/connector/connector.c index a8df9ecbf42b..b5cdd0b99736 100644 --- a/drivers/connector/connector.c +++ b/drivers/connector/connector.c @@ -42,6 +42,7 @@ MODULE_ALIAS_NET_PF_PROTO(PF_NETLINK, NETLINK_CONNECTOR); static struct cn_dev cdev; static int cn_already_initialized; +int sysctl_connector_debug = 0; /* * Sends mult (multiple) cn_msg at a time. diff --git a/include/linux/slab.h b/include/linux/slab.h index 1ab5fd97dec8..d6393413ef09 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -24,8 +24,6 @@ */ /* DEBUG: Perform (expensive) checks on alloc/free */ #define SLAB_CONSISTENCY_CHECKS ((slab_flags_t __force)0x00000100U) -/* Check if freelist is valid */ -#define SLAB_NO_FREELIST_CHECKS ((slab_flags_t __force)0x00000200U) /* DEBUG: Red zone objs in a cache */ #define SLAB_RED_ZONE ((slab_flags_t __force)0x00000400U) /* DEBUG: Poison objects */ @@ -113,8 +111,6 @@ #define SLAB_KASAN 0 #endif -#define SLAB_CONNECTOR_DEBUG ((slab_flags_t __force)0x10000000U) - /* The following flags affect the page allocator grouping pages by mobility */ /* Objects are reclaimable */ #define SLAB_RECLAIM_ACCOUNT ((slab_flags_t __force)0x00020000U) diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 665c9e2a8802..c921ee10615a 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1245,6 +1245,7 @@ static struct ctl_table kern_table[] = { { } }; +extern int sysctl_isolate_corrupted_freelist; static struct ctl_table vm_table[] = { { .procname = "overcommit_memory", @@ -1714,6 +1715,15 @@ static struct ctl_table vm_table[] = { .extra2 = (void *)&mmap_rnd_compat_bits_max, }, #endif + { + .procname = "isolate_corrupted_freelist", + .data = &sysctl_isolate_corrupted_freelist, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, { } }; @@ -1925,6 +1935,7 @@ static struct ctl_table fs_table[] = { { } }; +extern int sysctl_connector_debug; static struct ctl_table debug_table[] = { #ifdef CONFIG_SYSCTL_EXCEPTION_TRACE { @@ -1946,6 +1957,15 @@ static struct ctl_table debug_table[] = { .extra2 = &one, }, #endif + { + .procname = "connector-debug", + .data = &sysctl_connector_debug, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, { } }; diff --git a/mm/slub.c b/mm/slub.c index ee0d18699bde..2301c04353d0 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -1221,30 +1221,6 @@ static noinline int free_debug_processing( return ret; } -/* - * Check if the next object in freechain is - * valid, then isolate the corrupted freelist. - */ -int isolate_cnt = 0; -static bool isolate_corrupted_freelist(struct kmem_cache *s, struct page *page, - void **freelist, void *next) -{ - if (!(slub_debug & SLAB_NO_FREELIST_CHECKS) && - (!check_valid_pointer(s, page, next))) { - /* Need caller make sure freelist is not NULL */ - *freelist = NULL; - isolate_cnt++; - - slab_fix(s, "Freelist corrupt,isolate corrupted freechain."); - pr_info("freelist=%lx object=%lx, page->base=%lx, page->objects=%u, objsize=%u, s->offset=%d page:%llx\n", - (long)freelist, (long)next, (long)page_address(page), page->objects, s->size, s->offset, (u64)page); - - return true; - } - - return false; -} - static int __init setup_slub_debug(char *str) { slub_debug = DEBUG_DEFAULT_FLAGS; @@ -1291,14 +1267,6 @@ static int __init setup_slub_debug(char *str) case 'a': slub_debug |= SLAB_FAILSLAB; break; - case 'n': - slub_debug |= SLAB_NO_FREELIST_CHECKS; - pr_info("Freelist pointer check disabled."); - break; - case 'd': - slub_debug |= SLAB_CONNECTOR_DEBUG; - pr_info("Connector debug enabled.\n"); - break; case 'o': /* * Avoid enabling debugging on caches if its minimum @@ -1311,8 +1279,6 @@ static int __init setup_slub_debug(char *str) *str); } } - if (!(slub_debug & SLAB_NO_FREELIST_CHECKS)) - pr_info("Freelist pointer check enabled."); check_slabs: if (*str == ',') @@ -1380,12 +1346,33 @@ static bool freelist_corrupted(struct kmem_cache *s, struct page *page, { return false; } +#endif /* CONFIG_SLUB_DEBUG */ + +extern int sysctl_connector_debug; +/* + * Check if the next object in freechain is + * valid, then isolate the corrupted freelist. + */ +int isolate_cnt = 0; +int sysctl_isolate_corrupted_freelist = 1; static bool isolate_corrupted_freelist(struct kmem_cache *s, struct page *page, - void **freelist, void *nextfree) + void **freelist, void *next) { + if (sysctl_isolate_corrupted_freelist && + (!check_valid_pointer(s, page, next))) { + /* Need caller make sure freelist is not NULL */ + *freelist = NULL; + isolate_cnt++; + + slab_fix(s, "Freelist corrupt,isolate corrupted freechain."); + pr_info("freelist=%lx object=%lx, page->base=%lx, page->objects=%u, objsize=%u, s->offset=%d page:%llx\n", + (long)freelist, (long)next, (long)page_address(page), page->objects, s->size, s->offset, (u64)page); + + return true; + } + return false; } -#endif /* CONFIG_SLUB_DEBUG */ /* * Hooks for other subsystems that check memory allocations. In a typical @@ -2798,7 +2785,7 @@ static __always_inline void *slab_alloc_node(struct kmem_cache *s, if (unlikely(gfpflags & __GFP_ZERO) && object) memset(object, 0, s->object_size); - if ((slub_debug & SLAB_CONNECTOR_DEBUG) && + if (sysctl_connector_debug && unlikely(gfpflags & __GFP_CONNECTOR) && object) { if (s->object_size == 512) memset(object + 504, 0xad, 8); @@ -4015,7 +4002,7 @@ void kfree(const void *x) __free_pages(page, compound_order(page)); return; } - if ((slub_debug & SLAB_CONNECTOR_DEBUG) && + if (sysctl_connector_debug && unlikely(page->slab_cache->object_size == 512)) { u64 *tail = (u64 *)(x + 504); -- 2.25.1

1 0

[PATCH openEuler-1.0-LTS 00/82] bugfix for 20.03 at 202102
by Cheng Jian 11 Mar '21

11 Mar '21

bugfix for 20.03 at 202102 Alexey Gladkov (1): moduleparam: Save information about built-in modules in separate file Amir Goldstein (1): ovl: skip getxattr of security labels Aurelien Aptel (1): cifs: report error instead of invalid when revalidating a dentry fails Baptiste Lepers (1): udp: Prevent reuseport_select_sock from reading uninitialized socks Byron Stanoszek (1): tmpfs: restore functionality of nr_inodes=0 Chris Down (2): tmpfs: per-superblock i_ino support tmpfs: support 64-bit inums per-sb Christian Brauner (1): sysctl: handle overflow in proc_get_long Cong Wang (1): af_key: relax availability checks for skb size calculation Dave Wysochanski (2): SUNRPC: Move simple_get_bytes and simple_get_netobj into private header SUNRPC: Handle 0 length opaque XDR object data properly Dong Kai (1): livepatch/core: Fix jump_label_apply_nops called multi times Edwin Peer (1): net: watchdog: hold device global xmit lock during tx disable Eric Biggers (1): fs: fix lazytime expiration handling in __writeback_single_inode() Eric Dumazet (1): net_sched: gen_estimator: support large ewma log Eyal Birger (1): xfrm: fix disable_xfrm sysctl when used on xfrm interfaces Florian Westphal (1): netfilter: conntrack: skip identical origin tuple in same zone only Gaurav Kohli (1): tracing: Fix race in trace_open and buffer resize call Gustavo A. R. Silva (1): smb3: Fix out-of-bounds bug in SMB2_negotiate() Hugh Dickins (1): mm: thp: fix MADV_REMOVE deadlock on shmem THP Jakub Kicinski (1): net: sit: unregister_netdevice on newlink's error path Jan Kara (1): writeback: Drop I_DIRTY_TIME_EXPIRE Jozsef Kadlecsik (1): netfilter: xt_recent: Fix attempt to update deleted entry Liangyan (1): ovl: fix dentry leak in ovl_get_redirect Lin Feng (1): bfq-iosched: Revert "bfq: Fix computation of shallow depth" Marc Zyngier (1): genirq/msi: Activate Multi-MSI early when MSI_FLAG_ACTIVATE_EARLY is set Martin K. Petersen (2): scsi: sd: block: Fix regressions in read-only block device handling scsi: sd: block: Fix kabi change by 'scsi: sd: block: Fix regressions in read-only block device handling' Martin Willi (1): vrf: Fix fast path output packet handling with async Netfilter rules Masami Hiramatsu (1): tracing/kprobe: Fix to support kretprobe events on unloaded modules Miklos Szeredi (4): proc/mounts: add cursor ovl: perform vfs_getxattr() with mounter creds cap: fix conversions on getxattr ovl: expand warning in ovl_d_real() Mikulas Patocka (1): dm integrity: conditionally disable "recalculate" feature Ming Lei (4): scsi: core: Run queue in case of I/O resource contention failure scsi: core: Only re-run queue in scsi_end_request() if device queue is busy block: don't hold q->sysfs_lock in elevator_init_mq blk-mq: don't hold q->sysfs_lock in blk_mq_map_swqueue Muchun Song (4): mm: hugetlbfs: fix cannot migrate the fallocated HugeTLB page mm: hugetlb: fix a race between freeing and dissolving the page mm: hugetlb: fix a race between isolating and freeing page mm: hugetlb: remove VM_BUG_ON_PAGE from page_huge_active NeilBrown (1): net: fix iteration for sctp transport seq_files Ondrej Jirman (1): brcmfmac: Loading the correct firmware for brcm43456 Pablo Neira Ayuso (1): netfilter: nft_dynset: add timeout extension to template Pavel Begunkov (1): list: introduce list_for_each_continue() Pengcheng Yang (1): tcp: fix TLP timer not set when CA_STATE changes from DISORDER to OPEN Peter Zijlstra (2): kthread: Extract KTHREAD_IS_PER_CPU workqueue: Restrict affinity change to rescuer Roi Dayan (1): net/mlx5: Fix memory leak on flow table creation error flow Roman Gushchin (1): memblock: do not start bottom-up allocations with kernel_end Sabyrzhan Tasbolatov (1): net/rds: restrict iovecs length for RDS_CMSG_RDMA_ARGS Shmulik Ladkani (1): xfrm: Fix oops in xfrm_replay_advance_bmp Steven Rostedt (VMware) (3): fgraph: Initialize tracing_graph_pause at task creation tracing: Do not count ftrace events in top level enable output tracing: Check length before giving out the filter buffer Sven Auhagen (1): netfilter: flowtable: fix tcp and udp header checksum update Vadim Fedorenko (1): net: ip_tunnel: fix mtu calculation Wang Hai (1): Revert "mm/slub: fix a memory leak in sysfs_slab_add()" Wang ShaoBo (1): kretprobe: Avoid re-registration of the same kretprobe earlier Willem de Bruijn (1): esp: avoid unneeded kmap_atomic call Xiao Ni (1): md: Set prev_flush_start and flush_bio in an atomic way Yang Yingliang (3): sysctl/mm: fix compile error when CONFIG_SLUB is disabled config: disable config TMPFS_INODE64 by default connector: change GFP_CONNECTOR to bit-31 Ye Bin (3): scsi: sd: block: Fix read-only flag residuals when partition table change Revert "scsi: sd: block: Fix read-only flag residuals when partition table change" Revert "scsi: sg: fix memory leak in sg_build_indirect" Yonglong Liu (4): net: hns3: adds support for setting pf max tx rate via sysfs net: hns3: update hns3 version to 1.9.38.10 net: hns3: fix 'ret' may be used uninitialized problem net: hns3: update hns3 version to 1.9.38.11 Yufen Yu (1): scsi: fix kabi for scsi_device Zhang Xiaoxu (1): proc/mounts: Fix kabi broken liubo (3): etmem: add etmem-scan feature etmem: add etmem-swap feature config: Enable the config option of the etmem feature shiyongbang (3): gpu: hibmc: Fix erratic display during startup stage. gpu: hibmc: Use drm get pci dev api. gpu: hibmc: Fix stuck when switch GUI to text. zhangyi (F) (1): ext4: find old entry again if failed to rename whiteout .gitignore | 1 + Documentation/device-mapper/dm-integrity.txt | 7 + Documentation/dontdiff | 1 + Documentation/filesystems/tmpfs.txt | 17 + Documentation/kbuild/kbuild.txt | 5 + Makefile | 2 + arch/arm64/configs/hulk_defconfig | 3 + arch/arm64/configs/openeuler_defconfig | 2 + arch/x86/configs/hulk_defconfig | 2 + arch/x86/configs/openeuler_defconfig | 2 + block/bfq-iosched.c | 8 +- block/blk-mq.c | 7 - block/elevator.c | 14 +- block/genhd.c | 33 +- block/ioctl.c | 4 + block/partition-generic.c | 7 +- drivers/connector/connector.c | 6 +- .../gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c | 68 +- .../gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h | 2 + .../gpu/drm/hisilicon/hibmc/hibmc_drm_vdac.c | 1 + drivers/md/dm-integrity.c | 24 +- drivers/md/md.c | 2 + drivers/net/ethernet/hisilicon/hns3/Makefile | 3 +- drivers/net/ethernet/hisilicon/hns3/hnae3.h | 10 +- .../hns3/hns3_cae/hns3_cae_version.h | 2 +- .../net/ethernet/hisilicon/hns3/hns3_enet.h | 2 +- .../hns3_extension/hns3pf/hclge_main_it.c | 33 + .../hns3/hns3_extension/hns3pf/hclge_sysfs.c | 140 +++ .../hns3/hns3_extension/hns3pf/hclge_sysfs.h | 14 + .../hisilicon/hns3/hns3pf/hclge_main.c | 17 + .../hisilicon/hns3/hns3pf/hclge_main.h | 7 +- .../hisilicon/hns3/hns3vf/hclgevf_main.h | 2 +- .../net/ethernet/mellanox/mlx5/core/fs_core.c | 1 + drivers/net/vrf.c | 92 +- .../broadcom/brcm80211/brcmfmac/sdio.c | 4 +- drivers/scsi/scsi_lib.c | 59 +- drivers/scsi/sg.c | 6 +- fs/Kconfig | 21 + fs/cifs/dir.c | 22 +- fs/cifs/smb2pdu.h | 2 +- fs/ext4/inode.c | 2 +- fs/ext4/namei.c | 29 +- fs/fs-writeback.c | 36 +- fs/hugetlbfs/inode.c | 3 +- fs/mount.h | 17 +- fs/namespace.c | 119 +- fs/overlayfs/copy_up.c | 15 +- fs/overlayfs/dir.c | 2 +- fs/overlayfs/inode.c | 2 + fs/overlayfs/super.c | 13 +- fs/proc/Makefile | 2 + fs/proc/base.c | 4 + fs/proc/etmem_scan.c | 1046 +++++++++++++++++ fs/proc/etmem_scan.h | 132 +++ fs/proc/etmem_swap.c | 102 ++ fs/proc/internal.h | 2 + fs/proc/task_mmu.c | 117 ++ fs/proc_namespace.c | 4 +- fs/xfs/xfs_trans_inode.c | 4 +- include/asm-generic/vmlinux.lds.h | 1 + include/linux/fs.h | 16 +- include/linux/genhd.h | 5 + include/linux/gfp.h | 8 +- include/linux/hugetlb.h | 3 + include/linux/kprobes.h | 2 +- include/linux/kthread.h | 3 + include/linux/list.h | 10 + include/linux/mm_types.h | 18 + include/linux/module.h | 1 + include/linux/moduleparam.h | 12 +- include/linux/mount.h | 4 +- include/linux/msi.h | 6 + include/linux/netdevice.h | 2 + include/linux/shmem_fs.h | 3 + include/linux/sunrpc/xdr.h | 3 +- include/linux/swap.h | 5 + include/net/tcp.h | 2 +- include/scsi/scsi_device.h | 3 +- include/trace/events/writeback.h | 1 - init/init_task.c | 3 +- kernel/irq/msi.c | 44 +- kernel/kprobes.c | 36 +- kernel/kthread.c | 27 +- kernel/livepatch/core.c | 20 +- kernel/smpboot.c | 1 + kernel/sysctl.c | 6 + kernel/trace/ftrace.c | 2 - kernel/trace/ring_buffer.c | 4 + kernel/trace/trace.c | 2 +- kernel/trace/trace_events.c | 3 +- kernel/trace/trace_kprobe.c | 4 +- kernel/workqueue.c | 9 +- lib/Kconfig | 11 + mm/huge_memory.c | 37 +- mm/hugetlb.c | 48 +- mm/memblock.c | 49 +- mm/pagewalk.c | 1 + mm/shmem.c | 133 ++- mm/slub.c | 14 +- mm/vmscan.c | 112 ++ mm/vmstat.c | 4 + net/core/gen_estimator.c | 11 +- net/core/sock_reuseport.c | 2 +- net/ipv4/esp4.c | 7 +- net/ipv4/ip_tunnel.c | 16 +- net/ipv4/tcp_input.c | 10 +- net/ipv4/tcp_recovery.c | 5 +- net/ipv6/esp6.c | 7 +- net/ipv6/sit.c | 5 +- net/key/af_key.c | 6 +- net/netfilter/nf_conntrack_core.c | 3 +- net/netfilter/nf_flow_table_core.c | 4 +- net/netfilter/nft_dynset.c | 4 +- net/netfilter/xt_recent.c | 12 +- net/rds/rdma.c | 3 + net/sctp/proc.c | 16 +- net/sunrpc/auth_gss/auth_gss.c | 30 +- net/sunrpc/auth_gss/auth_gss_internal.h | 45 + net/sunrpc/auth_gss/gss_krb5_mech.c | 31 +- net/xfrm/xfrm_input.c | 2 +- net/xfrm/xfrm_policy.c | 4 +- scripts/link-vmlinux.sh | 3 + security/commoncap.c | 67 +- virt/kvm/kvm_main.c | 6 + 124 files changed, 2817 insertions(+), 466 deletions(-) create mode 100644 drivers/net/ethernet/hisilicon/hns3/hns3_extension/hns3pf/hclge_sysfs.c create mode 100644 drivers/net/ethernet/hisilicon/hns3/hns3_extension/hns3pf/hclge_sysfs.h create mode 100644 fs/proc/etmem_scan.c create mode 100644 fs/proc/etmem_scan.h create mode 100644 fs/proc/etmem_swap.c create mode 100644 net/sunrpc/auth_gss/auth_gss_internal.h -- 2.25.1

1 82

[PATCH 01/82] list: introduce list_for_each_continue()
by Cheng Jian 11 Mar '21

11 Mar '21

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 11

[PATCH openEuler-21.03 v1] park: Reserve park mem before kexec reserved
by sangyan＠huawei.com 10 Mar '21

10 Mar '21

From: Sang Yan <sangyan(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: N/A ------------------------------ reserve_crashkernel or reserve_quick_kexec may find one sutiable memory region and reserves it, which address of the region is not fixed. As a result, cpu park reserves memory could be failed while specified address used by crashkernel or quickkexec. So, move reserve_park_mem before reserve_crashkernel and reserve_quick_kexec. Signed-off-by: Sang Yan <sangyan(a)huawei.com> --- arch/arm64/mm/init.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index b343744..043a981 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -497,16 +497,16 @@ void __init arm64_memblock_init(void) else arm64_dma32_phys_limit = PHYS_MASK + 1; +#ifdef CONFIG_ARM64_CPU_PARK + reserve_park_mem(); +#endif + reserve_crashkernel(); #ifdef CONFIG_QUICK_KEXEC reserve_quick_kexec(); #endif -#ifdef CONFIG_ARM64_CPU_PARK - reserve_park_mem(); -#endif - reserve_pin_memory_res(); reserve_elfcorehdr(); -- 2.9.5

1 0

[PATCH openEuler-21.03 v4 1/2] arm64: Add memmap parameter and register pmem
by Zhuling 10 Mar '21

10 Mar '21

From: ZhuLing <zhuling8(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: NA Register pmem in arm64: Use memmap(memmap=nn[KMG]!ss[KMG]) reserve memory and e820(driver/nvdimm/e820.c) function to register persistent memory in arm64. when the kernel restart or update, the data in PMEM will not be lost and can be loaded faster. this is a general features. driver/nvdimm/e820.c: The function of this file is scan "iomem_resource" and take advantage of nvdimm resource discovery mechanism by registering a resource named "Persistent Memory (legacy)", this function doesn't depend on architecture. We will push the feature to linux kernel community and discuss to modify the file name. because people have a mistaken notion that the e820.c is depend on x86. If you want use this features, you need do as follows: 1.Reserve memory: add memmap to reserve memory in grub.cfg memmap=nn[KMG]!ss[KMG] exp:memmap=100K!0x1a0000000. 2.Insmod nd_e820.ko: modprobe nd_e820. 3.Check pmem device in /dev exp: /dev/pmem0 Signed-off-by: ZhuLing <zhuling8(a)huawei.com> --- arch/arm64/Kconfig | 21 +++++++++ arch/arm64/kernel/Makefile | 1 + arch/arm64/kernel/pmem.c | 35 ++++++++++++++ arch/arm64/kernel/setup.c | 10 ++++ arch/arm64/mm/init.c | 97 ++++++++++++++++++++++++++++++++++++++ drivers/nvdimm/Kconfig | 5 ++ drivers/nvdimm/Makefile | 2 +- 7 files changed, 170 insertions(+), 1 deletion(-) create mode 100644 arch/arm64/kernel/pmem.c diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index c451137ab..326f26d40 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -360,6 +360,27 @@ config ARM64_CPU_PARK config ARCH_HAS_CPU_RELAX def_bool y +config ARM64_PMEM_RESERVE + bool "Reserve memory for persistent storage" + default n + help + Use memmap=nn[KMG]!ss[KMG](memmap=100K!0x1a0000000) reserve + memory for persistent storage. + + Say y here to enable this feature. + +config ARM64_PMEM_LEGACY_DEVICE + bool "Create persistent storage" + depends on BLK_DEV + depends on LIBNVDIMM + select ARM64_PMEM_RESERVE + help + Use reserved memory for persistent storage when the kernel + restart or update. the data in PMEM will not be lost and + can be loaded faster. + + Say y if unsure. + source "arch/arm64/Kconfig.platforms" menu "Kernel Features" diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile index 967cb3c6d..be996f3c1 100644 --- a/arch/arm64/kernel/Makefile +++ b/arch/arm64/kernel/Makefile @@ -67,6 +67,7 @@ obj-$(CONFIG_ARM64_PTR_AUTH) += pointer_auth.o obj-$(CONFIG_SHADOW_CALL_STACK) += scs.o obj-$(CONFIG_ARM64_MTE) += mte.o obj-$(CONFIG_MPAM) += mpam/ +obj-$(CONFIG_ARM64_PMEM_LEGACY_DEVICE) += pmem.o obj-y += vdso/ probes/ obj-$(CONFIG_COMPAT_VDSO) += vdso32/ diff --git a/arch/arm64/kernel/pmem.c b/arch/arm64/kernel/pmem.c new file mode 100644 index 000000000..16eaf706f --- /dev/null +++ b/arch/arm64/kernel/pmem.c @@ -0,0 +1,35 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright(c) 2021 Huawei Technologies Co., Ltd + * + * Derived from x86 and arm64 implement PMEM. + */ +#include <linux/platform_device.h> +#include <linux/init.h> +#include <linux/ioport.h> +#include <linux/module.h> + +static int found(struct resource *res, void *data) +{ + return 1; +} + +static int __init register_e820_pmem(void) +{ + struct platform_device *pdev; + int rc; + + rc = walk_iomem_res_desc(IORES_DESC_PERSISTENT_MEMORY_LEGACY, + IORESOURCE_MEM, 0, -1, NULL, found); + if (rc <= 0) + return 0; + + /* + * See drivers/nvdimm/e820.c for the implementation, this is + * simply here to trigger the module to load on demand. + */ + pdev = platform_device_alloc("e820_pmem", -1); + + return platform_device_add(pdev); +} +device_initcall(register_e820_pmem); diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index 5e282d31a..84c71c88d 100644 --- a/arch/arm64/kernel/setup.c +++ b/arch/arm64/kernel/setup.c @@ -57,6 +57,10 @@ static int num_standard_resources; static struct resource *standard_resources; +#ifdef CONFIG_ARM64_PMEM_RESERVE +extern struct resource pmem_res; +#endif + phys_addr_t __fdt_pointer __initdata; /* @@ -270,6 +274,12 @@ static void __init request_standard_resources(void) request_resource(res, &pin_memory_resource); #endif } + +#ifdef CONFIG_ARM64_PMEM_RESERVE + if (pmem_res.end && pmem_res.start) + request_resource(&iomem_resource, &pmem_res); +#endif + } static int __init reserve_memblock_reserved_regions(void) diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index b3437440d..f22faea1a 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -66,6 +66,18 @@ EXPORT_SYMBOL(memstart_addr); phys_addr_t arm64_dma_phys_limit __ro_after_init; phys_addr_t arm64_dma32_phys_limit __ro_after_init; +static unsigned long long pmem_size, pmem_start; + +#ifdef CONFIG_ARM64_PMEM_RESERVE +struct resource pmem_res = { + .name = "Persistent Memory (legacy)", + .start = 0, + .end = 0, + .flags = IORESOURCE_MEM, + .desc = IORES_DESC_PERSISTENT_MEMORY_LEGACY +}; +#endif + #ifndef CONFIG_KEXEC_CORE static void __init reserve_crashkernel(void) { @@ -378,6 +390,87 @@ static int __init reserve_park_mem(void) } #endif +static int __init is_mem_valid(unsigned long long mem_size, unsigned long long mem_start) +{ + if (!memblock_is_region_memory(mem_start, mem_size)) { + pr_warn("cannot reserve mem: region is not memory!\n"); + return -EINVAL; + } + + if (memblock_is_region_reserved(mem_start, mem_size)) { + pr_warn("cannot reserve mem: region overlaps reserved memory!\n"); + return -EINVAL; + } + + if (!IS_ALIGNED(mem_start, SZ_2M)) { + pr_warn("cannot reserve mem: base address is not 2MB aligned!\n"); + return -EINVAL; + } + + return 0; +} + +static int __init parse_memmap_one(char *p) +{ + char *oldp; + phys_addr_t start_at, mem_size; + + if (!p) + return -EINVAL; + + oldp = p; + mem_size = memparse(p, &p); + if (p == oldp) + return -EINVAL; + + if (!mem_size) + return -EINVAL; + + mem_size = PAGE_ALIGN(mem_size); + + if (*p == '!') { + start_at = memparse(p+1, &p); + + if (is_mem_valid(mem_size, start_at) != 0) + return -EINVAL; + + pr_info("pmem reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + start_at, start_at + mem_size, mem_size >> 20); + pmem_start = start_at; + pmem_size = mem_size; + } else + pr_info("Unrecognized memmap option, please check the parameter.\n"); + + return *p == '\0' ? 0 : -EINVAL; +} + +static int __init parse_memmap_opt(char *str) +{ + while (str) { + char *k = strchr(str, ','); + + if (k) + *k++ = 0; + + parse_memmap_one(str); + str = k; + } + + return 0; +} +early_param("memmap", parse_memmap_opt); + +#ifdef CONFIG_ARM64_PMEM_RESERVE +static void __init reserve_pmem(void) +{ + memblock_remove(pmem_start, pmem_size); + pr_info("pmem reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + pmem_start, pmem_start + pmem_size, pmem_size >> 20); + pmem_res.start = pmem_start; + pmem_res.end = pmem_start + pmem_size - 1; +} +#endif + void __init arm64_memblock_init(void) { const s64 linear_region_size = BIT(vabits_actual - 1); @@ -511,6 +604,10 @@ void __init arm64_memblock_init(void) reserve_elfcorehdr(); +#ifdef CONFIG_ARM64_PMEM_RESERVE + reserve_pmem(); +#endif + high_memory = __va(memblock_end_of_DRAM() - 1) + 1; dma_contiguous_reserve(arm64_dma32_phys_limit); diff --git a/drivers/nvdimm/Kconfig b/drivers/nvdimm/Kconfig index b7d1eb38b..ce4de7526 100644 --- a/drivers/nvdimm/Kconfig +++ b/drivers/nvdimm/Kconfig @@ -132,3 +132,8 @@ config NVDIMM_TEST_BUILD infrastructure. endif + +config PMEM_LEGACY + tristate "Pmem_legacy" + select X86_PMEM_LEGACY if X86 + select ARM64_PMEM_LEGACY_DEVICE if ARM64 diff --git a/drivers/nvdimm/Makefile b/drivers/nvdimm/Makefile index 29203f3d3..6f8dc9242 100644 --- a/drivers/nvdimm/Makefile +++ b/drivers/nvdimm/Makefile @@ -3,7 +3,7 @@ obj-$(CONFIG_LIBNVDIMM) += libnvdimm.o obj-$(CONFIG_BLK_DEV_PMEM) += nd_pmem.o obj-$(CONFIG_ND_BTT) += nd_btt.o obj-$(CONFIG_ND_BLK) += nd_blk.o -obj-$(CONFIG_X86_PMEM_LEGACY) += nd_e820.o +obj-$(CONFIG_PMEM_LEGACY) += nd_e820.o obj-$(CONFIG_OF_PMEM) += of_pmem.o obj-$(CONFIG_VIRTIO_PMEM) += virtio_pmem.o nd_virtio.o -- 2.19.1

1 1

(no subject)
by 点击设置昵称 09 Mar '21

09 Mar '21

ok!

1 0

openEuler kernel sig meeting
by openEuler conference 08 Mar '21

08 Mar '21

4 7

[PATCH] kbuild: fix compile error in Makefile.lib
by fangyafenqidai＠163.com 08 Mar '21

08 Mar '21

From: Fang Yafen <yafen(a)iscas.ac.cn> Fix the following compile error when using bcm2711_defconfig (for RPi). /bin/sh: -c: line 0: syntax error near unexpected token `(' /bin/sh: -c: line 0: `set -e; echo ' DTCO arch/arm64/boot/dts/overlays/act-led.dtbo'; mkdir -p arch/arm64/boot/dts/overlays/ ; gcc -E -Wp,-MMD,arch/arm64/boot/dts/overlays/.act-led.dtbo.d.pre.tmp -nostdinc -I./scripts/dtc/include-prefixes -undef -D__DTS__ -x assembler-with-cpp -o arch/arm64/boot/dts/overlays/.act-led.dtbo.dts.tmp arch/arm64/boot/dts/overlays/act-led-overlay.dts ; ./scripts/dtc/dtc -@ -H epapr -O dtb -o arch/arm64/boot/dts/overlays/act-led.dtbo -b 0 -i arch/arm64/boot/dts/overlays/ -Wno-interrupt_provider -Wno-unit_address_vs_reg -Wno-unit_address_format -Wno-gpios_property -Wno-avoid_unnecessary_addr_size -Wno-alias_paths -Wno-graph_child_address -Wno-simple_bus_reg -Wno-unique_unit_address -Wno-pci_device_reg -Wno-interrupts_property ifeq (y,y) -Wno-label_is_string -Wno-reg_format -Wno-pci_device_bus_num -Wno-i2c_bus_reg -Wno-spi_bus_reg -Wno-avoid_default_addr_size endif -d arch/arm64/boot/dts/overlays/.act-led.dtbo.d.dtc.tmp arch/arm64/boot/dts/overlays/.act-led.dtbo.dts.tmp ; cat ...; rm -f arch/arm64/boot/dts/overlays/.act-led.dtbo.d' make[2]: *** [scripts/Makefile.lib;363: arch/arm64/boot/dts/overlays/act-led.dtbo] Error 1 make[2]: *** Waiting for unfinished jobs.... Related patches: ffa2d13ccc3 BCM2708: Add core Device Tree support 4894352ec98 kbuild: Silence unavoidable dtc overlay warnings a4a4d07f0cf kbuild: keep the original function for non-RPi Signed-off-by: Fang Yafen <yafen(a)iscas.ac.cn> --- scripts/Makefile.lib | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib index c23e3ae7ef40..a0e0e2543165 100644 --- a/scripts/Makefile.lib +++ b/scripts/Makefile.lib @@ -342,25 +342,25 @@ endef $(obj)/%.dt.yaml: $(src)/%.dts $(DTC) $(DT_TMP_SCHEMA) FORCE $(call if_changed_rule,dtc,yaml) +ifeq ($(CONFIG_OPENEULER_RASPBERRYPI),y) quiet_cmd_dtco = DTCO $@ cmd_dtco = mkdir -p $(dir ${dtc-tmp}) ; \ $(CPP) $(dtc_cpp_flags) -x assembler-with-cpp -o $(dtc-tmp) $< ; \ $(DTC) -@ -H epapr -O dtb -o $@ -b 0 \ -i $(dir $<) $(DTC_FLAGS) \ -Wno-interrupts_property \ -ifeq ($(CONFIG_OPENEULER_RASPBERRYPI),y) \ -Wno-label_is_string \ -Wno-reg_format \ -Wno-pci_device_bus_num \ -Wno-i2c_bus_reg \ -Wno-spi_bus_reg \ -Wno-avoid_default_addr_size \ -endif \ -d $(depfile).dtc.tmp $(dtc-tmp) ; \ cat $(depfile).pre.tmp $(depfile).dtc.tmp > $(depfile) $(obj)/%.dtbo: $(src)/%-overlay.dts FORCE $(call if_changed_dep,dtco) +endif dtc-tmp = $(subst $(comma),_,$(dot-target).dts.tmp) -- 2.27.0

2 1

[PATCH kernel-4.19 01/32] ascend: sharepool: don't enable the vmalloc to use hugepage default
by Yang Yingliang 08 Mar '21

08 Mar '21

From: Ding Tianhong <dingtianhong(a)huawei.com> ascend inclusion category: feature bugzilla: NA CVE: NA ------------------------------------------------- The commit 59a57a82fb2a ("mm/vmalloc: Hugepage vmalloc mappings") would enable the vmalloc for hugepage default when the alloc size is bigger than the PMD_SIZE, it looks like the transparent hugepage for mmap, the driver could not control the hugepage accurately and be break the logic, now the share pool already export the vmalloc_hugepage_xxx function to control the vmalloc hugepage allocation, it looks like the static hugepage for vmalloc, so disable the transparent hugepage function. This patch also fix the problem of breaking the kabi of vm_struct, the user could applied it for commercial version. Fixes: 59a57a82fb2a ("mm/vmalloc: Hugepage vmalloc mappings") 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 | 51 ++++++++++++++++++++++++++++---------- include/linux/vmalloc.h | 1 - mm/vmalloc.c | 47 ++++++++++++----------------------- 3 files changed, 54 insertions(+), 45 deletions(-) diff --git a/include/linux/share_pool.h b/include/linux/share_pool.h index c3120b7b24948..4a18c88d5a10e 100644 --- a/include/linux/share_pool.h +++ b/include/linux/share_pool.h @@ -211,15 +211,6 @@ static inline void sp_area_work_around(struct vm_unmapped_area_info *info) extern struct page *sp_alloc_pages(struct vm_struct *area, gfp_t mask, unsigned int page_order, int node); - -static inline void sp_free_pages(struct page *page, struct vm_struct *area) -{ - if (PageHuge(page)) - put_page(page); - else - __free_pages(page, area->page_order); -} - static inline bool sp_check_vm_share_pool(unsigned long vm_flags) { if (enable_ascend_share_pool && (vm_flags & VM_SHARE_POOL)) @@ -264,6 +255,30 @@ extern void *buff_vzalloc_hugepage_user(unsigned long size); void sp_exit_mm(struct mm_struct *mm); +static inline bool is_vmalloc_huge(unsigned long vm_flags) +{ + if (enable_ascend_share_pool && (vm_flags & VM_HUGE_PAGES)) + return true; + + return false; +} + +static inline bool is_vmalloc_sharepool(unsigned long vm_flags) +{ + if (enable_ascend_share_pool && (vm_flags & VM_SHAREPOOL)) + return true; + + return false; +} + +static inline void sp_free_pages(struct page *page, struct vm_struct *area) +{ + if (PageHuge(page)) + put_page(page); + else + __free_pages(page, is_vmalloc_huge(area->flags) ? PMD_SHIFT - PAGE_SHIFT : 0); +} + #else static inline int sp_group_add_task(int pid, int spg_id) @@ -400,10 +415,6 @@ static inline struct page *sp_alloc_pages(void *area, gfp_t mask, return NULL; } -static inline void sp_free_pages(struct page *page, struct vm_struct *area) -{ -} - static inline bool sp_check_vm_share_pool(unsigned long vm_flags) { return false; @@ -448,6 +459,20 @@ static inline void *buff_vzalloc_hugepage_user(unsigned long size) return NULL; } +static inline bool is_vmalloc_huge(struct vm_struct *vm) +{ + return NULL; +} + +static inline bool is_vmalloc_sharepool(struct vm_struct *vm) +{ + return NULL; +} + +static inline void sp_free_pages(struct page *page, struct vm_struct *area) +{ +} + #endif #endif /* LINUX_SHARE_POOL_H */ diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h index bb814f6418fd9..298eff5579b21 100644 --- a/include/linux/vmalloc.h +++ b/include/linux/vmalloc.h @@ -43,7 +43,6 @@ struct vm_struct { unsigned long size; unsigned long flags; struct page **pages; - unsigned int page_order; unsigned int nr_pages; phys_addr_t phys_addr; const void *caller; diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 37b4762871142..8c70131e0b078 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -2354,6 +2354,7 @@ struct vm_struct *remove_vm_area(const void *addr) static void __vunmap(const void *addr, int deallocate_pages) { struct vm_struct *area; + unsigned int page_order = 0; if (!addr) return; @@ -2369,13 +2370,14 @@ static void __vunmap(const void *addr, int deallocate_pages) return; } -#ifdef CONFIG_ASCEND_SHARE_POOL /* unmap a sharepool vm area will cause meamleak! */ - if (area->flags & VM_SHAREPOOL) { + if (is_vmalloc_sharepool(area->flags)) { WARN(1, KERN_ERR "Memory leak due to vfree() sharepool vm area (%p) !\n", addr); return; } -#endif + + if (is_vmalloc_huge(area->flags)) + page_order = PMD_SHIFT - PAGE_SHIFT; debug_check_no_locks_freed(area->addr, get_vm_area_size(area)); debug_check_no_obj_freed(area->addr, get_vm_area_size(area)); @@ -2384,14 +2386,14 @@ static void __vunmap(const void *addr, int deallocate_pages) if (deallocate_pages) { int i; - for (i = 0; i < area->nr_pages; i += 1U << area->page_order) { + for (i = 0; i < area->nr_pages; i += 1U << page_order) { struct page *page = area->pages[i]; BUG_ON(!page); if (sp_is_enabled()) sp_free_pages(page, area); else - __free_pages(page, area->page_order); + __free_pages(page, page_order); } kvfree(area->pages); @@ -2589,7 +2591,6 @@ static void *__vmalloc_area_node(struct vm_struct *area, gfp_t gfp_mask, area->pages = pages; area->nr_pages = nr_pages; - area->page_order = page_order; for (i = 0; i < area->nr_pages; i += 1U << page_order) { struct page *page; @@ -2657,27 +2658,17 @@ void *__vmalloc_node_range(unsigned long size, unsigned long align, if (!size || (size >> PAGE_SHIFT) > totalram_pages) goto fail; - if (vmap_allow_huge && (pgprot_val(prot) == pgprot_val(PAGE_KERNEL))) { - unsigned long size_per_node; - + if (vmap_allow_huge && (pgprot_val(prot) == pgprot_val(PAGE_KERNEL)) && is_vmalloc_huge(vm_flags)) { /* - * Try huge pages. Only try for PAGE_KERNEL allocations, - * others like modules don't yet expect huge pages in - * their allocations due to apply_to_page_range not - * supporting them. + * Alloc huge pages. Only valid for PAGE_KERNEL allocations and + * VM_HUGE_PAGES flags. */ - size_per_node = size; - if (node == NUMA_NO_NODE && !sp_is_enabled()) - size_per_node /= num_online_nodes(); - if (size_per_node >= PMD_SIZE) { - shift = PMD_SHIFT; - align = max(real_align, 1UL << shift); - size = ALIGN(real_size, 1UL << shift); - } + shift = PMD_SHIFT; + align = max(real_align, 1UL << shift); + size = ALIGN(real_size, 1UL << shift); } -again: size = PAGE_ALIGN(size); area = __get_vm_area_node(size, align, VM_ALLOC | VM_UNINITIALIZED | vm_flags, start, end, node, gfp_mask, caller); @@ -2706,12 +2697,6 @@ void *__vmalloc_node_range(unsigned long size, unsigned long align, return addr; fail: - if (shift > PAGE_SHIFT) { - shift = PAGE_SHIFT; - align = real_align; - size = real_size; - goto again; - } if (!area) { /* Warn for area allocation, page allocations already warn */ @@ -3776,7 +3761,7 @@ static int s_show(struct seq_file *m, void *p) seq_printf(m, " %pS", v->caller); if (v->nr_pages) - seq_printf(m, " pages=%d order=%d", v->nr_pages, v->page_order); + seq_printf(m, " pages=%d", v->nr_pages); if (v->phys_addr) seq_printf(m, " phys=%pa", &v->phys_addr); @@ -3796,8 +3781,8 @@ static int s_show(struct seq_file *m, void *p) if (is_vmalloc_addr(v->pages)) seq_puts(m, " vpages"); - if (sp_is_enabled()) - seq_printf(m, " order=%d", v->page_order); + if (is_vmalloc_huge(v->flags)) + seq_printf(m, " order=%d", PMD_SHIFT - PAGE_SHIFT); show_numa_info(m, v); seq_putc(m, '\n'); -- 2.25.1

1 31

[PATCH kernel-4.19] Certs: Fix a out-of-tree build Error
by Zhichang Yuan 08 Mar '21

08 Mar '21

openEuler inclusion category: bugfix bugzilla: https://bugzilla.openeuler.org/show_bug.cgi?id=6 CVE: NA When performing the out-of-tree kernel 4.19 building, some error happened saying: cannot touch 'certs/pubring.gpg'. The detail can be checked in Gitee issue #I39160 or Bug 6 of openeuler bugzilla. This patch will fix that build error. Signed-off-by: Zhichang Yuan <erik.yuan(a)arm.com> Signed-off-by: Steve Capper <Steve.Capper(a)arm.com> --- certs/Makefile | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/certs/Makefile b/certs/Makefile index 5053e3c86c97..5e56f9405758 100644 --- a/certs/Makefile +++ b/certs/Makefile @@ -5,11 +5,12 @@ obj-$(CONFIG_SYSTEM_TRUSTED_KEYRING) += system_keyring.o system_certificates.o ifdef CONFIG_PGP_PRELOAD_PUBLIC_KEYS -ifneq ($(shell ls certs/pubring.gpg 2> /dev/null), certs/pubring.gpg) -$(shell touch certs/pubring.gpg) -endif -$(obj)/system_certificates.o: certs/pubring.gpg +$(obj)/system_certificates.o: $(obj)/pubring.gpg + +$(obj)/pubring.gpg: + $(Q)touch $@ endif + obj-$(CONFIG_SYSTEM_BLACKLIST_KEYRING) += blacklist.o ifneq ($(CONFIG_SYSTEM_BLACKLIST_HASH_LIST),"") obj-$(CONFIG_SYSTEM_BLACKLIST_KEYRING) += blacklist_hashes.o -- 2.23.0 IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.

4 6

[PATCH kernel-4.19 00/50] backport stable-4.19.177
by Yang Yingliang 08 Mar '21

08 Mar '21

Alain Volmat (1): i2c: stm32f7: fix configuration of the digital filter Alexandre Belloni (1): ARM: dts: lpc32xx: Revert set default clock rate of HCLK PLL Alexandre Ghiti (1): riscv: virt_addr_valid must check the address belongs to linear mapping Amir Goldstein (1): ovl: skip getxattr of security labels Arun Easi (1): scsi: qla2xxx: Fix crash during driver load on big endian machines Borislav Petkov (1): x86/build: Disable CET instrumentation in the kernel for 32-bit too Bui Quang Minh (1): bpf: Check for integer overflow when using roundup_pow_of_two() Edwin Peer (1): net: watchdog: hold device global xmit lock during tx disable Fangrui Song (1): firmware_loader: align .builtin_fw to 8 Felipe Balbi (1): usb: dwc3: ulpi: fix checkpatch warning Florian Westphal (1): netfilter: conntrack: skip identical origin tuple in same zone only Greg Kroah-Hartman (1): Linux 4.19.177 Hans de Goede (1): platform/x86: hp-wmi: Disable tablet-mode reporting by default Jan Beulich (8): Xen/x86: don't bail early from clear_foreign_p2m_mapping() Xen/x86: also check kernel mapping in set_foreign_p2m_mapping() Xen/gntdev: correct dev_bus_addr handling in gntdev_map_grant_pages() Xen/gntdev: correct error checking in gntdev_map_grant_pages() xen-blkback: don't "handle" error by BUG() xen-netback: don't "handle" error by BUG() xen-scsiback: don't "handle" error by BUG() xen-blkback: fix error handling in xen_blkbk_map() Jozsef Kadlecsik (1): netfilter: xt_recent: Fix attempt to update deleted entry Juergen Gross (1): xen/netback: avoid race in xenvif_rx_ring_slots_available() Julien Grall (1): arm/xen: Don't probe xenbus as part of an early initcall Lai Jiangshan (1): kvm: check tlbs_dirty directly Lin Feng (1): bfq-iosched: Revert "bfq: Fix computation of shallow depth" Loic Poulain (1): net: qrtr: Fix port ID for control messages Lorenzo Bianconi (1): mt76: dma: fix a possible memory leak in mt76_add_fragment() Marc Zyngier (1): arm64: dts: rockchip: Fix PCIe DT properties on rk3399 Miklos Szeredi (3): ovl: perform vfs_getxattr() with mounter creds cap: fix conversions on getxattr ovl: expand warning in ovl_d_real() Mohammad Athari Bin Ismail (1): net: stmmac: set TxQ mode back to DCB after disabling CBS NeilBrown (1): net: fix iteration for sctp transport seq_files Norbert Slusarek (1): net/vmw_vsock: improve locking in vsock_connect_timeout() Paolo Bonzini (1): KVM: SEV: fix double locking due to incorrect backport Randy Dunlap (1): h8300: fix PREEMPTION build, TI_PRE_COUNT undefined Russell King (2): ARM: ensure the signal page contains defined contents ARM: kexec: fix oops after TLB are invalidated Sabyrzhan Tasbolatov (2): net/rds: restrict iovecs length for RDS_CMSG_RDMA_ARGS net/qrtr: restrict user-controlled length in qrtr_tun_write_iter() Serge Semin (1): usb: dwc3: ulpi: Replace CPU-based busyloop with Protocol-based one Stefano Garzarella (2): vsock/virtio: update credit only if socket is not closed vsock: fix locking in vsock_shutdown() Stefano Stabellini (1): xen/arm: don't ignore return errors from set_phys_to_machine Steven Rostedt (VMware) (2): tracing: Do not count ftrace events in top level enable output tracing: Check length before giving out the filter buffer Sven Auhagen (1): netfilter: flowtable: fix tcp and udp header checksum update Victor Lu (2): drm/amd/display: Fix dc_sink kref count in emulated_link_detect drm/amd/display: Free atomic state after drm_atomic_commit Makefile | 2 +- arch/arm/boot/dts/lpc32xx.dtsi | 3 - arch/arm/include/asm/kexec-internal.h | 12 ++++ arch/arm/kernel/asm-offsets.c | 5 ++ arch/arm/kernel/machine_kexec.c | 20 +++--- arch/arm/kernel/relocate_kernel.S | 38 +++-------- arch/arm/kernel/signal.c | 14 ++-- arch/arm/xen/enlighten.c | 2 - arch/arm/xen/p2m.c | 6 +- arch/arm64/boot/dts/rockchip/rk3399.dtsi | 2 +- arch/h8300/kernel/asm-offsets.c | 3 + arch/riscv/include/asm/page.h | 5 +- arch/x86/Makefile | 6 +- arch/x86/kvm/svm.c | 1 - arch/x86/xen/p2m.c | 15 ++--- block/bfq-iosched.c | 8 +-- drivers/block/xen-blkback/blkback.c | 30 +++++---- .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 18 +++-- drivers/i2c/busses/i2c-stm32f7.c | 11 ++- .../net/ethernet/stmicro/stmmac/stmmac_tc.c | 7 +- drivers/net/wireless/mediatek/mt76/dma.c | 8 ++- drivers/net/xen-netback/netback.c | 4 +- drivers/net/xen-netback/rx.c | 9 ++- drivers/platform/x86/hp-wmi.c | 14 ++-- drivers/scsi/qla2xxx/qla_tmpl.c | 9 +-- drivers/scsi/qla2xxx/qla_tmpl.h | 2 +- drivers/usb/dwc3/ulpi.c | 20 ++++-- drivers/xen/gntdev.c | 37 +++++----- drivers/xen/xen-scsiback.c | 4 +- drivers/xen/xenbus/xenbus.h | 1 - drivers/xen/xenbus/xenbus_probe.c | 2 +- fs/overlayfs/copy_up.c | 15 +++-- fs/overlayfs/inode.c | 2 + fs/overlayfs/super.c | 13 ++-- include/asm-generic/vmlinux.lds.h | 2 +- include/linux/netdevice.h | 2 + include/xen/grant_table.h | 1 + include/xen/xenbus.h | 2 - kernel/bpf/stackmap.c | 2 + kernel/trace/trace.c | 2 +- kernel/trace/trace_events.c | 3 +- net/netfilter/nf_conntrack_core.c | 3 +- net/netfilter/nf_flow_table_core.c | 4 +- net/netfilter/xt_recent.c | 12 +++- net/qrtr/qrtr.c | 2 +- net/qrtr/tun.c | 6 ++ net/rds/rdma.c | 3 + net/sctp/proc.c | 16 +++-- net/vmw_vsock/af_vsock.c | 13 ++-- net/vmw_vsock/hyperv_transport.c | 4 -- net/vmw_vsock/virtio_transport_common.c | 4 +- security/commoncap.c | 67 ++++++++++++------- virt/kvm/kvm_main.c | 3 +- 53 files changed, 295 insertions(+), 204 deletions(-) create mode 100644 arch/arm/include/asm/kexec-internal.h -- 2.25.1

1 50

[PATCH kernel-4.19] arm64/mpam: Fix compile warning
by Yang Yingliang 08 Mar '21

08 Mar '21

From: Wang ShaoBo <bobo.shaobowang(a)huawei.com> hulk inclusion category: bugfix bugzilla: 34278 CVE: NA ------------------------------------------------- Set dentry before goto error handling branch. fs/resctrlfs.c: In function ‘resctrl_mount’: fs/resctrlfs.c:419:9: warning: ‘dentry’ may be used uninitialized in this function [-Wmaybe-uninitialized] return dentry; ^~~~~~ Fixes: 1b70f98921b10 ("arm64/mpam: resctrl: Use resctrl_group_init_alloc() for default group") Signed-off-by: Wang ShaoBo <bobo.shaobowang(a)huawei.com> Reviewed-by: Jian Cheng <cj.chengjian(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- fs/resctrlfs.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/resctrlfs.c b/fs/resctrlfs.c index 190ad509714d5..df47b105609b9 100644 --- a/fs/resctrlfs.c +++ b/fs/resctrlfs.c @@ -351,8 +351,10 @@ static struct dentry *resctrl_mount(struct file_system_type *fs_type, } ret = resctrl_group_init_alloc(&resctrl_group_default); - if (ret < 0) + if (ret < 0) { + dentry = ERR_PTR(ret); goto out_schema; + } ret = resctrl_group_create_info_dir(resctrl_group_default.kn, &kn_info); if (ret) { -- 2.25.1

1 0

[PATCH kernel-4.19 00/25] backport stable-4.19.176
by Yang Yingliang 08 Mar '21

08 Mar '21

Cong Wang (1): af_key: relax availability checks for skb size calculation Dave Wysochanski (2): SUNRPC: Move simple_get_bytes and simple_get_netobj into private header SUNRPC: Handle 0 length opaque XDR object data properly David Collins (1): regulator: core: avoid regulator_resolve_supply() race condition Douglas Anderson (1): regulator: core: Clean enabling always-on regulators + their supplies Emmanuel Grumbach (1): iwlwifi: pcie: add a NULL check in iwl_pcie_txq_unmap Greg Kroah-Hartman (1): Linux 4.19.176 Johannes Berg (3): iwlwifi: mvm: take mutex for calling iwl_mvm_get_sync_time() iwlwifi: pcie: fix context info memory leak iwlwifi: mvm: guard against device removal in reprobe Mark Brown (1): regulator: Fix lockdep warning resolving supplies Masami Hiramatsu (1): tracing/kprobe: Fix to support kretprobe events on unloaded modules Ming Lei (2): block: don't hold q->sysfs_lock in elevator_init_mq blk-mq: don't hold q->sysfs_lock in blk_mq_map_swqueue Olliver Schinagl (1): regulator: core: enable power when setting up constraints Pan Bian (1): chtls: Fix potential resource leak Peter Gonda (1): Fix unsynchronized access to sev members through svm_register_enc_region Phillip Lougher (3): squashfs: add more sanity checks in id lookup squashfs: add more sanity checks in inode lookup squashfs: add more sanity checks in xattr id lookup Sibi Sankar (2): remoteproc: qcom_q6v5_mss: Validate modem blob firmware size before load remoteproc: qcom_q6v5_mss: Validate MBA firmware size before load Steven Rostedt (VMware) (1): fgraph: Initialize tracing_graph_pause at task creation Tobin C. Harding (1): lib/string: Add strscpy_pad() function Trond Myklebust (1): pNFS/NFSv4: Try to return invalid layout in pnfs_layout_process() Makefile | 2 +- arch/x86/kvm/svm.c | 18 ++-- block/blk-mq.c | 7 -- block/elevator.c | 14 ++-- drivers/crypto/chelsio/chtls/chtls_cm.c | 7 +- .../wireless/intel/iwlwifi/mvm/debugfs-vif.c | 3 + drivers/net/wireless/intel/iwlwifi/mvm/ops.c | 3 +- .../intel/iwlwifi/pcie/ctxt-info-gen3.c | 11 ++- drivers/net/wireless/intel/iwlwifi/pcie/tx.c | 5 ++ drivers/regulator/core.c | 84 +++++++++++++------ drivers/remoteproc/qcom_q6v5_pil.c | 11 ++- fs/nfs/pnfs.c | 8 +- fs/squashfs/export.c | 41 +++++++-- fs/squashfs/id.c | 40 +++++++-- fs/squashfs/squashfs_fs_sb.h | 1 + fs/squashfs/super.c | 6 +- fs/squashfs/xattr.h | 10 ++- fs/squashfs/xattr_id.c | 66 +++++++++++++-- include/linux/kprobes.h | 2 +- include/linux/string.h | 4 + include/linux/sunrpc/xdr.h | 3 +- init/init_task.c | 3 +- kernel/kprobes.c | 34 ++++++-- kernel/trace/ftrace.c | 2 - kernel/trace/trace_kprobe.c | 4 +- lib/string.c | 47 +++++++++-- net/key/af_key.c | 6 +- net/sunrpc/auth_gss/auth_gss.c | 30 +------ net/sunrpc/auth_gss/auth_gss_internal.h | 45 ++++++++++ net/sunrpc/auth_gss/gss_krb5_mech.c | 31 +------ 30 files changed, 375 insertions(+), 173 deletions(-) create mode 100644 net/sunrpc/auth_gss/auth_gss_internal.h -- 2.25.1

1 25

[PATCH kernel-4.19 1/5] scsi/hifc: only support 64 bit OS
by Yang Yingliang 08 Mar '21

08 Mar '21

From: Chenguangli <chenguangli2(a)huawei.com> driver inclusion category: feature bugzilla: NA ----------------------------------------------------------------------- This module is used to register the hifc driver FC capability to the SCSI layer. Signed-off-by: Chenguangli <chenguangli2(a)huawei.com> Reviewed-by: Zengweiliang <zengweiliang.zengweiliang(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/scsi/huawei/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/huawei/Kconfig b/drivers/scsi/huawei/Kconfig index a9fbdef9b4b38..1fe47e626f568 100644 --- a/drivers/scsi/huawei/Kconfig +++ b/drivers/scsi/huawei/Kconfig @@ -3,9 +3,10 @@ # config SCSI_HUAWEI_FC - tristate "Huawei devices" + tristate "Huawei Fibre Channel Adapter" depends on PCI && SCSI depends on SCSI_FC_ATTRS + depends on ARM64 || X86_64 default m ---help--- If you have a Fibre Channel PCI card belonging to this class, say Y. -- 2.25.1

1 4

[PATCH kernel-4.19 00/39] backport stable-4.19.175
by Yang Yingliang 08 Mar '21

08 Mar '21

Alexey Dobriyan (1): Input: i8042 - unbreak Pegatron C15B Arnd Bergmann (1): elfcore: fix building with clang Aurelien Aptel (1): cifs: report error instead of invalid when revalidating a dentry fails Benjamin Valentin (1): Input: xpad - sync supported devices with fork on GitHub Chenxin Jin (1): USB: serial: cp210x: add new VID/PID for supporting Teraoka AD2000 Christoph Schemmel (1): USB: serial: option: Adding support for Cinterion MV31 DENG Qingfang (1): net: dsa: mv88e6xxx: override existent unicast portvec in port_fdb_add Dan Carpenter (1): USB: gadget: legacy: fix an error code in eth_bind() Dave Hansen (1): x86/apic: Add extra serialization for non-serializing MSRs David Howells (1): rxrpc: Fix deadlock around release of dst cached on udp tunnel Felix Fietkau (1): mac80211: fix station rate table updates on assoc Fengnan Chang (1): mmc: core: Limit retries when analyse of SDIO tuples fails Gary Bisson (1): usb: dwc3: fix clock issue during resume in OTG mode Greg Kroah-Hartman (1): Linux 4.19.175 Gustavo A. R. Silva (1): smb3: Fix out-of-bounds bug in SMB2_negotiate() Heiko Stuebner (1): usb: dwc2: Fix endpoint direction check in ep_from_windex Hugh Dickins (1): mm: thp: fix MADV_REMOVE deadlock on shmem THP Jeremy Figgins (1): USB: usblp: don't call usb_set_interface if there's a single alt Josh Poimboeuf (1): x86/build: Disable CET instrumentation in the kernel Liangyan (1): ovl: fix dentry leak in ovl_get_redirect Marc Zyngier (1): genirq/msi: Activate Multi-MSI early when MSI_FLAG_ACTIVATE_EARLY is set Mathias Nyman (1): xhci: fix bounce buffer usage for non-sg list case Muchun Song (4): mm: hugetlbfs: fix cannot migrate the fallocated HugeTLB page mm: hugetlb: fix a race between freeing and dissolving the page mm: hugetlb: fix a race between isolating and freeing page mm: hugetlb: remove VM_BUG_ON_PAGE from page_huge_active Nadav Amit (1): iommu/vt-d: Do not use flush-queue when caching-mode is on Pho Tran (1): USB: serial: cp210x: add pid/vid for WSDA-200-USB Roman Gushchin (1): memblock: do not start bottom-up allocations with kernel_end Russell King (1): ARM: footbridge: fix dc21285 PCI configuration accessors Sean Christopherson (1): KVM: SVM: Treat SVM as unsupported when running as an SEV guest Stefan Chulski (1): net: mvpp2: TCAM entry enable should be written after SRAM data Thorsten Leemhuis (1): nvme-pci: avoid the deepest sleep state on Kingston A2000 SSDs Vadim Fedorenko (1): net: ip_tunnel: fix mtu calculation Wang ShaoBo (1): kretprobe: Avoid re-registration of the same kretprobe earlier Xiao Ni (1): md: Set prev_flush_start and flush_bio in an atomic way Xie He (1): net: lapb: Copy the skb before sending a packet Yoshihiro Shimoda (1): usb: renesas_usbhs: Clear pipe running flag in usbhs_pkt_pop() Zyta Szpak (1): arm64: dts: ls1046a: fix dcfg address range Makefile | 8 +-- arch/arm/mach-footbridge/dc21285.c | 12 ++--- .../arm64/boot/dts/freescale/fsl-ls1046a.dtsi | 2 +- arch/x86/Makefile | 3 ++ arch/x86/include/asm/apic.h | 10 ---- arch/x86/include/asm/barrier.h | 18 +++++++ arch/x86/kernel/apic/apic.c | 4 ++ arch/x86/kernel/apic/x2apic_cluster.c | 6 ++- arch/x86/kernel/apic/x2apic_phys.c | 6 ++- arch/x86/kvm/svm.c | 5 ++ drivers/input/joystick/xpad.c | 17 ++++++- drivers/input/serio/i8042-x86ia64io.h | 2 + drivers/iommu/intel-iommu.c | 6 +++ drivers/md/md.c | 2 + drivers/mmc/core/sdio_cis.c | 6 +++ drivers/net/dsa/mv88e6xxx/chip.c | 6 ++- .../net/ethernet/marvell/mvpp2/mvpp2_prs.c | 10 ++-- drivers/nvme/host/pci.c | 2 + drivers/usb/class/usblp.c | 19 ++++--- drivers/usb/dwc2/gadget.c | 8 +-- drivers/usb/dwc3/core.c | 2 +- drivers/usb/gadget/legacy/ether.c | 4 +- drivers/usb/host/xhci-ring.c | 31 +++++++----- drivers/usb/renesas_usbhs/fifo.c | 1 + drivers/usb/serial/cp210x.c | 2 + drivers/usb/serial/option.c | 6 +++ fs/afs/main.c | 6 +-- fs/cifs/dir.c | 22 ++++++++- fs/cifs/smb2pdu.h | 2 +- fs/hugetlbfs/inode.c | 3 +- fs/overlayfs/dir.c | 2 +- include/linux/elfcore.h | 22 +++++++++ include/linux/hugetlb.h | 3 ++ include/linux/msi.h | 6 +++ kernel/Makefile | 1 - kernel/elfcore.c | 26 ---------- kernel/irq/msi.c | 44 ++++++++--------- kernel/kprobes.c | 4 ++ mm/huge_memory.c | 37 ++++++++------ mm/hugetlb.c | 48 ++++++++++++++++-- mm/memblock.c | 49 +++---------------- net/ipv4/ip_tunnel.c | 16 +++--- net/lapb/lapb_out.c | 3 +- net/mac80211/driver-ops.c | 5 +- net/mac80211/rate.c | 3 +- net/rxrpc/af_rxrpc.c | 6 +-- 46 files changed, 307 insertions(+), 199 deletions(-) delete mode 100644 kernel/elfcore.c -- 2.25.1

1 39

Re: [PATCH] kbuild fix: keep the original function in Makefile.lib for non-RPi
by Zheng Zengkai 08 Mar '21

08 Mar '21

Hi Yafen, > ---------------------- [WARNING] checkpatch ---------------------- > > total: 0 errors, 1 warnings, 27 lines checked > > WARNING:COMMIT_LOG_LONG_LINE: Possible unwrapped commit description (prefer a maximum 75 chars per line) > #84: > /bin/sh: -c: line 0: `set -e; echo ' DTCO arch/arm64/boot/dts/overlays/act-led.dtbo'; mkdir -p arch/arm64/boot/dts/overlays/ ; gcc -E -Wp,-MMD,arch/arm64/boot/dts/overlays/.act-led.dtbo.d.pre.tmp -nostdinc -I./scripts/dtc/include-prefixes -undef -D__DTS__ -x assembler-with-cpp -o arch/arm64/boot/dts/overlays/.act-led.dtbo.dts.tmp arch/arm64/boot/dts/overlays/act-led-overlay.dts ; ./scripts/dtc/dtc -@ -H epapr -O dtb -o arch/arm64/boot/dts/overlays/act-led.dtbo -b 0 -i arch/arm64/boot/dts/overlays/ -Wno-interrupt_provider -Wno-unit_address_vs_reg -Wno-unit_address_format -Wno-gpios_property -Wno-avoid_unnecessary_addr_size -Wno-alias_paths -Wno-graph_child_address -Wno-simple_bus_reg -Wno-unique_unit_address -Wno-pci_device_reg -Wno-interrupts_property ifeq (y,y) -Wno-label_is_string -Wno-reg_format -Wno-pci_device_bus_num -Wno-i2c_bus_reg -Wno-spi_bus_reg -Wno-avoid_default_addr_size endif -d arch/arm64/boot/dts/overlays/.act-led.dtbo.d.dtc.tmp arch/arm64/boot/dts/overlays/.act-led.dtbo.dts.tmp ; cat ...; rm -f arch/arm64/boot/dts/overlays/.act-led.dtbo.d' > > total: 0 errors, 1 warnings, 27 lines checked > > NOTE: For some of the reported defects, checkpatch may be able to > mechanically convert to the typical style using --fix or --fix-inplace. > > kbuild-fix-keep-the-original-function-in-Makefile.lib-for-non-RPi.patch has style problems, please review. This patch has style problems, please review. Thanks!

2 1

[PATCH] kbuild fix: keep the original function in Makefile.lib for non-RPi
by fangyafenqidai＠163.com 08 Mar '21

08 Mar '21

From: Fang Yafen <yafen(a)iscas.ac.cn> Fix the following compile error when using bcm2711_defconfig (for RPi). /bin/sh: -c: line 0: syntax error near unexpected token `(' /bin/sh: -c: line 0: `set -e; echo ' DTCO arch/arm64/boot/dts/overlays/act-led.dtbo'; mkdir -p arch/arm64/boot/dts/overlays/ ; gcc -E -Wp,-MMD,arch/arm64/boot/dts/overlays/.act-led.dtbo.d.pre.tmp -nostdinc -I./scripts/dtc/include-prefixes -undef -D__DTS__ -x assembler-with-cpp -o arch/arm64/boot/dts/overlays/.act-led.dtbo.dts.tmp arch/arm64/boot/dts/overlays/act-led-overlay.dts ; ./scripts/dtc/dtc -@ -H epapr -O dtb -o arch/arm64/boot/dts/overlays/act-led.dtbo -b 0 -i arch/arm64/boot/dts/overlays/ -Wno-interrupt_provider -Wno-unit_address_vs_reg -Wno-unit_address_format -Wno-gpios_property -Wno-avoid_unnecessary_addr_size -Wno-alias_paths -Wno-graph_child_address -Wno-simple_bus_reg -Wno-unique_unit_address -Wno-pci_device_reg -Wno-interrupts_property ifeq (y,y) -Wno-label_is_string -Wno-reg_format -Wno-pci_device_bus_num -Wno-i2c_bus_reg -Wno-spi_bus_reg -Wno-avoid_default_addr_size endif -d arch/arm64/boot/dts/overlays/.act-led.dtbo.d.dtc.tmp arch/arm64/boot/dts/overlays/.act-led.dtbo.dts.tmp ; cat ...; rm -f arch/arm64/boot/dts/overlays/.act-led.dtbo.d' make[2]: *** [scripts/Makefile.lib;363: arch/arm64/boot/dts/overlays/act-led.dtbo] Error 1 make[2]: *** Waiting for unfinished jobs.... Related patches: ffa2d13ccc3 BCM2708: Add core Device Tree support 4894352ec98 kbuild: Silence unavoidable dtc overlay warnings a4a4d07f0cf kbuild: keep the original function for non-RPi Signed-off-by: Fang Yafen <yafen(a)iscas.ac.cn> --- scripts/Makefile.lib | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib index c23e3ae7ef40..a0e0e2543165 100644 --- a/scripts/Makefile.lib +++ b/scripts/Makefile.lib @@ -342,25 +342,25 @@ endef $(obj)/%.dt.yaml: $(src)/%.dts $(DTC) $(DT_TMP_SCHEMA) FORCE $(call if_changed_rule,dtc,yaml) +ifeq ($(CONFIG_OPENEULER_RASPBERRYPI),y) quiet_cmd_dtco = DTCO $@ cmd_dtco = mkdir -p $(dir ${dtc-tmp}) ; \ $(CPP) $(dtc_cpp_flags) -x assembler-with-cpp -o $(dtc-tmp) $< ; \ $(DTC) -@ -H epapr -O dtb -o $@ -b 0 \ -i $(dir $<) $(DTC_FLAGS) \ -Wno-interrupts_property \ -ifeq ($(CONFIG_OPENEULER_RASPBERRYPI),y) \ -Wno-label_is_string \ -Wno-reg_format \ -Wno-pci_device_bus_num \ -Wno-i2c_bus_reg \ -Wno-spi_bus_reg \ -Wno-avoid_default_addr_size \ -endif \ -d $(depfile).dtc.tmp $(dtc-tmp) ; \ cat $(depfile).pre.tmp $(depfile).dtc.tmp > $(depfile) $(obj)/%.dtbo: $(src)/%-overlay.dts FORCE $(call if_changed_dep,dtco) +endif dtc-tmp = $(subst $(comma),_,$(dot-target).dts.tmp) -- 2.23.0

1 0

[PATCH openEuler-21.03 v3 1/2] arm64: Add memmap parameter and register pmem
by zhuling8＠huawei.com 08 Mar '21

08 Mar '21

From: ZhuLing <zhuling8(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: NA Register pmem in arm64: Use memmap(memmap=nn[KMG]!ss[KMG]) reserve memory and e820(driver/nvdimm/e820.c) function to register persistent memory in arm64. when the kernel restart or update, the data in PMEM will not be lost and can be loaded faster. this is a general features. driver/nvdimm/e820.c: The function of this file is scan "iomem_resource" and take advantage of nvdimm resource discovery mechanism by registering a resource named "Persistent Memory (legacy)", this function doesn't depend on architecture. We will push the feature to linux kernel community and discuss to modify the file name. because people have a mistaken notion that the e820.c is depend on x86. If you want use this features, you need do as follows: 1.Reserve memory: add memmap to reserve memory in grub.cfg memmap=nn[KMG]!ss[KMG] exp:memmap=100K!0x1a0000000. 2.Insmod nd_e820.ko: modprobe nd_e820. 3.Check pmem device in /dev exp: /dev/pmem0 Signed-off-by: ZhuLing <zhuling8(a)huawei.com> Signed-off-by: Sang Yan <sangyan(a)huawei.com> Acked-by: Hanjun Guo <guohanjun(a)huawei.com> --- arch/arm64/Kconfig | 21 ++++++++++ arch/arm64/kernel/Makefile | 1 + arch/arm64/kernel/pmem.c | 35 +++++++++++++++++ arch/arm64/kernel/setup.c | 10 +++++ arch/arm64/mm/init.c | 97 ++++++++++++++++++++++++++++++++++++++++++++++ drivers/nvdimm/Kconfig | 5 +++ drivers/nvdimm/Makefile | 2 +- 7 files changed, 170 insertions(+), 1 deletion(-) create mode 100644 arch/arm64/kernel/pmem.c diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index c451137..326f26d 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -360,6 +360,27 @@ config ARM64_CPU_PARK config ARCH_HAS_CPU_RELAX def_bool y +config ARM64_PMEM_RESERVE + bool "Reserve memory for persistent storage" + default n + help + Use memmap=nn[KMG]!ss[KMG](memmap=100K!0x1a0000000) reserve + memory for persistent storage. + + Say y here to enable this feature. + +config ARM64_PMEM_LEGACY_DEVICE + bool "Create persistent storage" + depends on BLK_DEV + depends on LIBNVDIMM + select ARM64_PMEM_RESERVE + help + Use reserved memory for persistent storage when the kernel + restart or update. the data in PMEM will not be lost and + can be loaded faster. + + Say y if unsure. + source "arch/arm64/Kconfig.platforms" menu "Kernel Features" diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile index 967cb3c..be996f3 100644 --- a/arch/arm64/kernel/Makefile +++ b/arch/arm64/kernel/Makefile @@ -67,6 +67,7 @@ obj-$(CONFIG_ARM64_PTR_AUTH) += pointer_auth.o obj-$(CONFIG_SHADOW_CALL_STACK) += scs.o obj-$(CONFIG_ARM64_MTE) += mte.o obj-$(CONFIG_MPAM) += mpam/ +obj-$(CONFIG_ARM64_PMEM_LEGACY_DEVICE) += pmem.o obj-y += vdso/ probes/ obj-$(CONFIG_COMPAT_VDSO) += vdso32/ diff --git a/arch/arm64/kernel/pmem.c b/arch/arm64/kernel/pmem.c new file mode 100644 index 0000000..16eaf70 --- /dev/null +++ b/arch/arm64/kernel/pmem.c @@ -0,0 +1,35 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright(c) 2021 Huawei Technologies Co., Ltd + * + * Derived from x86 and arm64 implement PMEM. + */ +#include <linux/platform_device.h> +#include <linux/init.h> +#include <linux/ioport.h> +#include <linux/module.h> + +static int found(struct resource *res, void *data) +{ + return 1; +} + +static int __init register_e820_pmem(void) +{ + struct platform_device *pdev; + int rc; + + rc = walk_iomem_res_desc(IORES_DESC_PERSISTENT_MEMORY_LEGACY, + IORESOURCE_MEM, 0, -1, NULL, found); + if (rc <= 0) + return 0; + + /* + * See drivers/nvdimm/e820.c for the implementation, this is + * simply here to trigger the module to load on demand. + */ + pdev = platform_device_alloc("e820_pmem", -1); + + return platform_device_add(pdev); +} +device_initcall(register_e820_pmem); diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index 5e282d3..84c71c8 100644 --- a/arch/arm64/kernel/setup.c +++ b/arch/arm64/kernel/setup.c @@ -57,6 +57,10 @@ static int num_standard_resources; static struct resource *standard_resources; +#ifdef CONFIG_ARM64_PMEM_RESERVE +extern struct resource pmem_res; +#endif + phys_addr_t __fdt_pointer __initdata; /* @@ -270,6 +274,12 @@ static void __init request_standard_resources(void) request_resource(res, &pin_memory_resource); #endif } + +#ifdef CONFIG_ARM64_PMEM_RESERVE + if (pmem_res.end && pmem_res.start) + request_resource(&iomem_resource, &pmem_res); +#endif + } static int __init reserve_memblock_reserved_regions(void) diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index b343744..f22faea 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -66,6 +66,18 @@ EXPORT_SYMBOL(memstart_addr); phys_addr_t arm64_dma_phys_limit __ro_after_init; phys_addr_t arm64_dma32_phys_limit __ro_after_init; +static unsigned long long pmem_size, pmem_start; + +#ifdef CONFIG_ARM64_PMEM_RESERVE +struct resource pmem_res = { + .name = "Persistent Memory (legacy)", + .start = 0, + .end = 0, + .flags = IORESOURCE_MEM, + .desc = IORES_DESC_PERSISTENT_MEMORY_LEGACY +}; +#endif + #ifndef CONFIG_KEXEC_CORE static void __init reserve_crashkernel(void) { @@ -378,6 +390,87 @@ static int __init reserve_park_mem(void) } #endif +static int __init is_mem_valid(unsigned long long mem_size, unsigned long long mem_start) +{ + if (!memblock_is_region_memory(mem_start, mem_size)) { + pr_warn("cannot reserve mem: region is not memory!\n"); + return -EINVAL; + } + + if (memblock_is_region_reserved(mem_start, mem_size)) { + pr_warn("cannot reserve mem: region overlaps reserved memory!\n"); + return -EINVAL; + } + + if (!IS_ALIGNED(mem_start, SZ_2M)) { + pr_warn("cannot reserve mem: base address is not 2MB aligned!\n"); + return -EINVAL; + } + + return 0; +} + +static int __init parse_memmap_one(char *p) +{ + char *oldp; + phys_addr_t start_at, mem_size; + + if (!p) + return -EINVAL; + + oldp = p; + mem_size = memparse(p, &p); + if (p == oldp) + return -EINVAL; + + if (!mem_size) + return -EINVAL; + + mem_size = PAGE_ALIGN(mem_size); + + if (*p == '!') { + start_at = memparse(p+1, &p); + + if (is_mem_valid(mem_size, start_at) != 0) + return -EINVAL; + + pr_info("pmem reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + start_at, start_at + mem_size, mem_size >> 20); + pmem_start = start_at; + pmem_size = mem_size; + } else + pr_info("Unrecognized memmap option, please check the parameter.\n"); + + return *p == '\0' ? 0 : -EINVAL; +} + +static int __init parse_memmap_opt(char *str) +{ + while (str) { + char *k = strchr(str, ','); + + if (k) + *k++ = 0; + + parse_memmap_one(str); + str = k; + } + + return 0; +} +early_param("memmap", parse_memmap_opt); + +#ifdef CONFIG_ARM64_PMEM_RESERVE +static void __init reserve_pmem(void) +{ + memblock_remove(pmem_start, pmem_size); + pr_info("pmem reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + pmem_start, pmem_start + pmem_size, pmem_size >> 20); + pmem_res.start = pmem_start; + pmem_res.end = pmem_start + pmem_size - 1; +} +#endif + void __init arm64_memblock_init(void) { const s64 linear_region_size = BIT(vabits_actual - 1); @@ -511,6 +604,10 @@ void __init arm64_memblock_init(void) reserve_elfcorehdr(); +#ifdef CONFIG_ARM64_PMEM_RESERVE + reserve_pmem(); +#endif + high_memory = __va(memblock_end_of_DRAM() - 1) + 1; dma_contiguous_reserve(arm64_dma32_phys_limit); diff --git a/drivers/nvdimm/Kconfig b/drivers/nvdimm/Kconfig index b7d1eb3..ce4de75 100644 --- a/drivers/nvdimm/Kconfig +++ b/drivers/nvdimm/Kconfig @@ -132,3 +132,8 @@ config NVDIMM_TEST_BUILD infrastructure. endif + +config PMEM_LEGACY + tristate "Pmem_legacy" + select X86_PMEM_LEGACY if X86 + select ARM64_PMEM_LEGACY_DEVICE if ARM64 diff --git a/drivers/nvdimm/Makefile b/drivers/nvdimm/Makefile index 29203f3..6f8dc92 100644 --- a/drivers/nvdimm/Makefile +++ b/drivers/nvdimm/Makefile @@ -3,7 +3,7 @@ obj-$(CONFIG_LIBNVDIMM) += libnvdimm.o obj-$(CONFIG_BLK_DEV_PMEM) += nd_pmem.o obj-$(CONFIG_ND_BTT) += nd_btt.o obj-$(CONFIG_ND_BLK) += nd_blk.o -obj-$(CONFIG_X86_PMEM_LEGACY) += nd_e820.o +obj-$(CONFIG_PMEM_LEGACY) += nd_e820.o obj-$(CONFIG_OF_PMEM) += of_pmem.o obj-$(CONFIG_VIRTIO_PMEM) += virtio_pmem.o nd_virtio.o -- 2.9.5

1 1

[PATCH openEuler-21.03 v2 1/2] arm64: Add memmap parameter and register pmem
by Zhuling 08 Mar '21

08 Mar '21

From: ZhuLing <zhuling8(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: NA Register pmem in arm64: Use memmap(memmap=nn[KMG]!ss[KMG]) reserve memory and e820(driver/nvdimm/e820.c) function to register persistent memory in arm64. when the kernel restart or update, the data in PMEM will not be lost and can be loaded faster. this is a general features. driver/nvdimm/e820.c: The function of this file is scan "iomem_resource" and take advantage of nvdimm resource discovery mechanism by registering a resource named "Persistent Memory (legacy)", this function doesn't depend on architecture. We will push the feature to linux kernel community and discuss to modify the file name. because people have a mistaken notion that the e820.c is depend on x86. If you want use this features, you need do as follows: 1.Reserve memory: add memmap to reserve memory in grub.cfg memmap=nn[KMG]!ss[KMG] exp:memmap=100K!0x1a0000000. 2.Insmod nd_e820.ko: modprobe nd_e820. 3.Check pmem device in /dev exp: /dev/pmem0 Signed-off-by: ZhuLing <zhuling8(a)huawei.com> --- arch/arm64/Kconfig | 21 +++++++++ arch/arm64/kernel/Makefile | 1 + arch/arm64/kernel/pmem.c | 35 ++++++++++++++ arch/arm64/kernel/setup.c | 10 ++++ arch/arm64/mm/init.c | 97 ++++++++++++++++++++++++++++++++++++++ drivers/nvdimm/Kconfig | 5 ++ drivers/nvdimm/Makefile | 2 +- 7 files changed, 170 insertions(+), 1 deletion(-) create mode 100644 arch/arm64/kernel/pmem.c diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index c451137ab..326f26d40 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -360,6 +360,27 @@ config ARM64_CPU_PARK config ARCH_HAS_CPU_RELAX def_bool y +config ARM64_PMEM_RESERVE + bool "Reserve memory for persistent storage" + default n + help + Use memmap=nn[KMG]!ss[KMG](memmap=100K!0x1a0000000) reserve + memory for persistent storage. + + Say y here to enable this feature. + +config ARM64_PMEM_LEGACY_DEVICE + bool "Create persistent storage" + depends on BLK_DEV + depends on LIBNVDIMM + select ARM64_PMEM_RESERVE + help + Use reserved memory for persistent storage when the kernel + restart or update. the data in PMEM will not be lost and + can be loaded faster. + + Say y if unsure. + source "arch/arm64/Kconfig.platforms" menu "Kernel Features" diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile index 967cb3c6d..be996f3c1 100644 --- a/arch/arm64/kernel/Makefile +++ b/arch/arm64/kernel/Makefile @@ -67,6 +67,7 @@ obj-$(CONFIG_ARM64_PTR_AUTH) += pointer_auth.o obj-$(CONFIG_SHADOW_CALL_STACK) += scs.o obj-$(CONFIG_ARM64_MTE) += mte.o obj-$(CONFIG_MPAM) += mpam/ +obj-$(CONFIG_ARM64_PMEM_LEGACY_DEVICE) += pmem.o obj-y += vdso/ probes/ obj-$(CONFIG_COMPAT_VDSO) += vdso32/ diff --git a/arch/arm64/kernel/pmem.c b/arch/arm64/kernel/pmem.c new file mode 100644 index 000000000..16eaf706f --- /dev/null +++ b/arch/arm64/kernel/pmem.c @@ -0,0 +1,35 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright(c) 2021 Huawei Technologies Co., Ltd + * + * Derived from x86 and arm64 implement PMEM. + */ +#include <linux/platform_device.h> +#include <linux/init.h> +#include <linux/ioport.h> +#include <linux/module.h> + +static int found(struct resource *res, void *data) +{ + return 1; +} + +static int __init register_e820_pmem(void) +{ + struct platform_device *pdev; + int rc; + + rc = walk_iomem_res_desc(IORES_DESC_PERSISTENT_MEMORY_LEGACY, + IORESOURCE_MEM, 0, -1, NULL, found); + if (rc <= 0) + return 0; + + /* + * See drivers/nvdimm/e820.c for the implementation, this is + * simply here to trigger the module to load on demand. + */ + pdev = platform_device_alloc("e820_pmem", -1); + + return platform_device_add(pdev); +} +device_initcall(register_e820_pmem); diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index 5e282d31a..84c71c88d 100644 --- a/arch/arm64/kernel/setup.c +++ b/arch/arm64/kernel/setup.c @@ -57,6 +57,10 @@ static int num_standard_resources; static struct resource *standard_resources; +#ifdef CONFIG_ARM64_PMEM_RESERVE +extern struct resource pmem_res; +#endif + phys_addr_t __fdt_pointer __initdata; /* @@ -270,6 +274,12 @@ static void __init request_standard_resources(void) request_resource(res, &pin_memory_resource); #endif } + +#ifdef CONFIG_ARM64_PMEM_RESERVE + if (pmem_res.end && pmem_res.start) + request_resource(&iomem_resource, &pmem_res); +#endif + } static int __init reserve_memblock_reserved_regions(void) diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index b3437440d..f22faea1a 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -66,6 +66,18 @@ EXPORT_SYMBOL(memstart_addr); phys_addr_t arm64_dma_phys_limit __ro_after_init; phys_addr_t arm64_dma32_phys_limit __ro_after_init; +static unsigned long long pmem_size, pmem_start; + +#ifdef CONFIG_ARM64_PMEM_RESERVE +struct resource pmem_res = { + .name = "Persistent Memory (legacy)", + .start = 0, + .end = 0, + .flags = IORESOURCE_MEM, + .desc = IORES_DESC_PERSISTENT_MEMORY_LEGACY +}; +#endif + #ifndef CONFIG_KEXEC_CORE static void __init reserve_crashkernel(void) { @@ -378,6 +390,87 @@ static int __init reserve_park_mem(void) } #endif +static int __init is_mem_valid(unsigned long long mem_size, unsigned long long mem_start) +{ + if (!memblock_is_region_memory(mem_start, mem_size)) { + pr_warn("cannot reserve mem: region is not memory!\n"); + return -EINVAL; + } + + if (memblock_is_region_reserved(mem_start, mem_size)) { + pr_warn("cannot reserve mem: region overlaps reserved memory!\n"); + return -EINVAL; + } + + if (!IS_ALIGNED(mem_start, SZ_2M)) { + pr_warn("cannot reserve mem: base address is not 2MB aligned!\n"); + return -EINVAL; + } + + return 0; +} + +static int __init parse_memmap_one(char *p) +{ + char *oldp; + phys_addr_t start_at, mem_size; + + if (!p) + return -EINVAL; + + oldp = p; + mem_size = memparse(p, &p); + if (p == oldp) + return -EINVAL; + + if (!mem_size) + return -EINVAL; + + mem_size = PAGE_ALIGN(mem_size); + + if (*p == '!') { + start_at = memparse(p+1, &p); + + if (is_mem_valid(mem_size, start_at) != 0) + return -EINVAL; + + pr_info("pmem reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + start_at, start_at + mem_size, mem_size >> 20); + pmem_start = start_at; + pmem_size = mem_size; + } else + pr_info("Unrecognized memmap option, please check the parameter.\n"); + + return *p == '\0' ? 0 : -EINVAL; +} + +static int __init parse_memmap_opt(char *str) +{ + while (str) { + char *k = strchr(str, ','); + + if (k) + *k++ = 0; + + parse_memmap_one(str); + str = k; + } + + return 0; +} +early_param("memmap", parse_memmap_opt); + +#ifdef CONFIG_ARM64_PMEM_RESERVE +static void __init reserve_pmem(void) +{ + memblock_remove(pmem_start, pmem_size); + pr_info("pmem reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + pmem_start, pmem_start + pmem_size, pmem_size >> 20); + pmem_res.start = pmem_start; + pmem_res.end = pmem_start + pmem_size - 1; +} +#endif + void __init arm64_memblock_init(void) { const s64 linear_region_size = BIT(vabits_actual - 1); @@ -511,6 +604,10 @@ void __init arm64_memblock_init(void) reserve_elfcorehdr(); +#ifdef CONFIG_ARM64_PMEM_RESERVE + reserve_pmem(); +#endif + high_memory = __va(memblock_end_of_DRAM() - 1) + 1; dma_contiguous_reserve(arm64_dma32_phys_limit); diff --git a/drivers/nvdimm/Kconfig b/drivers/nvdimm/Kconfig index b7d1eb38b..ce4de7526 100644 --- a/drivers/nvdimm/Kconfig +++ b/drivers/nvdimm/Kconfig @@ -132,3 +132,8 @@ config NVDIMM_TEST_BUILD infrastructure. endif + +config PMEM_LEGACY + tristate "Pmem_legacy" + select X86_PMEM_LEGACY if X86 + select ARM64_PMEM_LEGACY_DEVICE if ARM64 diff --git a/drivers/nvdimm/Makefile b/drivers/nvdimm/Makefile index 29203f3d3..6f8dc9242 100644 --- a/drivers/nvdimm/Makefile +++ b/drivers/nvdimm/Makefile @@ -3,7 +3,7 @@ obj-$(CONFIG_LIBNVDIMM) += libnvdimm.o obj-$(CONFIG_BLK_DEV_PMEM) += nd_pmem.o obj-$(CONFIG_ND_BTT) += nd_btt.o obj-$(CONFIG_ND_BLK) += nd_blk.o -obj-$(CONFIG_X86_PMEM_LEGACY) += nd_e820.o +obj-$(CONFIG_PMEM_LEGACY) += nd_e820.o obj-$(CONFIG_OF_PMEM) += of_pmem.o obj-$(CONFIG_VIRTIO_PMEM) += virtio_pmem.o nd_virtio.o -- 2.19.1

3 4

openEuler 20.03 LTS SP1 SP2 兼容性 KABI 白名单初稿预审
by Xie XiuQi 08 Mar '21

08 Mar '21

各位好，根据前期沟通，OSV、驱动兼容性SIG都对内核接口兼容性提出了诉求。根据前期沟通的信息，以及收集的驱动接口使用情况，形成一份 20.03 LTS SP1 SP2 兼容性 KABI 白名单初稿，请大家评审。 KABI (Kernel Application Binary Interface) 兼容，即内核与驱动的二进制兼容。就是驱动不用重新编译，就可以在新内核上安装使用。如果驱动用到的接口都是兼容的，那么驱动就可以不用重新编译就可以在新版本安装使用。上游社区考虑到便于开发、和防止架构腐化，不考虑 KABI 的兼容性。业界的 KABI 兼容都是由 Linux 发行版来做的，而且兼容的接口越多、维护时间越长，维护成本也越高。 openEuler 通过多次与下游OSV和驱动团队交流，根据下游反馈的诉求， openEuler 20.03 LTS SP1、SP2 版本，提供一定范围的 KABI 兼容能力，目的是使常用的板卡驱动，在SP1、后续的 SP1 update 版本、SP2及SP2 update版本能兼容。 kabi 兼容列表白名单的主要输入是驱动（ko），我们根据下游的反馈，收集到部分常用驱动，作为兼容的目标，驱动列表如下。 ARM64 版本拟兼容的驱动（ko）列表 |amdgpu.ko bnx2.ko bnx2x.ko bnxt_en.ko bnxt_re.ko hclge.ko hclgevf.ko hifc.ko hinic.ko hnae3.ko hns3.ko i40e.ko ice.ko igb.ko ixgbe.ko ixgbevf.ko lpfc.ko megaraid_sas.ko mlx4_core.ko mlx4_ib.ko mlx5_core.ko mlx5_ib.ko mpt3sas.ko nouveau.ko nvme.ko qed.ko qede.ko qla2xxx.ko smartpqi.ko tg3.ko txgbe.ko | X86 版本拟兼容的驱动（ko）列表 |amdgpu.ko bnx2.ko bnx2x.ko bnxt_en.ko bnxt_re.ko hifc.ko hinic.ko i40e.ko ice.ko igb.ko ixgbe.ko ixgbevf.ko lpfc.ko megaraid_sas.ko mlx4_core.ko mlx4_ib.ko mlx5_core.ko mlx5_ib.ko mpt3sas.ko nouveau.ko nvme.ko qed.ko qede.ko qla2xxx.ko smartpqi.ko tg3.ko txgbe.ko | 我们收集了上表中驱动使用到接口，形成了一份 kabi 白名单列表初稿，请大家评审。重要的说明和提示： 1. 上述列表中多数驱动，还没有针对 openEuler SP1 的正式二进制版本，因此我们根据开源的版本或相近的版本收集了相关 kabi 列表，可能与板卡厂商的最终发布的版本存在少量差异。下游OSV或驱动团队如果发现有接口没在给出的列表中的，可以在评审过程中提出来。 2. 如果有新增 KABI 兼容性的诉求，需要给出需要兼容的 kabi 接口名称，以及使用到的驱动名称，以方便评估。 3. 评审反馈，可以在 issue 中反馈，https://gitee.com/openeuler/kernel/issues/I3ABVJ 也可以通过回复邮件反馈。 4. 收集评审反馈信息的时间为1周，截止到下周五（3月12日）下午17:00. --- openEuler kernel SIG， 2021-3-6 --- 附1：ARM64 平台 KABI 白名单列表初稿（1905个） acpi_bus_get_device acpi_check_dsm acpi_dev_found acpi_disabled acpi_dma_configure acpi_evaluate_dsm acpi_evaluate_object acpi_format_exception acpi_gbl_FADT acpi_get_devices acpi_get_handle acpi_get_name acpi_get_table acpi_gsi_to_irq acpi_handle_printk acpi_has_method acpi_lid_open acpi_os_map_memory acpi_os_unmap_memory acpi_register_gsi acpi_unregister_gsi add_timer add_wait_queue add_wait_queue_exclusive admin_timeout alloc_chrdev_region alloc_cpu_rmap __alloc_disk_node alloc_etherdev_mqs alloc_netdev_mqs alloc_pages_current __alloc_pages_nodemask __alloc_percpu __alloc_skb __alloc_workqueue_key __arch_clear_user __arch_copy_from_user __arch_copy_in_user __arch_copy_to_user arch_timer_read_counter arch_wb_cache_pmem arm64_const_caps_ready arp_tbl async_schedule _atomic_dec_and_lock atomic_notifier_call_chain atomic_notifier_chain_register atomic_notifier_chain_unregister attribute_container_find_class_device autoremove_wake_function backlight_device_register backlight_device_unregister backlight_force_update bdevname bdev_read_only bdget_disk _bin2bcd bio_add_page bio_alloc_bioset bio_clone_fast bio_endio bio_free_pages bio_init bio_put bioset_exit bioset_init __bitmap_and __bitmap_andnot __bitmap_clear __bitmap_complement __bitmap_equal bitmap_find_free_region bitmap_find_next_zero_area_off bitmap_free __bitmap_intersects __bitmap_or __bitmap_parse bitmap_parselist bitmap_print_to_pagebuf bitmap_release_region __bitmap_set __bitmap_weight __bitmap_xor bitmap_zalloc bit_wait blk_alloc_queue blk_check_plugged blk_cleanup_queue blkdev_get_by_path blkdev_issue_discard blkdev_issue_write_same blkdev_issue_zeroout blkdev_put blk_execute_rq blk_execute_rq_nowait blk_finish_plug blk_get_queue blk_get_request blk_mq_alloc_tag_set blk_mq_complete_request blk_mq_end_request blk_mq_free_request blk_mq_free_tag_set blk_mq_init_queue blk_mq_map_queues blk_mq_pci_map_queues blk_mq_quiesce_queue blk_mq_run_hw_queues blk_mq_start_request blk_mq_tagset_busy_iter blk_mq_tag_to_rq blk_mq_unique_tag blk_mq_unquiesce_queue blk_mq_update_nr_hw_queues blk_put_queue blk_put_request blk_queue_bounce_limit blk_queue_dma_alignment blk_queue_flag_clear blk_queue_flag_set blk_queue_io_min blk_queue_io_opt blk_queue_logical_block_size blk_queue_make_request blk_queue_max_discard_sectors blk_queue_max_hw_sectors blk_queue_max_segments blk_queue_max_segment_size blk_queue_max_write_same_sectors blk_queue_physical_block_size blk_queue_rq_timeout blk_queue_segment_boundary blk_queue_split blk_queue_stack_limits blk_queue_update_dma_alignment blk_queue_virt_boundary blk_queue_write_cache blk_rq_append_bio blk_rq_count_integrity_sg blk_rq_map_integrity_sg blk_rq_map_kern blk_rq_map_sg blk_rq_map_user_iov blk_rq_unmap_user blk_set_stacking_limits blk_start_plug blk_status_to_errno blk_verify_command blocking_notifier_call_chain blocking_notifier_chain_register blocking_notifier_chain_unregister bpf_prog_add bpf_prog_inc bpf_prog_put bpf_trace_run1 bpf_trace_run2 bpf_trace_run3 bpf_trace_run5 bpf_warn_invalid_xdp_action bsg_job_done btree_destroy btree_geo32 btree_geo64 btree_get_prev btree_init btree_insert btree_last btree_lookup btree_remove btree_update build_skb cache_line_size call_netdevice_notifiers call_rcu_sched call_srcu call_usermodehelper cancel_delayed_work cancel_delayed_work_sync cancel_work_sync capable cdev_add cdev_del cdev_init __chash_table_copy_in __chash_table_copy_out __check_object_size __class_create class_destroy __class_register class_unregister _cleanup_srcu_struct clk_get_rate commit_creds compat_alloc_user_space complete complete_all complete_and_exit completion_done component_add component_del _cond_resched console_lock console_unlock __const_udelay consume_skb cpu_bit_bitmap cpufreq_quick_get __cpuhp_remove_state __cpuhp_setup_state cpu_hwcap_keys cpu_hwcaps cpumask_local_spread cpumask_next cpumask_next_and cpu_number __cpu_online_mask __cpu_possible_mask __cpu_present_mask cpus_read_lock cpus_read_unlock crc32c crc32_le crc8 crc8_populate_msb crc_t10dif crypto_ahash_digest crypto_ahash_final crypto_ahash_setkey crypto_alloc_ahash crypto_destroy_tfm crypto_register_shash crypto_unregister_shash csum_ipv6_magic csum_partial csum_tcpudp_nofold _ctype dcb_getapp dcb_ieee_delapp dcb_ieee_getapp_mask dcb_ieee_setapp dcbnl_cee_notify dcb_setapp debugfs_create_atomic_t debugfs_create_dir debugfs_create_file debugfs_create_symlink debugfs_create_u16 debugfs_create_u32 debugfs_create_u64 debugfs_create_u8 debugfs_lookup debugfs_remove default_llseek default_wake_function __delay delayed_work_timer_fn del_gendisk del_timer del_timer_sync destroy_workqueue dev_add_pack dev_addr_add dev_addr_del dev_base_lock dev_close dev_driver_string _dev_err __dev_get_by_index dev_get_by_index dev_get_by_index_rcu dev_get_by_name dev_get_iflink dev_get_stats device_add_disk device_create device_create_file device_destroy device_link_add device_release_driver device_remove_file device_reprobe device_set_wakeup_capable device_set_wakeup_enable _dev_info __dev_kfree_skb_any __dev_kfree_skb_irq devlink_alloc devlink_free devlink_param_driverinit_value_get devlink_param_driverinit_value_set devlink_params_register devlink_params_unregister devlink_param_value_changed devlink_port_attrs_set devlink_port_register devlink_port_type_clear devlink_port_type_eth_set devlink_port_type_ib_set devlink_port_unregister devlink_region_create devlink_region_destroy devlink_region_shapshot_id_get devlink_region_snapshot_create devlink_register devlink_unregister dev_mc_add dev_mc_add_excl dev_mc_del devm_free_irq devm_hwmon_device_register_with_groups devm_ioremap devm_iounmap devm_kfree devm_kmalloc devm_kmemdup devm_mdiobus_alloc_size devm_request_threaded_irq _dev_notice dev_open dev_printk dev_queue_xmit dev_remove_pack dev_set_mac_address dev_set_mtu dev_set_promiscuity dev_trans_start dev_uc_add dev_uc_add_excl dev_uc_del _dev_warn disable_irq disable_irq_nosync dma_alloc_from_dev_coherent dma_common_get_sgtable dma_common_mmap dma_fence_add_callback dma_fence_array_create dma_fence_context_alloc dma_fence_free dma_fence_get_status dma_fence_init dma_fence_release dma_fence_signal dma_fence_signal_locked dma_fence_wait_any_timeout dma_fence_wait_timeout dma_get_required_mask dmam_alloc_coherent dmam_free_coherent dma_pool_alloc dma_pool_create dma_pool_destroy dma_pool_free dma_release_from_dev_coherent dmi_check_system dmi_get_system_info dmi_match __do_once_done __do_once_start do_wait_intr down downgrade_write down_interruptible down_read down_read_trylock down_timeout down_trylock down_write down_write_killable down_write_trylock dput dql_completed dql_reset drain_workqueue driver_create_file driver_for_each_device driver_remove_file drm_add_edid_modes drm_add_modes_noedid drm_atomic_add_affected_connectors drm_atomic_add_affected_planes drm_atomic_commit drm_atomic_get_connector_state drm_atomic_get_crtc_state drm_atomic_get_plane_state drm_atomic_helper_check drm_atomic_helper_check_modeset drm_atomic_helper_check_planes drm_atomic_helper_check_plane_state drm_atomic_helper_cleanup_planes drm_atomic_helper_commit drm_atomic_helper_commit_cleanup_done drm_atomic_helper_commit_hw_done __drm_atomic_helper_connector_destroy_state drm_atomic_helper_connector_destroy_state __drm_atomic_helper_connector_duplicate_state __drm_atomic_helper_connector_reset __drm_atomic_helper_crtc_destroy_state __drm_atomic_helper_crtc_duplicate_state drm_atomic_helper_disable_plane drm_atomic_helper_legacy_gamma_set drm_atomic_helper_page_flip __drm_atomic_helper_plane_destroy_state drm_atomic_helper_plane_destroy_state __drm_atomic_helper_plane_duplicate_state drm_atomic_helper_prepare_planes drm_atomic_helper_resume drm_atomic_helper_set_config drm_atomic_helper_setup_commit drm_atomic_helper_shutdown drm_atomic_helper_suspend drm_atomic_helper_swap_state drm_atomic_helper_update_legacy_modeset_state drm_atomic_helper_update_plane drm_atomic_helper_wait_for_dependencies drm_atomic_helper_wait_for_fences drm_atomic_helper_wait_for_flip_done drm_atomic_state_alloc drm_atomic_state_default_clear drm_atomic_state_default_release __drm_atomic_state_free drm_atomic_state_init drm_calc_vbltimestamp_from_scanoutpos drm_color_lut_extract drm_compat_ioctl drm_connector_attach_encoder drm_connector_cleanup drm_connector_init drm_connector_list_iter_begin drm_connector_list_iter_end drm_connector_list_iter_next drm_connector_register drm_connector_set_path_property drm_connector_unregister drm_connector_update_edid_property drm_crtc_accurate_vblank_count drm_crtc_add_crc_entry drm_crtc_arm_vblank_event drm_crtc_cleanup __drm_crtc_commit_free drm_crtc_enable_color_mgmt drm_crtc_force_disable_all drm_crtc_from_index drm_crtc_handle_vblank drm_crtc_helper_set_config drm_crtc_helper_set_mode drm_crtc_init drm_crtc_init_with_planes drm_crtc_send_vblank_event drm_crtc_vblank_count drm_crtc_vblank_get drm_crtc_vblank_off drm_crtc_vblank_on drm_crtc_vblank_put drm_cvt_mode drm_dbg drm_debugfs_create_files drm_detect_hdmi_monitor drm_detect_monitor_audio drm_dev_alloc drm_dev_put drm_dev_register drm_dev_unregister drm_dp_atomic_find_vcpi_slots drm_dp_atomic_release_vcpi_slots drm_dp_aux_register drm_dp_aux_unregister drm_dp_bw_code_to_link_rate drm_dp_calc_pbn_mode drm_dp_channel_eq_ok drm_dp_check_act_status drm_dp_clock_recovery_ok drm_dp_dpcd_read drm_dp_dpcd_read_link_status drm_dp_dpcd_write drm_dp_find_vcpi_slots drm_dp_get_adjust_request_pre_emphasis drm_dp_get_adjust_request_voltage drm_dp_link_rate_to_bw_code drm_dp_link_train_channel_eq_delay drm_dp_link_train_clock_recovery_delay drm_dp_mst_allocate_vcpi drm_dp_mst_deallocate_vcpi drm_dp_mst_detect_port drm_dp_mst_get_edid drm_dp_mst_hpd_irq drm_dp_mst_reset_vcpi_slots drm_dp_mst_topology_mgr_destroy drm_dp_mst_topology_mgr_init drm_dp_mst_topology_mgr_resume drm_dp_mst_topology_mgr_set_mst drm_dp_mst_topology_mgr_suspend drm_dp_update_payload_part1 drm_dp_update_payload_part2 drm_edid_header_is_valid drm_edid_is_valid drm_edid_to_sad drm_edid_to_speaker_allocation drm_encoder_cleanup drm_encoder_init drm_err drm_fb_helper_add_one_connector drm_fb_helper_alloc_fbi drm_fb_helper_blank drm_fb_helper_cfb_copyarea drm_fb_helper_cfb_fillrect drm_fb_helper_cfb_imageblit drm_fb_helper_check_var drm_fb_helper_debug_enter drm_fb_helper_debug_leave drm_fb_helper_fill_fix drm_fb_helper_fill_var drm_fb_helper_fini drm_fb_helper_hotplug_event drm_fb_helper_init drm_fb_helper_initial_config drm_fb_helper_ioctl drm_fb_helper_lastclose drm_fb_helper_output_poll_changed drm_fb_helper_pan_display drm_fb_helper_prepare drm_fb_helper_remove_one_connector drm_fb_helper_setcmap drm_fb_helper_set_par drm_fb_helper_set_suspend drm_fb_helper_set_suspend_unlocked drm_fb_helper_single_add_all_connectors drm_fb_helper_unregister_fbi drm_format_plane_cpp drm_framebuffer_cleanup drm_framebuffer_init drm_framebuffer_unregister_private drm_gem_dmabuf_kmap drm_gem_dmabuf_kunmap drm_gem_dmabuf_mmap drm_gem_dmabuf_release drm_gem_dmabuf_vmap drm_gem_dmabuf_vunmap drm_gem_fb_create_handle drm_gem_fb_destroy drm_gem_handle_create drm_gem_handle_delete drm_gem_map_attach drm_gem_map_detach drm_gem_map_dma_buf drm_gem_object_init drm_gem_object_lookup drm_gem_object_put_unlocked drm_gem_object_release drm_gem_prime_export drm_gem_prime_fd_to_handle drm_gem_prime_handle_to_fd drm_gem_prime_import drm_gem_private_object_init drm_gem_unmap_dma_buf drm_get_edid drm_get_edid_switcheroo drm_get_format_name drm_get_max_iomem drm_global_item_ref drm_global_item_unref drm_handle_vblank drm_hdmi_avi_infoframe_from_display_mode drm_hdmi_vendor_infoframe_from_display_mode drm_helper_connector_dpms drm_helper_disable_unused_functions drm_helper_hpd_irq_event drm_helper_mode_fill_fb_struct drm_helper_probe_single_connector_modes drm_helper_resume_force_mode drm_i2c_encoder_detect drm_i2c_encoder_init drm_i2c_encoder_mode_fixup drm_i2c_encoder_restore drm_i2c_encoder_save drm_invalid_op drm_ioctl drm_irq_install drm_irq_uninstall drm_is_current_master drm_kms_helper_hotplug_event drm_kms_helper_is_poll_worker drm_kms_helper_poll_disable drm_kms_helper_poll_enable drm_kms_helper_poll_fini drm_kms_helper_poll_init drm_match_cea_mode drm_mm_init drm_mm_insert_node_in_range drm_mm_print drm_mm_remove_node drm_mm_takedown drm_mode_config_cleanup drm_mode_config_init drm_mode_config_reset drm_mode_copy drm_mode_create_dvi_i_properties drm_mode_create_scaling_mode_property drm_mode_create_tv_properties drm_mode_crtc_set_gamma_size drm_mode_debug_printmodeline drm_mode_destroy drm_mode_duplicate drm_mode_equal drm_mode_get_hv_timing drm_mode_is_420_only drm_mode_legacy_fb_format drm_mode_object_find drm_mode_object_put drm_mode_probed_add drm_mode_set_crtcinfo drm_modeset_lock drm_modeset_lock_all drm_modeset_lock_all_ctx drm_mode_set_name drm_modeset_unlock drm_modeset_unlock_all drm_mode_vrefresh drm_object_attach_property drm_object_property_set_value drm_open drm_plane_cleanup drm_plane_create_alpha_property drm_plane_create_color_properties drm_plane_create_zpos_immutable_property drm_plane_create_zpos_property drm_plane_force_disable drm_plane_init drm_poll drm_primary_helper_destroy drm_primary_helper_funcs drm_prime_gem_destroy drm_prime_pages_to_sg drm_prime_sg_to_page_addr_arrays drm_printf __drm_printfn_seq_file drm_property_add_enum drm_property_create drm_property_create_enum drm_property_create_range __drm_puts_seq_file drm_read drm_release drm_scdc_read drm_scdc_write drm_sched_dependency_optimized drm_sched_entity_destroy drm_sched_entity_fini drm_sched_entity_flush drm_sched_entity_init drm_sched_entity_push_job drm_sched_entity_set_rq drm_sched_fini drm_sched_hw_job_reset drm_sched_init drm_sched_job_init drm_sched_job_recovery drm_send_event_locked drm_syncobj_create drm_syncobj_find drm_syncobj_find_fence drm_syncobj_free drm_syncobj_get_fd drm_syncobj_get_handle drm_syncobj_replace_fence drm_universal_plane_init drm_vblank_init drm_vma_node_allow drm_vma_node_is_allowed drm_vma_node_revoke dst_release dummy_dma_ops dump_stack __dynamic_dev_dbg __dynamic_netdev_dbg __dynamic_pr_debug elfcorehdr_addr emergency_restart enable_irq errno_to_blk_status ether_setup eth_get_headlen eth_platform_get_mac_address ethtool_convert_legacy_u32_to_link_mode ethtool_convert_link_mode_to_legacy_u32 ethtool_intersect_link_masks ethtool_op_get_link ethtool_op_get_ts_info eth_type_trans eth_validate_addr eventfd_ctx_fdget eventfd_ctx_put eventfd_signal event_triggers_call fasync_helper fc_attach_transport fc_block_scsi_eh fc_eh_timed_out fc_get_event_number fc_host_post_event fc_host_post_vendor_event fc_release_transport fc_remote_port_add fc_remote_port_delete fc_remote_port_rolechg fc_remove_host fc_vport_create fc_vport_terminate fd_install fget filemap_fault filp_close filp_open find_get_pid find_last_bit find_next_bit find_next_zero_bit find_pid_ns find_vma finish_wait firmware_request_nowarn fixed_size_llseek flow_keys_dissector flush_delayed_work flush_signals flush_work flush_workqueue force_sig fortify_panic fput free_fib_info free_irq free_irq_cpu_rmap free_netdev __free_pages free_pages free_percpu from_kgid from_kuid fs_bio_set gcd generate_random_uuid generic_end_io_acct generic_file_llseek generic_handle_irq generic_make_request generic_start_io_acct genlmsg_put genl_register_family genl_unregister_family genphy_read_status genphy_restart_aneg get_device __get_free_pages get_gendisk get_pid_task get_random_bytes __get_task_comm get_task_mm get_unused_fd_flags get_user_pages get_user_pages_remote get_zeroed_page gic_pmr_sync groups_alloc groups_free handle_simple_irq hdmi_avi_infoframe_pack hdmi_infoframe_pack hnae3_register_ae_algo hnae3_register_ae_dev hnae3_register_client hnae3_set_client_init_flag hnae3_unregister_ae_algo hnae3_unregister_ae_dev hnae3_unregister_client hrtimer_cancel hrtimer_forward hrtimer_init hrtimer_start_range_ns hrtimer_try_to_cancel __hw_addr_sync_dev __hw_addr_unsync_dev hwmon_device_register hwmon_device_register_with_groups hwmon_device_register_with_info hwmon_device_unregister i2c_add_adapter i2c_bit_add_bus i2c_bit_algo i2c_del_adapter i2c_new_device i2c_smbus_read_byte_data i2c_smbus_write_byte_data i2c_transfer i2c_unregister_device __ib_alloc_cq ib_alloc_device ib_alloc_odp_umem __ib_alloc_pd __ib_alloc_xrcd __ib_create_cq ib_create_qp ib_create_send_mad ib_dealloc_device ib_dealloc_pd ib_dereg_mr ib_destroy_cq ib_destroy_qp ib_dispatch_event ib_find_cached_pkey ib_free_cq ib_free_send_mad ib_get_cached_pkey ib_get_eth_speed ib_get_gids_from_rdma_hdr ib_get_rdma_header_version ib_modify_qp ib_modify_qp_is_ok ib_post_send_mad ib_process_cq_direct ib_query_pkey ib_query_port ib_query_qp ib_register_device ib_register_mad_agent ib_sa_cancel_query ib_sa_guid_info_rec_query ib_sa_register_client ib_sa_unregister_client ib_sg_to_pages ib_ud_header_init ib_ud_header_pack ib_ud_ip4_csum ib_umem_copy_from ib_umem_get ib_umem_odp_map_dma_pages ib_umem_odp_unmap_dma_pages ib_umem_page_count ib_umem_release ib_unregister_device ib_unregister_mad_agent ib_uverbs_get_ucontext ida_alloc_range ida_destroy ida_free idr_alloc idr_alloc_cyclic idr_alloc_u32 idr_destroy idr_find idr_for_each idr_get_next idr_get_next_ul idr_preload idr_remove idr_replace in4_pton in6_pton in_egroup_p __inet6_lookup_established __inet_lookup_established in_group_p init_net __init_rwsem init_srcu_struct init_task init_timer_key init_uts_ns init_wait_entry __init_waitqueue_head interval_tree_insert interval_tree_iter_first interval_tree_iter_next interval_tree_remove int_to_scsilun invalidate_partition iomem_resource iommu_get_domain_for_dev iommu_iova_to_phys iommu_map iommu_unmap __ioremap ioremap_cache io_schedule io_schedule_timeout __iounmap __iowrite32_copy __iowrite64_copy ip6_dst_hoplimit ip_compute_csum ipmi_create_user ipmi_destroy_user ipmi_free_recv_msg ipmi_poll_interface ipmi_request_settime ipmi_set_gets_events ipmi_set_my_address ipmi_smi_msg_received ipmi_unregister_smi ipmi_validate_addr ip_route_output_flow ip_send_check __ipv6_addr_type ipv6_ext_hdr ipv6_find_hdr ipv6_skip_exthdr ipv6_stub irq_cpu_rmap_add irq_create_mapping __irq_domain_add irq_domain_remove irq_find_mapping irq_get_irq_data irq_poll_complete irq_poll_disable irq_poll_enable irq_poll_init irq_poll_sched irq_set_affinity_hint irq_set_affinity_notifier irq_set_chip_and_handler_name irq_to_desc is_acpi_device_node jiffies jiffies_64 jiffies_to_msecs jiffies_to_timespec64 jiffies_to_usecs kallsyms_lookup_name kasprintf kernel_recvmsg kernel_sendmsg kernel_setsockopt kfree kfree_call_rcu kfree_const kfree_skb kgdb_active kgdb_breakpoint kill_fasync kimage_voffset __kmalloc kmalloc_caches __kmalloc_node kmalloc_order_trace kmem_cache_alloc kmem_cache_alloc_node kmem_cache_alloc_node_trace kmem_cache_alloc_trace kmem_cache_create kmem_cache_create_usercopy kmem_cache_destroy kmem_cache_free kmem_cache_shrink kmemdup kobject_add kobject_create_and_add kobject_del kobject_get kobject_init kobject_init_and_add kobject_put kobject_set_name kobject_uevent kobject_uevent_env krealloc kset_create_and_add kset_find_obj kset_register kset_unregister ksize kstrdup kstrdup_const kstrndup kstrtobool kstrtobool_from_user kstrtoint kstrtoint_from_user kstrtoll kstrtoll_from_user kstrtou16 kstrtouint kstrtouint_from_user kstrtoul_from_user kstrtoull kstrtoull_from_user kthread_bind kthread_create_on_node kthread_park kthread_should_stop kthread_stop kthread_unpark ktime_get ktime_get_coarse_real_ts64 ktime_get_raw ktime_get_raw_ts64 ktime_get_real_seconds ktime_get_real_ts64 ktime_get_seconds ktime_get_ts64 ktime_get_with_offset kvasprintf kvfree kvmalloc_node kzfree led_classdev_resume led_classdev_suspend led_classdev_unregister __list_add_valid __list_del_entry_valid llist_add_batch __ll_sc_atomic64_add __ll_sc_atomic64_add_return __ll_sc_atomic64_andnot __ll_sc_atomic64_fetch_add __ll_sc_atomic64_fetch_andnot __ll_sc_atomic64_fetch_andnot_release __ll_sc_atomic64_fetch_or __ll_sc_atomic64_fetch_or_acquire __ll_sc_atomic64_fetch_xor __ll_sc_atomic64_or __ll_sc_atomic64_sub __ll_sc_atomic64_sub_return __ll_sc_atomic_add __ll_sc_atomic_add_return __ll_sc_atomic_add_return_acquire __ll_sc_atomic_sub __ll_sc_atomic_sub_return __ll_sc_atomic_sub_return_release __ll_sc___cmpxchg_case_acq_4 __ll_sc___cmpxchg_case_mb_4 __ll_sc___cmpxchg_case_mb_8 __local_bh_enable_ip __lock_page lock_page_memcg lockref_get lock_sock_nested logic_inw logic_outw make_kgid make_kuid mark_page_accessed match_strdup match_string match_token _mcount mdev_dev mdev_from_dev mdev_get_drvdata mdev_parent_dev mdev_register_device mdev_register_driver mdev_set_drvdata mdev_unregister_device mdev_unregister_driver mdio45_probe mdiobus_alloc_size mdiobus_free mdiobus_get_phy mdiobus_read __mdiobus_register mdiobus_unregister mdiobus_write mdio_mii_ioctl memchr memchr_inv memcmp memcpy __memcpy_fromio __memcpy_toio memdup_user memdup_user_nul memmove memory_read_from_buffer memparse mempool_alloc mempool_alloc_slab mempool_create mempool_create_node mempool_destroy mempool_free mempool_free_slab mempool_kfree mempool_kmalloc memscan mem_section memset __memset_io memstart_addr memzero_explicit metadata_dst_alloc misc_deregister misc_register mlxfw_firmware_flash mmput __mmu_notifier_register mmu_notifier_register mmu_notifier_unregister mmu_notifier_unregister_no_release mod_delayed_work_on mod_timer mod_timer_pending __module_get module_layout module_put module_refcount __msecs_to_jiffies msleep msleep_interruptible __mutex_init mutex_lock mutex_lock_interruptible mutex_lock_killable mutex_trylock mutex_unlock __napi_alloc_skb napi_complete_done napi_consume_skb napi_disable napi_gro_flush napi_gro_receive napi_hash_del __napi_schedule __napi_schedule_irqoff napi_schedule_prep __ndelay ndo_dflt_bridge_getlink ndo_dflt_fdb_add neigh_destroy __neigh_event_send neigh_lookup netdev_alloc_frag __netdev_alloc_skb netdev_bind_sb_channel_queue netdev_crit netdev_err netdev_features_change netdev_info netdev_lower_get_next netdev_master_upper_dev_get netdev_master_upper_dev_get_rcu netdev_notice netdev_printk netdev_reset_tc netdev_rss_key_fill netdev_rx_handler_register netdev_rx_handler_unregister netdev_set_num_tc netdev_set_sb_channel netdev_set_tc_queue netdev_unbind_sb_channel netdev_update_features netdev_walk_all_upper_dev_rcu netdev_warn netif_carrier_off netif_carrier_on netif_device_attach netif_device_detach netif_get_num_default_rss_queues netif_napi_add netif_napi_del netif_receive_skb netif_rx netif_schedule_queue netif_set_real_num_rx_queues netif_set_real_num_tx_queues netif_set_xps_queue netif_tx_stop_all_queues netif_tx_wake_queue netlink_ack netlink_broadcast __netlink_kernel_create netlink_kernel_release netlink_unicast net_ratelimit nla_find nla_parse nla_put __nlmsg_put node_data __node_distance node_states node_to_cpumask_map no_llseek noop_llseek nr_cpu_ids nr_node_ids nsecs_to_jiffies ns_to_timespec ns_to_timespec64 ns_to_timeval numa_node nvme_alloc_request nvme_cancel_request nvme_change_ctrl_state nvme_cleanup_cmd nvme_complete_async_event nvme_complete_rq nvme_disable_ctrl nvme_enable_ctrl nvme_fc_register_localport nvme_fc_register_remoteport nvme_fc_set_remoteport_devloss nvme_fc_unregister_localport nvme_fc_unregister_remoteport nvme_init_ctrl nvme_init_identify nvme_io_timeout nvme_kill_queues nvme_remove_namespaces nvme_reset_ctrl nvme_reset_ctrl_sync nvme_set_queue_count nvme_setup_cmd nvme_shutdown_ctrl nvme_start_ctrl nvme_start_freeze nvme_start_queues nvme_stop_ctrl nvme_stop_queues nvme_submit_sync_cmd nvmet_fc_rcv_fcp_abort nvmet_fc_rcv_fcp_req nvmet_fc_rcv_ls_req nvmet_fc_register_targetport nvmet_fc_unregister_targetport nvme_unfreeze nvme_uninit_ctrl nvme_wait_freeze nvme_wait_freeze_timeout nvme_wq of_led_classdev_register on_each_cpu orderly_poweroff out_of_line_wait_on_bit override_creds __page_file_index __page_frag_cache_drain page_frag_free __page_mapcount page_mapped page_pool_alloc_pages page_pool_create page_pool_destroy __page_pool_put_page pagevec_lookup_range pagevec_lookup_range_tag __pagevec_release panic panic_notifier_list param_array_ops param_get_int param_ops_bool param_ops_byte param_ops_charp param_ops_int param_ops_long param_ops_short param_ops_string param_ops_uint param_ops_ullong param_ops_ulong param_ops_ushort param_set_bool param_set_int pci_alloc_irq_vectors_affinity pci_assign_unassigned_bus_resources pcibios_resource_to_bus pci_bus_read_config_dword pci_bus_resource_n pci_bus_type pci_cfg_access_lock pci_cfg_access_unlock pci_check_and_mask_intx pci_choose_state pci_cleanup_aer_uncorrect_error_status pci_clear_master pci_clear_mwi pci_d3cold_disable pci_dev_driver pci_dev_get pci_device_is_present pci_dev_present pci_dev_put pci_disable_device pci_disable_link_state pci_disable_msi pci_disable_msix pci_disable_pcie_error_reporting pci_disable_rom pci_disable_sriov pcie_bandwidth_available pcie_capability_clear_and_set_word pcie_capability_read_dword pcie_capability_read_word pcie_capability_write_word pcie_flr pcie_get_speed_cap pcie_get_width_cap pci_enable_atomic_ops_to_root pci_enable_device pci_enable_device_mem pci_enable_msi pci_enable_msix_range pci_enable_pcie_error_reporting pci_enable_rom pci_enable_sriov pci_enable_wake pcie_print_link_status pcie_relaxed_ordering_enabled pcie_set_readrq pci_find_capability pci_find_ext_capability pci_free_irq pci_free_irq_vectors pci_get_class pci_get_device pci_get_domain_bus_and_slot pci_get_slot pci_ignore_hotplug pci_intx pci_iomap pci_ioremap_bar pci_irq_get_affinity pci_irq_vector pci_map_rom pci_match_id pcim_enable_device pcim_iomap pcim_iomap_regions pcim_iomap_table pcim_iounmap pci_msi_mask_irq pci_msi_unmask_irq pci_num_vf pci_platform_rom pci_prepare_to_sleep pci_read_config_byte pci_read_config_dword pci_read_config_word pci_read_vpd __pci_register_driver pci_release_regions pci_release_resource pci_release_selected_regions pci_request_irq pci_request_regions pci_request_selected_regions pci_rescan_bus pci_resize_resource pci_restore_state pci_save_state pci_select_bars pci_set_master pci_set_mwi pci_set_power_state pci_sriov_configure_simple pci_sriov_get_totalvfs pci_sriov_set_totalvfs pci_stop_and_remove_bus_device pci_stop_and_remove_bus_device_locked pci_try_set_mwi pci_unmap_rom pci_unregister_driver pci_vfs_assigned pci_vpd_find_info_keyword pci_vpd_find_tag pci_wait_for_pending_transaction pci_wake_from_d3 pci_write_config_byte pci_write_config_dword pci_write_config_word pcix_set_mmrbc PDE_DATA __per_cpu_offset perf_trace_buf_alloc perf_trace_run_bpf_submit pfn_valid phy_attached_info phy_connect phy_connect_direct phy_disconnect phy_ethtool_ksettings_get phy_ethtool_ksettings_set phy_loopback phy_mii_ioctl phy_resume phy_start phy_start_aneg phy_stop phy_suspend pid_task pid_vnr pm_power_off pm_runtime_allow __pm_runtime_disable pm_runtime_enable pm_runtime_forbid __pm_runtime_idle __pm_runtime_resume pm_runtime_set_autosuspend_delay __pm_runtime_set_status __pm_runtime_suspend __pm_runtime_use_autosuspend pm_schedule_suspend power_supply_is_system_supplied prandom_bytes prandom_u32 prepare_creds prepare_to_wait prepare_to_wait_event prepare_to_wait_exclusive print_hex_dump printk __printk_ratelimit print_stack_trace proc_create_data proc_mkdir __pskb_copy_fclone pskb_expand_head __pskb_pull_tail ___pskb_trim ptp_clock_event ptp_clock_index ptp_clock_register ptp_clock_unregister ptp_find_pin __put_cred put_device put_disk __put_page __put_task_struct put_unused_fd qed_get_eth_ops qed_put_eth_ops queue_delayed_work_on queued_read_lock_slowpath queued_spin_lock_slowpath queued_write_lock_slowpath queue_work_on radix_tree_delete radix_tree_gang_lookup __radix_tree_insert radix_tree_lookup radix_tree_tagged raid_class_attach raid_class_release ___ratelimit raw_notifier_call_chain raw_notifier_chain_register raw_notifier_chain_unregister rb_erase __rb_erase_color rb_first rb_first_postorder __rb_insert_augmented rb_insert_color rb_next rb_next_postorder rb_replace_node rbt_ib_umem_for_each_in_range rbt_ib_umem_lookup rcu_barrier rdma_create_ah rdma_destroy_ah rdma_is_zero_gid rdma_port_get_link_layer rdma_query_ah rdma_query_gid rdma_restrack_get rdma_restrack_put rdma_roce_rescan_device read_cache_pages recalc_sigpending refcount_dec_and_mutex_lock refcount_dec_and_test_checked refcount_dec_checked refcount_inc_checked refcount_inc_not_zero_checked register_acpi_notifier register_blkdev __register_chrdev register_chrdev_region register_fib_notifier register_inet6addr_notifier register_inetaddr_notifier register_netdev register_netdevice_notifier register_netevent_notifier register_reboot_notifier release_firmware release_pages __release_region release_sock remap_pfn_range remove_conflicting_framebuffers remove_proc_entry remove_wait_queue request_firmware request_firmware_direct request_firmware_nowait __request_module __request_region request_threaded_irq reservation_object_add_excl_fence reservation_object_add_shared_fence reservation_object_get_fences_rcu reservation_object_reserve_shared reservation_object_wait_timeout_rcu reservation_ww_class reset_devices revert_creds rhashtable_destroy rhashtable_free_and_destroy rhashtable_init rhashtable_insert_slow rhashtable_walk_enter rhashtable_walk_exit rhashtable_walk_next rhashtable_walk_start_check rhashtable_walk_stop rhltable_init rht_bucket_nested rht_bucket_nested_insert round_jiffies round_jiffies_relative rps_may_expire_flow rtc_time64_to_tm rtnl_is_locked rtnl_lock rtnl_trylock rtnl_unlock sas_attach_transport sas_disable_tlr sas_enable_tlr sas_end_device_alloc sas_expander_alloc sas_is_tlr_enabled sas_phy_add sas_phy_alloc sas_phy_free sas_port_add sas_port_add_phy sas_port_alloc_num sas_port_delete sas_port_delete_phy sas_port_free sas_read_port_mode_page sas_release_transport sas_remove_host sas_rphy_add save_stack_trace save_stack_trace_tsk sbitmap_queue_clear __sbitmap_queue_get sched_clock sched_setscheduler schedule schedule_hrtimeout schedule_hrtimeout_range schedule_timeout schedule_timeout_interruptible schedule_timeout_uninterruptible scmd_printk scnprintf scsi_add_device scsi_add_host_with_dma scsi_block_requests scsi_build_sense_buffer scsi_change_queue_depth scsi_device_get scsi_device_lookup scsi_device_put scsi_device_set_state scsi_device_type scsi_dma_map scsi_dma_unmap __scsi_execute scsi_get_vpd_page scsi_host_alloc scsi_host_busy scsi_host_get scsi_host_lookup scsi_host_put scsi_internal_device_block_nowait scsi_internal_device_unblock_nowait scsi_is_fc_rport scsi_is_host_device scsi_is_sdev_device __scsi_iterate_devices scsilun_to_int scsi_normalize_sense scsi_print_command scsi_register_driver scsi_remove_device scsi_remove_host scsi_remove_target scsi_sanitize_inquiry_string scsi_scan_host scsi_unblock_requests sdev_prefix_printk secpath_dup security_d_instantiate send_sig seq_lseek seq_open seq_printf seq_putc seq_puts seq_read seq_release seq_write set_cpus_allowed_ptr set_current_groups set_device_ro set_disk_ro set_freezable set_normalized_timespec64 set_page_dirty set_page_dirty_lock set_user_nice sg_alloc_table_from_pages sg_copy_from_buffer sg_copy_to_buffer sg_free_table sg_init_table sg_miter_next sg_miter_start sg_miter_stop sg_next sigprocmask si_meminfo simple_attr_open simple_attr_read simple_attr_release simple_attr_write simple_open simple_read_from_buffer simple_strtol simple_strtoul simple_strtoull simple_write_to_buffer single_open single_release skb_add_rx_frag skb_checksum skb_checksum_help skb_clone skb_clone_tx_timestamp skb_copy skb_copy_bits skb_copy_expand skb_dequeue __skb_flow_dissect __skb_get_hash __skb_gso_segment skb_gso_validate_mac_len __skb_pad skb_pull skb_push skb_put skb_queue_purge skb_queue_tail skb_realloc_headroom skb_store_bits skb_trim skb_tstamp_tx skb_vlan_pop sk_free smp_call_function_many smp_call_function_single snprintf sock_create_kern sock_edemux sock_queue_err_skb sock_release softnet_data sort sprintf srcu_barrier __srcu_read_lock __srcu_read_unlock sscanf __stack_chk_fail __stack_chk_guard starget_for_each_device strcasecmp strcat strchr strcmp strcpy strcspn strim strlcat strlcpy strlen strncasecmp strncat strncmp strncpy strncpy_from_user strnlen strnstr strpbrk strrchr strsep strspn strstr submit_bio __sw_hweight32 __sw_hweight64 __sw_hweight8 swiotlb_nr_tbl switchdev_port_same_parent_id __symbol_put sync_file_create synchronize_irq synchronize_net synchronize_sched synchronize_srcu sysfs_add_file_to_group sysfs_create_bin_file sysfs_create_file_ns sysfs_create_group sysfs_remove_bin_file sysfs_remove_file_from_group sysfs_remove_file_ns sysfs_remove_group sysfs_streq system_state system_unbound_wq system_wq sys_tz task_active_pid_ns tasklet_init tasklet_kill __tasklet_schedule __task_pid_nr_ns tcf_block_cb_register tcf_block_cb_unregister tcp_gro_complete tcp_hashinfo tc_setup_cb_egdev_register tc_setup_cb_egdev_unregister time64_to_tm timecounter_cyc2time timecounter_init timecounter_read tls_get_record tls_validate_xmit_skb to_drm_sched_fence trace_define_field trace_event_buffer_commit trace_event_buffer_reserve trace_event_ignore_this_pid trace_event_raw_init trace_event_reg trace_handle_return __tracepoint_dma_fence_emit __tracepoint_xdp_exception trace_print_array_seq trace_print_flags_seq trace_raw_output_prep trace_seq_printf trace_seq_putc try_module_get try_wait_for_completion ttm_bo_add_to_lru ttm_bo_clean_mm ttm_bo_del_sub_from_lru ttm_bo_device_init ttm_bo_device_release ttm_bo_dma_acc_size ttm_bo_eviction_valuable ttm_bo_evict_mm ttm_bo_global_init ttm_bo_global_release ttm_bo_init ttm_bo_init_mm ttm_bo_init_reserved ttm_bo_kmap ttm_bo_kunmap ttm_bo_lock_delayed_workqueue ttm_bo_manager_func ttm_bo_mem_put ttm_bo_mem_space ttm_bo_mmap ttm_bo_move_accel_cleanup ttm_bo_move_memcpy ttm_bo_move_to_lru_tail ttm_bo_move_ttm ttm_bo_pipeline_move ttm_bo_put ttm_bo_unlock_delayed_workqueue ttm_bo_validate ttm_bo_wait ttm_dma_page_alloc_debugfs ttm_dma_populate ttm_dma_tt_fini ttm_dma_tt_init ttm_dma_unpopulate ttm_eu_backoff_reservation ttm_eu_fence_buffer_objects ttm_eu_reserve_buffers ttm_fbdev_mmap ttm_mem_global_init ttm_mem_global_release ttm_page_alloc_debugfs ttm_pool_populate ttm_pool_unpopulate ttm_populate_and_map_pages ttm_sg_tt_init ttm_tt_bind ttm_tt_set_placement_caching ttm_unmap_and_unpopulate_pages __udelay udp4_hwcsum uio_event_notify __uio_register_device uio_unregister_device unlock_page unlock_page_memcg unmap_mapping_range unregister_acpi_notifier unregister_blkdev __unregister_chrdev unregister_chrdev_region unregister_fib_notifier unregister_inet6addr_notifier unregister_inetaddr_notifier unregister_netdev unregister_netdevice_notifier unregister_netevent_notifier unregister_reboot_notifier unuse_mm up up_read up_write __usecs_to_jiffies use_mm usleep_range _uverbs_alloc uverbs_copy_to uverbs_destroy_def_handler uverbs_fd_class uverbs_get_flags32 uverbs_get_flags64 uverbs_idr_class vfree vga_client_register vlan_dev_real_dev vlan_dev_vlan_id vlan_dev_vlan_proto vmalloc __vmalloc vmalloc_node vmalloc_to_page vmap vm_insert_page vm_mmap vm_munmap vprintk vscnprintf vsnprintf vsprintf vunmap vzalloc vzalloc_node wait_for_completion wait_for_completion_interruptible wait_for_completion_interruptible_timeout wait_for_completion_io_timeout wait_for_completion_killable wait_for_completion_timeout wait_on_page_bit __wake_up wake_up_bit __wake_up_locked wake_up_process __warn_printk work_busy write_cache_pages ww_mutex_lock ww_mutex_lock_interruptible ww_mutex_unlock xdp_do_flush_map xdp_do_redirect xdp_return_frame xdp_return_frame_rx_napi xdp_rxq_info_is_reg xdp_rxq_info_reg xdp_rxq_info_reg_mem_model xdp_rxq_info_unreg xdp_rxq_info_unused xfrm_replay_seqhi xz_dec_end xz_dec_init xz_dec_run yield zap_vma_ptes zlib_inflate zlib_inflateEnd zlib_inflateInit2 zlib_inflate_workspacesize 附2：x86 平台 KABI 白名单列表初稿（2228个） acpi_bus_get_device acpi_bus_register_driver acpi_bus_unregister_driver acpi_check_dsm acpi_dev_found acpi_disabled acpi_dma_configure acpi_evaluate_dsm acpi_evaluate_integer acpi_evaluate_object acpi_format_exception acpi_gbl_FADT acpi_get_devices acpi_get_handle acpi_get_name acpi_get_table acpi_gsi_to_irq acpi_handle_printk acpi_has_method acpi_install_notify_handler acpi_lid_open acpi_os_map_memory acpi_os_unmap_memory acpi_register_gsi acpi_remove_notify_handler acpi_unregister_gsi acpi_video_get_edid acpi_walk_namespace address_space_init_once add_timer add_wait_queue add_wait_queue_exclusive admin_timeout alloc_chrdev_region alloc_cpumask_var alloc_cpu_rmap __alloc_disk_node alloc_etherdev_mqs alloc_netdev_mqs alloc_pages_current __alloc_pages_nodemask __alloc_percpu __alloc_skb __alloc_workqueue_key anon_inode_getfile apic arch_dma_alloc_attrs arch_io_free_memtype_wc arch_io_reserve_memtype_wc arch_phys_wc_add arch_phys_wc_del arch_wb_cache_pmem arp_tbl async_schedule _atomic_dec_and_lock atomic_notifier_call_chain atomic_notifier_chain_register atomic_notifier_chain_unregister attribute_container_find_class_device autoremove_wake_function backlight_device_register backlight_device_unregister backlight_force_update bdevname bdev_read_only bdget_disk _bin2bcd bio_add_page bio_alloc_bioset bio_clone_fast bio_endio bio_free_pages bio_init bio_put bioset_exit bioset_init __bitmap_and __bitmap_andnot __bitmap_clear __bitmap_complement __bitmap_equal bitmap_find_free_region bitmap_find_next_zero_area_off __bitmap_intersects __bitmap_or __bitmap_parse bitmap_parselist bitmap_print_to_pagebuf bitmap_release_region __bitmap_set __bitmap_shift_left __bitmap_shift_right __bitmap_subset __bitmap_weight __bitmap_xor bitmap_zalloc bit_wait blk_alloc_queue blk_check_plugged blk_cleanup_queue blkdev_get_by_path blkdev_issue_discard blkdev_issue_write_same blkdev_issue_zeroout blkdev_put blk_execute_rq blk_execute_rq_nowait blk_finish_plug blk_get_queue blk_get_request blk_init_tags blk_mq_alloc_tag_set blk_mq_complete_request blk_mq_end_request blk_mq_free_request blk_mq_free_tag_set blk_mq_init_queue blk_mq_map_queues blk_mq_pci_map_queues blk_mq_quiesce_queue blk_mq_run_hw_queues blk_mq_start_request blk_mq_tagset_busy_iter blk_mq_tag_to_rq blk_mq_unique_tag blk_mq_unquiesce_queue blk_mq_update_nr_hw_queues blk_put_queue blk_put_request blk_queue_bounce_limit blk_queue_dma_alignment blk_queue_flag_clear blk_queue_flag_set blk_queue_free_tags blk_queue_init_tags blk_queue_io_min blk_queue_io_opt blk_queue_logical_block_size blk_queue_make_request blk_queue_max_discard_sectors blk_queue_max_hw_sectors blk_queue_max_segments blk_queue_max_segment_size blk_queue_max_write_same_sectors blk_queue_physical_block_size blk_queue_rq_timeout blk_queue_segment_boundary blk_queue_split blk_queue_stack_limits blk_queue_update_dma_alignment blk_queue_virt_boundary blk_queue_write_cache blk_rq_append_bio blk_rq_count_integrity_sg blk_rq_map_integrity_sg blk_rq_map_kern blk_rq_map_sg blk_rq_map_user_iov blk_rq_unmap_user blk_set_stacking_limits blk_start_plug blk_status_to_errno blk_verify_command blocking_notifier_call_chain blocking_notifier_chain_register blocking_notifier_chain_unregister boot_cpu_data bpf_prog_add bpf_prog_inc bpf_prog_put bpf_trace_run1 bpf_trace_run2 bpf_trace_run3 bpf_trace_run5 bpf_warn_invalid_xdp_action bsg_job_done btree_destroy btree_geo32 btree_geo64 btree_get_prev btree_init btree_insert btree_last btree_lookup btree_remove btree_update build_skb bus_find_device_by_name __cachemode2pte_tbl call_netdevice_notifiers call_rcu_sched call_usermodehelper cancel_delayed_work cancel_delayed_work_sync cancel_work_sync capable cdev_add cdev_del cdev_device_add cdev_device_del cdev_init __chash_table_copy_in __chash_table_copy_out __check_object_size __class_create class_create_file_ns class_destroy __class_register class_remove_file_ns class_unregister _cleanup_srcu_struct clear_user clk_get_rate cm_class cnic_register_driver cnic_unregister_driver commit_creds compat_alloc_user_space complete complete_all complete_and_exit completion_done component_add component_del _cond_resched configfs_register_subsystem configfs_remove_default_groups configfs_unregister_subsystem config_group_init config_group_init_type_name config_item_put console_lock console_unlock __const_udelay consume_skb _copy_from_user _copy_to_user copy_user_enhanced_fast_string copy_user_generic_string copy_user_generic_unrolled cpu_bit_bitmap cpu_core_map cpufreq_get cpufreq_quick_get __cpuhp_remove_state __cpuhp_setup_state cpu_info cpu_khz cpumask_local_spread cpumask_next cpumask_next_and cpu_number __cpu_online_mask __cpu_possible_mask __cpu_present_mask cpu_sibling_map cpus_read_lock cpus_read_unlock crc32c crc32_le crc8 crc8_populate_msb crc_t10dif crypto_ahash_digest crypto_ahash_final crypto_ahash_setkey crypto_alloc_ahash crypto_destroy_tfm crypto_register_shash crypto_unregister_shash csum_ipv6_magic csum_partial _ctype current_task dca3_get_tag dca_add_requester dca_register_notify dca_remove_requester dca_unregister_notify dcb_getapp dcb_ieee_delapp dcb_ieee_getapp_mask dcb_ieee_setapp dcbnl_cee_notify dcbnl_ieee_notify dcb_setapp debugfs_create_atomic_t debugfs_create_dir debugfs_create_file debugfs_create_u32 debugfs_create_u64 debugfs_create_u8 debugfs_lookup debugfs_remove __default_kernel_pte_mask default_llseek default_wake_function __delay delayed_work_timer_fn del_gendisk del_timer del_timer_sync destroy_workqueue dev_add_pack dev_addr_add dev_addr_del dev_base_lock dev_close dev_driver_string _dev_err __dev_get_by_index dev_get_by_index dev_get_by_index_rcu dev_get_by_name dev_get_iflink dev_get_stats device_add_disk device_create device_create_file device_destroy device_initialize device_link_add device_release_driver device_remove_file device_reprobe device_set_wakeup_capable device_set_wakeup_enable _dev_info __dev_kfree_skb_any __dev_kfree_skb_irq devlink_alloc devlink_free devlink_param_driverinit_value_get devlink_param_driverinit_value_set devlink_params_register devlink_params_unregister devlink_param_value_changed devlink_port_attrs_set devlink_port_register devlink_port_type_clear devlink_port_type_ib_set devlink_port_unregister devlink_region_create devlink_region_destroy devlink_region_shapshot_id_get devlink_region_snapshot_create devlink_register devlink_unregister devmap_managed_key dev_mc_add dev_mc_add_excl dev_mc_del devm_free_irq devm_hwmon_device_register_with_groups devm_ioremap devm_iounmap devm_kfree devm_kmalloc devm_kmemdup devm_request_threaded_irq _dev_notice dev_printk dev_queue_xmit __dev_remove_pack dev_remove_pack dev_set_mac_address dev_set_mtu dev_set_name dev_set_promiscuity dev_trans_start dev_uc_add dev_uc_add_excl dev_uc_del _dev_warn disable_irq disable_irq_nosync dma_fence_add_callback dma_fence_array_create dma_fence_context_alloc dma_fence_free dma_fence_get_status dma_fence_init dma_fence_release dma_fence_signal dma_fence_signal_locked dma_fence_wait_any_timeout dma_fence_wait_timeout dma_get_required_mask dmam_alloc_coherent dmam_free_coherent dma_ops dma_pool_alloc dma_pool_create dma_pool_destroy dma_pool_free dmi_check_system dmi_get_system_info dmi_match __do_once_done __do_once_start do_wait_intr down downgrade_write down_interruptible down_read down_read_trylock down_timeout down_trylock down_write down_write_killable down_write_trylock dput dql_completed dql_reset drain_workqueue driver_create_file driver_for_each_device driver_remove_file drm_add_edid_modes drm_add_modes_noedid drm_atomic_add_affected_connectors drm_atomic_add_affected_planes drm_atomic_commit drm_atomic_get_connector_state drm_atomic_get_crtc_state drm_atomic_get_plane_state drm_atomic_helper_check drm_atomic_helper_check_modeset drm_atomic_helper_check_planes drm_atomic_helper_check_plane_state drm_atomic_helper_cleanup_planes drm_atomic_helper_commit drm_atomic_helper_commit_cleanup_done drm_atomic_helper_commit_hw_done __drm_atomic_helper_connector_destroy_state drm_atomic_helper_connector_destroy_state __drm_atomic_helper_connector_duplicate_state __drm_atomic_helper_connector_reset __drm_atomic_helper_crtc_destroy_state __drm_atomic_helper_crtc_duplicate_state drm_atomic_helper_disable_plane drm_atomic_helper_legacy_gamma_set drm_atomic_helper_page_flip __drm_atomic_helper_plane_destroy_state drm_atomic_helper_plane_destroy_state __drm_atomic_helper_plane_duplicate_state drm_atomic_helper_prepare_planes drm_atomic_helper_resume drm_atomic_helper_set_config drm_atomic_helper_setup_commit drm_atomic_helper_shutdown drm_atomic_helper_suspend drm_atomic_helper_swap_state drm_atomic_helper_update_legacy_modeset_state drm_atomic_helper_update_plane drm_atomic_helper_wait_for_dependencies drm_atomic_helper_wait_for_fences drm_atomic_helper_wait_for_flip_done drm_atomic_state_alloc drm_atomic_state_default_clear drm_atomic_state_default_release __drm_atomic_state_free drm_atomic_state_init drm_calc_vbltimestamp_from_scanoutpos drm_color_lut_extract drm_compat_ioctl drm_connector_attach_encoder drm_connector_cleanup drm_connector_init drm_connector_list_iter_begin drm_connector_list_iter_end drm_connector_list_iter_next drm_connector_register drm_connector_set_path_property drm_connector_unregister drm_connector_update_edid_property drm_crtc_accurate_vblank_count drm_crtc_add_crc_entry drm_crtc_arm_vblank_event drm_crtc_cleanup __drm_crtc_commit_free drm_crtc_enable_color_mgmt drm_crtc_force_disable_all drm_crtc_from_index drm_crtc_handle_vblank drm_crtc_helper_set_config drm_crtc_helper_set_mode drm_crtc_init drm_crtc_init_with_planes drm_crtc_send_vblank_event drm_crtc_vblank_count drm_crtc_vblank_get drm_crtc_vblank_off drm_crtc_vblank_on drm_crtc_vblank_put drm_cvt_mode drm_dbg drm_debug drm_debugfs_create_files drm_detect_hdmi_monitor drm_detect_monitor_audio drm_dev_alloc drm_dev_put drm_dev_register drm_dev_unref drm_dev_unregister drm_dp_atomic_find_vcpi_slots drm_dp_atomic_release_vcpi_slots drm_dp_aux_register drm_dp_aux_unregister drm_dp_bw_code_to_link_rate drm_dp_calc_pbn_mode drm_dp_channel_eq_ok drm_dp_check_act_status drm_dp_clock_recovery_ok drm_dp_dpcd_read drm_dp_dpcd_read_link_status drm_dp_dpcd_write drm_dp_find_vcpi_slots drm_dp_get_adjust_request_pre_emphasis drm_dp_get_adjust_request_voltage drm_dp_link_rate_to_bw_code drm_dp_link_train_channel_eq_delay drm_dp_link_train_clock_recovery_delay drm_dp_mst_allocate_vcpi drm_dp_mst_deallocate_vcpi drm_dp_mst_detect_port drm_dp_mst_get_edid drm_dp_mst_hpd_irq drm_dp_mst_reset_vcpi_slots drm_dp_mst_topology_mgr_destroy drm_dp_mst_topology_mgr_init drm_dp_mst_topology_mgr_resume drm_dp_mst_topology_mgr_set_mst drm_dp_mst_topology_mgr_suspend drm_dp_update_payload_part1 drm_dp_update_payload_part2 drm_edid_header_is_valid drm_edid_is_valid drm_edid_to_sad drm_edid_to_speaker_allocation drm_encoder_cleanup drm_encoder_init drm_err drm_fb_helper_add_one_connector drm_fb_helper_alloc_fbi drm_fb_helper_blank drm_fb_helper_cfb_copyarea drm_fb_helper_cfb_fillrect drm_fb_helper_cfb_imageblit drm_fb_helper_check_var drm_fb_helper_debug_enter drm_fb_helper_debug_leave drm_fb_helper_fill_fix drm_fb_helper_fill_var drm_fb_helper_fini drm_fb_helper_hotplug_event drm_fb_helper_init drm_fb_helper_initial_config drm_fb_helper_ioctl drm_fb_helper_lastclose drm_fb_helper_output_poll_changed drm_fb_helper_pan_display drm_fb_helper_prepare drm_fb_helper_remove_one_connector drm_fb_helper_setcmap drm_fb_helper_set_par drm_fb_helper_set_suspend drm_fb_helper_set_suspend_unlocked drm_fb_helper_single_add_all_connectors drm_fb_helper_unregister_fbi drm_format_plane_cpp drm_framebuffer_cleanup drm_framebuffer_init drm_framebuffer_unregister_private drm_gem_dmabuf_kmap drm_gem_dmabuf_kunmap drm_gem_dmabuf_mmap drm_gem_dmabuf_release drm_gem_dmabuf_vmap drm_gem_dmabuf_vunmap drm_gem_fb_create_handle drm_gem_fb_destroy drm_gem_handle_create drm_gem_handle_delete drm_gem_map_attach drm_gem_map_detach drm_gem_map_dma_buf drm_gem_object_free drm_gem_object_init drm_gem_object_lookup drm_gem_object_put_unlocked drm_gem_object_release drm_gem_prime_export drm_gem_prime_fd_to_handle drm_gem_prime_handle_to_fd drm_gem_prime_import drm_gem_private_object_init drm_gem_unmap_dma_buf drm_get_edid drm_get_edid_switcheroo drm_get_format_name drm_get_max_iomem drm_global_item_ref drm_global_item_unref drm_handle_vblank drm_hdmi_avi_infoframe_from_display_mode drm_hdmi_vendor_infoframe_from_display_mode drm_helper_connector_dpms drm_helper_disable_unused_functions drm_helper_hpd_irq_event drm_helper_mode_fill_fb_struct drm_helper_probe_single_connector_modes drm_helper_resume_force_mode drm_i2c_encoder_detect drm_i2c_encoder_init drm_i2c_encoder_mode_fixup drm_i2c_encoder_restore drm_i2c_encoder_save drm_invalid_op drm_ioctl drm_irq_install drm_irq_uninstall drm_is_current_master drm_kms_helper_hotplug_event drm_kms_helper_is_poll_worker drm_kms_helper_poll_disable drm_kms_helper_poll_enable drm_kms_helper_poll_fini drm_kms_helper_poll_init drm_match_cea_mode drm_mm_init drm_mm_insert_node_in_range drm_mm_print drm_mm_remove_node drm_mm_takedown drm_mode_config_cleanup drm_mode_config_init drm_mode_config_reset drm_mode_copy drm_mode_create_dvi_i_properties drm_mode_create_scaling_mode_property drm_mode_create_tv_properties drm_mode_crtc_set_gamma_size drm_mode_debug_printmodeline drm_mode_destroy drm_mode_duplicate drm_mode_equal drm_mode_get_hv_timing drm_mode_is_420_only drm_mode_legacy_fb_format drm_mode_object_find drm_mode_object_put drm_mode_probed_add drm_mode_set_crtcinfo drm_modeset_lock drm_modeset_lock_all drm_modeset_lock_all_ctx drm_mode_set_name drm_modeset_unlock drm_modeset_unlock_all drm_mode_vrefresh drm_object_attach_property drm_object_property_set_value drm_open drm_plane_cleanup drm_plane_create_alpha_property drm_plane_create_color_properties drm_plane_create_zpos_immutable_property drm_plane_create_zpos_property drm_plane_force_disable drm_plane_init drm_poll drm_primary_helper_destroy drm_primary_helper_funcs drm_prime_gem_destroy drm_prime_pages_to_sg drm_prime_sg_to_page_addr_arrays drm_printf __drm_printfn_seq_file drm_property_add_enum drm_property_create drm_property_create_enum drm_property_create_range __drm_puts_seq_file drm_read drm_release drm_scdc_read drm_scdc_write drm_sched_dependency_optimized drm_sched_entity_destroy drm_sched_entity_fini drm_sched_entity_flush drm_sched_entity_init drm_sched_entity_push_job drm_sched_entity_set_rq drm_sched_fini drm_sched_hw_job_reset drm_sched_init drm_sched_job_init drm_sched_job_recovery drm_send_event_locked drm_syncobj_create drm_syncobj_find drm_syncobj_find_fence drm_syncobj_free drm_syncobj_get_fd drm_syncobj_get_handle drm_syncobj_replace_fence drm_universal_plane_init drm_vblank_init drm_vma_node_allow drm_vma_node_is_allowed drm_vma_node_revoke dst_release dump_stack __dynamic_dev_dbg __dynamic_netdev_dbg __dynamic_pr_debug efi elfcorehdr_addr emergency_restart empty_zero_page enable_irq errno_to_blk_status ether_setup eth_get_headlen eth_platform_get_mac_address ethtool_convert_legacy_u32_to_link_mode ethtool_convert_link_mode_to_legacy_u32 __ethtool_get_link_ksettings ethtool_intersect_link_masks ethtool_op_get_link ethtool_op_get_ts_info eth_type_trans eth_validate_addr event_triggers_call ex_handler_default ex_handler_refcount fasync_helper fc_attach_transport fc_block_scsi_eh fc_disc_config fc_disc_init fc_eh_host_reset fc_eh_timed_out fc_elsct_init fc_elsct_send fc_exch_init fc_exch_mgr_alloc fc_exch_mgr_free fc_exch_mgr_list_clone fc_exch_recv fc_fabric_login fc_fabric_logoff _fc_frame_alloc fc_frame_alloc_fill fc_get_event_number fc_get_host_port_state fc_get_host_speed fc_get_host_stats fc_host_post_event fc_host_post_vendor_event fc_lport_bsg_request fc_lport_config fc_lport_destroy fc_lport_flogi_resp fc_lport_init fc_lport_logo_resp fc_lport_reset fcoe_check_wait_queue fcoe_clean_pending_queue fcoe_ctlr_destroy fcoe_ctlr_device_add fcoe_ctlr_device_delete fcoe_ctlr_els_send fcoe_ctlr_get_lesb fcoe_ctlr_init fcoe_ctlr_link_down fcoe_ctlr_link_up fcoe_ctlr_recv fcoe_ctlr_recv_flogi fcoe_fc_crc fcoe_fcf_get_selected fcoe_get_lesb fcoe_get_paged_crc_eof fcoe_get_wwn fcoe_link_speed_update fcoe_queue_timer fcoe_start_io fcoe_transport_attach fcoe_transport_detach fcoe_validate_vport_create fcoe_wwn_from_mac fcoe_wwn_to_str fc_release_transport fc_remote_port_add fc_remote_port_delete fc_remote_port_rolechg fc_remove_host fc_rport_terminate_io fc_set_mfs fc_set_rport_loss_tmo fc_slave_alloc fc_vport_create fc_vport_id_lookup fc_vport_setlink fc_vport_terminate __fdget fd_install __fentry__ fget __fib_lookup fib_table_lookup filemap_fault filp_close filp_open find_first_bit find_first_zero_bit find_get_pid find_last_bit find_next_bit find_next_zero_bit find_pid_ns find_vma finish_wait firmware_request_nowarn fixed_size_llseek flow_keys_dissector flush_delayed_work flush_signals flush_work flush_workqueue follow_pfn force_sig fortify_panic fput free_cpumask_var free_fib_info free_irq free_irq_cpu_rmap free_netdev __free_pages free_pages free_percpu from_kgid from_kuid fs_bio_set gcd generate_random_uuid generic_end_io_acct generic_handle_irq generic_make_request generic_start_io_acct genlmsg_put genl_register_family genl_unregister_family get_device __get_free_pages get_gendisk get_pid_task get_random_bytes __get_task_comm get_task_mm get_task_pid get_unused_fd_flags __get_user_2 __get_user_4 __get_user_8 get_user_pages get_user_pages_remote get_zeroed_page groups_alloc groups_free handle_simple_irq hdmi_avi_infoframe_pack hdmi_infoframe_pack hrtimer_cancel hrtimer_forward hrtimer_init hrtimer_start_range_ns hrtimer_try_to_cancel __hw_addr_sync_dev __hw_addr_unsync_dev hwmon_device_register hwmon_device_register_with_groups hwmon_device_register_with_info hwmon_device_unregister i2c_add_adapter i2c_bit_add_bus i2c_bit_algo i2c_del_adapter i2c_new_device i2c_smbus_read_byte_data i2c_smbus_write_byte_data i2c_transfer i2c_unregister_device __ib_alloc_cq ib_alloc_device ib_alloc_odp_umem __ib_alloc_pd ib_attach_mcast ib_cache_gid_parse_type_str ib_cache_gid_type_str ib_cancel_mad ib_cm_init_qp_attr ib_cm_insert_listen ib_cm_listen ib_cm_notify ibcm_reject_msg ib_copy_path_rec_from_user ib_copy_path_rec_to_user ib_copy_qp_attr_to_user ib_create_ah_from_wc ib_create_cm_id __ib_create_cq ib_create_qp ib_create_qp_security ib_create_send_mad ib_dealloc_device ib_dealloc_pd ib_dealloc_xrcd ib_destroy_cm_id ib_destroy_cq ib_destroy_qp ib_destroy_rwq_ind_table ib_destroy_wq ib_detach_mcast ib_dispatch_event ib_find_cached_pkey ib_free_cq ib_free_recv_mad ib_free_send_mad ib_get_cached_pkey ib_get_cached_port_state ib_get_eth_speed ib_get_gids_from_rdma_hdr ib_get_mad_data_offset ib_get_net_dev_by_params ib_get_rdma_header_version ib_get_rmpp_segment ib_init_ah_attr_from_path ib_init_ah_attr_from_wc ib_init_ah_from_mcmember ib_is_mad_class_rmpp ib_mad_kernel_rmpp_agent ib_modify_mad ib_modify_port ib_modify_qp ib_modify_qp_is_ok ib_modify_qp_with_udata ibnl_put_attr ibnl_put_msg ib_open_qp ib_post_send_mad ib_process_cq_direct ib_query_pkey ib_query_port ib_query_qp ib_query_srq ib_rdmacg_try_charge ib_rdmacg_uncharge ib_register_client ib_register_device ib_register_event_handler ib_register_mad_agent ib_response_mad ib_sa_cancel_query ib_sa_free_multicast ib_sa_get_mcmember_rec ib_sa_guid_info_rec_query ib_sa_join_multicast ib_sa_path_rec_get ib_sa_register_client ib_sa_sendonly_fullmem_support ib_sa_unregister_client ib_send_cm_apr ib_send_cm_drep ib_send_cm_dreq ib_send_cm_lap ib_send_cm_mra ib_send_cm_rej ib_send_cm_rep ib_send_cm_req ib_send_cm_rtu ib_send_cm_sidr_rep ib_send_cm_sidr_req ib_set_client_data ib_sg_to_pages ib_ud_header_init ib_ud_header_pack ib_ud_ip4_csum ib_umem_copy_from ib_umem_get ib_umem_odp_map_dma_pages ib_umem_odp_unmap_dma_pages ib_umem_page_count ib_umem_release ib_unregister_client ib_unregister_device ib_unregister_event_handler ib_unregister_mad_agent ib_uverbs_get_ucontext ib_wc_status_msg ida_alloc_range ida_destroy ida_free idr_alloc idr_alloc_cyclic idr_alloc_u32 idr_destroy idr_find idr_for_each idr_get_next idr_get_next_ul idr_preload idr_remove idr_replace igrab in6_dev_finish_destroy in_dev_finish_destroy in_egroup_p __inet6_lookup_established inet_get_local_port_range __inet_lookup_established in_group_p init_net __init_rwsem init_srcu_struct init_task init_timer_key init_uts_ns init_wait_entry __init_waitqueue_head interval_tree_insert interval_tree_iter_first interval_tree_iter_next interval_tree_remove int_to_scsilun invalidate_partition iomem_resource iommu_get_domain_for_dev iommu_iova_to_phys iommu_map iommu_unmap ioread16 ioread16be ioread32 ioread32be ioread8 ioremap_cache ioremap_nocache ioremap_wc io_schedule io_schedule_timeout iounmap iowrite16 iowrite32 iowrite32be __iowrite32_copy __iowrite64_copy iowrite8 ip6_dst_hoplimit ip_compute_csum ip_mc_dec_group ip_mc_inc_group ipmi_create_user ipmi_destroy_user ipmi_free_recv_msg ipmi_poll_interface ipmi_request_settime ipmi_set_gets_events ipmi_set_my_address ipmi_smi_msg_received ipmi_unregister_smi ipmi_validate_addr ip_route_output_flow ip_send_check ip_tos2prio iput __ipv6_addr_type ipv6_ext_hdr ipv6_find_hdr ipv6_skip_exthdr ipv6_stub irq_cpu_rmap_add irq_create_mapping __irq_domain_add irq_domain_remove irq_find_mapping irq_modify_status irq_poll_complete irq_poll_disable irq_poll_enable irq_poll_init irq_poll_sched irq_set_affinity_hint irq_set_affinity_notifier irq_set_chip_and_handler_name irq_to_desc is_acpi_device_node iscsi_boot_create_ethernet iscsi_boot_create_host_kset iscsi_boot_create_initiator iscsi_boot_create_target iscsi_boot_destroy_kset __iscsi_complete_pdu iscsi_complete_scsi_task iscsi_conn_bind iscsi_conn_failure iscsi_conn_get_param iscsi_conn_send_pdu iscsi_conn_setup iscsi_conn_start iscsi_conn_stop iscsi_conn_teardown iscsi_create_endpoint iscsi_create_iface iscsi_destroy_endpoint iscsi_destroy_iface iscsi_eh_abort iscsi_eh_device_reset iscsi_eh_recover_target iscsi_eh_session_reset iscsi_get_port_speed_name iscsi_get_port_state_name __iscsi_get_task iscsi_host_add iscsi_host_alloc iscsi_host_for_each_session iscsi_host_free iscsi_host_get_param iscsi_host_remove iscsi_itt_to_task iscsi_lookup_endpoint iscsi_offload_mesg __iscsi_put_task iscsi_put_task iscsi_queuecommand iscsi_register_transport iscsi_session_failure iscsi_session_get_param iscsi_session_recovery_timedout iscsi_session_setup iscsi_session_teardown iscsi_set_param iscsi_suspend_queue iscsi_target_alloc iscsi_unregister_transport is_uv_system iw_cm_accept iw_cm_connect iw_cm_disconnect iw_cm_init_qp_attr iw_cm_listen iw_cm_reject iwcm_reject_msg iw_create_cm_id iw_destroy_cm_id jiffies jiffies_64 jiffies_to_msecs jiffies_to_timespec64 jiffies_to_usecs kallsyms_lookup_name kasprintf kernel_fpu_begin kernel_fpu_end kernel_recvmsg kernel_sendmsg kernel_setsockopt kfree kfree_call_rcu kfree_const kfree_skb kgdb_active kgdb_breakpoint kill_fasync __kmalloc kmalloc_caches __kmalloc_node kmalloc_order_trace kmem_cache_alloc kmem_cache_alloc_node kmem_cache_alloc_node_trace kmem_cache_alloc_trace kmem_cache_create kmem_cache_create_usercopy kmem_cache_destroy kmem_cache_free kmem_cache_shrink kmemdup kobject_add kobject_create_and_add kobject_del kobject_get kobject_init kobject_init_and_add kobject_put kobject_set_name kobject_uevent kobject_uevent_env krealloc kset_create_and_add kset_find_obj kset_register kset_unregister ksize kstrdup kstrdup_const kstrndup kstrtobool kstrtobool_from_user kstrtoint kstrtoint_from_user kstrtoll kstrtoll_from_user kstrtou16 kstrtou8 kstrtouint kstrtouint_from_user kstrtoul_from_user kstrtoull kstrtoull_from_user kthread_bind kthread_create_on_node kthread_park kthread_should_stop kthread_stop kthread_unpark ktime_get ktime_get_coarse_real_ts64 ktime_get_raw ktime_get_raw_ts64 ktime_get_real_seconds ktime_get_real_ts64 ktime_get_seconds ktime_get_ts64 ktime_get_with_offset kvasprintf kvfree kvmalloc_node kzfree led_classdev_resume led_classdev_suspend led_classdev_unregister libfc_vport_create __list_add_valid __list_del_entry_valid llist_add_batch __local_bh_enable_ip __lock_page lock_page_memcg lockref_get lock_sock_nested make_kgid make_kuid mark_page_accessed match_strdup match_string match_token mdio45_probe mdiobus_alloc_size mdiobus_free mdiobus_get_phy __mdiobus_register mdiobus_unregister mdio_mii_ioctl memchr memchr_inv memcmp memcpy memdup_user memdup_user_nul memmove memory_read_from_buffer memparse mempool_alloc mempool_alloc_slab mempool_create mempool_create_node mempool_destroy mempool_free mempool_free_slab mempool_kfree mempool_kmalloc memscan mem_section memset memzero_explicit metadata_dst_alloc mfd_add_devices mfd_remove_devices misc_deregister misc_register mlxfw_firmware_flash __mmdrop mmput __mmu_notifier_register mmu_notifier_register mmu_notifier_unregister mmu_notifier_unregister_no_release mod_delayed_work_on mod_timer mod_timer_pending __module_get module_layout module_put module_refcount __msecs_to_jiffies msleep msleep_interruptible __mutex_init mutex_lock mutex_lock_interruptible mutex_lock_killable mutex_trylock mutex_unlock mxm_wmi_call_mxds mxm_wmi_call_mxmx mxm_wmi_supported __napi_alloc_skb napi_complete_done napi_consume_skb napi_disable napi_get_frags napi_gro_flush napi_gro_frags napi_gro_receive napi_hash_del __napi_schedule __napi_schedule_irqoff napi_schedule_prep __ndelay ndo_dflt_bridge_getlink ndo_dflt_fdb_add nd_tbl neigh_destroy __neigh_event_send neigh_lookup netdev_alloc_frag __netdev_alloc_skb netdev_bind_sb_channel_queue netdev_crit netdev_err netdev_features_change netdev_info netdev_lower_get_next netdev_master_upper_dev_get netdev_master_upper_dev_get_rcu netdev_notice netdev_printk netdev_reset_tc netdev_rss_key_fill netdev_rx_handler_register netdev_rx_handler_unregister netdev_set_num_tc netdev_set_sb_channel netdev_set_tc_queue netdev_unbind_sb_channel netdev_update_features netdev_walk_all_upper_dev_rcu netdev_warn netif_carrier_off netif_carrier_on netif_device_attach netif_device_detach netif_get_num_default_rss_queues netif_napi_add netif_napi_del netif_receive_skb netif_rx netif_schedule_queue netif_set_real_num_rx_queues netif_set_real_num_tx_queues netif_set_xps_queue netif_tx_stop_all_queues netif_tx_wake_queue netlink_broadcast netlink_unicast net_ratelimit nla_find nla_parse nla_put nla_validate node_data __node_distance node_states node_to_cpumask_map no_llseek nonseekable_open noop_llseek nr_cpu_ids nr_node_ids nsecs_to_jiffies ns_to_timespec ns_to_timespec64 ns_to_timeval numa_node nvme_alloc_request nvme_cancel_request nvme_change_ctrl_state nvme_cleanup_cmd nvme_complete_async_event nvme_complete_rq nvme_disable_ctrl nvme_enable_ctrl nvme_fc_register_localport nvme_fc_register_remoteport nvme_fc_set_remoteport_devloss nvme_fc_unregister_localport nvme_fc_unregister_remoteport nvme_init_ctrl nvme_init_identify nvme_io_timeout nvme_kill_queues nvme_remove_namespaces nvme_reset_ctrl nvme_reset_ctrl_sync nvme_set_queue_count nvme_setup_cmd nvme_shutdown_ctrl nvme_start_ctrl nvme_start_freeze nvme_start_queues nvme_stop_ctrl nvme_stop_queues nvme_submit_sync_cmd nvmet_fc_rcv_fcp_abort nvmet_fc_rcv_fcp_req nvmet_fc_rcv_ls_req nvmet_fc_register_targetport nvmet_fc_unregister_targetport nvme_unfreeze nvme_uninit_ctrl nvme_wait_freeze nvme_wait_freeze_timeout nvme_wq of_led_classdev_register on_each_cpu orderly_poweroff out_of_line_wait_on_bit out_of_line_wait_on_bit_lock override_creds __page_file_index __page_frag_cache_drain page_frag_free __page_mapcount page_mapped page_offset_base page_pool_alloc_pages page_pool_create page_pool_destroy __page_pool_put_page pagevec_lookup_range pagevec_lookup_range_tag __pagevec_release panic panic_notifier_list param_array_ops param_get_int param_ops_bool param_ops_byte param_ops_charp param_ops_int param_ops_long param_ops_short param_ops_string param_ops_uint param_ops_ullong param_ops_ulong param_ops_ushort param_set_bool param_set_int pat_enabled pci_alloc_irq_vectors_affinity pci_assign_unassigned_bus_resources pcibios_resource_to_bus pci_bus_resource_n pci_bus_type pci_cfg_access_lock pci_cfg_access_unlock pci_choose_state pci_cleanup_aer_uncorrect_error_status pci_clear_master pci_clear_mwi pci_d3cold_disable pci_dev_driver pci_dev_get pci_device_is_present pci_dev_present pci_dev_put pci_disable_device pci_disable_link_state pci_disable_msi pci_disable_msix pci_disable_pcie_error_reporting pci_disable_rom pci_disable_sriov pcie_bandwidth_available pcie_capability_clear_and_set_word pcie_capability_read_dword pcie_capability_read_word pcie_capability_write_word pcie_flr pcie_get_speed_cap pcie_get_width_cap pci_enable_atomic_ops_to_root pci_enable_device pci_enable_device_mem pci_enable_msi pci_enable_msix_range pci_enable_pcie_error_reporting pci_enable_rom pci_enable_sriov pci_enable_wake pcie_print_link_status pcie_relaxed_ordering_enabled pcie_set_readrq pci_find_capability pci_find_ext_capability pci_free_irq pci_free_irq_vectors pci_get_class pci_get_device pci_get_domain_bus_and_slot pci_get_slot pci_ignore_hotplug pci_intx pci_iomap pci_ioremap_bar pci_iounmap pci_irq_get_affinity pci_irq_vector pci_map_rom pcim_enable_device pcim_iomap_regions pcim_iomap_table pci_num_vf pci_platform_rom pci_prepare_to_sleep pci_read_config_byte pci_read_config_dword pci_read_config_word pci_read_vpd __pci_register_driver pci_release_regions pci_release_resource pci_release_selected_regions pci_request_irq pci_request_regions pci_request_selected_regions pci_rescan_bus pci_resize_resource pci_restore_state pci_save_state pci_select_bars pci_set_master pci_set_mwi pci_set_power_state pci_sriov_configure_simple pci_sriov_get_totalvfs pci_sriov_set_totalvfs pci_stop_and_remove_bus_device pci_stop_and_remove_bus_device_locked pci_try_set_mwi pci_unmap_rom pci_unregister_driver pci_vfs_assigned pci_vpd_find_info_keyword pci_vpd_find_tag pci_wait_for_pending_transaction pci_wake_from_d3 pci_walk_bus pci_write_config_byte pci_write_config_dword pci_write_config_word pcix_set_mmrbc PDE_DATA __per_cpu_offset perf_tp_event perf_trace_buf_alloc perf_trace_run_bpf_submit pgprot_writecombine phy_attached_info phy_connect phy_disconnect phy_ethtool_ksettings_get phy_ethtool_ksettings_set phy_ethtool_sset phy_mii_ioctl phys_base physical_mask phy_start phy_start_aneg phy_stop pid_task pid_vnr platform_bus_type pm_genpd_add_device pm_genpd_init pm_genpd_remove_device pm_power_off pm_runtime_allow __pm_runtime_disable pm_runtime_enable pm_runtime_forbid __pm_runtime_idle __pm_runtime_resume pm_runtime_set_autosuspend_delay __pm_runtime_set_status __pm_runtime_suspend __pm_runtime_use_autosuspend pm_schedule_suspend pm_vt_switch_required pm_vt_switch_unregister power_supply_is_system_supplied prandom_bytes prandom_seed prandom_u32 __preempt_count prepare_creds prepare_to_wait prepare_to_wait_event prepare_to_wait_exclusive print_hex_dump printk __printk_ratelimit print_stack_trace proc_create_data proc_dointvec proc_mkdir proc_mkdir_mode proc_remove proc_symlink __pskb_copy_fclone pskb_expand_head __pskb_pull_tail ___pskb_trim ptp_clock_event ptp_clock_index ptp_clock_register ptp_clock_unregister ptp_find_pin __put_cred put_device __put_devmap_managed_page put_disk __put_net __put_page put_pid __put_task_struct put_unused_fd __put_user_1 __put_user_2 __put_user_4 __put_user_8 pv_cpu_ops pv_irq_ops pv_lock_ops pv_mmu_ops qed_get_eth_ops qed_put_eth_ops queue_delayed_work_on queued_read_lock_slowpath queued_write_lock_slowpath queue_work_on radix_tree_delete radix_tree_gang_lookup __radix_tree_insert radix_tree_iter_delete radix_tree_lookup radix_tree_lookup_slot radix_tree_next_chunk __radix_tree_next_slot radix_tree_preload radix_tree_tagged raid_class_attach raid_class_release ___ratelimit raw_notifier_call_chain raw_notifier_chain_register raw_notifier_chain_unregister _raw_read_lock _raw_read_lock_bh _raw_read_lock_irq _raw_read_lock_irqsave _raw_read_unlock_bh _raw_read_unlock_irqrestore _raw_spin_lock _raw_spin_lock_bh _raw_spin_lock_irq _raw_spin_lock_irqsave _raw_spin_trylock _raw_spin_unlock_bh _raw_spin_unlock_irqrestore _raw_write_lock _raw_write_lock_bh _raw_write_lock_irq _raw_write_lock_irqsave _raw_write_unlock_bh _raw_write_unlock_irqrestore rb_erase __rb_erase_color rb_first rb_first_postorder __rb_insert_augmented rb_insert_color rb_next rb_next_postorder rb_replace_node rbt_ib_umem_for_each_in_range rbt_ib_umem_lookup rcu_barrier rdma_addr_cancel rdma_addr_size rdma_create_ah rdma_create_user_ah rdma_destroy_ah rdma_destroy_ah_attr rdma_find_gid rdma_find_gid_by_port rdma_get_gid_attr rdma_is_zero_gid rdma_move_ah_attr rdma_nl_multicast rdma_nl_register rdma_nl_unicast rdma_nl_unicast_wait rdma_nl_unregister rdma_node_get_transport rdma_port_get_link_layer rdma_put_gid_attr rdma_query_ah rdma_query_gid rdma_resolve_ip rdma_restrack_del rdma_roce_rescan_device rdma_set_cq_moderation rdma_translate_ip read_cache_pages recalc_sigpending refcount_dec_and_mutex_lock refcount_dec_and_test_checked refcount_inc_checked refcount_inc_not_zero_checked register_acpi_notifier register_blkdev __register_chrdev register_chrdev_region register_fib_notifier register_inet6addr_notifier register_inetaddr_notifier register_netdev register_netdevice_notifier register_netevent_notifier register_net_sysctl __register_nmi_handler register_pernet_subsys register_reboot_notifier release_firmware release_pages __release_region release_sock remap_pfn_range remove_conflicting_framebuffers remove_proc_entry remove_wait_queue request_firmware request_firmware_direct request_firmware_nowait __request_module __request_region request_threaded_irq reservation_object_add_excl_fence reservation_object_add_shared_fence reservation_object_get_fences_rcu reservation_object_reserve_shared reservation_object_wait_timeout_rcu reservation_ww_class reset_devices revert_creds rhashtable_destroy rhashtable_free_and_destroy rhashtable_init rhashtable_insert_slow rhashtable_walk_enter rhashtable_walk_exit rhashtable_walk_next rhashtable_walk_start_check rhashtable_walk_stop rhltable_init rht_bucket_nested rht_bucket_nested_insert ring_buffer_event_data roce_gid_type_mask_support round_jiffies round_jiffies_relative rps_may_expire_flow rt6_lookup rtc_time64_to_tm rtnl_is_locked rtnl_lock rtnl_trylock rtnl_unlock sas_attach_transport sas_disable_tlr sas_enable_tlr sas_end_device_alloc sas_expander_alloc sas_is_tlr_enabled sas_phy_add sas_phy_alloc sas_phy_free sas_port_add sas_port_add_phy sas_port_alloc_num sas_port_delete sas_port_delete_phy sas_port_free sas_read_port_mode_page sas_release_transport sas_remove_host sas_rphy_add save_stack_trace save_stack_trace_tsk sbitmap_queue_clear __sbitmap_queue_get sched_setscheduler schedule schedule_hrtimeout schedule_hrtimeout_range schedule_timeout schedule_timeout_interruptible schedule_timeout_uninterruptible scmd_printk scnprintf screen_info scsi_add_device scsi_add_host_with_dma scsi_block_requests scsi_build_sense_buffer scsi_change_queue_depth scsi_device_get scsi_device_lookup scsi_device_put scsi_device_set_state scsi_device_type scsi_dma_map scsi_dma_unmap __scsi_execute scsi_get_vpd_page scsi_host_alloc scsi_host_busy scsi_host_get scsi_host_lookup scsi_host_put scsi_internal_device_block_nowait scsi_internal_device_unblock_nowait scsi_is_fc_rport scsi_is_host_device scsi_is_sdev_device __scsi_iterate_devices scsilun_to_int scsi_normalize_sense scsi_print_command scsi_register_driver scsi_remove_device scsi_remove_host scsi_remove_target scsi_sanitize_inquiry_string scsi_scan_host scsi_track_queue_full scsi_unblock_requests sdev_prefix_printk secpath_dup security_d_instantiate send_sig seq_lseek seq_open seq_printf seq_putc seq_puts seq_read seq_release seq_write set_cpus_allowed_ptr set_current_groups set_device_ro set_disk_ro set_freezable set_memory_array_uc set_memory_array_wb set_memory_uc set_memory_wb set_memory_wc set_normalized_timespec set_normalized_timespec64 set_page_dirty set_page_dirty_lock set_user_nice sg_alloc_table_from_pages sg_copy_from_buffer sg_copy_to_buffer sg_free_table sg_init_table sg_miter_next sg_miter_start sg_miter_stop sg_next show_class_attr_string sigprocmask si_meminfo simple_open simple_read_from_buffer simple_strtol simple_strtoul simple_strtoull simple_write_to_buffer single_open single_release skb_add_rx_frag skb_checksum skb_checksum_help skb_clone skb_clone_tx_timestamp skb_copy skb_copy_bits skb_copy_expand skb_dequeue __skb_flow_dissect __skb_get_hash __skb_gso_segment skb_gso_validate_mac_len __skb_pad skb_pull skb_push skb_put skb_queue_purge skb_queue_tail skb_realloc_headroom skb_store_bits skb_trim skb_tstamp_tx skb_vlan_pop sme_active sme_me_mask smp_call_function_many smp_call_function_single snprintf sn_rtc_cycles_per_second sock_create_kern sock_edemux sock_release softnet_data sort sprintf __srcu_read_lock __srcu_read_unlock sscanf __stack_chk_fail starget_for_each_device strcasecmp strcat strchr strcmp strcpy strcspn strim strlcat strlcpy strlen strncasecmp strncat strncmp strncpy strncpy_from_user strnlen strnstr strpbrk strrchr strscpy strsep strspn strstr submit_bio __sw_hweight32 __sw_hweight64 swiotlb_nr_tbl switchdev_port_same_parent_id __symbol_get __symbol_put sync_file_create synchronize_irq synchronize_net synchronize_sched synchronize_srcu sysfs_add_file_to_group sysfs_create_bin_file sysfs_create_file_ns sysfs_create_group sysfs_format_mac sysfs_remove_bin_file sysfs_remove_file_from_group sysfs_remove_file_ns sysfs_remove_group sysfs_streq system_state system_unbound_wq system_wq sys_tz task_active_pid_ns tasklet_init tasklet_kill __tasklet_schedule __task_pid_nr_ns tcf_block_cb_register tcf_block_cb_unregister tcp_gro_complete tcp_hashinfo tc_setup_cb_egdev_register tc_setup_cb_egdev_unregister this_cpu_off time64_to_tm timecounter_cyc2time timecounter_init timecounter_read to_drm_sched_fence trace_define_field trace_event_buffer_commit trace_event_buffer_lock_reserve trace_event_buffer_reserve trace_event_ignore_this_pid trace_event_raw_init trace_event_reg trace_handle_return __tracepoint_dma_fence_emit __tracepoint_xdp_exception trace_print_flags_seq trace_raw_output_prep trace_seq_printf trace_seq_putc try_module_get try_wait_for_completion tsc_khz ttm_bo_add_to_lru ttm_bo_clean_mm ttm_bo_del_sub_from_lru ttm_bo_device_init ttm_bo_device_release ttm_bo_dma_acc_size ttm_bo_eviction_valuable ttm_bo_evict_mm ttm_bo_global_init ttm_bo_global_release ttm_bo_init ttm_bo_init_mm ttm_bo_init_reserved ttm_bo_kmap ttm_bo_kunmap ttm_bo_lock_delayed_workqueue ttm_bo_manager_func ttm_bo_mem_put ttm_bo_mem_space ttm_bo_mmap ttm_bo_move_accel_cleanup ttm_bo_move_memcpy ttm_bo_move_to_lru_tail ttm_bo_move_ttm ttm_bo_pipeline_move ttm_bo_put ttm_bo_unlock_delayed_workqueue ttm_bo_validate ttm_bo_wait ttm_dma_page_alloc_debugfs ttm_dma_populate ttm_dma_tt_fini ttm_dma_tt_init ttm_dma_unpopulate ttm_eu_backoff_reservation ttm_eu_fence_buffer_objects ttm_eu_reserve_buffers ttm_fbdev_mmap ttm_mem_global_init ttm_mem_global_release ttm_page_alloc_debugfs ttm_pool_populate ttm_pool_unpopulate ttm_populate_and_map_pages ttm_sg_tt_init ttm_tt_bind ttm_tt_set_placement_caching ttm_unmap_and_unpopulate_pages __udelay udp4_hwcsum uio_event_notify __uio_register_device uio_unregister_device unlock_page unlock_page_memcg unmap_mapping_range unregister_acpi_notifier unregister_blkdev __unregister_chrdev unregister_chrdev_region unregister_fib_notifier unregister_inet6addr_notifier unregister_inetaddr_notifier unregister_netdev unregister_netdevice_notifier unregister_netevent_notifier unregister_net_sysctl_table unregister_nmi_handler unregister_pernet_subsys unregister_reboot_notifier unuse_mm up up_read up_write __usecs_to_jiffies use_mm usleep_range __uv_cpu_info _uverbs_alloc uverbs_copy_to uverbs_destroy_def_handler uverbs_get_flags32 uverbs_get_flags64 uverbs_idr_class __uv_hub_info_list uv_possible_blades uv_setup_irq uv_teardown_irq vfree vga_client_register vgacon_text_force vga_set_legacy_decoding vga_switcheroo_client_fb_set vga_switcheroo_client_probe_defer vga_switcheroo_fini_domain_pm_ops vga_switcheroo_handler_flags vga_switcheroo_init_domain_pm_ops vga_switcheroo_lock_ddc vga_switcheroo_process_delayed_switch vga_switcheroo_register_client vga_switcheroo_register_handler vga_switcheroo_unlock_ddc vga_switcheroo_unregister_client vga_switcheroo_unregister_handler vga_tryget __virt_addr_valid vlan_dev_real_dev vlan_dev_vlan_id vlan_dev_vlan_proto vmalloc __vmalloc vmalloc_base vmalloc_node vmalloc_to_page vmap vmemmap_base vm_get_page_prot vm_insert_page vm_mmap vm_munmap vprintk vscnprintf vsnprintf vsprintf vunmap vzalloc vzalloc_node wait_for_completion wait_for_completion_interruptible wait_for_completion_interruptible_timeout wait_for_completion_io_timeout wait_for_completion_killable wait_for_completion_timeout wait_on_page_bit __wake_up wake_up_bit __wake_up_locked wake_up_process __warn_printk wmi_evaluate_method wmi_has_guid work_busy write_cache_pages ww_mutex_lock ww_mutex_lock_interruptible ww_mutex_unlock x86_cpu_to_apicid x86_dma_fallback_dev __x86_indirect_thunk_r10 __x86_indirect_thunk_r11 __x86_indirect_thunk_r12 __x86_indirect_thunk_r13 __x86_indirect_thunk_r14 __x86_indirect_thunk_r15 __x86_indirect_thunk_r8 __x86_indirect_thunk_r9 __x86_indirect_thunk_rax __x86_indirect_thunk_rbp __x86_indirect_thunk_rbx __x86_indirect_thunk_rcx __x86_indirect_thunk_rdi __x86_indirect_thunk_rdx __x86_indirect_thunk_rsi xdp_do_flush_map xdp_do_redirect xdp_return_frame xdp_return_frame_rx_napi xdp_rxq_info_is_reg xdp_rxq_info_reg xdp_rxq_info_reg_mem_model xdp_rxq_info_unreg xdp_rxq_info_unused xfrm_replay_seqhi xz_dec_end xz_dec_init xz_dec_run yield zalloc_cpumask_var zap_vma_ptes zlib_inflate zlib_inflateEnd zlib_inflateInit2 zlib_inflate_workspacesize

2 2

来自严亮的邮件
by 严亮 04 Mar '21

04 Mar '21

yanl1229(a)163.com

1 0

[PATCH openEuler-21.03 1/2] mm: add pin memory method for checkpoint add restore
by hejingxian 02 Mar '21

02 Mar '21

From: Jingxian He <hejingxian(a)huawei.com> Date: Mon, 1 Mar 2021 17:35:32 +0800 Subject: [PATCH openEuler-21.03 1/2] mm: add pin memory method for checkpoint add restore hulk inclusion category: feature bugzilla: 48159 CVE: N/A 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). By pin the memory data of tasks and collect the corresponding physical pages mapping info in checkpoint process, we can remap the physical pages to restore tasks after upgrading the kernel. This pin memory method can restore the task data within one second. The pin memory area info is saved in the reserved memblock, 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> Reviewed-by: Wenliang He <hewenliang4(a)huawei.com> Reviewed-by: Jing Xiangfeng <jingxiangfeng(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 2 + arch/arm64/kernel/setup.c | 9 + arch/arm64/mm/init.c | 60 +++ drivers/char/Kconfig | 6 + drivers/char/Makefile | 1 + drivers/char/pin_memory.c | 208 ++++++++ include/linux/crash_core.h | 5 + include/linux/pin_mem.h | 78 +++ kernel/crash_core.c | 11 + mm/Kconfig | 8 + mm/Makefile | 1 + mm/huge_memory.c | 61 +++ mm/memory.c | 59 ++ mm/pin_mem.c | 950 +++++++++++++++++++++++++++++++++ 14 files changed, 1459 insertions(+) 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/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index c5271e7..76fda68 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -1036,6 +1036,7 @@ CONFIG_FRAME_VECTOR=y # CONFIG_GUP_BENCHMARK is not set # CONFIG_READ_ONLY_THP_FOR_FS is not set CONFIG_ARCH_HAS_PTE_SPECIAL=y +CONFIG_PIN_MEMORY=y # end of Memory Management options CONFIG_NET=y @@ -3282,6 +3283,7 @@ CONFIG_TCG_TIS_ST33ZP24_SPI=y # CONFIG_RANDOM_TRUST_CPU is not set # CONFIG_RANDOM_TRUST_BOOTLOADER is not set +CONFIG_PIN_MEMORY_DEV=m # # I2C support diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index c1f1fb9..5e282d3 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_mem.h> +#endif static int num_standard_resources; static struct resource *standard_resources; @@ -260,6 +263,12 @@ static void __init request_standard_resources(void) quick_kexec_res.end <= res->end) request_resource(res, &quick_kexec_res); #endif +#ifdef CONFIG_PIN_MEMORY + if (pin_memory_resource.end && + pin_memory_resource.start >= res->start && + pin_memory_resource.end <= res->end) + request_resource(res, &pin_memory_resource); +#endif } } diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index f3e5a66..8ab5aac 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -42,6 +42,9 @@ #include <linux/sizes.h> #include <asm/tlb.h> #include <asm/alternative.h> +#ifdef CONFIG_PIN_MEMORY +#include <linux/pin_mem.h> +#endif #define ARM64_ZONE_DMA_BITS 30 @@ -78,6 +81,55 @@ static void __init reserve_crashkernel(void) */ #define MAX_USABLE_RANGES 2 +#ifdef CONFIG_PIN_MEMORY +struct resource pin_memory_resource = { + .name = "Pin memory", + .start = 0, + .end = 0, + .flags = IORESOURCE_MEM, + .desc = IORES_DESC_RESERVED +}; + +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) @@ -455,6 +507,8 @@ void __init arm64_memblock_init(void) reserve_park_mem(); #endif + reserve_pin_memory_res(); + reserve_elfcorehdr(); high_memory = __va(memblock_end_of_DRAM() - 1) + 1; @@ -583,6 +637,12 @@ 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 + 1)); +#endif + mem_init_print_info(NULL); /* diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index d229a2d..fbb94b8 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -496,3 +496,9 @@ 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" + default m + help + pin memory driver diff --git a/drivers/char/Makefile b/drivers/char/Makefile index ffce287..71d76fd 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -47,3 +47,4 @@ obj-$(CONFIG_PS3_FLASH) += ps3flash.o obj-$(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 0000000..f46e056 --- /dev/null +++ b/drivers/char/pin_memory.c @@ -0,0 +1,208 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2021. Huawei Technologies Co., Ltd. All rights reserved. + * Pin memory driver for checkpoint and restore. + */ +#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 <linux/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 _FINISH_PIN_MEM_DUMP 4 +#define _PIN_MEM_IOC_MAX_NR 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 FINISH_PIN_MEM_DUMP _IOW(PIN_MEM_MAGIC, _FINISH_PIN_MEM_DUMP, 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); + return -EFAULT; + } + 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(); + put_pid(pid_s); + return ret; + +fail: + rcu_read_unlock(); + put_pid(pid_s); + 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(); + put_pid(pid_s); + return 0; + +fault: + rcu_read_unlock(); + put_pid(pid_s); + return -EFAULT; +} + +static long pin_memory_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + long ret = 0; + + if (_IOC_TYPE(cmd) != PIN_MEM_MAGIC) + return -EINVAL; + if (_IOC_NR(cmd) > _PIN_MEM_IOC_MAX_NR) + 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; + case FINISH_PIN_MEM_DUMP: + ret = finish_pin_mem_dump(); + 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 fc0ef33..30f0df3 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -87,4 +87,9 @@ 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 0000000..bc8b03e --- /dev/null +++ b/include/linux/pin_mem.h @@ -0,0 +1,78 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (C) 2021. Huawei Technologies Co., Ltd. All rights reserved. + * Provide the pin memory method for checkpoint and restore task. + */ +#ifndef _LINUX_PIN_MEMORY_H +#define _LINUX_PIN_MEMORY_H + +#ifdef CONFIG_PIN_MEMORY +#include <linux/errno.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 PAGE_FLAGS_CHECK_RESERVED (1UL << PG_reserved) +#define SHA256_DIGEST_SIZE 32 +#define next_pme(pme) ((unsigned long *)(pme + 1) + pme->nr_pages) +#define PIN_MEM_DUMP_MAGIC 0xfeab000000001acd +struct page_map_entry { + unsigned long virt_addr; + unsigned int nr_pages; + unsigned int is_huge_page; + unsigned long redirect_start; + unsigned long phy_addr_array[0]; +}; + +struct page_map_info { + int pid; + int pid_reserved; + unsigned int entry_num; + int disable_free_page; + struct page_map_entry *pme; +}; + +struct pin_mem_dump_info { + char sha_digest[SHA256_DIGEST_SIZE]; + unsigned long magic; + unsigned int pin_pid_num; + struct page_map_info pmi_array[0]; +}; + +struct redirect_info { + unsigned int redirect_pages; + unsigned int redirect_index[0]; +}; + +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); +extern int finish_pin_mem_dump(void); + +/* 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 bfed474..2407de3 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -450,6 +450,17 @@ void __init reserve_crashkernel(void) } #endif /* CONFIG_ARCH_WANT_RESERVE_CRASH_KERNEL */ +#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 390165f..930dc13 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -859,4 +859,12 @@ 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. We can pin the memory data of tasks and collect + the corresponding physical pages mapping info in checkpoint, + and remap the physical pages to restore tasks in restore. endmenu diff --git a/mm/Makefile b/mm/Makefile index d73aed0..4963827 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -120,3 +120,4 @@ obj-$(CONFIG_MEMFD_CREATE) += memfd.o obj-$(CONFIG_MAPPING_DIRTY_HELPERS) += mapping_dirty_helpers.o obj-$(CONFIG_PTDUMP_CORE) += ptdump.o obj-$(CONFIG_PAGE_REPORTING) += page_reporting.o +obj-$(CONFIG_PIN_MEMORY) += pin_mem.o diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 0bc4a2c..8a11d30 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -2996,3 +2996,64 @@ void remove_migration_pmd(struct page_vma_mapped_walk *pvmw, struct page *new) update_mmu_cache_pmd(vma, address, pvmw->pmd); } #endif + +#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; + + 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_charge(page, vma->vm_mm, gfp)) { + put_page(page); + count_vm_event(THP_FAULT_FALLBACK); + count_vm_event(THP_FAULT_FALLBACK_CHARGE); + return VM_FAULT_FALLBACK; + } + cgroup_throttle_swaprate(page, gfp); + + pgtable = pte_alloc_one(vma->vm_mm); + 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); + lru_cache_add_inactive_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); + count_memcg_event_mm(vma->vm_mm, THP_FAULT_ALLOC); + } + + return 0; +unlock_release: + spin_unlock(ptl); +release: + if (pgtable) + pte_free(vma->vm_mm, pgtable); + put_page(page); + return ret; +} +#endif diff --git a/mm/memory.c b/mm/memory.c index 50632c4..7b7f1a7 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -5248,3 +5248,62 @@ void ptlock_free(struct page *page) kmem_cache_free(page_ptl_cachep, page->ptl); } #endif + +#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) +{ + pte_t entry; + spinlock_t *ptl; + pte_t *pte; + vm_fault_t ret = 0; + + if (pte_alloc(vma->vm_mm, pmd)) + 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_charge(page, vma->vm_mm, GFP_KERNEL)) + goto oom_free_page; + cgroup_throttle_swaprate(page, GFP_KERNEL); + + __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); + lru_cache_add_inactive_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: + put_page(page); + goto unlock; +oom_free_page: + put_page(page); +oom: + return VM_FAULT_OOM; +} +#endif diff --git a/mm/pin_mem.c b/mm/pin_mem.c new file mode 100644 index 0000000..0a143b6 --- /dev/null +++ b/mm/pin_mem.c @@ -0,0 +1,950 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2021. Huawei Technologies Co., Ltd. All rights reserved. + * Provide the pin memory method for checkpoint and restore task. + */ +#ifdef CONFIG_PIN_MEMORY +#include <linux/init.h> +#include <linux/slab.h> +#include <linux/time.h> +#include <linux/sched/cputime.h> +#include <linux/tick.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> +#include <linux/highmem.h> +#include <crypto/sha.h> + +#define MAX_PIN_PID_NUM 128 +static DEFINE_SPINLOCK(page_map_entry_lock); + +struct pin_mem_dump_info *pin_mem_dump_start; +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; +unsigned long redirect_space_size; +unsigned long redirect_space_start; +#define DEFAULT_REDIRECT_SPACE_SIZE 0x100000 + +static int __init setup_max_pin_pid_num(char *str) +{ + int ret = 0; + + 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); + +static int __init setup_redirect_space_size(char *str) +{ + if (!str) + goto out; + + redirect_space_size = memparse(str, NULL); +out: + if (!redirect_space_size) { + pr_warn("Unable to parse redirect space size, use the default value.\n"); + redirect_space_size = DEFAULT_REDIRECT_SPACE_SIZE; + } + return 0; +} +early_param("redirect_space_size", setup_redirect_space_size); + +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; + new->disable_free_page = 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; + + 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; + + 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) +{ + atomic_inc(&page->_refcount); + reserve_page_from_buddy(1, page); +} + +static void init_huge_pmd_pages(struct page *head_page) +{ + int i = 0; + struct page *page = head_page; + + __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_dtor = HUGETLB_PAGE_DTOR + 1; + page->compound_order = HPAGE_PMD_ORDER; + page++; + i++; + page->compound_head = (unsigned long)head_page + 1; + 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; + } +} + +static inline void reserve_user_huge_pmd_pages(struct page *page) +{ + atomic_inc(&page->_refcount); + reserve_page_from_buddy((1 << HPAGE_PMD_ORDER), page); + init_huge_pmd_pages(page); +} + +int reserve_user_map_pages_fail; + +void free_user_map_pages(unsigned int pid_index, unsigned int entry_index, unsigned int page_index) +{ + unsigned int i, j, index, order; + struct page_map_info *pmi; + struct page_map_entry *pme; + struct page *page; + unsigned long phy_addr; + + for (index = 0; index < pid_index; index++) { + pmi = &(user_space_reserve_start[index]); + pme = pmi->pme; + for (i = 0; i < pmi->entry_num; i++) { + for (j = 0; j < pme->nr_pages; j++) { + order = pme->is_huge_page ? HPAGE_PMD_ORDER : 0; + phy_addr = pme->phy_addr_array[j]; + if (phy_addr) { + page = phys_to_page(phy_addr); + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, order); + pme->phy_addr_array[j] = 0; + } + } + } + pme = (struct page_map_entry *)next_pme(pme); + } + } + pmi = &(user_space_reserve_start[index]); + pme = pmi->pme; + for (i = 0; i < entry_index; i++) { + for (j = 0; j < pme->nr_pages; j++) { + order = pme->is_huge_page ? HPAGE_PMD_ORDER : 0; + phy_addr = pme->phy_addr_array[j]; + if (phy_addr) { + page = phys_to_page(phy_addr); + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, order); + pme->phy_addr_array[j] = 0; + } + } + } + pme = (struct page_map_entry *)next_pme(pme); + } + for (j = 0; j < page_index; j++) { + order = pme->is_huge_page ? HPAGE_PMD_ORDER : 0; + phy_addr = pme->phy_addr_array[j]; + if (phy_addr) { + page = phys_to_page(phy_addr); + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, order); + pme->phy_addr_array[j] = 0; + } + } + } +} + +bool check_redirect_end_valid(struct redirect_info *redirect_start, + unsigned long max_redirect_page_num) +{ + unsigned long redirect_end; + + redirect_end = ((unsigned long)(redirect_start + 1) + + max_redirect_page_num * sizeof(unsigned int)); + if (redirect_end > redirect_space_start + redirect_space_size) + return false; + return false; +} + +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 phy_addr; + unsigned long redirect_pages = 0; + struct redirect_info *redirect_start = (struct redirect_info *)redirect_space_start; + + if (!user_space_reserve_start || !redirect_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++) { + redirect_pages = 0; + if (!check_redirect_end_valid(redirect_start, pme->nr_pages)) + redirect_start = NULL; + 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)) { + if ((page->flags & PAGE_FLAGS_CHECK_RESERVED) + && !pme->redirect_start) + pme->redirect_start = + (unsigned long)redirect_start; + if (redirect_start && + (page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + redirect_start->redirect_index[redirect_pages] = j; + redirect_pages++; + continue; + } else { + reserve_user_map_pages_fail = 1; + pr_warn("Page %pK refcount %d large than zero, no need reserve.\n", + page, atomic_read(&page->_refcount)); + goto free_pages; + } + } + 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 (redirect_pages && redirect_start) { + redirect_start->redirect_pages = redirect_pages; + redirect_start = (struct redirect_info *)( + (unsigned long)(redirect_start + 1) + + redirect_start->redirect_pages * sizeof(unsigned int)); + } + } + } + spin_unlock(&page_map_entry_lock); + return; +free_pages: + free_user_map_pages(index, i, j); + spin_unlock(&page_map_entry_lock); +} + + +int calculate_pin_mem_digest(struct pin_mem_dump_info *pmdi, char *digest) +{ + int i; + struct sha256_state sctx; + + if (!digest) + digest = pmdi->sha_digest; + sha256_init(&sctx); + sha256_update(&sctx, (unsigned char *)(&(pmdi->magic)), + sizeof(struct pin_mem_dump_info) - SHA256_DIGEST_SIZE); + for (i = 0; i < pmdi->pin_pid_num; i++) { + sha256_update(&sctx, (unsigned char *)(&(pmdi->pmi_array[i])), + sizeof(struct page_map_info)); + } + sha256_final(&sctx, digest); + return 0; +} + +static int check_sha_digest(struct pin_mem_dump_info *pmdi) +{ + int ret = 0; + char digest[SHA256_DIGEST_SIZE] = {0}; + + ret = calculate_pin_mem_digest(pmdi, digest); + if (ret) { + pr_warn("calculate pin mem digest fail:%d\n", ret); + return ret; + } + if (memcmp(pmdi->sha_digest, digest, SHA256_DIGEST_SIZE)) { + pr_warn("pin mem dump info sha256 digest match error!\n"); + return -EFAULT; + } + return ret; +} + +/* + * 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(struct pin_mem_dump_info *pmdi, unsigned long map_len) +{ + if (pin_mem_dump_start || !max_pin_pid_num) { + pr_warn("pin page map already init or max_pin_pid_num not set.\n"); + return; + } + if (map_len < sizeof(struct pin_mem_dump_info) + + max_pin_pid_num * sizeof(struct page_map_info) + redirect_space_size) { + pr_warn("pin memory reserved memblock too small.\n"); + return; + } + if ((pmdi->magic != PIN_MEM_DUMP_MAGIC) || (pmdi->pin_pid_num > max_pin_pid_num) || + check_sha_digest(pmdi)) + memset(pmdi, 0, sizeof(struct pin_mem_dump_info)); + pin_mem_dump_start = pmdi; + pin_pid_num = pmdi->pin_pid_num; + pr_info("pin_pid_num: %d\n", pin_pid_num); + pin_pid_num_addr = &(pmdi->pin_pid_num); + user_space_reserve_start = + (struct page_map_info *)pmdi->pmi_array; + page_map_entry_start = + (struct page_map_entry *)(user_space_reserve_start + max_pin_pid_num); + page_map_entry_end = (unsigned long)pmdi + map_len - redirect_space_size; + redirect_space_start = page_map_entry_end; + if (pin_pid_num > 0) + reserve_user_space_map_pages(); +} + +int finish_pin_mem_dump(void) +{ + int ret; + + pin_mem_dump_start->magic = PIN_MEM_DUMP_MAGIC; + memset(pin_mem_dump_start->sha_digest, 0, SHA256_DIGEST_SIZE); + ret = calculate_pin_mem_digest(pin_mem_dump_start, NULL); + if (ret) { + pr_warn("calculate pin mem digest fail:%d\n", ret); + return ret; + } + return ret; +} + +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->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 = (PAGE_ALIGN(next) - start_addr) / PAGE_SIZE; + res = get_user_pages_remote(task->mm, start_addr, 1, + FOLL_TOUCH | FOLL_GET, &tmp_page, NULL, NULL); + if (!res) { + pr_warn("Get user page 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->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->redirect_start = 0; + pme->is_huge_page = is_huge_page; + memset(pme->phy_addr_array, 0, nr_pages * sizeof(unsigned long)); + down_write(&mm->mmap_lock); + if (!is_huge_page) { + ret = collect_normal_pages(task, start_addr, end_addr, pme); + if (ret != COLLECT_PAGES_FAIL && !pme->nr_pages) { + if (ret == COLLECT_PAGES_FINISH) { + ret = 0; + up_write(&mm->mmap_lock); + 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 (ret != COLLECT_PAGES_FAIL && !pme->nr_pages) { + if (ret == COLLECT_PAGES_FINISH) { + ret = 0; + up_write(&mm->mmap_lock); + 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_lock); + 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, i; + pgd_t *pgd; + p4d_t *p4d; + pmd_t *pmd; + pud_t *pud; + struct page *page, *new; + unsigned long address; + unsigned long phy_addr; + unsigned int redirect_pages = 0; + struct redirect_info *redirect_start; + + redirect_start = (struct redirect_info *)pme->redirect_start; + 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_to_pfn(page) == my_zero_pfn(address)) { + pme->phy_addr_array[j] = 0; + continue; + } + pme->phy_addr_array[j] = 0; + if (redirect_start && (redirect_pages < redirect_start->redirect_pages) && + (j == redirect_start->redirect_index[redirect_pages])) { + new = alloc_zeroed_user_highpage_movable(vma, address); + if (!new) { + pr_warn("Redirect alloc page fail\n"); + continue; + } + copy_page(page_to_virt(new), phys_to_virt(phy_addr)); + page = new; + redirect_pages++; + } + page->mapping = NULL; + pgd = pgd_offset(mm, address); + p4d = p4d_alloc(mm, pgd, address); + if (!p4d) { + ret = VM_FAULT_OOM; + goto free; + } + pud = pud_alloc(mm, p4d, address); + if (!pud) { + ret = VM_FAULT_OOM; + goto free; + } + pmd = pmd_alloc(mm, pud, address); + if (!pmd) { + ret = VM_FAULT_OOM; + goto free; + } + ret = do_anon_page_remap(vma, address, pmd, page); + if (ret) + goto free; + } + return 0; +free: + for (i = j; i < pme->nr_pages; i++) { + phy_addr = pme->phy_addr_array[i]; + if (phy_addr) { + __free_page(phys_to_page(phy_addr)); + pme->phy_addr_array[i] = 0; + } + } + return ret; +} + +static inline gfp_t get_hugepage_gfpmask(struct vm_area_struct *vma) +{ + const bool vma_madvised = !!(vma->vm_flags & VM_HUGEPAGE); + + if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_DIRECT_FLAG, &transparent_hugepage_flags)) + return GFP_TRANSHUGE | (vma_madvised ? 0 : __GFP_NORETRY); + if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_FLAG, &transparent_hugepage_flags)) + return GFP_TRANSHUGE_LIGHT | __GFP_KSWAPD_RECLAIM; + if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_OR_MADV_FLAG, &transparent_hugepage_flags)) + return GFP_TRANSHUGE_LIGHT | (vma_madvised ? __GFP_DIRECT_RECLAIM : + __GFP_KSWAPD_RECLAIM); + if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG, &transparent_hugepage_flags)) + return GFP_TRANSHUGE_LIGHT | (vma_madvised ? __GFP_DIRECT_RECLAIM : + 0); + return GFP_TRANSHUGE_LIGHT; +} + +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, i; + pgd_t *pgd; + p4d_t *p4d; + pmd_t *pmd; + pud_t *pud; + gfp_t gfp; + struct page *page, *new; + unsigned long address; + unsigned long phy_addr; + unsigned int redirect_pages = 0; + struct redirect_info *redirect_start; + + redirect_start = (struct redirect_info *)pme->redirect_start; + 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 (is_huge_zero_page(page)) { + pme->phy_addr_array[j] = 0; + continue; + } + pme->phy_addr_array[j] = 0; + if (redirect_start && (redirect_pages < redirect_start->redirect_pages) && + (j == redirect_start->redirect_index[redirect_pages])) { + gfp = get_hugepage_gfpmask(vma); + new = alloc_hugepage_vma(gfp, vma, address, HPAGE_PMD_ORDER); + if (!new) { + pr_warn("Redirect alloc huge page fail\n"); + continue; + } + memcpy(page_to_virt(new), phys_to_virt(phy_addr), HPAGE_PMD_SIZE); + page = new; + redirect_pages++; + } + pgd = pgd_offset(mm, address); + p4d = p4d_alloc(mm, pgd, address); + if (!p4d) { + ret = VM_FAULT_OOM; + goto free; + } + pud = pud_alloc(mm, p4d, address); + if (!pud) { + ret = VM_FAULT_OOM; + goto free; + } + pmd = pmd_alloc(mm, pud, address); + if (!pmd) { + ret = VM_FAULT_OOM; + goto free; + } + ret = do_anon_huge_page_remap(vma, address, pmd, page); + if (ret) + goto free; + } + return 0; +free: + for (i = j; i < pme->nr_pages; i++) { + phy_addr = pme->phy_addr_array[i]; + if (phy_addr) { + page = phys_to_page(phy_addr); + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, HPAGE_PMD_ORDER); + pme->phy_addr_array[i] = 0; + } + } + } + return ret; +} + +static void free_unmap_pages(struct page_map_info *pmi, + struct page_map_entry *pme, + unsigned int index) +{ + unsigned int i, j; + unsigned long phy_addr; + unsigned int order; + struct page *page; + + pme = (struct page_map_entry *)(next_pme(pme)); + for (i = index; i < pmi->entry_num; i++) { + for (j = 0; j < pme->nr_pages; j++) { + phy_addr = pme->phy_addr_array[i]; + if (phy_addr) { + page = phys_to_page(phy_addr); + order = pme->is_huge_page ? HPAGE_PMD_ORDER : 0; + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, order); + pme->phy_addr_array[i] = 0; + } + } + } + pme = (struct page_map_entry *)(next_pme(pme)); + } +} + +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; + unsigned long flags; + + if (reserve_user_map_pages_fail) + return -EFAULT; + pmi = get_page_map_info(pid); + if (!pmi) + return -EFAULT; + + spin_lock_irqsave(&page_map_entry_lock, flags); + pmi->disable_free_page = true; + spin_unlock(&page_map_entry_lock); + down_write(&mm->mmap_lock); + 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 free; + } else { + ret = remap_huge_pmd_pages(mm, vma, pme); + if (ret < 0) + goto free; + } + pme = (struct page_map_entry *)(next_pme(pme)); + } else { + vma = vma->vm_next; + } + } + up_write(&mm->mmap_lock); + return 0; +free: + free_unmap_pages(pmi, pme, i); + up_write(&mm->mmap_lock); + 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((struct pin_mem_dump_info *)addr, map_size); +} +#else +void init_reserve_page_map(unsigned long map_addr, unsigned long map_size) +{ +} +#endif + +static void free_all_reserved_pages(void) +{ + unsigned int i, j, index, order; + struct page_map_info *pmi; + struct page_map_entry *pme; + struct page *page; + unsigned long phy_addr; + + if (!user_space_reserve_start || reserve_user_map_pages_fail) + return; + + for (index = 0; index < pin_pid_num; index++) { + pmi = &(user_space_reserve_start[index]); + if (pmi->disable_free_page) + continue; + pme = pmi->pme; + for (i = 0; i < pmi->entry_num; i++) { + for (j = 0; j < pme->nr_pages; j++) { + order = pme->is_huge_page ? HPAGE_PMD_ORDER : 0; + phy_addr = pme->phy_addr_array[j]; + if (phy_addr) { + page = phys_to_page(phy_addr); + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, order); + pme->phy_addr_array[j] = 0; + } + } + } + pme = (struct page_map_entry *)next_pme(pme); + } + } +} + +/* Clear all pin memory record. */ +void clear_pin_memory_record(void) +{ + unsigned long flags; + + spin_lock_irqsave(&page_map_entry_lock, flags); + free_all_reserved_pages(); + 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; + } + spin_unlock(&page_map_entry_lock); +} +EXPORT_SYMBOL_GPL(clear_pin_memory_record); + +#endif /* CONFIG_PIN_MEMORY */ -- 2.9.5

2 1

申请加入linux内核组
by 邓伟进 02 Mar '21

02 Mar '21

尊敬的linux内核组各位老师同学：你们好！我是北京拓林思软件有限公司的员工邓伟进。我对linux内核进程管理，文件系统和设备驱动有一定理解。熟悉bpf。目前正努力致力于研究学习bpf和linux内核进程管理，内存管理和文件系统。热爱操作系统和编程。严谨，并喜欢刨根究底的探究问题。这次在陈琪德老师鼓励下申请加入linux内核组，还望通过！非常感谢，我会争取能为社区做一些力所能及的工作。期待您的回复！邓伟进 2021.3.2

2 1

[PATCH OLK-5.10] arm64: enabled e820_pmem and add memmap parameter
by Zhuling 02 Mar '21

02 Mar '21

From: zhuling <zhuling8(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: NA Enabled e820_pmem in arm64: Use memmap=nn[KMG]!ss[KMG] reserve memory for persistent storage when the kernel restart or update. the data in PMEM will not be lost and can be loaded faster.this is a general features. if you want use this features, you need do as follows: 1.reserve memory: add memmap to reserve memory in grub.cfg memmap=nn[KMG]!ss[KMG] exp:memmap=100K!0x1a0000000. 2.insmod nd_e820.ko: modprobe nd_e820. 3.check pmem device in /dev exp: /dev/pmem0. Signed-off-by: zhuling <zhuling8(a)huawei.com> --- arch/arm64/Kconfig | 24 ++++++++++ arch/arm64/kernel/Makefile | 1 + arch/arm64/kernel/pmem.c | 35 ++++++++++++++ arch/arm64/kernel/setup.c | 6 +++ arch/arm64/mm/init.c | 98 ++++++++++++++++++++++++++++++++++++++ drivers/nvdimm/Makefile | 1 + include/linux/mm.h | 4 ++ 7 files changed, 169 insertions(+) create mode 100644 arch/arm64/kernel/pmem.c diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index b9c56543c..f1e05d9d2 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -1141,6 +1141,30 @@ config XEN_DOM0 def_bool y depends on XEN +config ARM64_PMEM_LEGACY_DEVICE + bool + +config ARM64_PMEM_RESERVE + bool "reserve memory for persistent storage" + default y + help + Use memmap=nn[KMG]!ss[KMG](memmap=100K!0x1a0000000) reserve memory for + persistent storage + + Say y here to enable this feature + +config ARM64_PMEM_LEGACY + tristate "create persistent storage" + depends on ARM64_PMEM_RESERVE + depends on BLK_DEV + select ARM64_PMEM_LEGACY_DEVICE + select LIBNVDIMM + help + Use reserved memory for persistent storage when the kernel restart + or update. the data in PMEM will not be lost and can be loaded faster. + + Say y if unsure. + config XEN bool "Xen guest support on ARM64" depends on ARM64 && OF diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile index 2621d5c2b..c363639b8 100644 --- a/arch/arm64/kernel/Makefile +++ b/arch/arm64/kernel/Makefile @@ -63,6 +63,7 @@ obj-$(CONFIG_ARM_SDE_INTERFACE) += sdei.o obj-$(CONFIG_ARM64_PTR_AUTH) += pointer_auth.o obj-$(CONFIG_SHADOW_CALL_STACK) += scs.o obj-$(CONFIG_ARM64_MTE) += mte.o +obj-$(CONFIG_ARM64_PMEM_LEGACY_DEVICE) += pmem.o obj-y += vdso/ probes/ obj-$(CONFIG_COMPAT_VDSO) += vdso32/ diff --git a/arch/arm64/kernel/pmem.c b/arch/arm64/kernel/pmem.c new file mode 100644 index 000000000..16eaf706f --- /dev/null +++ b/arch/arm64/kernel/pmem.c @@ -0,0 +1,35 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright(c) 2021 Huawei Technologies Co., Ltd + * + * Derived from x86 and arm64 implement PMEM. + */ +#include <linux/platform_device.h> +#include <linux/init.h> +#include <linux/ioport.h> +#include <linux/module.h> + +static int found(struct resource *res, void *data) +{ + return 1; +} + +static int __init register_e820_pmem(void) +{ + struct platform_device *pdev; + int rc; + + rc = walk_iomem_res_desc(IORES_DESC_PERSISTENT_MEMORY_LEGACY, + IORESOURCE_MEM, 0, -1, NULL, found); + if (rc <= 0) + return 0; + + /* + * See drivers/nvdimm/e820.c for the implementation, this is + * simply here to trigger the module to load on demand. + */ + pdev = platform_device_alloc("e820_pmem", -1); + + return platform_device_add(pdev); +} +device_initcall(register_e820_pmem); diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index 6aff30de8..7f506036d 100644 --- a/arch/arm64/kernel/setup.c +++ b/arch/arm64/kernel/setup.c @@ -255,6 +255,12 @@ static void __init request_standard_resources(void) request_resource(res, &crashk_res); #endif } + +#ifdef CONFIG_ARM64_PMEM_RESERVE + if (pmem_res.end && pmem_res.start) + request_resource(&iomem_resource, &pmem_res); +#endif + } static int __init reserve_memblock_reserved_regions(void) diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index 794f992cb..e4dc19145 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -63,6 +63,18 @@ EXPORT_SYMBOL(memstart_addr); phys_addr_t arm64_dma_phys_limit __ro_after_init; phys_addr_t arm64_dma32_phys_limit __ro_after_init; +static unsigned long long pmem_size, pmem_start; + +#ifdef CONFIG_ARM64_PMEM_RESERVE +struct resource pmem_res = { + .name = "Persistent Memory (legacy)", + .start = 0, + .end = 0, + .flags = IORESOURCE_MEM, + .desc = IORES_DESC_PERSISTENT_MEMORY_LEGACY +}; +#endif + #ifndef CONFIG_KEXEC_CORE static void __init reserve_crashkernel(void) { @@ -236,6 +248,88 @@ static void __init fdt_enforce_memory_region(void) memblock_add(usable_rgns[1].base, usable_rgns[1].size); } +static int __init is_mem_valid(unsigned long long mem_size, unsigned long long mem_start) +{ + if (!memblock_is_region_memory(mem_start, mem_size)) { + pr_warn("cannot reserve mem: region is not memory!\n"); + return -EINVAL; + } + + if (memblock_is_region_reserved(mem_start, mem_size)) { + pr_warn("cannot reserve mem: region overlaps reserved memory!\n"); + return -EINVAL; + } + + if (!IS_ALIGNED(mem_start, SZ_2M)) { + pr_warn("cannot reserve mem: base address is not 2MB aligned!\n"); + return -EINVAL; + } + + return 0; +} + +static int __init parse_memmap_one(char *p) +{ + char *oldp; + phys_addr_t start_at, mem_size; + int ret; + + if (!p) + return -EINVAL; + + oldp = p; + mem_size = memparse(p, &p); + if (p == oldp) + return -EINVAL; + + if (!mem_size) + return -EINVAL; + + mem_size = PAGE_ALIGN(mem_size); + + if (*p == '!') { + start_at = memparse(p+1, &p); + + if (is_mem_valid(mem_size, start_at) != 0) + return -EINVAL; + + pr_info("pmem reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + start_at, start_at + mem_size, mem_size >> 20); + pmem_start = start_at; + pmem_size = mem_size; + } else + pr_info("Unrecognized memmap option, please check the parameter.\n"); + + return *p == '\0' ? 0 : -EINVAL; +} + +static int __init parse_memmap_opt(char *str) +{ + while (str) { + char *k = strchr(str, ','); + + if (k) + *k++ = 0; + + parse_memmap_one(str); + str = k; + } + + return 0; +} +early_param("memmap", parse_memmap_opt); + +#ifdef CONFIG_ARM64_PMEM_RESERVE +static void __init reserve_pmem(void) +{ + memblock_remove(pmem_start, pmem_size); + pr_info("pmem reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + pmem_start, pmem_start + pmem_size, pmem_size >> 20); + pmem_res.start = pmem_start; + pmem_res.end = pmem_start + pmem_size - 1; +} +#endif + void __init arm64_memblock_init(void) { const s64 linear_region_size = BIT(vabits_actual - 1); @@ -359,6 +453,10 @@ void __init arm64_memblock_init(void) reserve_elfcorehdr(); +#ifdef CONFIG_ARM64_PMEM_RESERVE + reserve_pmem(); +#endif + high_memory = __va(memblock_end_of_DRAM() - 1) + 1; dma_contiguous_reserve(arm64_dma32_phys_limit); diff --git a/drivers/nvdimm/Makefile b/drivers/nvdimm/Makefile index 29203f3d3..b97760e9f 100644 --- a/drivers/nvdimm/Makefile +++ b/drivers/nvdimm/Makefile @@ -4,6 +4,7 @@ obj-$(CONFIG_BLK_DEV_PMEM) += nd_pmem.o obj-$(CONFIG_ND_BTT) += nd_btt.o obj-$(CONFIG_ND_BLK) += nd_blk.o obj-$(CONFIG_X86_PMEM_LEGACY) += nd_e820.o +obj-$(CONFIG_ARM64_PMEM_LEGACY) += nd_e820.o obj-$(CONFIG_OF_PMEM) += of_pmem.o obj-$(CONFIG_VIRTIO_PMEM) += virtio_pmem.o nd_virtio.o diff --git a/include/linux/mm.h b/include/linux/mm.h index cd5c31372..a5e50495e 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -45,6 +45,10 @@ extern int sysctl_page_lock_unfairness; void init_mm_internals(void); +#ifdef CONFIG_ARM64_PMEM_RESERVE +extern struct resource pmem_res; +#endif + #ifndef CONFIG_NEED_MULTIPLE_NODES /* Don't use mapnrs, do it properly */ extern unsigned long max_mapnr; -- 2.19.1

2 1

[PATCH openEuler-21.03 2/2] pid: add pid reserve method for checkpoint and recover
by hejingxian 02 Mar '21

02 Mar '21

From: Jingxian He <hejingxian(a)huawei.com> Date: Mon, 1 Mar 2021 17:44:59 +0800 Subject: [PATCH openEuler-21.03 2/2] pid: add pid reserve method for checkpoint and recover hulk inclusion category: feature bugzilla: 48159 CVE: N/A We record the pid of dump tasks in the reserved memory, and reserve the pids before init task start. In the recover process, free the reserved pids and realloc them for use. Signed-off-by: Jingxian He <hejingxian(a)huawei.com> Reviewed-by: Wenliang He <hewenliang4(a)huawei.com> Reviewed-by: Jing Xiangfeng <jingxiangfeng(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 1 + include/linux/pin_mem.h | 6 ++++ kernel/pid.c | 10 +++++++ mm/Kconfig | 10 +++++++ mm/pin_mem.c | 51 ++++++++++++++++++++++++++++++++++ 5 files changed, 78 insertions(+) diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 76fda68..de6db02 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -1037,6 +1037,7 @@ CONFIG_FRAME_VECTOR=y # CONFIG_READ_ONLY_THP_FOR_FS is not set CONFIG_ARCH_HAS_PTE_SPECIAL=y CONFIG_PIN_MEMORY=y +CONFIG_PID_RESERVE=y # end of Memory Management options CONFIG_NET=y diff --git a/include/linux/pin_mem.h b/include/linux/pin_mem.h index bc8b03e..a9fe2ef 100644 --- a/include/linux/pin_mem.h +++ b/include/linux/pin_mem.h @@ -74,5 +74,11 @@ extern struct resource pin_memory_resource; #endif extern void init_reserve_page_map(unsigned long map_addr, unsigned long map_size); +#ifdef CONFIG_PID_RESERVE +extern bool is_need_reserve_pids(void); +extern void free_reserved_pid(struct idr *idr, int pid); +extern void reserve_pids(struct idr *idr, int pid_max); +#endif + #endif /* CONFIG_PIN_MEMORY */ #endif /* _LINUX_PIN_MEMORY_H */ diff --git a/kernel/pid.c b/kernel/pid.c index 4856818..32ab9ef 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -44,6 +44,9 @@ #include <linux/idr.h> #include <net/sock.h> #include <uapi/linux/pidfd.h> +#ifdef CONFIG_PID_RESERVE +#include <linux/pin_mem.h> +#endif struct pid init_struct_pid = { .count = REFCOUNT_INIT(1), @@ -209,6 +212,9 @@ struct pid *alloc_pid(struct pid_namespace *ns, pid_t *set_tid, spin_lock_irq(&pidmap_lock); if (tid) { +#ifdef CONFIG_PID_RESERVE + free_reserved_pid(&tmp->idr, tid); +#endif nr = idr_alloc(&tmp->idr, NULL, tid, tid + 1, GFP_ATOMIC); /* @@ -621,6 +627,10 @@ void __init pid_idr_init(void) init_pid_ns.pid_cachep = KMEM_CACHE(pid, SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_ACCOUNT); +#ifdef CONFIG_PID_RESERVE + if (is_need_reserve_pids()) + reserve_pids(&init_pid_ns.idr, pid_max); +#endif } static struct file *__pidfd_fget(struct task_struct *task, int fd) diff --git a/mm/Kconfig b/mm/Kconfig index 930dc13..e27d2c6 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -868,3 +868,13 @@ config PIN_MEMORY the corresponding physical pages mapping info in checkpoint, and remap the physical pages to restore tasks in restore. endmenu + +config PID_RESERVE + bool "Support for reserve pid" + depends on PIN_MEMORY + help + Say y here to enable the pid reserved feature for checkpoint. + and restore. + We record the pid of dump task in the reserve memory, + and reserve the pids before init task start. In restore process, + free the reserved pids and realloc them for use. diff --git a/mm/pin_mem.c b/mm/pin_mem.c index 0a143b6..a040853 100644 --- a/mm/pin_mem.c +++ b/mm/pin_mem.c @@ -947,4 +947,55 @@ void clear_pin_memory_record(void) } EXPORT_SYMBOL_GPL(clear_pin_memory_record); +#ifdef CONFIG_PID_RESERVE +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; + } + } +} +#endif /* CONFIG_PID_RESERVE */ + #endif /* CONFIG_PIN_MEMORY */ -- 2.9.5

1 0

[PATCH openEuler-21.03 1/2] mm: add pin memory method for checkpoint add restore
by hejingxian＠huawei.com 02 Mar '21

02 Mar '21

From: Jingxian He <hejingxian(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: N/A 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). By pin the memory data of tasks and collect the corresponding physical pages mapping info in checkpoint process, we can remap the physical pages to restore tasks after upgrading the kernel. This pin memory method can restore the task data within one second. The pin memory area info is saved in the reserved memblock, 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> Reviewed-by: Wenliang He <hewenliang4(a)huawei.com> Reviewed-by: Jing Xiangfeng <jingxiangfeng(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 2 + arch/arm64/kernel/setup.c | 9 + arch/arm64/mm/init.c | 60 +++ drivers/char/Kconfig | 6 + drivers/char/Makefile | 1 + drivers/char/pin_memory.c | 208 ++++++++ include/linux/crash_core.h | 5 + include/linux/pin_mem.h | 78 +++ kernel/crash_core.c | 11 + mm/Kconfig | 8 + mm/Makefile | 1 + mm/huge_memory.c | 61 +++ mm/memory.c | 59 ++ mm/pin_mem.c | 950 +++++++++++++++++++++++++++++++++ 14 files changed, 1459 insertions(+) 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/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index c5271e7..76fda68 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -1036,6 +1036,7 @@ CONFIG_FRAME_VECTOR=y # CONFIG_GUP_BENCHMARK is not set # CONFIG_READ_ONLY_THP_FOR_FS is not set CONFIG_ARCH_HAS_PTE_SPECIAL=y +CONFIG_PIN_MEMORY=y # end of Memory Management options CONFIG_NET=y @@ -3282,6 +3283,7 @@ CONFIG_TCG_TIS_ST33ZP24_SPI=y # CONFIG_RANDOM_TRUST_CPU is not set # CONFIG_RANDOM_TRUST_BOOTLOADER is not set +CONFIG_PIN_MEMORY_DEV=m # # I2C support diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index c1f1fb9..5e282d3 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_mem.h> +#endif static int num_standard_resources; static struct resource *standard_resources; @@ -260,6 +263,12 @@ static void __init request_standard_resources(void) quick_kexec_res.end <= res->end) request_resource(res, &quick_kexec_res); #endif +#ifdef CONFIG_PIN_MEMORY + if (pin_memory_resource.end && + pin_memory_resource.start >= res->start && + pin_memory_resource.end <= res->end) + request_resource(res, &pin_memory_resource); +#endif } } diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index f3e5a66..8ab5aac 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -42,6 +42,9 @@ #include <linux/sizes.h> #include <asm/tlb.h> #include <asm/alternative.h> +#ifdef CONFIG_PIN_MEMORY +#include <linux/pin_mem.h> +#endif #define ARM64_ZONE_DMA_BITS 30 @@ -78,6 +81,55 @@ static void __init reserve_crashkernel(void) */ #define MAX_USABLE_RANGES 2 +#ifdef CONFIG_PIN_MEMORY +struct resource pin_memory_resource = { + .name = "Pin memory", + .start = 0, + .end = 0, + .flags = IORESOURCE_MEM, + .desc = IORES_DESC_RESERVED +}; + +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) @@ -455,6 +507,8 @@ void __init arm64_memblock_init(void) reserve_park_mem(); #endif + reserve_pin_memory_res(); + reserve_elfcorehdr(); high_memory = __va(memblock_end_of_DRAM() - 1) + 1; @@ -583,6 +637,12 @@ 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 + 1)); +#endif + mem_init_print_info(NULL); /* diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index d229a2d..fbb94b8 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -496,3 +496,9 @@ 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" + default m + help + pin memory driver diff --git a/drivers/char/Makefile b/drivers/char/Makefile index ffce287..71d76fd 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -47,3 +47,4 @@ obj-$(CONFIG_PS3_FLASH) += ps3flash.o obj-$(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 0000000..f46e056 --- /dev/null +++ b/drivers/char/pin_memory.c @@ -0,0 +1,208 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * 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 <linux/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 _FINISH_PIN_MEM_DUMP 4 +#define _PIN_MEM_IOC_MAX_NR 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 FINISH_PIN_MEM_DUMP _IOW(PIN_MEM_MAGIC, _FINISH_PIN_MEM_DUMP, 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); + return -EFAULT; + } + 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(); + put_pid(pid_s); + return ret; + +fail: + rcu_read_unlock(); + put_pid(pid_s); + 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(); + put_pid(pid_s); + return 0; + +fault: + rcu_read_unlock(); + put_pid(pid_s); + return -EFAULT; +} + +static long pin_memory_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + long ret = 0; + + if (_IOC_TYPE(cmd) != PIN_MEM_MAGIC) + return -EINVAL; + if (_IOC_NR(cmd) > _PIN_MEM_IOC_MAX_NR) + 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; + case FINISH_PIN_MEM_DUMP: + ret = finish_pin_mem_dump(); + 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 fc0ef33..30f0df3 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -87,4 +87,9 @@ 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 0000000..bc8b03e --- /dev/null +++ b/include/linux/pin_mem.h @@ -0,0 +1,78 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * 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/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 PAGE_FLAGS_CHECK_RESERVED (1UL << PG_reserved) +#define SHA256_DIGEST_SIZE 32 +#define next_pme(pme) ((unsigned long *)(pme + 1) + pme->nr_pages) +#define PIN_MEM_DUMP_MAGIC 0xfeab000000001acd +struct page_map_entry { + unsigned long virt_addr; + unsigned int nr_pages; + unsigned int is_huge_page; + unsigned long redirect_start; + unsigned long phy_addr_array[0]; +}; + +struct page_map_info { + int pid; + int pid_reserved; + unsigned int entry_num; + int disable_free_page; + struct page_map_entry *pme; +}; + +struct pin_mem_dump_info { + char sha_digest[SHA256_DIGEST_SIZE]; + unsigned long magic; + unsigned int pin_pid_num; + struct page_map_info pmi_array[0]; +}; + +struct redirect_info { + unsigned int redirect_pages; + unsigned int redirect_index[0]; +}; + +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); +extern int finish_pin_mem_dump(void); + +/* 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 bfed474..2407de3 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -450,6 +450,17 @@ void __init reserve_crashkernel(void) } #endif /* CONFIG_ARCH_WANT_RESERVE_CRASH_KERNEL */ +#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 390165f..930dc13 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -859,4 +859,12 @@ 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. We can pin the memory data of tasks and collect + the corresponding physical pages mapping info in checkpoint, + and remap the physical pages to restore tasks in restore. endmenu diff --git a/mm/Makefile b/mm/Makefile index d73aed0..4963827 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -120,3 +120,4 @@ obj-$(CONFIG_MEMFD_CREATE) += memfd.o obj-$(CONFIG_MAPPING_DIRTY_HELPERS) += mapping_dirty_helpers.o obj-$(CONFIG_PTDUMP_CORE) += ptdump.o obj-$(CONFIG_PAGE_REPORTING) += page_reporting.o +obj-$(CONFIG_PIN_MEMORY) += pin_mem.o diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 0bc4a2c..8a11d30 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -2996,3 +2996,64 @@ void remove_migration_pmd(struct page_vma_mapped_walk *pvmw, struct page *new) update_mmu_cache_pmd(vma, address, pvmw->pmd); } #endif + +#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; + + 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_charge(page, vma->vm_mm, gfp)) { + put_page(page); + count_vm_event(THP_FAULT_FALLBACK); + count_vm_event(THP_FAULT_FALLBACK_CHARGE); + return VM_FAULT_FALLBACK; + } + cgroup_throttle_swaprate(page, gfp); + + pgtable = pte_alloc_one(vma->vm_mm); + 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); + lru_cache_add_inactive_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); + count_memcg_event_mm(vma->vm_mm, THP_FAULT_ALLOC); + } + + return 0; +unlock_release: + spin_unlock(ptl); +release: + if (pgtable) + pte_free(vma->vm_mm, pgtable); + put_page(page); + return ret; +} +#endif diff --git a/mm/memory.c b/mm/memory.c index 50632c4..7b7f1a7 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -5248,3 +5248,62 @@ void ptlock_free(struct page *page) kmem_cache_free(page_ptl_cachep, page->ptl); } #endif + +#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) +{ + pte_t entry; + spinlock_t *ptl; + pte_t *pte; + vm_fault_t ret = 0; + + if (pte_alloc(vma->vm_mm, pmd)) + 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_charge(page, vma->vm_mm, GFP_KERNEL)) + goto oom_free_page; + cgroup_throttle_swaprate(page, GFP_KERNEL); + + __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); + lru_cache_add_inactive_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: + put_page(page); + goto unlock; +oom_free_page: + put_page(page); +oom: + return VM_FAULT_OOM; +} +#endif diff --git a/mm/pin_mem.c b/mm/pin_mem.c new file mode 100644 index 0000000..0a143b6 --- /dev/null +++ b/mm/pin_mem.c @@ -0,0 +1,950 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * 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/init.h> +#include <linux/slab.h> +#include <linux/time.h> +#include <linux/sched/cputime.h> +#include <linux/tick.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> +#include <linux/highmem.h> +#include <crypto/sha.h> + +#define MAX_PIN_PID_NUM 128 +static DEFINE_SPINLOCK(page_map_entry_lock); + +struct pin_mem_dump_info *pin_mem_dump_start; +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; +unsigned long redirect_space_size; +unsigned long redirect_space_start; +#define DEFAULT_REDIRECT_SPACE_SIZE 0x100000 + +static int __init setup_max_pin_pid_num(char *str) +{ + int ret = 0; + + 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); + +static int __init setup_redirect_space_size(char *str) +{ + if (!str) + goto out; + + redirect_space_size = memparse(str, NULL); +out: + if (!redirect_space_size) { + pr_warn("Unable to parse redirect space size, use the default value.\n"); + redirect_space_size = DEFAULT_REDIRECT_SPACE_SIZE; + } + return 0; +} +early_param("redirect_space_size", setup_redirect_space_size); + +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; + new->disable_free_page = 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; + + 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; + + 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) +{ + atomic_inc(&page->_refcount); + reserve_page_from_buddy(1, page); +} + +static void init_huge_pmd_pages(struct page *head_page) +{ + int i = 0; + struct page *page = head_page; + + __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_dtor = HUGETLB_PAGE_DTOR + 1; + page->compound_order = HPAGE_PMD_ORDER; + page++; + i++; + page->compound_head = (unsigned long)head_page + 1; + 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; + } +} + +static inline void reserve_user_huge_pmd_pages(struct page *page) +{ + atomic_inc(&page->_refcount); + reserve_page_from_buddy((1 << HPAGE_PMD_ORDER), page); + init_huge_pmd_pages(page); +} + +int reserve_user_map_pages_fail; + +void free_user_map_pages(unsigned int pid_index, unsigned int entry_index, unsigned int page_index) +{ + unsigned int i, j, index, order; + struct page_map_info *pmi; + struct page_map_entry *pme; + struct page *page; + unsigned long phy_addr; + + for (index = 0; index < pid_index; index++) { + pmi = &(user_space_reserve_start[index]); + pme = pmi->pme; + for (i = 0; i < pmi->entry_num; i++) { + for (j = 0; j < pme->nr_pages; j++) { + order = pme->is_huge_page ? HPAGE_PMD_ORDER : 0; + phy_addr = pme->phy_addr_array[j]; + if (phy_addr) { + page = phys_to_page(phy_addr); + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, order); + pme->phy_addr_array[j] = 0; + } + } + } + pme = (struct page_map_entry *)next_pme(pme); + } + } + pmi = &(user_space_reserve_start[index]); + pme = pmi->pme; + for (i = 0; i < entry_index; i++) { + for (j = 0; j < pme->nr_pages; j++) { + order = pme->is_huge_page ? HPAGE_PMD_ORDER : 0; + phy_addr = pme->phy_addr_array[j]; + if (phy_addr) { + page = phys_to_page(phy_addr); + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, order); + pme->phy_addr_array[j] = 0; + } + } + } + pme = (struct page_map_entry *)next_pme(pme); + } + for (j = 0; j < page_index; j++) { + order = pme->is_huge_page ? HPAGE_PMD_ORDER : 0; + phy_addr = pme->phy_addr_array[j]; + if (phy_addr) { + page = phys_to_page(phy_addr); + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, order); + pme->phy_addr_array[j] = 0; + } + } + } +} + +bool check_redirect_end_valid(struct redirect_info *redirect_start, + unsigned long max_redirect_page_num) +{ + unsigned long redirect_end; + + redirect_end = ((unsigned long)(redirect_start + 1) + + max_redirect_page_num * sizeof(unsigned int)); + if (redirect_end > redirect_space_start + redirect_space_size) + return false; + return false; +} + +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 phy_addr; + unsigned long redirect_pages = 0; + struct redirect_info *redirect_start = (struct redirect_info *)redirect_space_start; + + if (!user_space_reserve_start || !redirect_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++) { + redirect_pages = 0; + if (!check_redirect_end_valid(redirect_start, pme->nr_pages)) + redirect_start = NULL; + 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)) { + if ((page->flags & PAGE_FLAGS_CHECK_RESERVED) + && !pme->redirect_start) + pme->redirect_start = + (unsigned long)redirect_start; + if (redirect_start && + (page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + redirect_start->redirect_index[redirect_pages] = j; + redirect_pages++; + continue; + } else { + reserve_user_map_pages_fail = 1; + pr_warn("Page %pK refcount %d large than zero, no need reserve.\n", + page, atomic_read(&page->_refcount)); + goto free_pages; + } + } + 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 (redirect_pages && redirect_start) { + redirect_start->redirect_pages = redirect_pages; + redirect_start = (struct redirect_info *)( + (unsigned long)(redirect_start + 1) + + redirect_start->redirect_pages * sizeof(unsigned int)); + } + } + } + spin_unlock(&page_map_entry_lock); + return; +free_pages: + free_user_map_pages(index, i, j); + spin_unlock(&page_map_entry_lock); +} + + +int calculate_pin_mem_digest(struct pin_mem_dump_info *pmdi, char *digest) +{ + int i; + struct sha256_state sctx; + + if (!digest) + digest = pmdi->sha_digest; + sha256_init(&sctx); + sha256_update(&sctx, (unsigned char *)(&(pmdi->magic)), + sizeof(struct pin_mem_dump_info) - SHA256_DIGEST_SIZE); + for (i = 0; i < pmdi->pin_pid_num; i++) { + sha256_update(&sctx, (unsigned char *)(&(pmdi->pmi_array[i])), + sizeof(struct page_map_info)); + } + sha256_final(&sctx, digest); + return 0; +} + +static int check_sha_digest(struct pin_mem_dump_info *pmdi) +{ + int ret = 0; + char digest[SHA256_DIGEST_SIZE] = {0}; + + ret = calculate_pin_mem_digest(pmdi, digest); + if (ret) { + pr_warn("calculate pin mem digest fail:%d\n", ret); + return ret; + } + if (memcmp(pmdi->sha_digest, digest, SHA256_DIGEST_SIZE)) { + pr_warn("pin mem dump info sha256 digest match error!\n"); + return -EFAULT; + } + return ret; +} + +/* + * 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(struct pin_mem_dump_info *pmdi, unsigned long map_len) +{ + if (pin_mem_dump_start || !max_pin_pid_num) { + pr_warn("pin page map already init or max_pin_pid_num not set.\n"); + return; + } + if (map_len < sizeof(struct pin_mem_dump_info) + + max_pin_pid_num * sizeof(struct page_map_info) + redirect_space_size) { + pr_warn("pin memory reserved memblock too small.\n"); + return; + } + if ((pmdi->magic != PIN_MEM_DUMP_MAGIC) || (pmdi->pin_pid_num > max_pin_pid_num) || + check_sha_digest(pmdi)) + memset(pmdi, 0, sizeof(struct pin_mem_dump_info)); + pin_mem_dump_start = pmdi; + pin_pid_num = pmdi->pin_pid_num; + pr_info("pin_pid_num: %d\n", pin_pid_num); + pin_pid_num_addr = &(pmdi->pin_pid_num); + user_space_reserve_start = + (struct page_map_info *)pmdi->pmi_array; + page_map_entry_start = + (struct page_map_entry *)(user_space_reserve_start + max_pin_pid_num); + page_map_entry_end = (unsigned long)pmdi + map_len - redirect_space_size; + redirect_space_start = page_map_entry_end; + if (pin_pid_num > 0) + reserve_user_space_map_pages(); +} + +int finish_pin_mem_dump(void) +{ + int ret; + + pin_mem_dump_start->magic = PIN_MEM_DUMP_MAGIC; + memset(pin_mem_dump_start->sha_digest, 0, SHA256_DIGEST_SIZE); + ret = calculate_pin_mem_digest(pin_mem_dump_start, NULL); + if (ret) { + pr_warn("calculate pin mem digest fail:%d\n", ret); + return ret; + } + return ret; +} + +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->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 = (PAGE_ALIGN(next) - start_addr) / PAGE_SIZE; + res = get_user_pages_remote(task->mm, start_addr, 1, + FOLL_TOUCH | FOLL_GET, &tmp_page, NULL, NULL); + if (!res) { + pr_warn("Get user page 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->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->redirect_start = 0; + pme->is_huge_page = is_huge_page; + memset(pme->phy_addr_array, 0, nr_pages * sizeof(unsigned long)); + down_write(&mm->mmap_lock); + if (!is_huge_page) { + ret = collect_normal_pages(task, start_addr, end_addr, pme); + if (ret != COLLECT_PAGES_FAIL && !pme->nr_pages) { + if (ret == COLLECT_PAGES_FINISH) { + ret = 0; + up_write(&mm->mmap_lock); + 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 (ret != COLLECT_PAGES_FAIL && !pme->nr_pages) { + if (ret == COLLECT_PAGES_FINISH) { + ret = 0; + up_write(&mm->mmap_lock); + 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_lock); + 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, i; + pgd_t *pgd; + p4d_t *p4d; + pmd_t *pmd; + pud_t *pud; + struct page *page, *new; + unsigned long address; + unsigned long phy_addr; + unsigned int redirect_pages = 0; + struct redirect_info *redirect_start; + + redirect_start = (struct redirect_info *)pme->redirect_start; + 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_to_pfn(page) == my_zero_pfn(address)) { + pme->phy_addr_array[j] = 0; + continue; + } + pme->phy_addr_array[j] = 0; + if (redirect_start && (redirect_pages < redirect_start->redirect_pages) && + (j == redirect_start->redirect_index[redirect_pages])) { + new = alloc_zeroed_user_highpage_movable(vma, address); + if (!new) { + pr_warn("Redirect alloc page fail\n"); + continue; + } + copy_page(page_to_virt(new), phys_to_virt(phy_addr)); + page = new; + redirect_pages++; + } + page->mapping = NULL; + pgd = pgd_offset(mm, address); + p4d = p4d_alloc(mm, pgd, address); + if (!p4d) { + ret = VM_FAULT_OOM; + goto free; + } + pud = pud_alloc(mm, p4d, address); + if (!pud) { + ret = VM_FAULT_OOM; + goto free; + } + pmd = pmd_alloc(mm, pud, address); + if (!pmd) { + ret = VM_FAULT_OOM; + goto free; + } + ret = do_anon_page_remap(vma, address, pmd, page); + if (ret) + goto free; + } + return 0; +free: + for (i = j; i < pme->nr_pages; i++) { + phy_addr = pme->phy_addr_array[i]; + if (phy_addr) { + __free_page(phys_to_page(phy_addr)); + pme->phy_addr_array[i] = 0; + } + } + return ret; +} + +static inline gfp_t get_hugepage_gfpmask(struct vm_area_struct *vma) +{ + const bool vma_madvised = !!(vma->vm_flags & VM_HUGEPAGE); + + if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_DIRECT_FLAG, &transparent_hugepage_flags)) + return GFP_TRANSHUGE | (vma_madvised ? 0 : __GFP_NORETRY); + if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_FLAG, &transparent_hugepage_flags)) + return GFP_TRANSHUGE_LIGHT | __GFP_KSWAPD_RECLAIM; + if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_OR_MADV_FLAG, &transparent_hugepage_flags)) + return GFP_TRANSHUGE_LIGHT | (vma_madvised ? __GFP_DIRECT_RECLAIM : + __GFP_KSWAPD_RECLAIM); + if (test_bit(TRANSPARENT_HUGEPAGE_DEFRAG_REQ_MADV_FLAG, &transparent_hugepage_flags)) + return GFP_TRANSHUGE_LIGHT | (vma_madvised ? __GFP_DIRECT_RECLAIM : + 0); + return GFP_TRANSHUGE_LIGHT; +} + +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, i; + pgd_t *pgd; + p4d_t *p4d; + pmd_t *pmd; + pud_t *pud; + gfp_t gfp; + struct page *page, *new; + unsigned long address; + unsigned long phy_addr; + unsigned int redirect_pages = 0; + struct redirect_info *redirect_start; + + redirect_start = (struct redirect_info *)pme->redirect_start; + 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 (is_huge_zero_page(page)) { + pme->phy_addr_array[j] = 0; + continue; + } + pme->phy_addr_array[j] = 0; + if (redirect_start && (redirect_pages < redirect_start->redirect_pages) && + (j == redirect_start->redirect_index[redirect_pages])) { + gfp = get_hugepage_gfpmask(vma); + new = alloc_hugepage_vma(gfp, vma, address, HPAGE_PMD_ORDER); + if (!new) { + pr_warn("Redirect alloc huge page fail\n"); + continue; + } + memcpy(page_to_virt(new), phys_to_virt(phy_addr), HPAGE_PMD_SIZE); + page = new; + redirect_pages++; + } + pgd = pgd_offset(mm, address); + p4d = p4d_alloc(mm, pgd, address); + if (!p4d) { + ret = VM_FAULT_OOM; + goto free; + } + pud = pud_alloc(mm, p4d, address); + if (!pud) { + ret = VM_FAULT_OOM; + goto free; + } + pmd = pmd_alloc(mm, pud, address); + if (!pmd) { + ret = VM_FAULT_OOM; + goto free; + } + ret = do_anon_huge_page_remap(vma, address, pmd, page); + if (ret) + goto free; + } + return 0; +free: + for (i = j; i < pme->nr_pages; i++) { + phy_addr = pme->phy_addr_array[i]; + if (phy_addr) { + page = phys_to_page(phy_addr); + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, HPAGE_PMD_ORDER); + pme->phy_addr_array[i] = 0; + } + } + } + return ret; +} + +static void free_unmap_pages(struct page_map_info *pmi, + struct page_map_entry *pme, + unsigned int index) +{ + unsigned int i, j; + unsigned long phy_addr; + unsigned int order; + struct page *page; + + pme = (struct page_map_entry *)(next_pme(pme)); + for (i = index; i < pmi->entry_num; i++) { + for (j = 0; j < pme->nr_pages; j++) { + phy_addr = pme->phy_addr_array[i]; + if (phy_addr) { + page = phys_to_page(phy_addr); + order = pme->is_huge_page ? HPAGE_PMD_ORDER : 0; + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, order); + pme->phy_addr_array[i] = 0; + } + } + } + pme = (struct page_map_entry *)(next_pme(pme)); + } +} + +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; + unsigned long flags; + + if (reserve_user_map_pages_fail) + return -EFAULT; + pmi = get_page_map_info(pid); + if (!pmi) + return -EFAULT; + + spin_lock_irqsave(&page_map_entry_lock, flags); + pmi->disable_free_page = true; + spin_unlock(&page_map_entry_lock); + down_write(&mm->mmap_lock); + 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 free; + } else { + ret = remap_huge_pmd_pages(mm, vma, pme); + if (ret < 0) + goto free; + } + pme = (struct page_map_entry *)(next_pme(pme)); + } else { + vma = vma->vm_next; + } + } + up_write(&mm->mmap_lock); + return 0; +free: + free_unmap_pages(pmi, pme, i); + up_write(&mm->mmap_lock); + 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((struct pin_mem_dump_info *)addr, map_size); +} +#else +void init_reserve_page_map(unsigned long map_addr, unsigned long map_size) +{ +} +#endif + +static void free_all_reserved_pages(void) +{ + unsigned int i, j, index, order; + struct page_map_info *pmi; + struct page_map_entry *pme; + struct page *page; + unsigned long phy_addr; + + if (!user_space_reserve_start || reserve_user_map_pages_fail) + return; + + for (index = 0; index < pin_pid_num; index++) { + pmi = &(user_space_reserve_start[index]); + if (pmi->disable_free_page) + continue; + pme = pmi->pme; + for (i = 0; i < pmi->entry_num; i++) { + for (j = 0; j < pme->nr_pages; j++) { + order = pme->is_huge_page ? HPAGE_PMD_ORDER : 0; + phy_addr = pme->phy_addr_array[j]; + if (phy_addr) { + page = phys_to_page(phy_addr); + if (!(page->flags & PAGE_FLAGS_CHECK_RESERVED)) { + __free_pages(page, order); + pme->phy_addr_array[j] = 0; + } + } + } + pme = (struct page_map_entry *)next_pme(pme); + } + } +} + +/* Clear all pin memory record. */ +void clear_pin_memory_record(void) +{ + unsigned long flags; + + spin_lock_irqsave(&page_map_entry_lock, flags); + free_all_reserved_pages(); + 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; + } + spin_unlock(&page_map_entry_lock); +} +EXPORT_SYMBOL_GPL(clear_pin_memory_record); + +#endif /* CONFIG_PIN_MEMORY */ -- 2.9.5

1 1

[PATCH OLK-5.10 v1] arm64: Declare var of local_cpu_stop only on PARK
by sangyan＠huawei.com 01 Mar '21

01 Mar '21

From: Sang Yan <sangyan(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: N/A Fix compile warning: unused variable 'ops' 'cpu' while CONFIG_ARM64_CPU_PARK=n. Put declaration of 'ops' and 'cpu' under CONFIG_ARM64_CPU_PARK in local_cpu_stop. Signed-off-by: Sang Yan <sangyan(a)huawei.com> --- arch/arm64/kernel/smp.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c index 644bbd7..d7b750a 100644 --- a/arch/arm64/kernel/smp.c +++ b/arch/arm64/kernel/smp.c @@ -1024,8 +1024,10 @@ void arch_irq_work_raise(void) static void local_cpu_stop(void) { +#ifdef CONFIG_ARM64_CPU_PARK int cpu; const struct cpu_operations *ops = NULL; +#endif set_cpu_online(smp_processor_id(), false); -- 2.9.5

1 0

[PATCH OLK-5.10 v2 1/2] arm64: smp: Add support for cpu park
by sangyan＠huawei.com 26 Feb '21

26 Feb '21

From: Sang Yan <sangyan(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: N/A Introducing a feature of CPU PARK in order to save time of cpus down and up during kexec, which may cost 250ms of per cpu's down and 30ms of up. As a result, for 128 cores, it costs more than 30 seconds to down and up cpus during kexec. Think about 256 cores and more. CPU PARK is a state that cpu power-on and staying in spin loop, polling for exit chances, such as writing exit address. Reserving a block of memory, to fill with cpu park text section, exit address and park-magic-flag of each cpu. In implementation, reserved one page for one cpu core. Cpus going to park state instead of down in machine_shutdown(). Cpus going out of park state in smp_init instead of brought up. One of cpu park sections in pre-reserved memory blocks,: +--------------+ + exit address + +--------------+ + park magic + +--------------+ + park codes + + . + + . + + . + +--------------+ Signed-off-by: Sang Yan <sangyan(a)huawei.com> Reviewed-by: Jing Xiangfeng <jingxiangfeng(a)huawei.com> --- arch/arm64/Kconfig | 12 ++ arch/arm64/include/asm/kexec.h | 6 + arch/arm64/include/asm/smp.h | 15 +++ arch/arm64/kernel/Makefile | 1 + arch/arm64/kernel/cpu-park.S | 59 ++++++++++ arch/arm64/kernel/machine_kexec.c | 2 +- arch/arm64/kernel/process.c | 4 + arch/arm64/kernel/smp.c | 230 ++++++++++++++++++++++++++++++++++++++ arch/arm64/mm/init.c | 55 +++++++++ 9 files changed, 383 insertions(+), 1 deletion(-) create mode 100644 arch/arm64/kernel/cpu-park.S diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index b9c5654..0885668 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -345,6 +345,18 @@ config KASAN_SHADOW_OFFSET default 0xeffffff900000000 if ARM64_VA_BITS_36 && KASAN_SW_TAGS default 0xffffffffffffffff +config ARM64_CPU_PARK + bool "Support CPU PARK on kexec" + depends on SMP + depends on KEXEC_CORE + help + This enables support for CPU PARK feature in + order to save time of cpu down to up. + CPU park is a state through kexec, spin loop + instead of cpu die before jumping to new kernel, + jumping out from loop to new kernel entry in + smp_init. + source "arch/arm64/Kconfig.platforms" menu "Kernel Features" diff --git a/arch/arm64/include/asm/kexec.h b/arch/arm64/include/asm/kexec.h index 79909ae..a133889 100644 --- a/arch/arm64/include/asm/kexec.h +++ b/arch/arm64/include/asm/kexec.h @@ -36,6 +36,11 @@ #define CRASH_ADDR_HIGH_MAX MEMBLOCK_ALLOC_ACCESSIBLE +#ifdef CONFIG_ARM64_CPU_PARK +/* CPU park state flag: "park" */ +#define PARK_MAGIC 0x7061726b +#endif + #ifndef __ASSEMBLY__ /** @@ -104,6 +109,7 @@ static inline void crash_post_resume(void) {} #ifdef CONFIG_KEXEC_CORE extern void __init reserve_crashkernel(void); #endif +void machine_kexec_mask_interrupts(void); #ifdef CONFIG_KEXEC_FILE #define ARCH_HAS_KIMAGE_ARCH diff --git a/arch/arm64/include/asm/smp.h b/arch/arm64/include/asm/smp.h index 2e7f529..8c5d2d6 100644 --- a/arch/arm64/include/asm/smp.h +++ b/arch/arm64/include/asm/smp.h @@ -145,6 +145,21 @@ bool cpus_are_stuck_in_kernel(void); extern void crash_smp_send_stop(void); extern bool smp_crash_stop_failed(void); +#ifdef CONFIG_ARM64_CPU_PARK +#define PARK_SECTION_SIZE 1024 +struct cpu_park_info { + /* Physical address of reserved park memory. */ + unsigned long start; + /* park reserve mem len should be PARK_SECTION_SIZE * NR_CPUS */ + unsigned long len; + /* Virtual address of reserved park memory. */ + unsigned long start_v; +}; +extern struct cpu_park_info park_info; +extern void enter_cpu_park(unsigned long text, unsigned long exit); +extern void do_cpu_park(unsigned long exit); +extern int kexec_smp_send_park(void); +#endif #endif /* ifndef __ASSEMBLY__ */ diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile index 2621d5c..60478d2 100644 --- a/arch/arm64/kernel/Makefile +++ b/arch/arm64/kernel/Makefile @@ -54,6 +54,7 @@ obj-$(CONFIG_RANDOMIZE_BASE) += kaslr.o obj-$(CONFIG_HIBERNATION) += hibernate.o hibernate-asm.o obj-$(CONFIG_KEXEC_CORE) += machine_kexec.o relocate_kernel.o \ cpu-reset.o +obj-$(CONFIG_ARM64_CPU_PARK) += cpu-park.o obj-$(CONFIG_KEXEC_FILE) += machine_kexec_file.o kexec_image.o obj-$(CONFIG_ARM64_RELOC_TEST) += arm64-reloc-test.o arm64-reloc-test-y := reloc_test_core.o reloc_test_syms.o diff --git a/arch/arm64/kernel/cpu-park.S b/arch/arm64/kernel/cpu-park.S new file mode 100644 index 0000000..10c685c --- /dev/null +++ b/arch/arm64/kernel/cpu-park.S @@ -0,0 +1,59 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * CPU park routines + * + * Copyright (C) 2020 Huawei Technologies., 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. + */ + +#include <linux/linkage.h> +#include <asm/assembler.h> +#include <asm/kexec.h> +#include <asm/sysreg.h> +#include <asm/virt.h> + +.text +.pushsection .idmap.text, "awx" + +/* cpu park helper in idmap section */ +SYM_CODE_START(enter_cpu_park) + /* Clear sctlr_el1 flags. */ + mrs x12, sctlr_el1 + mov_q x13, SCTLR_ELx_FLAGS + bic x12, x12, x13 + pre_disable_mmu_workaround + msr sctlr_el1, x12 /* disable mmu */ + isb + + mov x18, x0 + mov x0, x1 /* secondary_entry addr */ + br x18 /* call do_cpu_park of each cpu */ +SYM_CODE_END(enter_cpu_park) + +.popsection + +SYM_CODE_START(do_cpu_park) + ldr x18, =PARK_MAGIC /* magic number "park" */ + add x1, x0, #8 + str x18, [x1] /* set on-park flag */ + dc civac, x1 /* flush cache of "park" */ + dsb nsh + isb + +.Lloop: + wfe + isb + ldr x19, [x0] + cmp x19, #0 /* test secondary_entry */ + b.eq .Lloop + + ic iallu /* invalidate the local I-cache */ + dsb nsh + isb + + br x19 /* jump to secondary_entry */ +SYM_CODE_END(do_cpu_park) + diff --git a/arch/arm64/kernel/machine_kexec.c b/arch/arm64/kernel/machine_kexec.c index a0b144c..f47ce96 100644 --- a/arch/arm64/kernel/machine_kexec.c +++ b/arch/arm64/kernel/machine_kexec.c @@ -213,7 +213,7 @@ void machine_kexec(struct kimage *kimage) BUG(); /* Should never get here. */ } -static void machine_kexec_mask_interrupts(void) +void machine_kexec_mask_interrupts(void) { unsigned int i; struct irq_desc *desc; diff --git a/arch/arm64/kernel/process.c b/arch/arm64/kernel/process.c index 73e3b32..10cffee 100644 --- a/arch/arm64/kernel/process.c +++ b/arch/arm64/kernel/process.c @@ -146,6 +146,10 @@ void arch_cpu_idle_dead(void) */ void machine_shutdown(void) { +#ifdef CONFIG_ARM64_CPU_PARK + if (kexec_smp_send_park() == 0) + return; +#endif smp_shutdown_nonboot_cpus(reboot_cpu); } diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c index 18e9727..dea67d0 100644 --- a/arch/arm64/kernel/smp.c +++ b/arch/arm64/kernel/smp.c @@ -32,6 +32,7 @@ #include <linux/irq_work.h> #include <linux/kernel_stat.h> #include <linux/kexec.h> + #include <linux/kvm_host.h> #include <asm/alternative.h> @@ -93,6 +94,167 @@ static inline int op_cpu_kill(unsigned int cpu) } #endif +#ifdef CONFIG_ARM64_CPU_PARK +struct cpu_park_section { + unsigned long exit; /* exit address of park look */ + unsigned long magic; /* maigc represent park state */ + char text[0]; /* text section of park */ +}; + +static int mmap_cpu_park_mem(void) +{ + if (!park_info.start) + return -ENOMEM; + + if (park_info.start_v) + return 0; + + park_info.start_v = (unsigned long)__ioremap(park_info.start, + park_info.len, + PAGE_KERNEL_EXEC); + if (!park_info.start_v) { + pr_warn("map park memory failed."); + return -ENOMEM; + } + + return 0; +} + +static inline unsigned long cpu_park_section_v(unsigned int cpu) +{ + return park_info.start_v + PARK_SECTION_SIZE * (cpu - 1); +} + +static inline unsigned long cpu_park_section_p(unsigned int cpu) +{ + return park_info.start + PARK_SECTION_SIZE * (cpu - 1); +} + +/* + * Write the secondary_entry to exit section of park state. + * Then the secondary cpu will jump straight into the kernel + * by the secondary_entry. + */ +static int write_park_exit(unsigned int cpu) +{ + struct cpu_park_section *park_section; + unsigned long *park_exit; + unsigned long *park_text; + + if (mmap_cpu_park_mem() != 0) + return -EPERM; + + park_section = (struct cpu_park_section *)cpu_park_section_v(cpu); + park_exit = &park_section->exit; + park_text = (unsigned long *)park_section->text; + pr_debug("park_text 0x%lx : 0x%lx, do_cpu_park text 0x%lx : 0x%lx", + (unsigned long)park_text, *park_text, + (unsigned long)do_cpu_park, + *(unsigned long *)do_cpu_park); + + /* + * Test first 8 bytes to determine + * whether needs to write cpu park exit. + */ + if (*park_text == *(unsigned long *)do_cpu_park) { + writeq_relaxed(__pa_symbol(secondary_entry), park_exit); + __flush_dcache_area((__force void *)park_exit, + sizeof(unsigned long)); + flush_icache_range((unsigned long)park_exit, + (unsigned long)(park_exit + 1)); + sev(); + dsb(sy); + isb(); + + pr_debug("Write cpu %u secondary entry 0x%lx to 0x%lx.", + cpu, *park_exit, (unsigned long)park_exit); + pr_info("Boot cpu %u from PARK state.", cpu); + return 0; + } + + return -EPERM; +} + +/* Install cpu park sections for the specific cpu. */ +static int install_cpu_park(unsigned int cpu) +{ + struct cpu_park_section *park_section; + unsigned long *park_exit; + unsigned long *park_magic; + unsigned long park_text_len; + + park_section = (struct cpu_park_section *)cpu_park_section_v(cpu); + pr_debug("Install cpu park on cpu %u park exit 0x%lx park text 0x%lx", + cpu, (unsigned long)park_section, + (unsigned long)(park_section->text)); + + park_exit = &park_section->exit; + park_magic = &park_section->magic; + park_text_len = PARK_SECTION_SIZE - sizeof(struct cpu_park_section); + + *park_exit = 0UL; + *park_magic = 0UL; + memcpy((void *)park_section->text, do_cpu_park, park_text_len); + __flush_dcache_area((void *)park_section, PARK_SECTION_SIZE); + + return 0; +} + +static int uninstall_cpu_park(unsigned int cpu) +{ + unsigned long park_section; + + if (mmap_cpu_park_mem() != 0) + return -EPERM; + + park_section = cpu_park_section_v(cpu); + memset((void *)park_section, 0, PARK_SECTION_SIZE); + __flush_dcache_area((void *)park_section, PARK_SECTION_SIZE); + + return 0; +} + +static int cpu_wait_park(unsigned int cpu) +{ + long timeout; + struct cpu_park_section *park_section; + + volatile unsigned long *park_magic; + + park_section = (struct cpu_park_section *)cpu_park_section_v(cpu); + park_magic = &park_section->magic; + + timeout = USEC_PER_SEC; + while (*park_magic != PARK_MAGIC && timeout--) + udelay(1); + + if (timeout > 0) + pr_debug("cpu %u park done.", cpu); + else + pr_err("cpu %u park failed.", cpu); + + return *park_magic == PARK_MAGIC; +} + +static void cpu_park(unsigned int cpu) +{ + unsigned long park_section_p; + unsigned long park_exit_phy; + unsigned long do_park; + typeof(enter_cpu_park) *park; + + park_section_p = cpu_park_section_p(cpu); + park_exit_phy = park_section_p; + pr_debug("Go to park cpu %u exit address 0x%lx", cpu, park_exit_phy); + + do_park = park_section_p + sizeof(struct cpu_park_section); + park = (void *)__pa_symbol(enter_cpu_park); + + cpu_install_idmap(); + park(do_park, park_exit_phy); + unreachable(); +} +#endif /* * Boot a secondary CPU, and assign it the specified idle task. @@ -102,6 +264,10 @@ static int boot_secondary(unsigned int cpu, struct task_struct *idle) { const struct cpu_operations *ops = get_cpu_ops(cpu); +#ifdef CONFIG_ARM64_CPU_PARK + if (write_park_exit(cpu) == 0) + return 0; +#endif if (ops->cpu_boot) return ops->cpu_boot(cpu); @@ -131,6 +297,9 @@ int __cpu_up(unsigned int cpu, struct task_struct *idle) return ret; } +#ifdef CONFIG_ARM64_CPU_PARK + uninstall_cpu_park(cpu); +#endif /* * CPU was successfully started, wait for it to come online or * time out. @@ -844,10 +1013,32 @@ void arch_irq_work_raise(void) static void local_cpu_stop(void) { +#ifdef CONFIG_ARM64_CPU_PARK + int cpu; + const struct cpu_operations *ops = NULL; +#endif + set_cpu_online(smp_processor_id(), false); local_daif_mask(); sdei_mask_local_cpu(); + +#ifdef CONFIG_ARM64_CPU_PARK + /* + * Go to cpu park state. + * Otherwise go to cpu die. + */ + cpu = smp_processor_id(); + if (kexec_in_progress && park_info.start_v) { + machine_kexec_mask_interrupts(); + cpu_park(cpu); + + ops = get_cpu_ops(cpu); + if (ops && ops->cpu_die) + ops->cpu_die(cpu); + } +#endif + cpu_park_loop(); } @@ -1053,6 +1244,45 @@ void smp_send_stop(void) sdei_mask_local_cpu(); } +#ifdef CONFIG_ARM64_CPU_PARK +int kexec_smp_send_park(void) +{ + unsigned long cpu; + + if (WARN_ON(!kexec_in_progress)) { + pr_crit("%s called not in kexec progress.", __func__); + return -EPERM; + } + + if (mmap_cpu_park_mem() != 0) { + pr_info("no cpuparkmem, goto normal way."); + return -EPERM; + } + + local_irq_disable(); + + if (num_online_cpus() > 1) { + cpumask_t mask; + + cpumask_copy(&mask, cpu_online_mask); + cpumask_clear_cpu(smp_processor_id(), &mask); + + for_each_cpu(cpu, &mask) + install_cpu_park(cpu); + smp_cross_call(&mask, IPI_CPU_STOP); + + /* Wait for other CPUs to park */ + for_each_cpu(cpu, &mask) + cpu_wait_park(cpu); + pr_info("smp park other cpus done\n"); + } + + sdei_mask_local_cpu(); + + return 0; +} +#endif + #ifdef CONFIG_KEXEC_CORE void crash_smp_send_stop(void) { diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index 794f992..d01259c 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -236,6 +236,57 @@ static void __init fdt_enforce_memory_region(void) memblock_add(usable_rgns[1].base, usable_rgns[1].size); } +#ifdef CONFIG_ARM64_CPU_PARK +struct cpu_park_info park_info = { + .start = 0, + .len = PARK_SECTION_SIZE * NR_CPUS, + .start_v = 0, +}; + +static int __init parse_park_mem(char *p) +{ + if (!p) + return 0; + + park_info.start = PAGE_ALIGN(memparse(p, NULL)); + if (park_info.start == 0) + pr_info("cpu park mem params[%s]", p); + + return 0; +} +early_param("cpuparkmem", parse_park_mem); + +static int __init reserve_park_mem(void) +{ + if (park_info.start == 0 || park_info.len == 0) + return 0; + + park_info.start = PAGE_ALIGN(park_info.start); + park_info.len = PAGE_ALIGN(park_info.len); + + if (!memblock_is_region_memory(park_info.start, park_info.len)) { + pr_warn("cannot reserve park mem: region is not memory!"); + goto out; + } + + if (memblock_is_region_reserved(park_info.start, park_info.len)) { + pr_warn("cannot reserve park mem: region overlaps reserved memory!"); + goto out; + } + + memblock_remove(park_info.start, park_info.len); + pr_info("cpu park mem reserved: 0x%016lx - 0x%016lx (%ld MB)", + park_info.start, park_info.start + park_info.len, + park_info.len >> 20); + + return 0; +out: + park_info.start = 0; + park_info.len = 0; + return -EINVAL; +} +#endif + void __init arm64_memblock_init(void) { const s64 linear_region_size = BIT(vabits_actual - 1); @@ -357,6 +408,10 @@ void __init arm64_memblock_init(void) reserve_crashkernel(); +#ifdef CONFIG_ARM64_CPU_PARK + reserve_park_mem(); +#endif + reserve_elfcorehdr(); high_memory = __va(memblock_end_of_DRAM() - 1) + 1; -- 2.9.5

1 1

[PATCH openEuler-1.0-LTS 00/65] MPAM bugfix @ 20210224
by Cheng Jian 24 Feb '21

24 Feb '21

MPAM bugfix @ 20210224 James Morse (10): arm64/mpam: Add mpam driver discovery phase and kbuild boiler plate cacheinfo: Provide a helper to find a cacheinfo leaf arm64/mpam: Probe supported partid/pmg ranges from devices arm64/mpam: Supplement MPAM MSC register layout definitions arm64/mpam: Probe the features resctrl supports arm64/mpam: Reset controls when CPUs come online arm64/mpam: Summarize feature support during mpam_enable() arm64/mpam: resctrl: Re-synchronise resctrl's view of online CPUs drivers: base: cacheinfo: Add helper to search cacheinfo by of_node arm64/mpam: Enabling registering and logging error interrupts Wang ShaoBo (55): arm64/mpam: Preparing for MPAM refactoring arm64/mpam: Add helper for getting mpam sysprops arm64/mpam: Allocate mpam component configuration arrays arm64/mpam: Pick MPAM resources and events for resctrl_res exported arm64/mpam: Init resctrl resources' info from resctrl_res selected arm64/mpam: resctrl: Handle cpuhp and resctrl_dom allocation arm64/mpam: Implement helpers for handling configuration and monitoring arm64/mpam: Migrate old MSCs' discovery process to new branch arm64/mpam: Add helper for getting MSCs' configuration arm64/mpam: Probe partid,pmg and feature capabilities' ranges from classes arm64/mpam: resctrl: Rebuild configuration and monitoring pipeline arm64/mpam: resctrl: Append schemata CDP definitions arm64/mpam: resctrl: Supplement cdpl2,cdpl3 for mount options arm64/mpam: resctrl: Add helpers for init and destroy schemata list arm64/mpam: resctrl: Use resctrl_group_init_alloc() to init schema list arm64/mpam: resctrl: Write and read schemata by schema_list arm64/mpam: Support cdp in mpam_sched_in() arm64/mpam: resctrl: Update resources reset process arm64/mpam: resctrl: Update closid alloc and free process with bitmap arm64/mpam: resctrl: Move ctrlmon sysfile write/read function to mpam_ctrlmon.c arm64/mpam: Support cdp on allocating monitors arm64/mpam: resctrl: Support cdp on monitoring data arm64/mpam: Clean up header files and rearrange declarations arm64/mpam: resctrl: Remove ctrlmon sysfile arm64/mpam: resctrl: Remove unnecessary CONFIG_ARM64 arm64/mpam: Implement intpartid narrowing process arm64/mpam: Using software-defined id for rdtgroup instead of 32-bit integer arm64/mpam: resctrl: collect child mon group's monitor data arm64/mpam: resctrl: Support cpus' monitoring for mon group arm64/mpam: resctrl: Support priority and hardlimit(Memory bandwidth) configuration arm64/mpam: Store intpri and dspri for mpam device reset arm64/mpam: Squash default priority from mpam device to class arm64/mpam: Restore extend ctrls' max width for checking schemata input arm64/mpam: Re-plan intpartid narrowing process arm64/mpam: Add hook-events id for ctrl features arm64/mpam: Integrate monitor data for Memory Bandwidth if cdp enabled arm64/mpam: Fix MPAM_ESR intPARTID_range error arm64/mpam: Separate internal and downstream priority event arm64/mpam: Remap reqpartid,pmg to rmid and intpartid to closid arm64/mpam: Add wait queue for monitor alloc and free arm64/mpam: Add resctrl_ctrl_feature structure to manage ctrl features arm64/mpam: resctrl: Export resource's properties to info directory arm64/mpam: Split header files into suitable location arm64/mpam: resctrl: Add rmid file in resctrl sysfs arm64/mpam: Filter schema control type with ctrl features arm64/mpam: Simplify mpamid cdp mapping process arm64/mpam: Set per-cpu's closid to none zero for cdp ACPI/MPAM: Use acpi_map_pxm_to_node() to get node id for memory node arm64/mpam: Supplement additional useful ctrl features for mount options arm64/mpam: resctrl: Add proper error handling to resctrl_mount() arm64/mpam: resctrl: Use resctrl_group_init_alloc() for default group arm64/mpam: resctrl: Allow setting register MPAMCFG_MBW_MIN to 0 arm64/mpam: resctrl: Refresh cpu mask for handling cpuhp arm64/mpam: Sort domains when cpu online arm64/mpam: Fix compile warning arch/arm64/include/asm/mpam.h | 324 +--- arch/arm64/include/asm/mpam_resource.h | 129 -- arch/arm64/include/asm/mpam_sched.h | 8 - arch/arm64/include/asm/resctrl.h | 514 +++++- arch/arm64/kernel/Makefile | 2 +- arch/arm64/kernel/mpam.c | 1499 ---------------- arch/arm64/kernel/mpam/Makefile | 3 + arch/arm64/kernel/mpam/mpam_ctrlmon.c | 961 ++++++++++ arch/arm64/kernel/mpam/mpam_device.c | 1706 ++++++++++++++++++ arch/arm64/kernel/mpam/mpam_device.h | 140 ++ arch/arm64/kernel/mpam/mpam_internal.h | 345 ++++ arch/arm64/kernel/mpam/mpam_mon.c | 334 ++++ arch/arm64/kernel/mpam/mpam_resctrl.c | 2240 ++++++++++++++++++++++++ arch/arm64/kernel/mpam/mpam_resource.h | 228 +++ arch/arm64/kernel/mpam/mpam_setup.c | 608 +++++++ arch/arm64/kernel/mpam_ctrlmon.c | 623 ------- arch/arm64/kernel/mpam_mon.c | 124 -- drivers/acpi/arm64/mpam.c | 87 +- drivers/base/cacheinfo.c | 38 + fs/resctrlfs.c | 396 +++-- include/linux/arm_mpam.h | 118 ++ include/linux/cacheinfo.h | 36 + include/linux/resctrlfs.h | 30 - 23 files changed, 7521 insertions(+), 2972 deletions(-) delete mode 100644 arch/arm64/include/asm/mpam_resource.h delete mode 100644 arch/arm64/kernel/mpam.c create mode 100644 arch/arm64/kernel/mpam/Makefile create mode 100644 arch/arm64/kernel/mpam/mpam_ctrlmon.c create mode 100644 arch/arm64/kernel/mpam/mpam_device.c create mode 100644 arch/arm64/kernel/mpam/mpam_device.h create mode 100644 arch/arm64/kernel/mpam/mpam_internal.h create mode 100644 arch/arm64/kernel/mpam/mpam_mon.c create mode 100644 arch/arm64/kernel/mpam/mpam_resctrl.c create mode 100644 arch/arm64/kernel/mpam/mpam_resource.h create mode 100644 arch/arm64/kernel/mpam/mpam_setup.c delete mode 100644 arch/arm64/kernel/mpam_ctrlmon.c delete mode 100644 arch/arm64/kernel/mpam_mon.c create mode 100644 include/linux/arm_mpam.h -- 2.25.1

1 65

[PATCH kernel-4.19] drivers: TCM: add GMJS(Nationz Tech) TCM driver.
by Xie XiuQi 23 Feb '21

23 Feb '21

From: Li ZhiGang <lizhigang(a)kylinos.cn> Nationz Tech TCM are used for trusted computing, the chip attached via SPI or LPC. We have a brief verify/test with this driver on KunPeng920 + openEuler system, with externally compiled module. Signed-off-by: Li ZhiGang <lizhigang(a)kylinos.cn> --- drivers/staging/Kconfig | 2 + drivers/staging/Makefile | 2 + drivers/staging/gmjstcm/Kconfig | 21 + drivers/staging/gmjstcm/Makefile | 5 + drivers/staging/gmjstcm/tcm.c | 949 ++++++++++++++++++++++++++ drivers/staging/gmjstcm/tcm.h | 123 ++++ drivers/staging/gmjstcm/tcm_tis_spi.c | 868 +++++++++++++++++++++++ 7 files changed, 1970 insertions(+) create mode 100644 drivers/staging/gmjstcm/Kconfig create mode 100644 drivers/staging/gmjstcm/Makefile create mode 100644 drivers/staging/gmjstcm/tcm.c create mode 100644 drivers/staging/gmjstcm/tcm.h create mode 100644 drivers/staging/gmjstcm/tcm_tis_spi.c diff --git a/drivers/staging/Kconfig b/drivers/staging/Kconfig index 1abf76be2aa8..d51fa4f4e7ca 100644 --- a/drivers/staging/Kconfig +++ b/drivers/staging/Kconfig @@ -126,4 +126,6 @@ source "drivers/staging/axis-fifo/Kconfig" source "drivers/staging/erofs/Kconfig" +source "drivers/staging/gmjstcm/Kconfig" + endif # STAGING diff --git a/drivers/staging/Makefile b/drivers/staging/Makefile index ab0cbe8815b1..6d41915dad5b 100644 --- a/drivers/staging/Makefile +++ b/drivers/staging/Makefile @@ -53,3 +53,5 @@ obj-$(CONFIG_SOC_MT7621) += mt7621-dts/ obj-$(CONFIG_STAGING_GASKET_FRAMEWORK) += gasket/ obj-$(CONFIG_XIL_AXIS_FIFO) += axis-fifo/ obj-$(CONFIG_EROFS_FS) += erofs/ +obj-$(CONFIG_GMJS_TCM) += gmjstcm/ + diff --git a/drivers/staging/gmjstcm/Kconfig b/drivers/staging/gmjstcm/Kconfig new file mode 100644 index 000000000000..5b5397ae1832 --- /dev/null +++ b/drivers/staging/gmjstcm/Kconfig @@ -0,0 +1,21 @@ +menu "GMJS TCM support" + +config GMJS_TCM + bool + +config GMJS_TCM_CORE + tristate "GMJS TCM core support" + depends on ARM64 || MIPS + default m + select GMJS_TCM + help + GMJS TCM core support. + +config GMJS_TCM_SPI + tristate "GMJS TCM support on SPI interface" + depends on GMJS_TCM_CORE && SPI_MASTER + default m + help + GMJS TCM support on SPI interface. + +endmenu diff --git a/drivers/staging/gmjstcm/Makefile b/drivers/staging/gmjstcm/Makefile new file mode 100644 index 000000000000..601c78e44793 --- /dev/null +++ b/drivers/staging/gmjstcm/Makefile @@ -0,0 +1,5 @@ + +obj-$(CONFIG_GMJS_TCM_CORE) += tcm_core.o +tcm_core-objs := tcm.o +obj-$(CONFIG_GMJS_TCM_SPI) += tcm_tis_spi.o + diff --git a/drivers/staging/gmjstcm/tcm.c b/drivers/staging/gmjstcm/tcm.c new file mode 100644 index 000000000000..5c41bfa8b423 --- /dev/null +++ b/drivers/staging/gmjstcm/tcm.c @@ -0,0 +1,949 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2009 Nationz Technologies Inc. + * + * Description: Exprot symbol for tcm_tis module + * + * Major Function: public write read register function etc. + * + */ + +#include <linux/sched.h> +#include <linux/poll.h> +#include <linux/spinlock.h> +#include <linux/timer.h> +#include "tcm.h" + +/* + * const var + */ +enum tcm_const { + TCM_MINOR = 224, /* officially assigned */ + TCM_BUFSIZE = 2048, /* Buffer Size */ + TCM_NUM_DEVICES = 256, /* Max supporting tcm device number */ +}; + +/* + * CMD duration + */ +enum tcm_duration { + TCM_SHORT = 0, + TCM_MEDIUM = 1, + TCM_LONG = 2, + TCM_UNDEFINED, +}; + +/* Max Total of Command Number */ +#define TCM_MAX_ORDINAL 88 /*243*/ + +static LIST_HEAD(tcm_chip_list); +static DEFINE_SPINLOCK(driver_lock); /* spin lock */ +static DECLARE_BITMAP(dev_mask, TCM_NUM_DEVICES); + +typedef struct tagTCM_Command { + u8 ordinal; + u8 DURATION; +} TCM_Command; + +static const TCM_Command TCM_Command_List[TCM_MAX_ORDINAL + 1] = { + {/*TCM_ORD_ActivateIdentity, */122, 1}, + {/*TCM_ORD_CertifyKey, */50, 1}, + {/*TCM_ORD_CertifyKeyM, */51, 1}, + {/*TCM_ORD_ChangeAuth, */12, 1}, + {/*TCM_ORD_ChangeAuthOwner, */16, 0}, + {/*TCM_ORD_ContinueSelfTeSt, */83, 2}, + {/*TCM_ORD_CreateCounter, */220, 0}, + {/*TCM_ORD_CreateWrapKey, */31, 2}, + {/*TCM_ORD_DiSableForceClear, */94, 0}, + {/*TCM_ORD_DiSableOwnerClear, */92, 0}, + {/*TCM_ORD_EStabliShTranSport, */230, 0}, + {/*TCM_ORD_ExecuteTranSport, */231, 2}, + {/*TCM_ORD_Extend, */20, 0}, + {/*TCM_ORD_FieldUpgrade, */170, 2}, + {/*TCM_ORD_FluShSpecific, */186, 0}, + {/*TCM_ORD_ForceClear, */93, 0}, + {/*TCM_ORD_GetAuditDigeSt, */133, 0}, + {/*TCM_ORD_GetAuditDigeStSigned, */134, 1}, + {/*TCM_ORD_GetCapability, */101, 0}, + {/*TCM_ORD_GetPubKey, */33, 0}, + {/*TCM_ORD_GetRandoM, */70, 0}, + {/*TCM_ORD_GetTeStReSult, */84, 0}, + {/*TCM_ORD_GetTickS, */241, 0}, + {/*TCM_ORD_IncreMentCounter, */221, 0}, + {/*TCM_ORD_LoadContext, */185, 1}, + {/*TCM_ORD_MakeIdentity, */121, 2}, + {/*TCM_ORD_NV_DefineSpace, */204, 0}, + {/*TCM_ORD_NV_ReadValue, */207, 0}, + {/*TCM_ORD_NV_ReadValueAuth, */208, 0}, + {/*TCM_ORD_NV_WriteValue, */205, 0}, + {/*TCM_ORD_NV_WriteValueAuth, */206, 0}, + {/*TCM_ORD_OwnerClear, */91, 0}, + {/*TCM_ORD_OwnerReadInternalPub, */129, 0}, + {/*TCM_ORD_OwnerSetDiSable, */110, 0}, + {/*TCM_ORD_PCR_ReSet, */200, 0}, + {/*TCM_ORD_PcrRead, */21, 0}, + {/*TCM_ORD_PhySicalDiSable, */112, 0}, + {/*TCM_ORD_PhySicalEnable, */111, 0}, + {/*TCM_ORD_PhySicalSetDeactivated, */114, 0}, + {/*TCM_ORD_Quote, */22, 1}, + {/*TCM_ORD_QuoteM, */62, 1}, + {/*TCM_ORD_ReadCounter, */222, 0}, + {/*TCM_ORD_ReadPubek, */124, 0}, + {/*TCM_ORD_ReleaSeCounter, */223, 0}, + {/*TCM_ORD_ReleaSeCounterOwner, */224, 0}, + {/*TCM_ORD_ReleaSeTranSportSigned, */232, 1}, + {/*TCM_ORD_ReSetLockValue, */64, 0}, + {/*TCM_ORD_RevokeTruSt, */128, 0}, + {/*TCM_ORD_SaveContext, */184, 1}, + {/*TCM_ORD_SaveState, */152, 1}, + {/*TCM_ORD_Seal, */23, 1}, + {/*TCM_ORD_Sealx, */61, 1}, + {/*TCM_ORD_SelfTeStFull, */80, 2}, + {/*TCM_ORD_SetCapability, */63, 0}, + {/*TCM_ORD_SetOperatorAuth, */116, 0}, + {/*TCM_ORD_SetOrdinalAuditStatuS, */141, 0}, + {/*TCM_ORD_SetOwnerInStall, */113, 0}, + {/*TCM_ORD_SetTeMpDeactivated, */115, 0}, + {/*TCM_ORD_Sign, */60, 1}, + {/*TCM_ORD_Startup, */153, 0}, + {/*TCM_ORD_TakeOwnerShip, */13, 1}, + {/*TCM_ORD_TickStaMpBlob, */242, 1}, + {/*TCM_ORD_UnSeal, */24, 1}, + {/*TSC_ORD_PhySicalPreSence, */10, 0}, + {/*TSC_ORD_ReSetEStabliShMentBit, */11, 0}, + {/*TCM_ORD_WrapKey, */189, 2}, + {/*TCM_ORD_APcreate, */191, 0}, + {/*TCM_ORD_APTerMinate, */192, 0}, + {/*TCM_ORD_CreateMigratedBlob, */193, 1}, + {/*TCM_ORD_ConvertMigratedBlob, */194, 1}, + {/*TCM_ORD_AuthorizeMigrationKey, */195, 0}, + {/*TCM_ORD_SMS4Encrypt, */197, 1}, + {/*TCM_ORD_SMS4Decrypt, */198, 1}, + {/*TCM_ORD_ReadEKCert, */199, 1}, + {/*TCM_ORD_WriteEKCert, */233, 1}, + {/*TCM_ORD_SCHStart, */234, 0}, + {/*TCM_ORD_SCHUpdata, */235, 0}, + {/*TCM_ORD_SCHCoMplete, */236, 0}, + {/*TCM_ORD_SCHCoMpleteExtend, */237, 0}, + {/*TCM_ORD_ECCDecrypt, */238, 1}, + {/*TCM_ORD_LoadKey, */239, 1}, + {/*TCM_ORD_CreateEndorSeMentKeyPair, */120, 2}, + {/*TCM_ORD_CreateRevocableEK, */127, 2}, + {/*TCM_ORD_ReleaSeECCExchangeSeSSion, */174, 1}, + {/*TCM_ORD_CreateECCExchangeSeSSion, */175, 1}, + {/*TCM_ORD_GetKeyECCExchangeSeSSion, */176, 1}, + {/*TCM_ORD_ActivatePEK, */217, 1}, + {/*TCM_ORD_ActivatePEKCert, */218, 1}, + {0, 0} +}; + +static void user_reader_timeout(struct timer_list *t) +{ + struct tcm_chip *chip = from_timer(chip, t, user_read_timer); + + schedule_work(&chip->work); +} + +static void timeout_work(struct work_struct *work) +{ + struct tcm_chip *chip = container_of(work, struct tcm_chip, work); + + mutex_lock(&chip->buffer_mutex); + atomic_set(&chip->data_pending, 0); + memset(chip->data_buffer, 0, TCM_BUFSIZE); + mutex_unlock(&chip->buffer_mutex); +} + +unsigned long tcm_calc_ordinal_duration(struct tcm_chip *chip, + u32 ordinal) +{ + int duration_idx = TCM_UNDEFINED; + int duration = 0; + int i = 0; + + for (i = 0; i < TCM_MAX_ORDINAL; i++) { + if (ordinal == TCM_Command_List[i].ordinal) { + duration_idx = TCM_Command_List[i].DURATION; + break; + } + } + + if (duration_idx != TCM_UNDEFINED) + duration = chip->vendor.duration[duration_idx]; + if (duration <= 0) + return 2 * 60 * HZ; + else + return duration; +} +EXPORT_SYMBOL_GPL(tcm_calc_ordinal_duration); + +/* + * Internal kernel interface to transmit TCM commands + * buff format: TAG(2 bytes) + Total Size(4 bytes ) + + * Command Ordinal(4 bytes ) + ...... + */ +static ssize_t tcm_transmit(struct tcm_chip *chip, const char *buf, + size_t bufsiz) +{ + ssize_t rc = 0; + u32 count = 0, ordinal = 0; + unsigned long stop = 0; + + count = be32_to_cpu(*((__be32 *)(buf + 2))); /* buff size */ + ordinal = be32_to_cpu(*((__be32 *)(buf + 6))); /* command ordinal */ + + if (count == 0) + return -ENODATA; + if (count > bufsiz) { /* buff size err ,invalid buff stream */ + dev_err(chip->dev, "invalid count value %x, %zx\n", + count, bufsiz); + return -E2BIG; + } + + mutex_lock(&chip->tcm_mutex); /* enter mutex */ + + rc = chip->vendor.send(chip, (u8 *)buf, count); + if (rc < 0) { + dev_err(chip->dev, "%s: tcm_send: error %zd\n", + __func__, rc); + goto out; + } + + if (chip->vendor.irq) + goto out_recv; + + stop = jiffies + tcm_calc_ordinal_duration(chip, + ordinal); /* cmd duration */ + do { + u8 status = chip->vendor.status(chip); + + if ((status & chip->vendor.req_complete_mask) == + chip->vendor.req_complete_val) + goto out_recv; + + if ((status == chip->vendor.req_canceled)) { + dev_err(chip->dev, "Operation Canceled\n"); + rc = -ECANCELED; + goto out; + } + + msleep(TCM_TIMEOUT); /* CHECK */ + rmb(); + } while (time_before(jiffies, stop)); + /* time out */ + chip->vendor.cancel(chip); + dev_err(chip->dev, "Operation Timed out\n"); + rc = -ETIME; + goto out; + +out_recv: + rc = chip->vendor.recv(chip, (u8 *)buf, bufsiz); + if (rc < 0) + dev_err(chip->dev, "%s: tcm_recv: error %zd\n", + __func__, rc); +out: + mutex_unlock(&chip->tcm_mutex); + return rc; +} + +#define TCM_DIGEST_SIZE 32 +#define TCM_ERROR_SIZE 10 +#define TCM_RET_CODE_IDX 6 +#define TCM_GET_CAP_RET_SIZE_IDX 10 +#define TCM_GET_CAP_RET_UINT32_1_IDX 14 +#define TCM_GET_CAP_RET_UINT32_2_IDX 18 +#define TCM_GET_CAP_RET_UINT32_3_IDX 22 +#define TCM_GET_CAP_RET_UINT32_4_IDX 26 +#define TCM_GET_CAP_PERM_DISABLE_IDX 16 +#define TCM_GET_CAP_PERM_INACTIVE_IDX 18 +#define TCM_GET_CAP_RET_BOOL_1_IDX 14 +#define TCM_GET_CAP_TEMP_INACTIVE_IDX 16 + +#define TCM_CAP_IDX 13 +#define TCM_CAP_SUBCAP_IDX 21 + +enum tcm_capabilities { + TCM_CAP_FLAG = 4, + TCM_CAP_PROP = 5, +}; + +enum tcm_sub_capabilities { + TCM_CAP_PROP_PCR = 0x1, /* tcm 0x101 */ + TCM_CAP_PROP_MANUFACTURER = 0x3, /* tcm 0x103 */ + TCM_CAP_FLAG_PERM = 0x8, /* tcm 0x108 */ + TCM_CAP_FLAG_VOL = 0x9, /* tcm 0x109 */ + TCM_CAP_PROP_OWNER = 0x11, /* tcm 0x101 */ + TCM_CAP_PROP_TIS_TIMEOUT = 0x15, /* tcm 0x115 */ + TCM_CAP_PROP_TIS_DURATION = 0x20, /* tcm 0x120 */ +}; + +/* + * This is a semi generic GetCapability command for use + * with the capability type TCM_CAP_PROP or TCM_CAP_FLAG + * and their associated sub_capabilities. + */ + +static const u8 tcm_cap[] = { + 0, 193, /* TCM_TAG_RQU_COMMAND 0xc1*/ + 0, 0, 0, 22, /* length */ + 0, 0, 128, 101, /* TCM_ORD_GetCapability */ + 0, 0, 0, 0, /* TCM_CAP_<TYPE> */ + 0, 0, 0, 4, /* TCM_CAP_SUB_<TYPE> size */ + 0, 0, 1, 0 /* TCM_CAP_SUB_<TYPE> */ +}; + +static ssize_t transmit_cmd(struct tcm_chip *chip, u8 *data, int len, + char *desc) +{ + int err = 0; + + len = tcm_transmit(chip, data, len); + if (len < 0) + return len; + if (len == TCM_ERROR_SIZE) { + err = be32_to_cpu(*((__be32 *)(data + TCM_RET_CODE_IDX))); + dev_dbg(chip->dev, "A TCM error (%d) occurred %s\n", err, desc); + return err; + } + return 0; +} + +/* + * Get default timeouts value form tcm by GetCapability with TCM_CAP_PROP_TIS_TIMEOUT prop + */ +void tcm_get_timeouts(struct tcm_chip *chip) +{ + u8 data[max_t(int, ARRAY_SIZE(tcm_cap), 30)]; + ssize_t rc = 0; + u32 timeout = 0; + + memcpy(data, tcm_cap, sizeof(tcm_cap)); + data[TCM_CAP_IDX] = TCM_CAP_PROP; + data[TCM_CAP_SUBCAP_IDX] = TCM_CAP_PROP_TIS_TIMEOUT; + + rc = transmit_cmd(chip, data, sizeof(data), + "attempting to determine the timeouts"); + if (rc) + goto duration; + + if (be32_to_cpu(*((__be32 *)(data + TCM_GET_CAP_RET_SIZE_IDX))) != + 4 * sizeof(u32)) + goto duration; + + /* Don't overwrite default if value is 0 */ + timeout = be32_to_cpu(*((__be32 *)(data + TCM_GET_CAP_RET_UINT32_1_IDX))); + if (timeout) + chip->vendor.timeout_a = msecs_to_jiffies(timeout); + timeout = be32_to_cpu(*((__be32 *)(data + TCM_GET_CAP_RET_UINT32_2_IDX))); + if (timeout) + chip->vendor.timeout_b = msecs_to_jiffies(timeout); + timeout = be32_to_cpu(*((__be32 *)(data + TCM_GET_CAP_RET_UINT32_3_IDX))); + if (timeout) + chip->vendor.timeout_c = msecs_to_jiffies(timeout); + timeout = be32_to_cpu(*((__be32 *)(data + TCM_GET_CAP_RET_UINT32_4_IDX))); + if (timeout) + chip->vendor.timeout_d = msecs_to_jiffies(timeout); + +duration: + memcpy(data, tcm_cap, sizeof(tcm_cap)); + data[TCM_CAP_IDX] = TCM_CAP_PROP; + data[TCM_CAP_SUBCAP_IDX] = TCM_CAP_PROP_TIS_DURATION; + + rc = transmit_cmd(chip, data, sizeof(data), + "attempting to determine the durations"); + if (rc) + return; + + if (be32_to_cpu(*((__be32 *)(data + TCM_GET_CAP_RET_SIZE_IDX))) != + 3 * sizeof(u32)) + return; + + chip->vendor.duration[TCM_SHORT] = + msecs_to_jiffies(be32_to_cpu(*((__be32 *)(data + + TCM_GET_CAP_RET_UINT32_1_IDX)))); + chip->vendor.duration[TCM_MEDIUM] = + msecs_to_jiffies(be32_to_cpu(*((__be32 *)(data + + TCM_GET_CAP_RET_UINT32_2_IDX)))); + chip->vendor.duration[TCM_LONG] = + msecs_to_jiffies(be32_to_cpu(*((__be32 *)(data + + TCM_GET_CAP_RET_UINT32_3_IDX)))); +} +EXPORT_SYMBOL_GPL(tcm_get_timeouts); + +ssize_t tcm_show_enabled(struct device *dev, struct device_attribute *attr, + char *buf) +{ + u8 data[max_t(int, ARRAY_SIZE(tcm_cap), 35)]; + ssize_t rc = 0; + struct tcm_chip *chip = dev_get_drvdata(dev); + + if (chip == NULL) + return -ENODEV; + + memcpy(data, tcm_cap, sizeof(tcm_cap)); + data[TCM_CAP_IDX] = TCM_CAP_FLAG; + data[TCM_CAP_SUBCAP_IDX] = TCM_CAP_FLAG_PERM; + + rc = transmit_cmd(chip, data, sizeof(data), + "attemtping to determine the permanent state"); + if (rc) + return 0; + if (data[TCM_GET_CAP_PERM_DISABLE_IDX]) + return sprintf(buf, "disable\n"); + else + return sprintf(buf, "enable\n"); +} +EXPORT_SYMBOL_GPL(tcm_show_enabled); + +ssize_t tcm_show_active(struct device *dev, struct device_attribute *attr, + char *buf) +{ + u8 data[max_t(int, ARRAY_SIZE(tcm_cap), 35)]; + ssize_t rc = 0; + struct tcm_chip *chip = dev_get_drvdata(dev); + + if (chip == NULL) + return -ENODEV; + + memcpy(data, tcm_cap, sizeof(tcm_cap)); + data[TCM_CAP_IDX] = TCM_CAP_FLAG; + data[TCM_CAP_SUBCAP_IDX] = TCM_CAP_FLAG_PERM; + + rc = transmit_cmd(chip, data, sizeof(data), + "attemtping to determine the permanent state"); + if (rc) + return 0; + if (data[TCM_GET_CAP_PERM_INACTIVE_IDX]) + return sprintf(buf, "deactivated\n"); + else + return sprintf(buf, "activated\n"); +} +EXPORT_SYMBOL_GPL(tcm_show_active); + +ssize_t tcm_show_owned(struct device *dev, struct device_attribute *attr, + char *buf) +{ + u8 data[sizeof(tcm_cap)]; + ssize_t rc = 0; + struct tcm_chip *chip = dev_get_drvdata(dev); + + if (chip == NULL) + return -ENODEV; + + memcpy(data, tcm_cap, sizeof(tcm_cap)); + data[TCM_CAP_IDX] = TCM_CAP_PROP; + data[TCM_CAP_SUBCAP_IDX] = TCM_CAP_PROP_OWNER; + + rc = transmit_cmd(chip, data, sizeof(data), + "attempting to determine the owner state"); + if (rc) + return 0; + if (data[TCM_GET_CAP_RET_BOOL_1_IDX]) + return sprintf(buf, "Owner installed\n"); + else + return sprintf(buf, "Owner have not installed\n"); +} +EXPORT_SYMBOL_GPL(tcm_show_owned); + +ssize_t tcm_show_temp_deactivated(struct device *dev, + struct device_attribute *attr, char *buf) +{ + u8 data[sizeof(tcm_cap)]; + ssize_t rc = 0; + struct tcm_chip *chip = dev_get_drvdata(dev); + + if (chip == NULL) + return -ENODEV; + + memcpy(data, tcm_cap, sizeof(tcm_cap)); + data[TCM_CAP_IDX] = TCM_CAP_FLAG; + data[TCM_CAP_SUBCAP_IDX] = TCM_CAP_FLAG_VOL; + + rc = transmit_cmd(chip, data, sizeof(data), + "attempting to determine the temporary state"); + if (rc) + return 0; + if (data[TCM_GET_CAP_TEMP_INACTIVE_IDX]) + return sprintf(buf, "Temp deactivated\n"); + else + return sprintf(buf, "activated\n"); +} +EXPORT_SYMBOL_GPL(tcm_show_temp_deactivated); + +static const u8 pcrread[] = { + 0, 193, /* TCM_TAG_RQU_COMMAND */ + 0, 0, 0, 14, /* length */ + 0, 0, 128, 21, /* TCM_ORD_PcrRead */ + 0, 0, 0, 0 /* PCR index */ +}; + +ssize_t tcm_show_pcrs(struct device *dev, struct device_attribute *attr, + char *buf) +{ + u8 data[1024]; + ssize_t rc = 0; + int i = 0, j = 0, num_pcrs = 0; + __be32 index = 0; + char *str = buf; + struct tcm_chip *chip = dev_get_drvdata(dev); + + if (chip == NULL) + return -ENODEV; + + memcpy(data, tcm_cap, sizeof(tcm_cap)); + data[TCM_CAP_IDX] = TCM_CAP_PROP; + data[TCM_CAP_SUBCAP_IDX] = TCM_CAP_PROP_PCR; + + rc = transmit_cmd(chip, data, sizeof(data), + "attempting to determine the number of PCRS"); + if (rc) + return 0; + + num_pcrs = be32_to_cpu(*((__be32 *)(data + 14))); + for (i = 0; i < num_pcrs; i++) { + memcpy(data, pcrread, sizeof(pcrread)); + index = cpu_to_be32(i); + memcpy(data + 10, &index, 4); + rc = transmit_cmd(chip, data, sizeof(data), + "attempting to read a PCR"); + if (rc) + goto out; + str += sprintf(str, "PCR-%02d: ", i); + for (j = 0; j < TCM_DIGEST_SIZE; j++) + str += sprintf(str, "%02X ", *(data + 10 + j)); + str += sprintf(str, "\n"); + memset(data, 0, 1024); + } +out: + return str - buf; +} +EXPORT_SYMBOL_GPL(tcm_show_pcrs); + +#define READ_PUBEK_RESULT_SIZE 128 +static const u8 readpubek[] = { + 0, 193, /* TCM_TAG_RQU_COMMAND */ + 0, 0, 0, 42, /* length */ + 0, 0, 128, 124, /* TCM_ORD_ReadPubek */ + 0, 0, 0, 0, 0, 0, 0, 0, /* NONCE */ + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0 +}; + +ssize_t tcm_show_pubek(struct device *dev, struct device_attribute *attr, + char *buf) +{ + u8 data[READ_PUBEK_RESULT_SIZE] = {0}; + ssize_t err = 0; + int i = 0, rc = 0; + char *str = buf; + struct tcm_chip *chip = dev_get_drvdata(dev); + + if (chip == NULL) + return -ENODEV; + + memcpy(data, readpubek, sizeof(readpubek)); + + err = transmit_cmd(chip, data, sizeof(data), + "attempting to read the PUBEK"); + if (err) + goto out; + + str += sprintf(str, "PUBEK:"); + for (i = 0 ; i < 65 ; i++) { + if ((i) % 16 == 0) + str += sprintf(str, "\n"); + str += sprintf(str, "%02X ", data[i+10]); + } + + str += sprintf(str, "\n"); +out: + rc = str - buf; + return rc; +} +EXPORT_SYMBOL_GPL(tcm_show_pubek); + +#define CAP_VERSION_1_1 6 +#define CAP_VERSION_1_2 0x1A +#define CAP_VERSION_IDX 13 +static const u8 cap_version[] = { + 0, 193, /* TCM_TAG_RQU_COMMAND */ + 0, 0, 0, 18, /* length */ + 0, 0, 128, 101, /* TCM_ORD_GetCapability */ + 0, 0, 0, 0, + 0, 0, 0, 0 +}; + +ssize_t tcm_show_caps(struct device *dev, struct device_attribute *attr, + char *buf) +{ + u8 data[max_t(int, max(ARRAY_SIZE(tcm_cap), ARRAY_SIZE(cap_version)), 30)]; + ssize_t rc = 0; + char *str = buf; + struct tcm_chip *chip = dev_get_drvdata(dev); + + if (chip == NULL) + return -ENODEV; + + memcpy(data, tcm_cap, sizeof(tcm_cap)); + data[TCM_CAP_IDX] = TCM_CAP_PROP; + data[TCM_CAP_SUBCAP_IDX] = TCM_CAP_PROP_MANUFACTURER; + + rc = transmit_cmd(chip, data, sizeof(data), + "attempting to determine the manufacturer"); + if (rc) + return 0; + + str += sprintf(str, "Manufacturer: 0x%x\n", + be32_to_cpu(*((__be32 *)(data + TCM_GET_CAP_RET_UINT32_1_IDX)))); + + memcpy(data, cap_version, sizeof(cap_version)); + data[CAP_VERSION_IDX] = CAP_VERSION_1_1; + rc = transmit_cmd(chip, data, sizeof(data), + "attempting to determine the 1.1 version"); + if (rc) + goto out; + + str += sprintf(str, "Firmware version: %02X.%02X.%02X.%02X\n", + (int)data[14], (int)data[15], (int)data[16], + (int)data[17]); + +out: + return str - buf; +} +EXPORT_SYMBOL_GPL(tcm_show_caps); + +ssize_t tcm_store_cancel(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct tcm_chip *chip = dev_get_drvdata(dev); + + if (chip == NULL) + return 0; + + chip->vendor.cancel(chip); + return count; +} +EXPORT_SYMBOL_GPL(tcm_store_cancel); + +/* + * Device file system interface to the TCM + * when App call file open in usr space ,this func will respone + */ +int tcm_open(struct inode *inode, struct file *file) +{ + int rc = 0, minor = iminor(inode); + struct tcm_chip *chip = NULL, *pos = NULL; + + spin_lock(&driver_lock); + + list_for_each_entry(pos, &tcm_chip_list, list) { + if (pos->vendor.miscdev.minor == minor) { + chip = pos; + break; + } + } + + if (chip == NULL) { + rc = -ENODEV; + goto err_out; + } + + if (chip->num_opens) { + dev_dbg(chip->dev, "Another process owns this TCM\n"); + rc = -EBUSY; + goto err_out; + } + + chip->num_opens++; + get_device(chip->dev); + + spin_unlock(&driver_lock); + + chip->data_buffer = kmalloc(TCM_BUFSIZE * sizeof(u8), GFP_KERNEL); + if (chip->data_buffer == NULL) { + chip->num_opens--; + put_device(chip->dev); + return -ENOMEM; + } + + atomic_set(&chip->data_pending, 0); + + file->private_data = chip; + return 0; + +err_out: + spin_unlock(&driver_lock); + return rc; +} +EXPORT_SYMBOL_GPL(tcm_open); + +int tcm_release(struct inode *inode, struct file *file) +{ + struct tcm_chip *chip = file->private_data; + + spin_lock(&driver_lock); + file->private_data = NULL; + chip->num_opens--; + del_singleshot_timer_sync(&chip->user_read_timer); + flush_work(&chip->work); + atomic_set(&chip->data_pending, 0); + put_device(chip->dev); + kfree(chip->data_buffer); + spin_unlock(&driver_lock); + return 0; +} +EXPORT_SYMBOL_GPL(tcm_release); + +ssize_t tcm_write(struct file *file, const char __user *buf, + size_t size, loff_t *off) +{ + struct tcm_chip *chip = file->private_data; + int in_size = size, out_size; + + /* + * cannot perform a write until the read has cleared + * either via tcm_read or a user_read_timer timeout + */ + while (atomic_read(&chip->data_pending) != 0) + msleep(TCM_TIMEOUT); + + mutex_lock(&chip->buffer_mutex); + + if (in_size > TCM_BUFSIZE) + in_size = TCM_BUFSIZE; + + if (copy_from_user(chip->data_buffer, (void __user *)buf, in_size)) { + mutex_unlock(&chip->buffer_mutex); + return -EFAULT; + } + + /* atomic tcm command send and result receive */ + out_size = tcm_transmit(chip, chip->data_buffer, TCM_BUFSIZE); + + if (out_size >= 0) { + atomic_set(&chip->data_pending, out_size); + mutex_unlock(&chip->buffer_mutex); + + /* Set a timeout by which the reader must come claim the result */ + mod_timer(&chip->user_read_timer, jiffies + (60 * HZ)); + } else + mutex_unlock(&chip->buffer_mutex); + + return in_size; +} +EXPORT_SYMBOL_GPL(tcm_write); + +ssize_t tcm_read(struct file *file, char __user *buf, + size_t size, loff_t *off) +{ + struct tcm_chip *chip = file->private_data; + int ret_size = 0; + + del_singleshot_timer_sync(&chip->user_read_timer); + flush_work(&chip->work); + ret_size = atomic_read(&chip->data_pending); + atomic_set(&chip->data_pending, 0); + if (ret_size > 0) { /* relay data */ + if (size < ret_size) + ret_size = size; + + mutex_lock(&chip->buffer_mutex); + if (copy_to_user(buf, chip->data_buffer, ret_size)) + ret_size = -EFAULT; + mutex_unlock(&chip->buffer_mutex); + } + + return ret_size; +} +EXPORT_SYMBOL_GPL(tcm_read); + +void tcm_remove_hardware(struct device *dev) +{ + struct tcm_chip *chip = dev_get_drvdata(dev); + + if (chip == NULL) { + dev_err(dev, "No device data found\n"); + return; + } + + spin_lock(&driver_lock); + list_del(&chip->list); + spin_unlock(&driver_lock); + + dev_set_drvdata(dev, NULL); + misc_deregister(&chip->vendor.miscdev); + kfree(chip->vendor.miscdev.name); + + sysfs_remove_group(&dev->kobj, chip->vendor.attr_group); + /* tcm_bios_log_teardown(chip->bios_dir); */ + + clear_bit(chip->dev_num, dev_mask); + kfree(chip); + put_device(dev); +} +EXPORT_SYMBOL_GPL(tcm_remove_hardware); + +static u8 savestate[] = { + 0, 193, /* TCM_TAG_RQU_COMMAND */ + 0, 0, 0, 10, /* blob length (in bytes) */ + 0, 0, 128, 152 /* TCM_ORD_SaveState */ +}; + +/* + * We are about to suspend. Save the TCM state + * so that it can be restored. + */ +int tcm_pm_suspend(struct device *dev, pm_message_t pm_state) +{ + struct tcm_chip *chip = dev_get_drvdata(dev); + + if (chip == NULL) + return -ENODEV; + + tcm_transmit(chip, savestate, sizeof(savestate)); + return 0; +} +EXPORT_SYMBOL_GPL(tcm_pm_suspend); + +int tcm_pm_suspend_p(struct device *dev) +{ + struct tcm_chip *chip = dev_get_drvdata(dev); + + if (chip == NULL) + return -ENODEV; + + tcm_transmit(chip, savestate, sizeof(savestate)); + return 0; +} +EXPORT_SYMBOL_GPL(tcm_pm_suspend_p); + +void tcm_startup(struct tcm_chip *chip) +{ + u8 start_up[] = { + 0, 193, /* TCM_TAG_RQU_COMMAND */ + 0, 0, 0, 12, /* blob length (in bytes) */ + 0, 0, 128, 153, /* TCM_ORD_SaveState */ + 0, 1 + }; + if (chip == NULL) + return; + tcm_transmit(chip, start_up, sizeof(start_up)); +} +EXPORT_SYMBOL_GPL(tcm_startup); + +/* + * Resume from a power safe. The BIOS already restored + * the TCM state. + */ +int tcm_pm_resume(struct device *dev) +{ + u8 start_up[] = { + 0, 193, /* TCM_TAG_RQU_COMMAND */ + 0, 0, 0, 12, /* blob length (in bytes) */ + 0, 0, 128, 153, /* TCM_ORD_SaveState */ + 0, 1 + }; + struct tcm_chip *chip = dev_get_drvdata(dev); + /* dev_info(chip->dev ,"--call tcm_pm_resume\n"); */ + if (chip == NULL) + return -ENODEV; + + tcm_transmit(chip, start_up, sizeof(start_up)); + return 0; +} +EXPORT_SYMBOL_GPL(tcm_pm_resume); + +/* + * Called from tcm_<specific>.c probe function only for devices + * the driver has determined it should claim. Prior to calling + * this function the specific probe function has called pci_enable_device + * upon errant exit from this function specific probe function should call + * pci_disable_device + */ +struct tcm_chip *tcm_register_hardware(struct device *dev, + const struct tcm_vendor_specific *entry) +{ + int rc; +#define DEVNAME_SIZE 7 + + char *devname = NULL; + struct tcm_chip *chip = NULL; + + /* Driver specific per-device data */ + chip = kzalloc(sizeof(*chip), GFP_KERNEL); + if (chip == NULL) { + dev_err(dev, "chip kzalloc err\n"); + return NULL; + } + + mutex_init(&chip->buffer_mutex); + mutex_init(&chip->tcm_mutex); + INIT_LIST_HEAD(&chip->list); + + INIT_WORK(&chip->work, timeout_work); + timer_setup(&chip->user_read_timer, user_reader_timeout, 0); + + memcpy(&chip->vendor, entry, sizeof(struct tcm_vendor_specific)); + + chip->dev_num = find_first_zero_bit(dev_mask, TCM_NUM_DEVICES); + + if (chip->dev_num >= TCM_NUM_DEVICES) { + dev_err(dev, "No available tcm device numbers\n"); + kfree(chip); + return NULL; + } else if (chip->dev_num == 0) + chip->vendor.miscdev.minor = TCM_MINOR; + else + chip->vendor.miscdev.minor = MISC_DYNAMIC_MINOR; + + set_bit(chip->dev_num, dev_mask); + + devname = kmalloc(DEVNAME_SIZE, GFP_KERNEL); + scnprintf(devname, DEVNAME_SIZE, "%s%d", "tcm", chip->dev_num); + chip->vendor.miscdev.name = devname; + + /* chip->vendor.miscdev.dev = dev; */ + + chip->dev = get_device(dev); + + if (misc_register(&chip->vendor.miscdev)) { + dev_err(chip->dev, + "unable to misc_register %s, minor %d\n", + chip->vendor.miscdev.name, + chip->vendor.miscdev.minor); + put_device(dev); + clear_bit(chip->dev_num, dev_mask); + kfree(chip); + kfree(devname); + return NULL; + } + + spin_lock(&driver_lock); + dev_set_drvdata(dev, chip); + list_add(&chip->list, &tcm_chip_list); + spin_unlock(&driver_lock); + + rc = sysfs_create_group(&dev->kobj, chip->vendor.attr_group); + /* chip->bios_dir = tcm_bios_log_setup(devname); */ + + return chip; +} +EXPORT_SYMBOL_GPL(tcm_register_hardware); + +static int __init tcm_init_module(void) +{ + return 0; +} + +static void __exit tcm_exit_module(void) +{ +} + +module_init(tcm_init_module); +module_exit(tcm_exit_module); + +MODULE_AUTHOR("Nationz Technologies Inc."); +MODULE_DESCRIPTION("TCM Driver"); +MODULE_VERSION("1.1.1.0"); +MODULE_LICENSE("GPL"); diff --git a/drivers/staging/gmjstcm/tcm.h b/drivers/staging/gmjstcm/tcm.h new file mode 100644 index 000000000000..b8cafe78d590 --- /dev/null +++ b/drivers/staging/gmjstcm/tcm.h @@ -0,0 +1,123 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2009 Nationz Technologies Inc. + * + */ +#include <linux/module.h> +#include <linux/types.h> +#include <linux/pci.h> +#include <linux/delay.h> +#include <linux/fs.h> +#include <linux/miscdevice.h> +#include <linux/platform_device.h> +#include <linux/io.h> + +struct device; +struct tcm_chip; + +enum tcm_timeout { + TCM_TIMEOUT = 5, +}; + +/* TCM addresses */ +enum tcm_addr { + TCM_SUPERIO_ADDR = 0x2E, + TCM_ADDR = 0x4E, +}; + +extern ssize_t tcm_show_pubek(struct device *, struct device_attribute *attr, + char *); +extern ssize_t tcm_show_pcrs(struct device *, struct device_attribute *attr, + char *); +extern ssize_t tcm_show_caps(struct device *, struct device_attribute *attr, + char *); +extern ssize_t tcm_store_cancel(struct device *, struct device_attribute *attr, + const char *, size_t); +extern ssize_t tcm_show_enabled(struct device *, struct device_attribute *attr, + char *); +extern ssize_t tcm_show_active(struct device *, struct device_attribute *attr, + char *); +extern ssize_t tcm_show_owned(struct device *, struct device_attribute *attr, + char *); +extern ssize_t tcm_show_temp_deactivated(struct device *, + struct device_attribute *attr, char *); + +struct tcm_vendor_specific { + const u8 req_complete_mask; + const u8 req_complete_val; + const u8 req_canceled; + void __iomem *iobase; /* ioremapped address */ + void __iomem *iolbc; + unsigned long base; /* TCM base address */ + + int irq; + + int region_size; + int have_region; + + int (*recv)(struct tcm_chip *, u8 *, size_t); + int (*send)(struct tcm_chip *, u8 *, size_t); + void (*cancel)(struct tcm_chip *); + u8 (*status)(struct tcm_chip *); + struct miscdevice miscdev; + struct attribute_group *attr_group; + struct list_head list; + int locality; + unsigned long timeout_a, timeout_b, timeout_c, timeout_d; /* jiffies */ + unsigned long duration[3]; /* jiffies */ + + wait_queue_head_t read_queue; + wait_queue_head_t int_queue; +}; + +struct tcm_chip { + struct device *dev; /* Device stuff */ + + int dev_num; /* /dev/tcm# */ + int num_opens; /* only one allowed */ + int time_expired; + + /* Data passed to and from the tcm via the read/write calls */ + u8 *data_buffer; + atomic_t data_pending; + struct mutex buffer_mutex; + + struct timer_list user_read_timer; /* user needs to claim result */ + struct work_struct work; + struct mutex tcm_mutex; /* tcm is processing */ + + struct tcm_vendor_specific vendor; + + struct dentry **bios_dir; + + struct list_head list; +}; + +#define to_tcm_chip(n) container_of(n, struct tcm_chip, vendor) + +static inline int tcm_read_index(int base, int index) +{ + outb(index, base); + return inb(base+1) & 0xFF; +} + +static inline void tcm_write_index(int base, int index, int value) +{ + outb(index, base); + outb(value & 0xFF, base+1); +} +extern void tcm_startup(struct tcm_chip *); +extern void tcm_get_timeouts(struct tcm_chip *); +extern unsigned long tcm_calc_ordinal_duration(struct tcm_chip *, u32); +extern struct tcm_chip *tcm_register_hardware(struct device *, + const struct tcm_vendor_specific *); +extern int tcm_open(struct inode *, struct file *); +extern int tcm_release(struct inode *, struct file *); +extern ssize_t tcm_write(struct file *, const char __user *, size_t, + loff_t *); +extern ssize_t tcm_read(struct file *, char __user *, size_t, loff_t *); +extern void tcm_remove_hardware(struct device *); +extern int tcm_pm_suspend(struct device *, pm_message_t); +extern int tcm_pm_suspend_p(struct device *); +extern int tcm_pm_resume(struct device *); + diff --git a/drivers/staging/gmjstcm/tcm_tis_spi.c b/drivers/staging/gmjstcm/tcm_tis_spi.c new file mode 100644 index 000000000000..db30a5b4c47d --- /dev/null +++ b/drivers/staging/gmjstcm/tcm_tis_spi.c @@ -0,0 +1,868 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2020 Kylin Tech. Co., Ltd. + */ + +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/interrupt.h> +#include <linux/wait.h> +#include <linux/acpi.h> +#include <linux/spi/spi.h> + +#include "tcm.h" + +#if !defined(CONFIG_KYLINOS_SERVER) && !defined(CONFIG_KYLINOS_DESKTOP) +static int is_ft_all(void) { + return 0; +} +#endif + +#define TCM_HEADER_SIZE 10 + +static bool tcm_debug; +module_param_named(debug, tcm_debug, bool, 0600); +MODULE_PARM_DESC(debug, "Turn TCM debugging mode on and off"); + +#define tcm_dbg(fmt, args...) \ +{ \ + if (tcm_debug) \ + pr_err(fmt, ## args); \ +} + +enum tis_access { + TCM_ACCESS_VALID = 0x80, + TCM_ACCESS_ACTIVE_LOCALITY = 0x20, + TCM_ACCESS_REQUEST_PENDING = 0x04, + TCM_ACCESS_REQUEST_USE = 0x02, +}; + +enum tis_status { + TCM_STS_VALID = 0x80, + TCM_STS_COMMAND_READY = 0x40, + TCM_STS_GO = 0x20, + TCM_STS_DATA_AVAIL = 0x10, + TCM_STS_DATA_EXPECT = 0x08, +}; + +enum tis_int_flags { + TCM_GLOBAL_INT_ENABLE = 0x80000000, + TCM_INTF_BURST_COUNT_STATIC = 0x100, + TCM_INTF_CMD_READY_INT = 0x080, + TCM_INTF_INT_EDGE_FALLING = 0x040, + TCM_INTF_INT_EDGE_RISING = 0x020, + TCM_INTF_INT_LEVEL_LOW = 0x010, + TCM_INTF_INT_LEVEL_HIGH = 0x008, + TCM_INTF_LOCALITY_CHANGE_INT = 0x004, + TCM_INTF_STS_VALID_INT = 0x002, + TCM_INTF_DATA_AVAIL_INT = 0x001, +}; + +enum tis_defaults { + TIS_SHORT_TIMEOUT = 750, /* ms */ + TIS_LONG_TIMEOUT = 2000, /* 2 sec */ +}; + +#define TCM_ACCESS(l) (0x0000 | ((l) << 12)) +#define TCM_INT_ENABLE(l) (0x0008 | ((l) << 12)) /* interperet */ +#define TCM_INT_VECTOR(l) (0x000C | ((l) << 12)) +#define TCM_INT_STATUS(l) (0x0010 | ((l) << 12)) +#define TCM_INTF_CAPS(l) (0x0014 | ((l) << 12)) +#define TCM_STS(l) (0x0018 | ((l) << 12)) +#define TCM_DATA_FIFO(l) (0x0024 | ((l) << 12)) + +#define TCM_DID_VID(l) (0x0F00 | ((l) << 12)) +#define TCM_RID(l) (0x0F04 | ((l) << 12)) + +#define TIS_MEM_BASE_huawei 0x3fed40000LL + +#define MAX_SPI_FRAMESIZE 64 + +// +#define _CPU_FT2000A4 +#define REUSE_CONF_REG_BASE 0x28180208 +#define REUSE_GPIO1_A5_BASE 0x28005000 + +static void *__iomem reuse_conf_reg; +static void *__iomem gpio1_a5; + +// +static LIST_HEAD(tis_chips); +static DEFINE_SPINLOCK(tis_lock); + +struct chip_data { + u8 cs; + u8 tmode; + u8 type; + u8 poll_mode; + u16 clk_div; + u32 speed_hz; + void (*cs_control)(u32 command); +}; + +struct tcm_tis_spi_phy { + struct spi_device *spi_device; + struct completion ready; + u8 *iobuf; +}; + +int tcm_tis_spi_transfer(struct device *dev, u32 addr, u16 len, + u8 *in, const u8 *out) +{ + struct tcm_tis_spi_phy *phy = dev_get_drvdata(dev); + int ret = 0; + struct spi_message m; + struct spi_transfer spi_xfer; + u8 transfer_len; + + tcm_dbg("TCM-dbg: %s, addr: 0x%x, len: %x, %s\n", + __func__, addr, len, (in) ? "in" : "out"); + + spi_bus_lock(phy->spi_device->master); + + /* set gpio1_a5 to LOW */ + if (is_ft_all() && (phy->spi_device->chip_select == 0)) { + iowrite32(0x0, gpio1_a5); + } + + while (len) { + transfer_len = min_t(u16, len, MAX_SPI_FRAMESIZE); + + phy->iobuf[0] = (in ? 0x80 : 0) | (transfer_len - 1); + phy->iobuf[1] = 0xd4; + phy->iobuf[2] = addr >> 8; + phy->iobuf[3] = addr; + + memset(&spi_xfer, 0, sizeof(spi_xfer)); + spi_xfer.tx_buf = phy->iobuf; + spi_xfer.rx_buf = phy->iobuf; + spi_xfer.len = 4; + spi_xfer.cs_change = 1; + + spi_message_init(&m); + spi_message_add_tail(&spi_xfer, &m); + ret = spi_sync_locked(phy->spi_device, &m); + if (ret < 0) + goto exit; + + spi_xfer.cs_change = 0; + spi_xfer.len = transfer_len; + spi_xfer.delay_usecs = 5; + + if (in) { + spi_xfer.tx_buf = NULL; + } else if (out) { + spi_xfer.rx_buf = NULL; + memcpy(phy->iobuf, out, transfer_len); + out += transfer_len; + } + + spi_message_init(&m); + spi_message_add_tail(&spi_xfer, &m); + reinit_completion(&phy->ready); + ret = spi_sync_locked(phy->spi_device, &m); + if (ret < 0) + goto exit; + + if (in) { + memcpy(in, phy->iobuf, transfer_len); + in += transfer_len; + } + + len -= transfer_len; + } + +exit: + /* set gpio1_a5 to HIGH */ + if (is_ft_all() && (phy->spi_device->chip_select == 0)) { + iowrite32(0x20, gpio1_a5); + } + + spi_bus_unlock(phy->spi_device->master); + tcm_dbg("TCM-dbg: ret: %d\n", ret); + return ret; +} + +static int tcm_tis_read8(struct device *dev, + u32 addr, u16 len, u8 *result) +{ + return tcm_tis_spi_transfer(dev, addr, len, result, NULL); +} + +static int tcm_tis_write8(struct device *dev, + u32 addr, u16 len, u8 *value) +{ + return tcm_tis_spi_transfer(dev, addr, len, NULL, value); +} + +static int tcm_tis_readb(struct device *dev, u32 addr, u8 *value) +{ + return tcm_tis_read8(dev, addr, sizeof(u8), value); +} + +static int tcm_tis_writeb(struct device *dev, u32 addr, u8 value) +{ + return tcm_tis_write8(dev, addr, sizeof(u8), &value); +} + +static int tcm_tis_readl(struct device *dev, u32 addr, u32 *result) +{ + int rc; + __le32 result_le; + + rc = tcm_tis_read8(dev, addr, sizeof(u32), (u8 *)&result_le); + tcm_dbg("TCM-dbg: result_le: 0x%x\n", result_le); + if (!rc) + *result = le32_to_cpu(result_le); + + return rc; +} + +static int tcm_tis_writel(struct device *dev, u32 addr, u32 value) +{ + int rc; + __le32 value_le; + + value_le = cpu_to_le32(value); + rc = tcm_tis_write8(dev, addr, sizeof(u32), (u8 *)&value_le); + + return rc; +} + +static int request_locality(struct tcm_chip *chip, int l); +static void release_locality(struct tcm_chip *chip, int l, int force); +static void cleanup_tis(void) +{ + int ret; + u32 inten; + struct tcm_vendor_specific *i, *j; + struct tcm_chip *chip; + + spin_lock(&tis_lock); + list_for_each_entry_safe(i, j, &tis_chips, list) { + chip = to_tcm_chip(i); + ret = tcm_tis_readl(chip->dev, + TCM_INT_ENABLE(chip->vendor.locality), &inten); + if (ret < 0) + return; + + tcm_tis_writel(chip->dev, TCM_INT_ENABLE(chip->vendor.locality), + ~TCM_GLOBAL_INT_ENABLE & inten); + release_locality(chip, chip->vendor.locality, 1); + } + spin_unlock(&tis_lock); +} + +static void tcm_tis_init(struct tcm_chip *chip) +{ + int ret; + u8 rid; + u32 vendor, intfcaps; + + ret = tcm_tis_readl(chip->dev, TCM_DID_VID(0), &vendor); + + if ((vendor & 0xffff) != 0x19f5 && (vendor & 0xffff) != 0x1B4E) + pr_info("there is no Nationz TCM on you computer\n"); + + ret = tcm_tis_readb(chip->dev, TCM_RID(0), &rid); + if (ret < 0) + return; + pr_info("kylin: 2019-09-21 1.2 TCM (device-id 0x%X, rev-id %d)\n", + vendor >> 16, rid); + + /* Figure out the capabilities */ + ret = tcm_tis_readl(chip->dev, + TCM_INTF_CAPS(chip->vendor.locality), &intfcaps); + if (ret < 0) + return; + + if (request_locality(chip, 0) != 0) + pr_err("tcm request_locality err\n"); + + atomic_set(&chip->data_pending, 0); +} + +static void tcm_handle_err(struct tcm_chip *chip) +{ + cleanup_tis(); + tcm_tis_init(chip); +} + +static bool check_locality(struct tcm_chip *chip, int l) +{ + int ret; + u8 access; + + ret = tcm_tis_readb(chip->dev, TCM_ACCESS(l), &access); + tcm_dbg("TCM-dbg: access: 0x%x\n", access); + if (ret < 0) + return false; + + if ((access & (TCM_ACCESS_ACTIVE_LOCALITY | TCM_ACCESS_VALID)) == + (TCM_ACCESS_ACTIVE_LOCALITY | TCM_ACCESS_VALID)) { + chip->vendor.locality = l; + return true; + } + + return false; +} + +static int request_locality(struct tcm_chip *chip, int l) +{ + unsigned long stop; + + if (check_locality(chip, l)) + return l; + + tcm_tis_writeb(chip->dev, TCM_ACCESS(l), TCM_ACCESS_REQUEST_USE); + + /* wait for burstcount */ + stop = jiffies + chip->vendor.timeout_a; + do { + if (check_locality(chip, l)) + return l; + msleep(TCM_TIMEOUT); + } while (time_before(jiffies, stop)); + + return -1; +} + +static void release_locality(struct tcm_chip *chip, int l, int force) +{ + int ret; + u8 access; + + ret = tcm_tis_readb(chip->dev, TCM_ACCESS(l), &access); + if (ret < 0) + return; + if (force || (access & (TCM_ACCESS_REQUEST_PENDING | TCM_ACCESS_VALID)) == + (TCM_ACCESS_REQUEST_PENDING | TCM_ACCESS_VALID)) + tcm_tis_writeb(chip->dev, + TCM_ACCESS(l), TCM_ACCESS_ACTIVE_LOCALITY); +} + +static u8 tcm_tis_status(struct tcm_chip *chip) +{ + int ret; + u8 status; + + ret = tcm_tis_readb(chip->dev, + TCM_STS(chip->vendor.locality), &status); + tcm_dbg("TCM-dbg: status: 0x%x\n", status); + if (ret < 0) + return 0; + + return status; +} + +static void tcm_tis_ready(struct tcm_chip *chip) +{ + /* this causes the current command to be aboreted */ + tcm_tis_writeb(chip->dev, TCM_STS(chip->vendor.locality), + TCM_STS_COMMAND_READY); +} + +static int get_burstcount(struct tcm_chip *chip) +{ + int ret; + unsigned long stop; + u8 tmp, tmp1; + int burstcnt = 0; + + /* wait for burstcount */ + /* which timeout value, spec has 2 answers (c & d) */ + stop = jiffies + chip->vendor.timeout_d; + do { + ret = tcm_tis_readb(chip->dev, + TCM_STS(chip->vendor.locality) + 1, + &tmp); + tcm_dbg("TCM-dbg: burstcnt: 0x%x\n", burstcnt); + if (ret < 0) + return -EINVAL; + ret = tcm_tis_readb(chip->dev, + (TCM_STS(chip->vendor.locality) + 2), + &tmp1); + tcm_dbg("TCM-dbg: burstcnt: 0x%x\n", burstcnt); + if (ret < 0) + return -EINVAL; + + burstcnt = tmp | (tmp1 << 8); + if (burstcnt) + return burstcnt; + msleep(TCM_TIMEOUT); + } while (time_before(jiffies, stop)); + + return -EBUSY; +} + +static int wait_for_stat(struct tcm_chip *chip, u8 mask, + unsigned long timeout, + wait_queue_head_t *queue) +{ + unsigned long stop; + u8 status; + + /* check current status */ + status = tcm_tis_status(chip); + if ((status & mask) == mask) + return 0; + + stop = jiffies + timeout; + do { + msleep(TCM_TIMEOUT); + status = tcm_tis_status(chip); + if ((status & mask) == mask) + return 0; + } while (time_before(jiffies, stop)); + + return -ETIME; +} + +static int recv_data(struct tcm_chip *chip, u8 *buf, size_t count) +{ + int ret; + int size = 0, burstcnt; + + while (size < count && wait_for_stat(chip, + TCM_STS_DATA_AVAIL | TCM_STS_VALID, + chip->vendor.timeout_c, + &chip->vendor.read_queue) == 0) { + burstcnt = get_burstcount(chip); + + if (burstcnt < 0) { + dev_err(chip->dev, "Unable to read burstcount\n"); + return burstcnt; + } + + for (; burstcnt > 0 && size < count; burstcnt--) { + ret = tcm_tis_readb(chip->dev, + TCM_DATA_FIFO(chip->vendor.locality), + &buf[size]); + tcm_dbg("TCM-dbg: buf[%d]: 0x%x\n", size, buf[size]); + size++; + } + } + + return size; +} + +static int tcm_tis_recv(struct tcm_chip *chip, u8 *buf, size_t count) +{ + int size = 0; + int expected, status; + unsigned long stop; + + if (count < TCM_HEADER_SIZE) { + dev_err(chip->dev, "read size is to small: %d\n", (u32)(count)); + size = -EIO; + goto out; + } + + /* read first 10 bytes, including tag, paramsize, and result */ + size = recv_data(chip, buf, TCM_HEADER_SIZE); + if (size < TCM_HEADER_SIZE) { + dev_err(chip->dev, "Unable to read header\n"); + goto out; + } + + expected = be32_to_cpu(*(__be32 *)(buf + 2)); + if (expected > count) { + dev_err(chip->dev, "Expected data count\n"); + size = -EIO; + goto out; + } + + size += recv_data(chip, &buf[TCM_HEADER_SIZE], + expected - TCM_HEADER_SIZE); + if (size < expected) { + dev_err(chip->dev, "Unable to read remainder of result\n"); + size = -ETIME; + goto out; + } + + wait_for_stat(chip, TCM_STS_VALID, chip->vendor.timeout_c, + &chip->vendor.int_queue); + + stop = jiffies + chip->vendor.timeout_c; + do { + msleep(TCM_TIMEOUT); + status = tcm_tis_status(chip); + if ((status & TCM_STS_DATA_AVAIL) == 0) + break; + + } while (time_before(jiffies, stop)); + + status = tcm_tis_status(chip); + if (status & TCM_STS_DATA_AVAIL) { /* retry? */ + dev_err(chip->dev, "Error left over data\n"); + size = -EIO; + goto out; + } + +out: + tcm_tis_ready(chip); + release_locality(chip, chip->vendor.locality, 0); + if (size < 0) + tcm_handle_err(chip); + return size; +} + +/* + * If interrupts are used (signaled by an irq set in the vendor structure) + * tcm.c can skip polling for the data to be available as the interrupt is + * waited for here + */ +static int tcm_tis_send(struct tcm_chip *chip, u8 *buf, size_t len) +{ + int rc, status, burstcnt; + size_t count = 0; + u32 ordinal; + unsigned long stop; + int send_again = 0; + +tcm_tis_send_again: + count = 0; + if (request_locality(chip, 0) < 0) { + dev_err(chip->dev, "send, tcm is busy\n"); + return -EBUSY; + } + status = tcm_tis_status(chip); + + if ((status & TCM_STS_COMMAND_READY) == 0) { + tcm_tis_ready(chip); + if (wait_for_stat(chip, TCM_STS_COMMAND_READY, + chip->vendor.timeout_b, &chip->vendor.int_queue) < 0) { + dev_err(chip->dev, "send, tcm wait time out1\n"); + rc = -ETIME; + goto out_err; + } + } + + while (count < len - 1) { + burstcnt = get_burstcount(chip); + if (burstcnt < 0) { + dev_err(chip->dev, "Unable to read burstcount\n"); + rc = burstcnt; + goto out_err; + } + for (; burstcnt > 0 && count < len - 1; burstcnt--) { + tcm_tis_writeb(chip->dev, + TCM_DATA_FIFO(chip->vendor.locality), buf[count]); + count++; + } + + wait_for_stat(chip, TCM_STS_VALID, chip->vendor.timeout_c, + &chip->vendor.int_queue); + } + + /* write last byte */ + tcm_tis_writeb(chip->dev, + TCM_DATA_FIFO(chip->vendor.locality), buf[count]); + + wait_for_stat(chip, TCM_STS_VALID, + chip->vendor.timeout_c, &chip->vendor.int_queue); + stop = jiffies + chip->vendor.timeout_c; + do { + msleep(TCM_TIMEOUT); + status = tcm_tis_status(chip); + if ((status & TCM_STS_DATA_EXPECT) == 0) + break; + + } while (time_before(jiffies, stop)); + + if ((status & TCM_STS_DATA_EXPECT) != 0) { + dev_err(chip->dev, "send, tcm expect data\n"); + rc = -EIO; + goto out_err; + } + + /* go and do it */ + tcm_tis_writeb(chip->dev, TCM_STS(chip->vendor.locality), TCM_STS_GO); + + ordinal = be32_to_cpu(*((__be32 *)(buf + 6))); + if (wait_for_stat(chip, TCM_STS_DATA_AVAIL | TCM_STS_VALID, + tcm_calc_ordinal_duration(chip, ordinal), + &chip->vendor.read_queue) < 0) { + dev_err(chip->dev, "send, tcm wait time out2\n"); + rc = -ETIME; + goto out_err; + } + + return len; + +out_err: + tcm_tis_ready(chip); + release_locality(chip, chip->vendor.locality, 0); + tcm_handle_err(chip); + if (send_again++ < 3) { + goto tcm_tis_send_again; + } + + dev_err(chip->dev, "kylin send, err: %d\n", rc); + return rc; +} + +static struct file_operations tis_ops = { + .owner = THIS_MODULE, + .llseek = no_llseek, + .open = tcm_open, + .read = tcm_read, + .write = tcm_write, + .release = tcm_release, +}; + +static DEVICE_ATTR(pubek, S_IRUGO, tcm_show_pubek, NULL); +static DEVICE_ATTR(pcrs, S_IRUGO, tcm_show_pcrs, NULL); +static DEVICE_ATTR(enabled, S_IRUGO, tcm_show_enabled, NULL); +static DEVICE_ATTR(active, S_IRUGO, tcm_show_active, NULL); +static DEVICE_ATTR(owned, S_IRUGO, tcm_show_owned, NULL); +static DEVICE_ATTR(temp_deactivated, S_IRUGO, tcm_show_temp_deactivated, + NULL); +static DEVICE_ATTR(caps, S_IRUGO, tcm_show_caps, NULL); +static DEVICE_ATTR(cancel, S_IWUSR | S_IWGRP, NULL, tcm_store_cancel); + +static struct attribute *tis_attrs[] = { + &dev_attr_pubek.attr, + &dev_attr_pcrs.attr, + &dev_attr_enabled.attr, + &dev_attr_active.attr, + &dev_attr_owned.attr, + &dev_attr_temp_deactivated.attr, + &dev_attr_caps.attr, + &dev_attr_cancel.attr, NULL, +}; + +static struct attribute_group tis_attr_grp = { + .attrs = tis_attrs +}; + +static struct tcm_vendor_specific tcm_tis = { + .status = tcm_tis_status, + .recv = tcm_tis_recv, + .send = tcm_tis_send, + .cancel = tcm_tis_ready, + .req_complete_mask = TCM_STS_DATA_AVAIL | TCM_STS_VALID, + .req_complete_val = TCM_STS_DATA_AVAIL | TCM_STS_VALID, + .req_canceled = TCM_STS_COMMAND_READY, + .attr_group = &tis_attr_grp, + .miscdev = { + .fops = &tis_ops, + }, +}; + +static struct tcm_chip *chip; +static int tcm_tis_spi_probe(struct spi_device *spi) +{ + int ret; + u8 revid; + u32 vendor, intfcaps; + struct tcm_tis_spi_phy *phy; + struct chip_data *spi_chip; + + pr_info("TCM(ky): __func__(v=%d) ..\n", + 10); + + tcm_dbg("TCM-dbg: %s/%d, enter\n", __func__, __LINE__); + phy = devm_kzalloc(&spi->dev, sizeof(struct tcm_tis_spi_phy), + GFP_KERNEL); + if (!phy) + return -ENOMEM; + + phy->iobuf = devm_kmalloc(&spi->dev, MAX_SPI_FRAMESIZE, GFP_KERNEL); + if (!phy->iobuf) + return -ENOMEM; + + phy->spi_device = spi; + init_completion(&phy->ready); + + tcm_dbg("TCM-dbg: %s/%d\n", __func__, __LINE__); + /* init spi dev */ + spi->chip_select = 0; /* cs0 */ + spi->mode = SPI_MODE_0; + spi->bits_per_word = 8; + spi->max_speed_hz = spi->max_speed_hz ? : 24000000; + spi_setup(spi); + + spi_chip = spi_get_ctldata(spi); + if (!spi_chip) { + pr_err("There was wrong in spi master\n"); + return -ENODEV; + } + /* tcm does not support interrupt mode, we use poll mode instead. */ + spi_chip->poll_mode = 1; + + tcm_dbg("TCM-dbg: %s/%d\n", __func__, __LINE__); + /* regiter tcm hw */ + chip = tcm_register_hardware(&spi->dev, &tcm_tis); + if (!chip) { + dev_err(chip->dev, "tcm register hardware err\n"); + return -ENODEV; + } + + dev_set_drvdata(chip->dev, phy); + + /** + * phytium2000a4 spi controller's clk clk level is unstable, + * so it is solved by using the low level of gpio output. + **/ + if (is_ft_all() && (spi->chip_select == 0)) { + /* reuse conf reg base */ + reuse_conf_reg = ioremap(REUSE_CONF_REG_BASE, 0x10); + if (!reuse_conf_reg) { + dev_err(&spi->dev, "Failed to ioremap reuse conf reg\n"); + ret = -ENOMEM; + goto out_err; + } + + /* gpio1 a5 base addr */ + gpio1_a5 = ioremap(REUSE_GPIO1_A5_BASE, 0x10); + if (!gpio1_a5) { + dev_err(&spi->dev, "Failed to ioremap gpio1 a5\n"); + ret = -ENOMEM; + goto out_err; + } + + /* reuse cs0 to gpio1_a5 */ + iowrite32((ioread32(reuse_conf_reg) | 0xFFFF0) & 0xFFF9004F, + reuse_conf_reg); + /* set gpio1 a5 to output */ + iowrite32(0x20, gpio1_a5 + 0x4); + } + + tcm_dbg("TCM-dbg: %s/%d\n", + __func__, __LINE__); + ret = tcm_tis_readl(chip->dev, TCM_DID_VID(0), &vendor); + if (ret < 0) + goto out_err; + + tcm_dbg("TCM-dbg: %s/%d, vendor: 0x%x\n", + __func__, __LINE__, vendor); + if ((vendor & 0xffff) != 0x19f5 && (vendor & 0xffff) != 0x1B4E) { + dev_err(chip->dev, "there is no Nationz TCM on you computer\n"); + goto out_err; + } + + ret = tcm_tis_readb(chip->dev, TCM_RID(0), &revid); + tcm_dbg("TCM-dbg: %s/%d, revid: 0x%x\n", + __func__, __LINE__, revid); + if (ret < 0) + goto out_err; + dev_info(chip->dev, "kylin: 2019-09-21 1.2 TCM " + "(device-id 0x%X, rev-id %d)\n", + vendor >> 16, revid); + + /* Default timeouts */ + chip->vendor.timeout_a = msecs_to_jiffies(TIS_SHORT_TIMEOUT); + chip->vendor.timeout_b = msecs_to_jiffies(TIS_LONG_TIMEOUT); + chip->vendor.timeout_c = msecs_to_jiffies(TIS_SHORT_TIMEOUT); + chip->vendor.timeout_d = msecs_to_jiffies(TIS_SHORT_TIMEOUT); + + tcm_dbg("TCM-dbg: %s/%d\n", + __func__, __LINE__); + /* Figure out the capabilities */ + ret = tcm_tis_readl(chip->dev, + TCM_INTF_CAPS(chip->vendor.locality), &intfcaps); + if (ret < 0) + goto out_err; + + tcm_dbg("TCM-dbg: %s/%d, intfcaps: 0x%x\n", + __func__, __LINE__, intfcaps); + if (request_locality(chip, 0) != 0) { + dev_err(chip->dev, "tcm request_locality err\n"); + ret = -ENODEV; + goto out_err; + } + + INIT_LIST_HEAD(&chip->vendor.list); + spin_lock(&tis_lock); + list_add(&chip->vendor.list, &tis_chips); + spin_unlock(&tis_lock); + + tcm_get_timeouts(chip); + tcm_startup(chip); + + tcm_dbg("TCM-dbg: %s/%d, exit\n", __func__, __LINE__); + return 0; + +out_err: + if (is_ft_all()) { + if (reuse_conf_reg) + iounmap(reuse_conf_reg); + if (gpio1_a5) + iounmap(gpio1_a5); + } + tcm_dbg("TCM-dbg: %s/%d, error\n", __func__, __LINE__); + dev_set_drvdata(chip->dev, chip); + tcm_remove_hardware(chip->dev); + + return ret; +} + +static int tcm_tis_spi_remove(struct spi_device *dev) +{ + if (is_ft_all()) { + if (reuse_conf_reg) + iounmap(reuse_conf_reg); + if (gpio1_a5) + iounmap(gpio1_a5); + } + + dev_info(&dev->dev, "%s\n", __func__); + dev_set_drvdata(chip->dev, chip); + tcm_remove_hardware(&dev->dev); + + return 0; +} + +static const struct acpi_device_id tcm_tis_spi_acpi_match[] = { + {"TCMS0001", 0}, + {"SMO0768", 0}, + {"ZIC0601", 0}, + {} +}; +MODULE_DEVICE_TABLE(acpi, tcm_tis_spi_acpi_match); + +static const struct spi_device_id tcm_tis_spi_id_table[] = { + {"SMO0768", 0}, + {"ZIC0601", 0}, + {} +}; +MODULE_DEVICE_TABLE(spi, tcm_tis_spi_id_table); + +static struct spi_driver tcm_tis_spi_drv = { + .driver = { + .name = "tcm_tis_spi", + .acpi_match_table = ACPI_PTR(tcm_tis_spi_acpi_match), + }, + .id_table = tcm_tis_spi_id_table, + .probe = tcm_tis_spi_probe, + .remove = tcm_tis_spi_remove, +}; + +#if 1 +module_spi_driver(tcm_tis_spi_drv); +#else/*0*/ + +static int __init __spi_driver_init(void) +{ + pr_info("TCM(ky): __init __func__(ver=%2d)\n", + 10); + return spi_register_driver(&tcm_tis_spi_drv); +} + +static void __exit __spi_driver_exit(void) +{ + pr_info("TCM(ky): __exit __func__\n"); + spi_unregister_driver(&tcm_tis_spi_drv); +} + +module_init(__spi_driver_init); +module_exit(__spi_driver_exit); +#endif/*0*/ + +MODULE_AUTHOR("xiongxin<xiongxin(a)tj.kylinos.cn>"); +MODULE_DESCRIPTION("TCM Driver Base Spi"); +MODULE_VERSION("6.1.0.2"); +MODULE_LICENSE("GPL"); -- 2.23.0

1 1

[PATCH kernel-4.19 v1] drivers: TCM: add GMJS(Nationz Tech) TCM driver.
by 李志刚 23 Feb '21

23 Feb '21

1 0

[PATCH OLK-5.10 1/2] arm64: smp: Add support for cpu park
by sangyan＠huawei.com 23 Feb '21

23 Feb '21

From: Sang Yan <sangyan(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: N/A Introducing a feature of CPU PARK in order to save time of cpus down and up during kexec, which may cost 250ms of per cpu's down and 30ms of up. As a result, for 128 cores, it costs more than 30 seconds to down and up cpus during kexec. Think about 256 cores and more. CPU PARK is a state that cpu power-on and staying in spin loop, polling for exit chances, such as writing exit address. Reserving a block of memory, to fill with cpu park text section, exit address and park-magic-flag of each cpu. In implementation, reserved one page for one cpu core. Cpus going to park state instead of down in machine_shutdown(). Cpus going out of park state in smp_init instead of brought up. One of cpu park sections in pre-reserved memory blocks,: +--------------+ + exit address + +--------------+ + park magic + +--------------+ + park codes + + . + + . + + . + +--------------+ Signed-off-by: Sang Yan <sangyan(a)huawei.com> --- arch/arm64/Kconfig | 12 ++ arch/arm64/include/asm/kexec.h | 6 + arch/arm64/include/asm/smp.h | 15 +++ arch/arm64/kernel/Makefile | 1 + arch/arm64/kernel/cpu-park.S | 59 ++++++++++ arch/arm64/kernel/machine_kexec.c | 2 +- arch/arm64/kernel/process.c | 4 + arch/arm64/kernel/smp.c | 229 ++++++++++++++++++++++++++++++++++++++ arch/arm64/mm/init.c | 55 +++++++++ 9 files changed, 382 insertions(+), 1 deletion(-) create mode 100644 arch/arm64/kernel/cpu-park.S diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index b9c5654..0885668 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -345,6 +345,18 @@ config KASAN_SHADOW_OFFSET default 0xeffffff900000000 if ARM64_VA_BITS_36 && KASAN_SW_TAGS default 0xffffffffffffffff +config ARM64_CPU_PARK + bool "Support CPU PARK on kexec" + depends on SMP + depends on KEXEC_CORE + help + This enables support for CPU PARK feature in + order to save time of cpu down to up. + CPU park is a state through kexec, spin loop + instead of cpu die before jumping to new kernel, + jumping out from loop to new kernel entry in + smp_init. + source "arch/arm64/Kconfig.platforms" menu "Kernel Features" diff --git a/arch/arm64/include/asm/kexec.h b/arch/arm64/include/asm/kexec.h index 79909ae..a133889 100644 --- a/arch/arm64/include/asm/kexec.h +++ b/arch/arm64/include/asm/kexec.h @@ -36,6 +36,11 @@ #define CRASH_ADDR_HIGH_MAX MEMBLOCK_ALLOC_ACCESSIBLE +#ifdef CONFIG_ARM64_CPU_PARK +/* CPU park state flag: "park" */ +#define PARK_MAGIC 0x7061726b +#endif + #ifndef __ASSEMBLY__ /** @@ -104,6 +109,7 @@ static inline void crash_post_resume(void) {} #ifdef CONFIG_KEXEC_CORE extern void __init reserve_crashkernel(void); #endif +void machine_kexec_mask_interrupts(void); #ifdef CONFIG_KEXEC_FILE #define ARCH_HAS_KIMAGE_ARCH diff --git a/arch/arm64/include/asm/smp.h b/arch/arm64/include/asm/smp.h index 2e7f529..8c5d2d6 100644 --- a/arch/arm64/include/asm/smp.h +++ b/arch/arm64/include/asm/smp.h @@ -145,6 +145,21 @@ bool cpus_are_stuck_in_kernel(void); extern void crash_smp_send_stop(void); extern bool smp_crash_stop_failed(void); +#ifdef CONFIG_ARM64_CPU_PARK +#define PARK_SECTION_SIZE 1024 +struct cpu_park_info { + /* Physical address of reserved park memory. */ + unsigned long start; + /* park reserve mem len should be PARK_SECTION_SIZE * NR_CPUS */ + unsigned long len; + /* Virtual address of reserved park memory. */ + unsigned long start_v; +}; +extern struct cpu_park_info park_info; +extern void enter_cpu_park(unsigned long text, unsigned long exit); +extern void do_cpu_park(unsigned long exit); +extern int kexec_smp_send_park(void); +#endif #endif /* ifndef __ASSEMBLY__ */ diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile index 2621d5c..60478d2 100644 --- a/arch/arm64/kernel/Makefile +++ b/arch/arm64/kernel/Makefile @@ -54,6 +54,7 @@ obj-$(CONFIG_RANDOMIZE_BASE) += kaslr.o obj-$(CONFIG_HIBERNATION) += hibernate.o hibernate-asm.o obj-$(CONFIG_KEXEC_CORE) += machine_kexec.o relocate_kernel.o \ cpu-reset.o +obj-$(CONFIG_ARM64_CPU_PARK) += cpu-park.o obj-$(CONFIG_KEXEC_FILE) += machine_kexec_file.o kexec_image.o obj-$(CONFIG_ARM64_RELOC_TEST) += arm64-reloc-test.o arm64-reloc-test-y := reloc_test_core.o reloc_test_syms.o diff --git a/arch/arm64/kernel/cpu-park.S b/arch/arm64/kernel/cpu-park.S new file mode 100644 index 0000000..10c685c --- /dev/null +++ b/arch/arm64/kernel/cpu-park.S @@ -0,0 +1,59 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * CPU park routines + * + * Copyright (C) 2020 Huawei Technologies., 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. + */ + +#include <linux/linkage.h> +#include <asm/assembler.h> +#include <asm/kexec.h> +#include <asm/sysreg.h> +#include <asm/virt.h> + +.text +.pushsection .idmap.text, "awx" + +/* cpu park helper in idmap section */ +SYM_CODE_START(enter_cpu_park) + /* Clear sctlr_el1 flags. */ + mrs x12, sctlr_el1 + mov_q x13, SCTLR_ELx_FLAGS + bic x12, x12, x13 + pre_disable_mmu_workaround + msr sctlr_el1, x12 /* disable mmu */ + isb + + mov x18, x0 + mov x0, x1 /* secondary_entry addr */ + br x18 /* call do_cpu_park of each cpu */ +SYM_CODE_END(enter_cpu_park) + +.popsection + +SYM_CODE_START(do_cpu_park) + ldr x18, =PARK_MAGIC /* magic number "park" */ + add x1, x0, #8 + str x18, [x1] /* set on-park flag */ + dc civac, x1 /* flush cache of "park" */ + dsb nsh + isb + +.Lloop: + wfe + isb + ldr x19, [x0] + cmp x19, #0 /* test secondary_entry */ + b.eq .Lloop + + ic iallu /* invalidate the local I-cache */ + dsb nsh + isb + + br x19 /* jump to secondary_entry */ +SYM_CODE_END(do_cpu_park) + diff --git a/arch/arm64/kernel/machine_kexec.c b/arch/arm64/kernel/machine_kexec.c index a0b144c..f47ce96 100644 --- a/arch/arm64/kernel/machine_kexec.c +++ b/arch/arm64/kernel/machine_kexec.c @@ -213,7 +213,7 @@ void machine_kexec(struct kimage *kimage) BUG(); /* Should never get here. */ } -static void machine_kexec_mask_interrupts(void) +void machine_kexec_mask_interrupts(void) { unsigned int i; struct irq_desc *desc; diff --git a/arch/arm64/kernel/process.c b/arch/arm64/kernel/process.c index 73e3b32..10cffee 100644 --- a/arch/arm64/kernel/process.c +++ b/arch/arm64/kernel/process.c @@ -146,6 +146,10 @@ void arch_cpu_idle_dead(void) */ void machine_shutdown(void) { +#ifdef CONFIG_ARM64_CPU_PARK + if (kexec_smp_send_park() == 0) + return; +#endif smp_shutdown_nonboot_cpus(reboot_cpu); } diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c index 18e9727..bc475d5 100644 --- a/arch/arm64/kernel/smp.c +++ b/arch/arm64/kernel/smp.c @@ -32,6 +32,8 @@ #include <linux/irq_work.h> #include <linux/kernel_stat.h> #include <linux/kexec.h> +#include <linux/console.h> + #include <linux/kvm_host.h> #include <asm/alternative.h> @@ -93,6 +95,167 @@ static inline int op_cpu_kill(unsigned int cpu) } #endif +#ifdef CONFIG_ARM64_CPU_PARK +struct cpu_park_section { + unsigned long exit; /* exit address of park look */ + unsigned long magic; /* maigc represent park state */ + char text[0]; /* text section of park */ +}; + +static int mmap_cpu_park_mem(void) +{ + if (!park_info.start) + return -ENOMEM; + + if (park_info.start_v) + return 0; + + park_info.start_v = (unsigned long)__ioremap(park_info.start, + park_info.len, + PAGE_KERNEL_EXEC); + if (!park_info.start_v) { + pr_warn("map park memory failed."); + return -ENOMEM; + } + + return 0; +} + +static inline unsigned long cpu_park_section_v(unsigned int cpu) +{ + return park_info.start_v + PARK_SECTION_SIZE * (cpu - 1); +} + +static inline unsigned long cpu_park_section_p(unsigned int cpu) +{ + return park_info.start + PARK_SECTION_SIZE * (cpu - 1); +} + +/* + * Write the secondary_entry to exit section of park state. + * Then the secondary cpu will jump straight into the kernel + * by the secondary_entry. + */ +static int write_park_exit(unsigned int cpu) +{ + struct cpu_park_section *park_section; + unsigned long *park_exit; + unsigned long *park_text; + + if (mmap_cpu_park_mem() != 0) + return -EPERM; + + park_section = (struct cpu_park_section *)cpu_park_section_v(cpu); + park_exit = &park_section->exit; + park_text = (unsigned long *)park_section->text; + pr_debug("park_text 0x%lx : 0x%lx, do_cpu_park text 0x%lx : 0x%lx", + (unsigned long)park_text, *park_text, + (unsigned long)do_cpu_park, + *(unsigned long *)do_cpu_park); + + /* + * Test first 8 bytes to determine + * whether needs to write cpu park exit. + */ + if (*park_text == *(unsigned long *)do_cpu_park) { + writeq_relaxed(__pa_symbol(secondary_entry), park_exit); + __flush_dcache_area((__force void *)park_exit, + sizeof(unsigned long)); + flush_icache_range((unsigned long)park_exit, + (unsigned long)(park_exit + 1)); + sev(); + dsb(sy); + isb(); + + pr_debug("Write cpu %u secondary entry 0x%lx to 0x%lx.", + cpu, *park_exit, (unsigned long)park_exit); + pr_info("Boot cpu %u from PARK state.", cpu); + return 0; + } + + return -EPERM; +} + +/* Install cpu park sections for the specific cpu. */ +static int install_cpu_park(unsigned int cpu) +{ + struct cpu_park_section *park_section; + unsigned long *park_exit; + unsigned long *park_magic; + unsigned long park_text_len; + + park_section = (struct cpu_park_section *)cpu_park_section_v(cpu); + pr_debug("Install cpu park on cpu %u park exit 0x%lx park text 0x%lx", + cpu, (unsigned long)park_section, + (unsigned long)(park_section->text)); + + park_exit = &park_section->exit; + park_magic = &park_section->magic; + park_text_len = PARK_SECTION_SIZE - sizeof(struct cpu_park_section); + + *park_exit = 0UL; + *park_magic = 0UL; + memcpy((void *)park_section->text, do_cpu_park, park_text_len); + __flush_dcache_area((void *)park_section, PARK_SECTION_SIZE); + + return 0; +} + +static int uninstall_cpu_park(unsigned int cpu) +{ + unsigned long park_section; + + if (mmap_cpu_park_mem() != 0) + return -EPERM; + + park_section = cpu_park_section_v(cpu); + memset((void *)park_section, 0, PARK_SECTION_SIZE); + __flush_dcache_area((void *)park_section, PARK_SECTION_SIZE); + + return 0; +} + +static int cpu_wait_park(unsigned int cpu) +{ + long timeout; + struct cpu_park_section *park_section; + + volatile unsigned long *park_magic; + + park_section = (struct cpu_park_section *)cpu_park_section_v(cpu); + park_magic = &park_section->magic; + + timeout = USEC_PER_SEC; + while (*park_magic != PARK_MAGIC && timeout--) + udelay(1); + + if (timeout > 0) + pr_debug("cpu %u park done.", cpu); + else + pr_err("cpu %u park failed.", cpu); + + return *park_magic == PARK_MAGIC; +} + +static void cpu_park(unsigned int cpu) +{ + unsigned long park_section_p; + unsigned long park_exit_phy; + unsigned long do_park; + typeof(enter_cpu_park) *park; + + park_section_p = cpu_park_section_p(cpu); + park_exit_phy = park_section_p; + pr_debug("Go to park cpu %u exit address 0x%lx", cpu, park_exit_phy); + + do_park = park_section_p + sizeof(struct cpu_park_section); + park = (void *)__pa_symbol(enter_cpu_park); + + cpu_install_idmap(); + park(do_park, park_exit_phy); + unreachable(); +} +#endif /* * Boot a secondary CPU, and assign it the specified idle task. @@ -102,6 +265,10 @@ static int boot_secondary(unsigned int cpu, struct task_struct *idle) { const struct cpu_operations *ops = get_cpu_ops(cpu); +#ifdef CONFIG_ARM64_CPU_PARK + if (write_park_exit(cpu) == 0) + return 0; +#endif if (ops->cpu_boot) return ops->cpu_boot(cpu); @@ -131,6 +298,9 @@ int __cpu_up(unsigned int cpu, struct task_struct *idle) return ret; } +#ifdef CONFIG_ARM64_CPU_PARK + uninstall_cpu_park(cpu); +#endif /* * CPU was successfully started, wait for it to come online or * time out. @@ -844,10 +1014,30 @@ void arch_irq_work_raise(void) static void local_cpu_stop(void) { + int cpu; + const struct cpu_operations *ops = NULL; + set_cpu_online(smp_processor_id(), false); local_daif_mask(); sdei_mask_local_cpu(); + +#ifdef CONFIG_ARM64_CPU_PARK + /* + * Go to cpu park state. + * Otherwise go to cpu die. + */ + cpu = smp_processor_id(); + if (kexec_in_progress && park_info.start_v) { + machine_kexec_mask_interrupts(); + cpu_park(cpu); + + ops = get_cpu_ops(cpu); + if (ops && ops->cpu_die) + ops->cpu_die(cpu); + } +#endif + cpu_park_loop(); } @@ -1053,6 +1243,45 @@ void smp_send_stop(void) sdei_mask_local_cpu(); } +#ifdef CONFIG_ARM64_CPU_PARK +int kexec_smp_send_park(void) +{ + unsigned long cpu; + + if (WARN_ON(!kexec_in_progress)) { + pr_crit("%s called not in kexec progress.", __func__); + return -EPERM; + } + + if (mmap_cpu_park_mem() != 0) { + pr_info("no cpuparkmem, goto normal way."); + return -EPERM; + } + + local_irq_disable(); + + if (num_online_cpus() > 1) { + cpumask_t mask; + + cpumask_copy(&mask, cpu_online_mask); + cpumask_clear_cpu(smp_processor_id(), &mask); + + for_each_cpu(cpu, &mask) + install_cpu_park(cpu); + smp_cross_call(&mask, IPI_CPU_STOP); + + /* Wait for other CPUs to park */ + for_each_cpu(cpu, &mask) + cpu_wait_park(cpu); + pr_info("smp park other cpus done\n"); + } + + sdei_mask_local_cpu(); + + return 0; +} +#endif + #ifdef CONFIG_KEXEC_CORE void crash_smp_send_stop(void) { diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index 794f992..d01259c 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -236,6 +236,57 @@ static void __init fdt_enforce_memory_region(void) memblock_add(usable_rgns[1].base, usable_rgns[1].size); } +#ifdef CONFIG_ARM64_CPU_PARK +struct cpu_park_info park_info = { + .start = 0, + .len = PARK_SECTION_SIZE * NR_CPUS, + .start_v = 0, +}; + +static int __init parse_park_mem(char *p) +{ + if (!p) + return 0; + + park_info.start = PAGE_ALIGN(memparse(p, NULL)); + if (park_info.start == 0) + pr_info("cpu park mem params[%s]", p); + + return 0; +} +early_param("cpuparkmem", parse_park_mem); + +static int __init reserve_park_mem(void) +{ + if (park_info.start == 0 || park_info.len == 0) + return 0; + + park_info.start = PAGE_ALIGN(park_info.start); + park_info.len = PAGE_ALIGN(park_info.len); + + if (!memblock_is_region_memory(park_info.start, park_info.len)) { + pr_warn("cannot reserve park mem: region is not memory!"); + goto out; + } + + if (memblock_is_region_reserved(park_info.start, park_info.len)) { + pr_warn("cannot reserve park mem: region overlaps reserved memory!"); + goto out; + } + + memblock_remove(park_info.start, park_info.len); + pr_info("cpu park mem reserved: 0x%016lx - 0x%016lx (%ld MB)", + park_info.start, park_info.start + park_info.len, + park_info.len >> 20); + + return 0; +out: + park_info.start = 0; + park_info.len = 0; + return -EINVAL; +} +#endif + void __init arm64_memblock_init(void) { const s64 linear_region_size = BIT(vabits_actual - 1); @@ -357,6 +408,10 @@ void __init arm64_memblock_init(void) reserve_crashkernel(); +#ifdef CONFIG_ARM64_CPU_PARK + reserve_park_mem(); +#endif + reserve_elfcorehdr(); high_memory = __va(memblock_end_of_DRAM() - 1) + 1; -- 2.9.5

2 3

[PATCH OLK-5.10 1/3] kexec: Add quick kexec support for kernel
by sangyan＠huawei.com 22 Feb '21

22 Feb '21

From: Sang Yan <sangyan(a)huawei.com> hulk inclusion category: feature bugzilla: 48159 CVE: N/A In normal kexec, relocating kernel may cost 5 ~ 10 seconds, to copy all segments from vmalloced memory to kernel boot memory, because of disabled mmu. We introduce quick kexec to save time of copying memory as above, just like kdump(kexec on crash), by using reserved memory "Quick Kexec". Constructing quick kimage as the same as crash kernel, then simply copy all segments of kimage to reserved memroy. We also add this support in syscall kexec_load using flags of KEXEC_QUICK. Signed-off-by: Sang Yan <sangyan(a)huawei.com> --- arch/Kconfig | 10 ++++++++++ include/linux/ioport.h | 1 + include/linux/kexec.h | 11 ++++++++++- include/uapi/linux/kexec.h | 1 + kernel/kexec.c | 10 ++++++++++ kernel/kexec_core.c | 42 +++++++++++++++++++++++++++++++++--------- 6 files changed, 65 insertions(+), 10 deletions(-) diff --git a/arch/Kconfig b/arch/Kconfig index 2592b4b..7811eee 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -18,6 +18,16 @@ config KEXEC_CORE select CRASH_CORE bool +config QUICK_KEXEC + bool "Support for quick kexec" + depends on KEXEC_CORE + help + It uses pre-reserved memory to accelerate kexec, just like + crash kexec, loads new kernel and initrd to reserved memory, + and boots new kernel on that memory. It will save the time + of relocating kernel. + + config KEXEC_ELF bool diff --git a/include/linux/ioport.h b/include/linux/ioport.h index 5135d4b..84a716f 100644 --- a/include/linux/ioport.h +++ b/include/linux/ioport.h @@ -139,6 +139,7 @@ enum { IORES_DESC_DEVICE_PRIVATE_MEMORY = 6, IORES_DESC_RESERVED = 7, IORES_DESC_SOFT_RESERVED = 8, + IORES_DESC_QUICK_KEXEC = 9, }; /* diff --git a/include/linux/kexec.h b/include/linux/kexec.h index f301f2f..7fff410 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -269,9 +269,10 @@ struct kimage { unsigned long control_page; /* Flags to indicate special processing */ - unsigned int type : 1; + unsigned int type : 2; #define KEXEC_TYPE_DEFAULT 0 #define KEXEC_TYPE_CRASH 1 +#define KEXEC_TYPE_QUICK 2 unsigned int preserve_context : 1; /* If set, we are using file mode kexec syscall */ unsigned int file_mode:1; @@ -331,6 +332,11 @@ extern int kexec_load_disabled; #define KEXEC_FLAGS (KEXEC_ON_CRASH | KEXEC_PRESERVE_CONTEXT) #endif +#ifdef CONFIG_QUICK_KEXEC +#undef KEXEC_FLAGS +#define KEXEC_FLAGS (KEXEC_ON_CRASH | KEXEC_QUICK) +#endif + /* List of defined/legal kexec file flags */ #define KEXEC_FILE_FLAGS (KEXEC_FILE_UNLOAD | KEXEC_FILE_ON_CRASH | \ KEXEC_FILE_NO_INITRAMFS) @@ -338,6 +344,9 @@ extern int kexec_load_disabled; /* Location of a reserved region to hold the crash kernel. */ extern note_buf_t __percpu *crash_notes; +#ifdef CONFIG_QUICK_KEXEC +extern struct resource quick_kexec_res; +#endif /* flag to track if kexec reboot is in progress */ extern bool kexec_in_progress; diff --git a/include/uapi/linux/kexec.h b/include/uapi/linux/kexec.h index 05669c8..d891d80 100644 --- a/include/uapi/linux/kexec.h +++ b/include/uapi/linux/kexec.h @@ -12,6 +12,7 @@ /* kexec flags for different usage scenarios */ #define KEXEC_ON_CRASH 0x00000001 #define KEXEC_PRESERVE_CONTEXT 0x00000002 +#define KEXEC_QUICK 0x00000004 #define KEXEC_ARCH_MASK 0xffff0000 /* diff --git a/kernel/kexec.c b/kernel/kexec.c index c82c6c0..4acc909 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -44,6 +44,9 @@ static int kimage_alloc_init(struct kimage **rimage, unsigned long entry, int ret; struct kimage *image; bool kexec_on_panic = flags & KEXEC_ON_CRASH; +#ifdef CONFIG_QUICK_KEXEC + bool kexec_on_quick = flags & KEXEC_QUICK; +#endif if (kexec_on_panic) { /* Verify we have a valid entry point */ @@ -69,6 +72,13 @@ static int kimage_alloc_init(struct kimage **rimage, unsigned long entry, image->type = KEXEC_TYPE_CRASH; } +#ifdef CONFIG_QUICK_KEXEC + if (kexec_on_quick) { + image->control_page = quick_kexec_res.start; + image->type = KEXEC_TYPE_QUICK; + } +#endif + ret = sanity_check_segment_list(image); if (ret) goto out_free_image; diff --git a/kernel/kexec_core.c b/kernel/kexec_core.c index 2ca8875..c7e2aa2 100644 --- a/kernel/kexec_core.c +++ b/kernel/kexec_core.c @@ -53,6 +53,17 @@ note_buf_t __percpu *crash_notes; /* Flag to indicate we are going to kexec a new kernel */ bool kexec_in_progress = false; +/* Resource for quick kexec */ +#ifdef CONFIG_QUICK_KEXEC +struct resource quick_kexec_res = { + .name = "Quick kexec", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_SYSTEM_RAM, + .desc = IORES_DESC_QUICK_KEXEC +}; +#endif + int kexec_should_crash(struct task_struct *p) { /* @@ -396,8 +407,9 @@ static struct page *kimage_alloc_normal_control_pages(struct kimage *image, return pages; } -static struct page *kimage_alloc_crash_control_pages(struct kimage *image, - unsigned int order) +static struct page *kimage_alloc_special_control_pages(struct kimage *image, + unsigned int order, + unsigned long end) { /* Control pages are special, they are the intermediaries * that are needed while we copy the rest of the pages @@ -427,7 +439,7 @@ static struct page *kimage_alloc_crash_control_pages(struct kimage *image, size = (1 << order) << PAGE_SHIFT; hole_start = (image->control_page + (size - 1)) & ~(size - 1); hole_end = hole_start + size - 1; - while (hole_end <= crashk_res.end) { + while (hole_end <= end) { unsigned long i; cond_resched(); @@ -462,7 +474,6 @@ static struct page *kimage_alloc_crash_control_pages(struct kimage *image, return pages; } - struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order) { @@ -473,8 +484,15 @@ struct page *kimage_alloc_control_pages(struct kimage *image, pages = kimage_alloc_normal_control_pages(image, order); break; case KEXEC_TYPE_CRASH: - pages = kimage_alloc_crash_control_pages(image, order); + pages = kimage_alloc_special_control_pages(image, order, + crashk_res.end); + break; +#ifdef CONFIG_QUICK_KEXEC + case KEXEC_TYPE_QUICK: + pages = kimage_alloc_special_control_pages(image, order, + quick_kexec_res.end); break; +#endif } return pages; @@ -830,11 +848,12 @@ static int kimage_load_normal_segment(struct kimage *image, return result; } -static int kimage_load_crash_segment(struct kimage *image, +static int kimage_load_special_segment(struct kimage *image, struct kexec_segment *segment) { - /* For crash dumps kernels we simply copy the data from - * user space to it's destination. + /* + * For crash dumps kernels and quick kexec kernels + * we simply copy the data from user space to it's destination. * We do things a page at a time for the sake of kmap. */ unsigned long maddr; @@ -908,8 +927,13 @@ int kimage_load_segment(struct kimage *image, result = kimage_load_normal_segment(image, segment); break; case KEXEC_TYPE_CRASH: - result = kimage_load_crash_segment(image, segment); + result = kimage_load_special_segment(image, segment); break; +#ifdef CONFIG_QUICK_KEXEC + case KEXEC_TYPE_QUICK: + result = kimage_load_special_segment(image, segment); + break; +#endif } return result; -- 2.9.5

2 3

[PATCH openEuler-1.0-LTS 000/316] bugfix for 20.03
by Cheng Jian 22 Feb '21

22 Feb '21

bugfix for 20.03 @ 2021/02/22 Aichun Li (7): netpoll: remove dev argument from netpoll_send_skb_on_dev() netpoll: move netpoll_send_skb() out of line netpoll: netpoll_send_skb() returns transmit status netpoll: accept NULL np argument in netpoll_send_skb() bonding: add an option to specify a delay between peer notifications bonding: fix value exported by Netlink for peer_notif_delay bonding: add documentation for peer_notif_delay Akilesh Kailash (1): dm snapshot: flush merged data before committing metadata Al Viro (2): don't dump the threads that had been already exiting when zapped. dump_common_audit_data(): fix racy accesses to ->d_name Aleksandr Nogikh (1): netem: fix zero division in tabledist Alexander Duyck (1): tcp: Set INET_ECN_xmit configuration in tcp_reinit_congestion_control Alexander Lobakin (1): skbuff: back tiny skbs with kmalloc() in __netdev_alloc_skb() too Alexey Dobriyan (2): proc: change ->nlink under proc_subdir_lock proc: fix lookup in /proc/net subdirectories after setns(2) Alexey Kardashevskiy (1): serial_core: Check for port state when tty is in error state Amit Cohen (1): mlxsw: core: Fix use-after-free in mlxsw_emad_trans_finish() Andy Shevchenko (2): device property: Keep secondary firmware node secondary by type device property: Don't clear secondary pointer for shared primary firmware node Antoine Tenart (6): netfilter: bridge: reset skb->pkt_type after NF_INET_POST_ROUTING traversal net: ip6_gre: set dev->hard_header_len when using header_ops net-sysfs: take the rtnl lock when storing xps_cpus net-sysfs: take the rtnl lock when accessing xps_cpus_map and num_tc net-sysfs: take the rtnl lock when storing xps_rxqs net-sysfs: take the rtnl lock when accessing xps_rxqs_map and num_tc Ard Biesheuvel (3): efivarfs: revert "fix memory leak in efivarfs_create()" arm64, mm, efi: Account for GICv3 LPI tables in static memblock reserve table efi/arm: Revert "Defer persistent reservations until after paging_init()" 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 Axel Lin (1): ASoC: msm8916-wcd-digital: Select REGMAP_MMIO to fix build error Aya Levin (1): net: ipv6: Validate GSO SKB before finish IPv6 processing Bart Van Assche (1): scsi: scsi_transport_spi: Set RQF_PM for domain validation commands Bharat Gooty (1): PCI: iproc: Fix out-of-bound array accesses Bixuan Cui (2): mmap: fix a compiling error for 'MAP_CHECKNODE' powerpc: fix a compiling error for 'access_ok' Bjorn Helgaas (1): PCI: Bounds-check command-line resource alignment requests Björn Töpel (1): ixgbe: avoid premature Rx buffer reuse Boqun Feng (1): fcntl: Fix potential deadlock in send_sig{io, urg}() 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 Calum Mackay (1): lockd: don't use interval-based rebinding over TCP Chen Zhou (1): selinux: Fix error return code in sel_ib_pkey_sid_slow() Chenguangli (1): scsi/hifc:Fix the bug that the system may be oops during unintall hifc module. Cheng Lin (1): nfs_common: need lock during iterate through the list 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 Chunguang Xu (1): ext4: fix a memory leak of ext4_free_data Chunyan Zhang (1): tick/common: Touch watchdog in tick_unfreeze() on all CPUs Colin Ian King (1): PCI: Fix overflow in command-line resource alignment requests Cong Wang (1): erspan: fix version 1 check in gre_parse_header() Damien Le Moal (1): null_blk: Fix zone size initialization Dan Carpenter (1): futex: Don't enable IRQs unconditionally in put_pi_state() Daniel Scally (1): Revert "ACPI / resources: Use AE_CTRL_TERMINATE to terminate resources walks" 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" Dave Wysochanski (1): NFS4: Fix use-after-free in trace_event_raw_event_nfs4_set_lock Dexuan Cui (1): ACPI: scan: Harden acpi_device_add() against device ID overflows Dinghao Liu (4): ext4: fix error handling code in add_new_gdb net/mlx5e: Fix memleak in mlx5e_create_l2_table_groups net/mlx5e: Fix two double free cases netfilter: nf_nat: Fix memleak in nf_nat_init Dongdong Wang (1): lwt: Disable BH too in run_lwt_bpf() 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 Eran Ben Elisha (1): net/mlx5: Fix wrong address reclaim when command interface is down Eric Auger (1): vfio/pci: Move dummy_resources_list init in vfio_pci_probe() Eric Biggers (1): ext4: fix leaking sysfs kobject after failed mount Eric Dumazet (4): tcp: select sane initial rcvq_space.space for big MSS net: avoid 32 x truesize under-estimation for tiny skbs net_sched: avoid shift-out-of-bounds in tcindex_set_parms() net_sched: reject silly cell_log in qdisc_get_rtab() Fang Lijun (3): arm64/ascend: mm: Add MAP_CHECKNODE flag to check node hugetlb arm64/ascend: mm: Fix arm32 compile warnings arm64/ascend: mm: Fix hugetlb check node error Fangrui Song (1): arm64: Change .weak to SYM_FUNC_START_WEAK_PI for arch/arm64/lib/mem*.S Florian Fainelli (1): net: Have netpoll bring-up DSA management interface Florian Westphal (4): netfilter: nf_tables: avoid false-postive lockdep splat netfilter: xt_RATEEST: reject non-null terminated string from userspace net: ip: always refragment ip defragmented packets net: fix pmtu check in nopmtudisc mode 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() Guillaume Nault (4): ipv4: Fix tos mask in inet_rtm_getroute() ipv4: Ignore ECN bits for fib lookups in fib_compute_spec_dst() netfilter: rpfilter: mask ecn bits before fib lookup udp: mask TOS bits in udp_v4_early_demux() Hanjun Guo (1): clocksource/drivers/arch_timer: Fix vdso_fix compile error for arm32 Hannes Reinecke (1): dm: avoid filesystem lookup in dm_get_dev_t() Hans de Goede (1): ACPI: scan: Make acpi_bus_get_device() clear return pointer on error Heiner Kallweit (1): net: bridge: add missing counters to ndo_get_stats64 callback Hoang Le (1): tipc: fix NULL deref in tipc_link_xmit() Huang Shijie (1): lib/genalloc: fix the overflow when size is too big Huang Ying (1): mm: fix a race during THP splitting Hugh Dickins (2): mlock: fix unevictable_pgs event counts on THP mm: fix check_move_unevictable_pages() on THP Hui Wang (1): ACPI: PNP: compare the string length in the matching_id() Hyeongseok Kim (1): dm verity: skip verity work if I/O error when system is shutting down Ido Schimmel (2): mlxsw: core: Fix memory leak on module removal mlxsw: core: Use variable timeout for EMAD retries Ilya Dryomov (1): libceph: clear con->out_msg on Policy::stateful_server faults Jakub Kicinski (1): net: vlan: avoid leaks on register_vlan_dev() failures Jamie Iles (1): bonding: wait for sysfs kobject destruction before freeing struct slave Jan Kara (8): ext4: Detect already used quota file early ext4: fix bogus warning in ext4_update_dx_flag() ext4: Protect superblock modifications with a buffer lock ext4: fix deadlock with fs freezing and EA inodes ext4: don't remount read-only with errors=continue on reboot quota: Don't overflow quota file offsets bfq: Fix computation of shallow depth ext4: fix superblock checksum failure when setting password salt Jann Horn (1): mm, slub: consider rest of partial list if acquire_slab() fails Jason A. Donenfeld (3): netfilter: use actual socket sk rather than skb sk when routing harder net: introduce skb_list_walk_safe for skb segment walking net: skbuff: disambiguate argument and member for skb_list_walk_safe helper Jeff Dike (1): virtio_net: Fix recursive call to cpus_read_lock() Jens Axboe (1): proc: don't allow async path resolution of /proc/self components Jesper Dangaard Brouer (1): netfilter: conntrack: fix reading nf_conntrack_buckets Jessica Yu (1): module: delay kobject uevent until after module init call Jiri Olsa (2): perf python scripting: Fix printable strings in python3 scripts perf tools: Add missing swap for ino_generation Johannes Thumshirn (1): block: factor out requeue handling from dispatch code 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() Jubin Zhong (1): PCI: Fix pci_slot_release() NULL pointer dereference Kaixu Xia (1): ext4: correctly report "not supported" for {usr, grp}jquota when !CONFIG_QUOTA Keqian Zhu (1): clocksource/drivers/arm_arch_timer: Correct fault programming of CNTKCTL_EL1.EVNTI Kirill Tkhai (1): mm: move nr_deactivate accounting to shrink_active_list() Lang Dai (1): uio: free uio id after uio file node is freed Lecopzer Chen (2): kasan: fix unaligned address is unhandled in kasan_remove_zero_shadow kasan: fix incorrect arguments passing in kasan_add_zero_shadow Lee Duncan (1): scsi: libiscsi: Fix NOP race condition 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 Leon Romanovsky (1): net/mlx5: Properly convey driver version to firmware Lijie (1): config: enable CONFIG_NVME_MULTIPATH by default Liu Shixin (3): config: set default value of CONFIG_TEST_FREE_PAGES mm: memcontrol: add struct mem_cgroup_extension mm: fix kabi broken Lorenzo Pieralisi (1): asm-generic/io.h: Fix !CONFIG_GENERIC_IOMAP pci_iounmap() implementation Lu Jialin (1): fs: fix files.usage bug when move tasks Luc Van Oostenryck (1): xsk: Fix xsk_poll()'s return type 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 (2): arm64: Run ARCH_WORKAROUND_1 enabling code on all CPUs genirq/irqdomain: Don't try to free an interrupt that has no mapping Mark Rutland (3): arm64: syscall: exit userspace before unmasking exceptions arm64: module: rework special section handling arm64: module/ftrace: intialize PLT at load time Martin Wilck (1): scsi: core: Fix VPD LUN ID designator priorities Mateusz Nosek (1): futex: Fix incorrect should_fail_futex() handling Matteo Croce (4): Revert "kernel/reboot.c: convert simple_strtoul to kstrtoint" reboot: fix overflow parsing reboot cpu number ipv6: create multicast route with RTPROT_KERNEL ipv6: set multicast flag on the multicast route Matthew Wilcox (Oracle) (1): mm/page_alloc.c: fix freeing non-compound pages Maurizio Lombardi (2): scsi: target: remove boilerplate code scsi: target: fix hang when multiple threads try to destroy the same iscsi session Maxim Mikityanskiy (1): net/tls: Protect from calling tls_dev_del for TLS RX twice Miaohe Lin (1): mm/hugetlb: fix potential missing huge page size info Michael Schaller (1): efivarfs: Replace invalid slashes with exclamation marks in dentries. Mike Christie (1): scsi: target: iscsi: Fix cmd abort fabric stop race Mike Galbraith (1): futex: Handle transient "ownerless" rtmutex state correctly Miklos Szeredi (1): fuse: fix page dereference after free Mikulas Patocka (3): dm integrity: fix the maximum number of arguments dm integrity: fix flush with external metadata device dm integrity: fix a crash if "recalculate" used without "internal_hash" Ming Lei (2): scsi: core: Don't start concurrent async scan on same host block: fix use-after-free in disk_part_iter_next Minwoo Im (1): nvme: free sq/cq dbbuf pointers when dbbuf set fails Miroslav Benes (1): module: set MODULE_STATE_GOING state when a module fails to load Moshe Shemesh (2): net/mlx4_en: Avoid scheduling restart task if it is already running net/mlx4_en: Handle TX error CQE Naoya Horiguchi (1): mm, hwpoison: double-check page count in __get_any_page() Naveen N. Rao (1): ftrace: Fix updating FTRACE_FL_TRAMP Neal Cardwell (1): tcp: fix cwnd-limited bug for TSO deferral where we send nothing NeilBrown (1): NFS: switch nfsiod to be an UNBOUND workqueue. 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 Paul Moore (1): selinux: fix inode_doinit_with_dentry() LABEL_INVALID error handling Paulo Alcantara (1): cifs: fix potential use-after-free in cifs_echo_request() Peng Liu (1): sched/deadline: Fix sched_dl_global_validate() Peter Zijlstra (2): serial: pl011: Fix lockdep splat when handling magic-sysrq interrupt perf: Fix get_recursion_context() Petr Malat (1): sctp: Fix COMM_LOST/CANT_STR_ASSOC err reporting on big-endian platforms Qian Cai (1): mm/swapfile: do not sleep with a spin lock held Qiujun Huang (2): ring-buffer: Return 0 on success from ring_buffer_resize() tracing: Fix out of bounds write in get_trace_buf Rafael J. Wysocki (1): driver core: Extend device_is_dependent() Randy Dunlap (1): net: sched: prevent invalid Scell_log shift count Ritika Srivastava (2): block: Return blk_status_t instead of errno codes block: better deal with the delayed not supported case in blk_cloned_rq_check_limits Ronnie Sahlberg (1): cifs: handle -EINTR in cifs_setattr Ryan Sharpelletti (1): tcp: only postpone PROBE_RTT if RTT is < current min_rtt estimate Sami Tolvanen (1): arm64: lse: fix LSE atomics with LLVM's integrated assembler Sean Tranchetti (1): net: ipv6: fib: flush exceptions when purging route Shakeel Butt (2): mm: swap: fix vmstats for huge pages mm: swap: memcg: fix memcg stats for huge pages 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) (4): ring-buffer: Fix recursion protection transitions between interrupt context ftrace: Fix recursion check for NMI test ftrace: Handle tracing when switching between context tracing: Fix userstacktrace option for instances Subash Abhinov Kasiviswanathan (2): netfilter: x_tables: Switch synchronization to RCU netfilter: x_tables: Update remaining dereference to RCU Sven Eckelmann (2): vxlan: Add needed_headroom for lower device vxlan: Copy needed_tailroom from lowerdev Sylwester Dziedziuch (2): i40e: Fix removing driver while bare-metal VFs pass traffic i40e: Fix Error I40E_AQ_RC_EINVAL when removing VFs Takashi Iwai (1): libata: transport: Use scnprintf() for avoiding potential buffer overflow Tariq Toukan (1): net: Disable NETIF_F_HW_TLS_RX when RXCSUM is disabled Thomas Gleixner (12): sched: Reenable interrupts in do_sched_yield() futex: Move futex exit handling into futex code futex: Replace PF_EXITPIDONE with a state exit/exec: Seperate mm_release() futex: Split futex_mm_release() for exit/exec futex: Set task::futex_state to DEAD right after handling futex exit futex: Mark the begin of futex exit explicitly futex: Sanitize exit state handling futex: Provide state handling for exec() as well futex: Add mutex around futex exit futex: Provide distinct return value when owner is exiting futex: Prevent exit livelock Tianyue Ren (1): selinux: fix error initialization in inode_doinit_with_dentry() Trond Myklebust (5): SUNRPC: xprt_load_transport() needs to support the netid "rdma6" NFSv4: Fix a pNFS layout related use-after-free race when freeing the inode pNFS: Mark layout for return if return-on-close was not sent NFS/pNFS: Fix a leak of the layout 'plh_outstanding' counter NFS: nfs_igrab_and_active must first reference the superblock Tung Nguyen (1): tipc: fix memory leak caused by tipc_buf_append() Tyler Hicks (1): tpm: efi: Don't create binary_bios_measurements file for an empty log Uwe Kleine-König (1): spi: fix resource leak for drivers without .remove callback Vadim Fedorenko (1): net/tls: missing received data after fast remote close Valentin Schneider (1): arm64: topology: Stop using MPIDR for topology information Vamshi K Sthambamkadi (1): efivarfs: fix memory leak in efivarfs_create() Vasily Averin (2): netfilter: ipset: fix shift-out-of-bounds in htable_bits() net: drop bogus skb with CHECKSUM_PARTIAL and offset beyond end of trimmed packet Vincenzo Frascino (1): arm64: lse: Fix LSE atomics with LLVM Vladyslav Tarasiuk (1): net/mlx5: Disable QoS when min_rates on all VFs are zero Wang Hai (4): 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() tipc: fix memory leak in tipc_topsrv_start() ipv6: addrlabel: fix possible memory leak in ip6addrlbl_net_init Wang Wensheng (1): sbsa_gwdt: Add WDIOF_PRETIMEOUT flag to watchdog_info at defination Wei Li (1): irqchip/gic-v3: Fix compiling error on ARM32 with GICv3 Wei Yang (1): mm: thp: don't need care deferred split queue in memcg charge move path Weilong Chen (1): hugetlbfs: Add dependency with ascend memory features Wengang Wang (1): ocfs2: initialize ip_next_orphan Will Deacon (3): arm64: psci: Avoid printing in cpu_psci_cpu_die() arm64: pgtable: Fix pte_accessible() arm64: pgtable: Ensure dirty bit is preserved across pte_wrprotect() Willem de Bruijn (1): sock: set sk_err to ee_errno on dequeue from errq Wu Bo (1): scsi: libiscsi: fix task hung when iscsid deamon exited Xie XiuQi (1): cputime: fix undefined reference to get_idle_time when CONFIG_PROC_FS disabled Xin Long (1): sctp: change to hold/put transport for proto_unreach_timer Xiongfeng Wang (1): arm64: fix compile error when CONFIG_HOTPLUG_CPU is disabled Xiubo Li (1): nbd: make the config put is called before the notifying the waiter Xu Qiang (4): NMI: Enable arm-pmu interrupt as NMI in Acensed. irqchip/gic-v3-its: Unconditionally save/restore the ITS state on suspend. irqchip/irq-gic-v3: Add workaround bindings in device tree to init ts core GICR. Document: In the binding document, add enable-init-all-GICR field description. Yang Shi (6): mm: list_lru: set shrinker map bit when child nr_items is not zero mm: thp: extract split_queue_* into a struct mm: move mem_cgroup_uncharge out of __page_cache_release() mm: shrinker: make shrinker not depend on memcg kmem mm: thp: make deferred split shrinker memcg aware mm: vmscan: protect shrinker idr replace with CONFIG_MEMCG Yang Yingliang (5): arm64: arch_timer: only do cntvct workaround on VDSO path on D05 armv7 fix compile error Kconfig: disable KTASK by default futex: sched: fix kabi broken in task_struct futex: sched: fix UAF when free futex_exit_mutex in free_task() 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() Yu Kuai (2): blk-cgroup: prevent rcu_sched detected stalls warnings in blkg_destroy_all() blk-throttle: don't check whether or not lower limit is valid if CONFIG_BLK_DEV_THROTTLING_LOW is off Yufen Yu (2): bdi: fix compiler error in bdi_get_dev_name() scsi: do quiesce for enclosure driver Yunfeng Ye (1): workqueue: Kick a worker based on the actual activation of delayed works Yunjian Wang (2): net: hns: fix return value check in __lb_other_process() vhost_net: fix ubuf refcount incorrectly when sendmsg fails 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 Changzhong (2): ah6: fix error return code in ah6_input() net: bridge: vlan: fix error return code in __vlan_add() Zhao Heming (2): md/cluster: block reshape with remote resync job md/cluster: fix deadlock when node is doing resync job Zhengyuan Liu (3): arm64/mm: return cpu_all_mask when node is NUMA_NO_NODE hifc: remove unnecessary __init specifier mmap: fix a compiling error for 'MAP_PA32BIT' Zhou Guanghui (2): memcg/ascend: Check sysctl oom config for memcg oom memcg/ascend: enable kmem cgroup by default for ascend Zqiang (1): kthread_worker: prevent queuing delayed work from timer_fn when it is being canceled chenmaodong (1): fix virtio_gpu use-after-free while creating dumb j.nixdorf(a)avm.de (1): net: sunrpc: interpret the return value of kstrtou32 correctly lijinlin (1): ext4: add ext3 report error to userspace by netlink miaoyubo (1): KVM: Enable PUD huge mappings only on 1620 yangerkun (1): ext4: fix bug for rename with RENAME_WHITEOUT zhuoliang zhang (1): net: xfrm: fix a race condition during allocing spi .../interrupt-controller/arm,gic-v3.txt | 4 + Documentation/networking/bonding.txt | 16 +- arch/Kconfig | 7 + arch/alpha/include/uapi/asm/mman.h | 2 + arch/arm/kernel/perf_event_v7.c | 16 +- arch/arm64/Kconfig | 2 + arch/arm64/configs/euleros_defconfig | 1 + arch/arm64/configs/hulk_defconfig | 1 + arch/arm64/configs/openeuler_defconfig | 3 +- arch/arm64/include/asm/atomic_lse.h | 76 ++- arch/arm64/include/asm/cpufeature.h | 4 + arch/arm64/include/asm/lse.h | 6 +- arch/arm64/include/asm/memory.h | 11 + arch/arm64/include/asm/numa.h | 3 + arch/arm64/include/asm/pgtable.h | 34 +- arch/arm64/kernel/cpu_errata.c | 8 + arch/arm64/kernel/cpufeature.c | 2 +- arch/arm64/kernel/ftrace.c | 50 +- arch/arm64/kernel/module.c | 47 +- arch/arm64/kernel/psci.c | 5 +- arch/arm64/kernel/setup.c | 5 +- arch/arm64/kernel/syscall.c | 2 +- arch/arm64/kernel/topology.c | 43 +- arch/arm64/lib/memcpy.S | 3 +- arch/arm64/lib/memmove.S | 3 +- arch/arm64/lib/memset.S | 3 +- arch/arm64/mm/init.c | 17 +- arch/arm64/mm/numa.c | 6 +- arch/mips/include/uapi/asm/mman.h | 2 + arch/parisc/include/uapi/asm/mman.h | 2 + arch/powerpc/include/asm/uaccess.h | 4 +- arch/powerpc/include/uapi/asm/mman.h | 2 + arch/sparc/include/uapi/asm/mman.h | 2 + arch/x86/configs/hulk_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + arch/xtensa/include/uapi/asm/mman.h | 2 + block/bfq-iosched.c | 8 +- block/blk-cgroup.c | 30 +- block/blk-core.c | 39 +- block/blk-mq.c | 29 +- block/blk-throttle.c | 6 + block/genhd.c | 9 +- drivers/acpi/acpi_pnp.c | 3 + drivers/acpi/internal.h | 2 +- drivers/acpi/numa.c | 2 +- drivers/acpi/resource.c | 2 +- drivers/acpi/scan.c | 17 +- drivers/ata/libata-transport.c | 10 +- drivers/base/core.c | 21 +- drivers/block/nbd.c | 3 +- drivers/block/null_blk_zoned.c | 20 +- drivers/char/Kconfig | 2 +- drivers/char/random.c | 1 - drivers/char/tpm/eventlog/efi.c | 5 + drivers/clocksource/arm_arch_timer.c | 36 +- drivers/firmware/efi/efi.c | 4 - drivers/firmware/efi/libstub/arm-stub.c | 3 - drivers/gpu/drm/virtio/virtgpu_gem.c | 4 +- drivers/irqchip/irq-gic-v3-its.c | 30 +- drivers/irqchip/irq-gic-v3.c | 18 +- drivers/md/dm-bufio.c | 6 + drivers/md/dm-integrity.c | 58 ++- drivers/md/dm-snap.c | 24 + drivers/md/dm-table.c | 15 +- drivers/md/dm-verity-target.c | 12 +- drivers/md/md-cluster.c | 67 +-- drivers/md/md.c | 14 +- drivers/mtd/hisilicon/sfc/hrd_sfc_driver.c | 4 +- drivers/net/bonding/bond_main.c | 92 ++-- drivers/net/bonding/bond_netlink.c | 14 + drivers/net/bonding/bond_options.c | 71 ++- drivers/net/bonding/bond_procfs.c | 2 + drivers/net/bonding/bond_sysfs.c | 13 + drivers/net/bonding/bond_sysfs_slave.c | 18 +- .../net/ethernet/hisilicon/hns/hns_ethtool.c | 4 + drivers/net/ethernet/intel/i40e/i40e.h | 4 + drivers/net/ethernet/intel/i40e/i40e_main.c | 32 +- .../ethernet/intel/i40e/i40e_virtchnl_pf.c | 30 +- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 24 +- .../net/ethernet/mellanox/mlx4/en_netdev.c | 21 +- drivers/net/ethernet/mellanox/mlx4/en_tx.c | 40 +- drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 12 +- .../net/ethernet/mellanox/mlx5/core/en_fs.c | 3 + .../net/ethernet/mellanox/mlx5/core/eswitch.c | 15 +- .../net/ethernet/mellanox/mlx5/core/main.c | 6 +- .../ethernet/mellanox/mlx5/core/pagealloc.c | 21 +- drivers/net/ethernet/mellanox/mlxsw/core.c | 8 +- drivers/net/geneve.c | 3 +- drivers/net/macvlan.c | 5 +- drivers/net/virtio_net.c | 12 +- drivers/net/vxlan.c | 3 + drivers/nvme/host/pci.c | 15 + drivers/pci/controller/pcie-iproc.c | 10 +- drivers/pci/pci.c | 14 +- drivers/pci/slot.c | 6 +- drivers/scsi/huawei/hifc/unf_common.h | 2 +- drivers/scsi/huawei/hifc/unf_scsi.c | 23 + drivers/scsi/libiscsi.c | 33 +- drivers/scsi/scsi_lib.c | 126 +++-- drivers/scsi/scsi_scan.c | 11 +- drivers/scsi/scsi_transport_spi.c | 27 +- drivers/spi/spi.c | 19 +- drivers/target/iscsi/iscsi_target.c | 96 ++-- drivers/target/iscsi/iscsi_target.h | 1 - drivers/target/iscsi/iscsi_target_configfs.c | 5 +- drivers/target/iscsi/iscsi_target_login.c | 5 +- drivers/tty/serial/amba-pl011.c | 11 +- drivers/tty/serial/serial_core.c | 4 + drivers/uio/uio.c | 12 +- drivers/vfio/pci/vfio_pci.c | 3 +- drivers/vhost/net.c | 6 +- drivers/watchdog/sbsa_gwdt.c | 6 +- fs/cifs/cifs_unicode.c | 8 +- fs/cifs/connect.c | 2 + fs/cifs/inode.c | 13 +- fs/efivarfs/inode.c | 2 + fs/efivarfs/super.c | 3 + fs/exec.c | 17 +- fs/ext4/ext4.h | 6 +- fs/ext4/ext4_jbd2.c | 1 - fs/ext4/file.c | 1 + fs/ext4/inline.c | 1 + fs/ext4/inode.c | 31 +- fs/ext4/ioctl.c | 3 + fs/ext4/mballoc.c | 1 + fs/ext4/namei.c | 23 +- fs/ext4/resize.c | 8 +- fs/ext4/super.c | 32 +- fs/ext4/xattr.c | 1 + fs/fcntl.c | 10 +- fs/filescontrol.c | 73 +-- fs/fuse/dev.c | 28 +- fs/hugetlbfs/inode.c | 2 +- fs/libfs.c | 6 +- fs/lockd/host.c | 20 +- fs/nfs/inode.c | 2 +- fs/nfs/internal.h | 12 +- fs/nfs/nfs4proc.c | 2 +- fs/nfs/nfs4super.c | 2 +- fs/nfs/pnfs.c | 40 +- fs/nfs/pnfs.h | 5 + fs/nfs_common/grace.c | 6 +- fs/ocfs2/super.c | 1 + fs/proc/generic.c | 55 ++- fs/proc/internal.h | 7 + fs/proc/proc_net.c | 16 - fs/proc/self.c | 7 + fs/quota/quota_tree.c | 8 +- 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/asm-generic/io.h | 39 +- include/linux/backing-dev.h | 1 + include/linux/blkdev.h | 1 + include/linux/compat.h | 2 - include/linux/dm-bufio.h | 1 + include/linux/efi.h | 7 - include/linux/futex.h | 39 +- include/linux/huge_mm.h | 9 + include/linux/hugetlb.h | 10 +- include/linux/if_team.h | 5 +- include/linux/memblock.h | 3 - include/linux/memcontrol.h | 32 +- include/linux/mm.h | 2 + include/linux/mm_types.h | 1 + include/linux/mman.h | 15 + include/linux/mmzone.h | 8 + include/linux/netfilter/x_tables.h | 5 +- include/linux/netfilter_ipv4.h | 2 +- include/linux/netfilter_ipv6.h | 2 +- include/linux/netpoll.h | 10 +- include/linux/prandom.h | 36 +- include/linux/proc_fs.h | 8 +- include/linux/sched.h | 5 +- include/linux/sched/cputime.h | 5 + include/linux/sched/mm.h | 6 +- include/linux/shrinker.h | 7 +- include/linux/skbuff.h | 5 + include/linux/sunrpc/xprt.h | 1 + include/linux/time64.h | 4 + include/net/bond_options.h | 1 + include/net/bonding.h | 14 +- include/net/ip_tunnels.h | 7 +- include/net/red.h | 4 +- include/net/tls.h | 6 + include/scsi/libiscsi.h | 3 + include/target/iscsi/iscsi_target_core.h | 2 +- include/uapi/asm-generic/mman.h | 1 + include/uapi/linux/if_link.h | 1 + init/Kconfig | 2 +- kernel/events/internal.h | 2 +- kernel/exit.c | 35 +- kernel/fail_function.c | 5 +- kernel/fork.c | 61 ++- kernel/futex.c | 291 +++++++++-- kernel/irq/irqdomain.c | 11 +- kernel/kthread.c | 3 +- kernel/module.c | 6 +- kernel/reboot.c | 28 +- kernel/sched/core.c | 6 +- kernel/sched/cputime.c | 6 + kernel/sched/deadline.c | 5 +- kernel/sched/sched.h | 42 +- kernel/signal.c | 19 +- kernel/time/itimer.c | 4 - kernel/time/tick-common.c | 2 + kernel/time/timer.c | 7 - kernel/trace/ftrace.c | 22 +- kernel/trace/ring_buffer.c | 66 ++- kernel/trace/trace.c | 9 +- kernel/trace/trace.h | 32 +- kernel/trace/trace_selftest.c | 9 +- kernel/workqueue.c | 13 +- lib/Kconfig.debug | 9 + lib/Makefile | 1 + 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/genalloc.c | 25 +- lib/random32.c | 462 +++++++++++------- lib/scatterlist.c | 2 +- lib/test_free_pages.c | 42 ++ mm/huge_memory.c | 167 +++++-- mm/hugetlb.c | 5 +- mm/kasan/kasan_init.c | 23 +- mm/list_lru.c | 10 +- mm/memblock.c | 11 +- mm/memcontrol.c | 34 +- mm/memory-failure.c | 6 + mm/mempolicy.c | 6 +- mm/mlock.c | 25 +- mm/mmap.c | 27 +- mm/page_alloc.c | 9 + mm/slub.c | 2 +- mm/swap.c | 37 +- mm/swapfile.c | 4 +- mm/vmscan.c | 64 +-- net/8021q/vlan.c | 3 +- net/8021q/vlan_dev.c | 5 +- net/bridge/br_device.c | 1 + net/bridge/br_netfilter_hooks.c | 7 +- net/bridge/br_private.h | 5 +- net/bridge/br_vlan.c | 4 +- net/ceph/messenger.c | 5 + net/core/dev.c | 5 + net/core/devlink.c | 6 +- net/core/lwt_bpf.c | 8 +- net/core/net-sysfs.c | 65 ++- net/core/netpoll.c | 51 +- net/core/skbuff.c | 23 +- net/dsa/slave.c | 5 +- net/ipv4/fib_frontend.c | 2 +- net/ipv4/gre_demux.c | 2 +- net/ipv4/inet_diag.c | 4 +- net/ipv4/ip_output.c | 2 +- net/ipv4/ip_tunnel.c | 10 +- net/ipv4/netfilter.c | 12 +- net/ipv4/netfilter/arp_tables.c | 16 +- net/ipv4/netfilter/ip_tables.c | 16 +- net/ipv4/netfilter/ipt_SYNPROXY.c | 2 +- net/ipv4/netfilter/ipt_rpfilter.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/route.c | 7 +- net/ipv4/syncookies.c | 9 +- net/ipv4/tcp.c | 2 + net/ipv4/tcp_bbr.c | 2 +- net/ipv4/tcp_cong.c | 5 + net/ipv4/tcp_input.c | 6 +- net/ipv4/tcp_output.c | 9 +- net/ipv4/udp.c | 3 +- net/ipv6/addrconf.c | 3 +- net/ipv6/addrlabel.c | 26 +- net/ipv6/ah6.c | 3 +- net/ipv6/ip6_fib.c | 5 +- net/ipv6/ip6_gre.c | 16 +- net/ipv6/ip6_output.c | 40 +- net/ipv6/netfilter.c | 6 +- net/ipv6/netfilter/ip6_tables.c | 16 +- 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/ipset/ip_set_hash_gen.h | 20 +- net/netfilter/ipvs/ip_vs_core.c | 4 +- net/netfilter/nf_conntrack_standalone.c | 3 + net/netfilter/nf_nat_core.c | 1 + net/netfilter/nf_tables_api.c | 3 +- net/netfilter/x_tables.c | 49 +- net/netfilter/xt_RATEEST.c | 3 + net/sched/cls_tcindex.c | 8 +- net/sched/sch_api.c | 3 +- net/sched/sch_choke.c | 2 +- net/sched/sch_generic.c | 3 + net/sched/sch_gred.c | 2 +- net/sched/sch_netem.c | 9 +- net/sched/sch_red.c | 2 +- net/sched/sch_sfq.c | 2 +- net/sctp/input.c | 4 +- net/sctp/sm_sideeffect.c | 8 +- net/sctp/transport.c | 2 +- net/sunrpc/addr.c | 2 +- net/sunrpc/xprt.c | 65 ++- net/sunrpc/xprtrdma/module.c | 1 + net/sunrpc/xprtrdma/transport.c | 1 + net/sunrpc/xprtsock.c | 4 + net/tipc/link.c | 9 +- net/tipc/msg.c | 5 +- net/tipc/topsrv.c | 10 +- net/tls/tls_device.c | 5 +- net/tls/tls_sw.c | 6 + net/xdp/xsk.c | 8 +- net/xfrm/xfrm_state.c | 8 +- security/lsm_audit.c | 7 +- security/selinux/hooks.c | 16 +- security/selinux/ibpkey.c | 4 +- sound/soc/codecs/Kconfig | 1 + tools/include/uapi/linux/if_link.h | 1 + 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 + virt/kvm/arm/mmu.c | 7 + 344 files changed, 3424 insertions(+), 1701 deletions(-) create mode 100644 lib/test_free_pages.c -- 2.25.1

1 316

[PATCH 001/316] blk-cgroup: prevent rcu_sched detected stalls warnings in blkg_destroy_all()
by Cheng Jian 22 Feb '21

22 Feb '21

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> Signed-off-by: Cheng Jian <cj.chengjian(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

[PATCH 001/316] blk-cgroup: prevent rcu_sched detected stalls warnings in blkg_destroy_all()
by Cheng Jian 22 Feb '21

22 Feb '21

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> Signed-off-by: Cheng Jian <cj.chengjian(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 4

openEuler kernel sig meeting
by openEuler conference 19 Feb '21

19 Feb '21

2 1

Re: [Dev] openEuler kernel sig meeting
by gwei3 19 Feb '21

19 Feb '21

申报议题： 2. 讨论设立sig安全联络员及如何缩短sig所属项目的CVE响应时间魏刚 ---Original--- From: "openEuler conference"<public(a)openeuler.io> Date: Thu, Feb 18, 2021 17:12 PM To: "dev"<dev(a)openeuler.org>gt;;"kernel"<kernel(a)openeuler.org>gt;; Subject: [Dev] openEuler kernel sig meeting 您好！ openEuler Kernel SIG 邀请您参加 2021-02-19 10:00 召开的ZOOM会议会议主题：openEuler kernel sig meeting 会议内容：1. 树莓派补丁合入openEuler 5.10 内核讨论会议链接：https://zoom.us/j/92887086207?pwd=emdmSVVFekh5eXRSczNNTngrcWYzUT09 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 2021-02-19 10:00, The subject of the conference is openEuler kernel sig meeting, Summary: 1. 树莓派补丁合入openEuler 5.10 内核讨论 You can join the meeting at https://zoom.us/j/92887086207?pwd=emdmSVVFekh5eXRSczNNTngrcWYzUT09. More information

1 0

[PATCH kernel-4.19 00/18] backport stable-4.19-174
by Yang Yingliang 09 Feb '21

09 Feb '21

Arnold Gozum (1): platform/x86: intel-vbtn: Support for tablet mode on Dell Inspiron 7352 Brian King (1): scsi: ibmvfc: Set default timeout to avoid crash during migration Christian Brauner (1): sysctl: handle overflow in proc_get_long Eric Dumazet (1): net_sched: gen_estimator: support large ewma log Felix Fietkau (1): mac80211: fix fast-rx encryption check Greg Kroah-Hartman (1): Linux 4.19.174 Hans de Goede (1): platform/x86: touchscreen_dmi: Add swap-x-y quirk for Goodix touchscreen on Estar Beauty HD tablet Javed Hasan (1): scsi: libfc: Avoid invoking response handler twice if ep is already completed Josh Poimboeuf (1): objtool: Don't fail on missing symbol table Lijun Pan (1): ibmvnic: Ensure that CRQ entry read are correctly ordered Martin Wilck (1): scsi: scsi_transport_srp: Don't block target in failfast state Michael Ellerman (1): selftests/powerpc: Only test lwm/stmw on big endian Pan Bian (1): net: dsa: bcm_sf2: put device node before return Peter Zijlstra (3): x86: __always_inline __{rd,wr}msr() kthread: Extract KTHREAD_IS_PER_CPU workqueue: Restrict affinity change to rescuer Rafael J. Wysocki (1): ACPI: thermal: Do not call acpi_thermal_check() directly Tony Lindgren (1): phy: cpcap-usb: Fix warning for missing regulator_disable Makefile | 2 +- arch/x86/include/asm/msr.h | 4 +- drivers/acpi/thermal.c | 55 +++++++++++++------ drivers/net/dsa/bcm_sf2.c | 8 ++- drivers/net/ethernet/ibm/ibmvnic.c | 6 ++ drivers/phy/motorola/phy-cpcap-usb.c | 19 +++++-- drivers/platform/x86/intel-vbtn.c | 6 ++ drivers/platform/x86/touchscreen_dmi.c | 18 ++++++ drivers/scsi/ibmvscsi/ibmvfc.c | 4 +- drivers/scsi/libfc/fc_exch.c | 16 +++++- drivers/scsi/scsi_transport_srp.c | 9 ++- include/linux/kthread.h | 3 + kernel/kthread.c | 27 ++++++++- kernel/smpboot.c | 1 + kernel/sysctl.c | 2 + kernel/workqueue.c | 9 +-- net/core/gen_estimator.c | 11 ++-- net/mac80211/rx.c | 2 + tools/objtool/elf.c | 7 ++- .../powerpc/alignment/alignment_handler.c | 5 +- 20 files changed, 168 insertions(+), 46 deletions(-) -- 2.25.1

1 18

[PATCH kernel-4.19 1/3] mmap: fix a compiling error for 'MAP_CHECKNODE'
by Yang Yingliang 09 Feb '21

09 Feb '21

From: Bixuan Cui <cuibixuan(a)huawei.com> hulk inclusion category: bugfix bugzilla: NA CVE: NA MAP_CHECKNODE was defined in uapi/asm-generic/mman.h, that was not automatically included by mm/mmap.c when building on platforms such as mips, and result in following compiling error: mm/mmap.c: In function ‘__do_mmap’: mm/mmap.c:1581:14: error: ‘MAP_CHECKNODE’ undeclared (first use in this function) if (flags & MAP_CHECKNODE) ^ mm/mmap.c:1581:14: note: each undeclared identifier is reported only once for each function it appears in scripts/Makefile.build:303: recipe for target 'mm/mmap.o' failed Fixes: 56a22a261008 ("arm64/ascend: mm: Add MAP_CHECKNODE flag to check node hugetlb") 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> --- arch/alpha/include/uapi/asm/mman.h | 1 + arch/mips/include/uapi/asm/mman.h | 1 + arch/parisc/include/uapi/asm/mman.h | 1 + arch/powerpc/include/uapi/asm/mman.h | 1 + arch/sparc/include/uapi/asm/mman.h | 1 + arch/xtensa/include/uapi/asm/mman.h | 1 + 6 files changed, 6 insertions(+) diff --git a/arch/alpha/include/uapi/asm/mman.h b/arch/alpha/include/uapi/asm/mman.h index b3acfc00c8ec5..1c7ce2716ad37 100644 --- a/arch/alpha/include/uapi/asm/mman.h +++ b/arch/alpha/include/uapi/asm/mman.h @@ -34,6 +34,7 @@ #define MAP_HUGETLB 0x100000 /* create a huge page mapping */ #define MAP_FIXED_NOREPLACE 0x200000/* MAP_FIXED which doesn't unmap underlying mapping */ #define MAP_PA32BIT 0x400000 /* physical address is within 4G */ +#define MAP_CHECKNODE 0x800000 /* hugetlb numa node check */ #define MS_ASYNC 1 /* sync memory asynchronously */ #define MS_SYNC 2 /* synchronous memory sync */ diff --git a/arch/mips/include/uapi/asm/mman.h b/arch/mips/include/uapi/asm/mman.h index 72a00c746e781..4570a54ac1d90 100644 --- a/arch/mips/include/uapi/asm/mman.h +++ b/arch/mips/include/uapi/asm/mman.h @@ -52,6 +52,7 @@ #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 /* physical address is within 4G */ +#define MAP_CHECKNODE 0x800000 /* hugetlb numa node check */ /* * Flags for msync diff --git a/arch/parisc/include/uapi/asm/mman.h b/arch/parisc/include/uapi/asm/mman.h index 9e989d649e854..06857eb1bee8f 100644 --- a/arch/parisc/include/uapi/asm/mman.h +++ b/arch/parisc/include/uapi/asm/mman.h @@ -28,6 +28,7 @@ #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 /* physical address is within 4G */ +#define MAP_CHECKNODE 0x800000 /* hugetlb numa node check */ #define MS_SYNC 1 /* synchronous memory sync */ #define MS_ASYNC 2 /* sync memory asynchronously */ diff --git a/arch/powerpc/include/uapi/asm/mman.h b/arch/powerpc/include/uapi/asm/mman.h index 95f884ada96f1..24354f792b00d 100644 --- a/arch/powerpc/include/uapi/asm/mman.h +++ b/arch/powerpc/include/uapi/asm/mman.h @@ -30,6 +30,7 @@ #define MAP_STACK 0x20000 /* give out an address that is best suited for process/thread stacks */ #define MAP_HUGETLB 0x40000 /* create a huge page mapping */ #define MAP_PA32BIT 0x400000 /* physical address is within 4G */ +#define MAP_CHECKNODE 0x800000 /* hugetlb numa node check */ /* Override any generic PKEY permission defines */ #define PKEY_DISABLE_EXECUTE 0x4 diff --git a/arch/sparc/include/uapi/asm/mman.h b/arch/sparc/include/uapi/asm/mman.h index 0d1881b8f30d1..214abe17f44a0 100644 --- a/arch/sparc/include/uapi/asm/mman.h +++ b/arch/sparc/include/uapi/asm/mman.h @@ -27,6 +27,7 @@ #define MAP_STACK 0x20000 /* give out an address that is best suited for process/thread stacks */ #define MAP_HUGETLB 0x40000 /* create a huge page mapping */ #define MAP_PA32BIT 0x400000 /* physical address is within 4G */ +#define MAP_CHECKNODE 0x800000 /* hugetlb numa node check */ #endif /* _UAPI__SPARC_MMAN_H__ */ diff --git a/arch/xtensa/include/uapi/asm/mman.h b/arch/xtensa/include/uapi/asm/mman.h index f584a590bb001..2c9d705602382 100644 --- a/arch/xtensa/include/uapi/asm/mman.h +++ b/arch/xtensa/include/uapi/asm/mman.h @@ -59,6 +59,7 @@ #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 /* physical address is within 4G */ +#define MAP_CHECKNODE 0x800000 /* hugetlb numa node check */ #ifdef CONFIG_MMAP_ALLOW_UNINITIALIZED # define MAP_UNINITIALIZED 0x4000000 /* For anonymous mmap, memory could be * uninitialized */ -- 2.25.1

1 2

[PATCH kernel-4.19 00/38] backport stable-4.19.173
by Yang Yingliang 09 Feb '21

09 Feb '21

Andrea Righi (1): leds: trigger: fix potential deadlock with libata Baoquan He (1): kernel: kexec: remove the lock operation of system_transition_mutex Bartosz Golaszewski (1): iommu/vt-d: Don't dereference iommu_device if IOMMU_API is not built Claudiu Beznea (1): drivers: soc: atmel: add null entry at the end of at91_soc_allowed_list[] Dan Carpenter (1): can: dev: prevent potential information leak in can_fill_info() David Woodhouse (2): xen: Fix XenStore initialisation for XS_LOCAL iommu/vt-d: Gracefully handle DMAR units with no supported address widths Eyal Birger (1): xfrm: fix disable_xfrm sysctl when used on xfrm interfaces Giacinto Cifelli (1): net: usb: qmi_wwan: added support for Thales Cinterion PLSx3 modem family Greg Kroah-Hartman (1): Linux 4.19.173 Ivan Vecera (1): team: protect features update by RCU to avoid deadlock Jay Zhou (1): KVM: x86: get smi pending status correctly Johannes Berg (4): wext: fix NULL-ptr-dereference with cfg80211's lack of commit() iwlwifi: pcie: use jiffies for memory read spin time limit iwlwifi: pcie: reschedule in long-running memory reads mac80211: pause TX while changing interface type Josef Bacik (1): nbd: freeze the queue while we're adding connections Kai-Heng Feng (1): ACPI: sysfs: Prefer "compatible" modalias Kamal Heib (1): RDMA/cxgb4: Fix the reported max_recv_sge value Koen Vandeputte (1): ARM: dts: imx6qdl-gw52xx: fix duplicate regulator naming Laurent Badel (1): PM: hibernate: flush swap writer after marking Like Xu (1): KVM: x86/pmu: Fix HW_REF_CPU_CYCLES event pseudo-encoding in intel_arch_events[] Lorenzo Bianconi (2): mt7601u: fix kernel crash unplugging the device mt7601u: fix rx buffer refcounting Max Krummenacher (1): ARM: imx: build suspend-imx6.S with arm instruction set Pablo Neira Ayuso (1): netfilter: nft_dynset: add timeout extension to template Pan Bian (2): NFC: fix resource leak when target index is invalid NFC: fix possible resource leak Pengcheng Yang (1): tcp: fix TLP timer not set when CA_STATE changes from DISORDER to OPEN Roger Pau Monne (2): xen/privcmd: allow fetching resource sizes xen-blkfront: allow discard-* nodes to be optional Roi Dayan (1): net/mlx5: Fix memory leak on flow table creation error flow Sean Young (1): media: rc: ensure that uevent can be read directly after rc device register Shmulik Ladkani (1): xfrm: Fix oops in xfrm_replay_advance_bmp Sudeep Holla (1): drivers: soc: atmel: Avoid calling at91_soc_init on non AT91 SoCs Takashi Iwai (1): ALSA: hda/via: Apply the workaround generically for Clevo machines Takeshi Misawa (1): rxrpc: Fix memory leak in rxrpc_lookup_local Trond Myklebust (1): pNFS/NFSv4: Fix a layout segment leak in pnfs_layout_process() Makefile | 2 +- arch/arm/boot/dts/imx6qdl-gw52xx.dtsi | 2 +- arch/arm/mach-imx/suspend-imx6.S | 1 + arch/x86/kvm/pmu_intel.c | 2 +- arch/x86/kvm/x86.c | 5 +++ drivers/acpi/device_sysfs.c | 20 +++------ drivers/block/nbd.c | 8 ++++ drivers/block/xen-blkfront.c | 20 +++------ drivers/infiniband/hw/cxgb4/qp.c | 2 +- drivers/iommu/dmar.c | 45 +++++++++++++------ drivers/leds/led-triggers.c | 10 +++-- drivers/media/rc/rc-main.c | 4 +- drivers/net/can/dev.c | 2 +- .../net/ethernet/mellanox/mlx5/core/fs_core.c | 1 + drivers/net/team/team.c | 6 +-- drivers/net/usb/qmi_wwan.c | 1 + .../net/wireless/intel/iwlwifi/pcie/trans.c | 14 +++--- drivers/net/wireless/mediatek/mt7601u/dma.c | 5 +-- drivers/soc/atmel/soc.c | 13 ++++++ drivers/xen/privcmd.c | 25 ++++++++--- drivers/xen/xenbus/xenbus_probe.c | 31 +++++++++++++ fs/nfs/pnfs.c | 1 + include/linux/intel-iommu.h | 2 + include/net/tcp.h | 2 +- kernel/kexec_core.c | 2 - kernel/power/swap.c | 2 +- net/ipv4/tcp_input.c | 10 +++-- net/ipv4/tcp_recovery.c | 5 ++- net/mac80211/ieee80211_i.h | 1 + net/mac80211/iface.c | 6 +++ net/netfilter/nft_dynset.c | 4 +- net/nfc/netlink.c | 1 + net/nfc/rawsock.c | 2 +- net/rxrpc/call_accept.c | 1 + net/wireless/wext-core.c | 5 ++- net/xfrm/xfrm_input.c | 2 +- net/xfrm/xfrm_policy.c | 4 +- sound/pci/hda/patch_via.c | 2 +- 38 files changed, 183 insertions(+), 88 deletions(-) -- 2.25.1

1 38

[PATCH openEuler-1.0-LTS 1/2] nbd: Fix memory leak in nbd_add_socket
by Yang Yingliang 07 Feb '21

07 Feb '21

From: Zheng Bin <zhengbin13(a)huawei.com> stable inclusion from linux-4.19.133 commit 306290bad932950adccfbf99c27b70c105c2c183 -------------------------------- [ Upstream commit 579dd91ab3a5446b148e7f179b6596b270dace46 ] When adding first socket to nbd, if nsock's allocation failed, the data structure member "config->socks" was reallocated, but the data structure member "config->num_connections" was not updated. A memory leak will occur then because the function "nbd_config_put" will free "config->socks" only when "config->num_connections" is not zero. Fixes: 03bf73c315ed ("nbd: prevent memory leak") Reported-by: syzbot+934037347002901b8d2a(a)syzkaller.appspotmail.com Signed-off-by: Zheng Bin <zhengbin13(a)huawei.com> Reviewed-by: Eric Biggers <ebiggers(a)google.com> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/block/nbd.c | 25 +++++++++++++++---------- 1 file changed, 15 insertions(+), 10 deletions(-) diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 49f98b9b2777..97fcb040056c 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -993,25 +993,26 @@ static int nbd_add_socket(struct nbd_device *nbd, unsigned long arg, test_bit(NBD_RT_BOUND, &config->runtime_flags))) { dev_err(disk_to_dev(nbd->disk), "Device being setup by another task"); - sockfd_put(sock); - return -EBUSY; + err = -EBUSY; + goto put_socket; + } + + nsock = kzalloc(sizeof(*nsock), GFP_KERNEL); + if (!nsock) { + err = -ENOMEM; + goto put_socket; } socks = krealloc(config->socks, (config->num_connections + 1) * sizeof(struct nbd_sock *), GFP_KERNEL); if (!socks) { - sockfd_put(sock); - return -ENOMEM; + kfree(nsock); + err = -ENOMEM; + goto put_socket; } config->socks = socks; - nsock = kzalloc(sizeof(struct nbd_sock), GFP_KERNEL); - if (!nsock) { - sockfd_put(sock); - return -ENOMEM; - } - nsock->fallback_index = -1; nsock->dead = false; mutex_init(&nsock->tx_lock); @@ -1023,6 +1024,10 @@ static int nbd_add_socket(struct nbd_device *nbd, unsigned long arg, atomic_inc(&config->live_connections); return 0; + +put_socket: + sockfd_put(sock); + return err; } static int nbd_reconnect_socket(struct nbd_device *nbd, unsigned long arg) -- 2.25.1

1 1

[PATCH openEuler-1.0-LTS 1/7] futex: Ensure the correct return value from futex_lock_pi()
by Yang Yingliang 04 Feb '21

04 Feb '21

From: Thomas Gleixner <tglx(a)linutronix.de> stable inclusion from linux-4.19.172 commit 72f38fffa4758b878f819f8a47761b3f03443f36 CVE: CVE-2021-3347 -------------------------------- commit 12bb3f7f1b03d5913b3f9d4236a488aa7774dfe9 upstream In case that futex_lock_pi() was aborted by a signal or a timeout and the task returned without acquiring the rtmutex, but is the designated owner of the futex due to a concurrent futex_unlock_pi() fixup_owner() is invoked to establish consistent state. In that case it invokes fixup_pi_state_owner() which in turn tries to acquire the rtmutex again. If that succeeds then it does not propagate this success to fixup_owner() and futex_lock_pi() returns -EINTR or -ETIMEOUT despite having the futex locked. Return success from fixup_pi_state_owner() in all cases where the current task owns the rtmutex and therefore the futex and propagate it correctly through fixup_owner(). Fixup the other callsite which does not expect a positive return value. Fixes: c1e2f0eaf015 ("futex: Avoid violating the 10th rule of futex") Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Acked-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> 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: Xiu Jianfeng <xiujianfeng(a)huawei.com> Reviewed-by: Wei Li <liwei391(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- kernel/futex.c | 32 ++++++++++++++++---------------- 1 file changed, 16 insertions(+), 16 deletions(-) diff --git a/kernel/futex.c b/kernel/futex.c index 28b321e23053..ca9389bd110f 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -2414,8 +2414,8 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, } if (__rt_mutex_futex_trylock(&pi_state->pi_mutex)) { - /* We got the lock after all, nothing to fix. */ - ret = 0; + /* We got the lock. pi_state is correct. Tell caller. */ + ret = 1; goto out_unlock; } @@ -2431,7 +2431,7 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, * We raced against a concurrent self; things are * already fixed up. Nothing to do. */ - ret = 0; + ret = 1; goto out_unlock; } newowner = argowner; @@ -2477,7 +2477,7 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, raw_spin_unlock(&newowner->pi_lock); raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock); - return 0; + return argowner == current; /* * In order to reschedule or handle a page fault, we need to drop the @@ -2519,7 +2519,7 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, * Check if someone else fixed it for us: */ if (pi_state->owner != oldowner) { - ret = 0; + ret = argowner == current; goto out_unlock; } @@ -2552,8 +2552,6 @@ static long futex_wait_restart(struct restart_block *restart); */ static int fixup_owner(u32 __user *uaddr, struct futex_q *q, int locked) { - int ret = 0; - if (locked) { /* * Got the lock. We might not be the anticipated owner if we @@ -2564,8 +2562,8 @@ static int fixup_owner(u32 __user *uaddr, struct futex_q *q, int locked) * stable state, anything else needs more attention. */ if (q->pi_state->owner != current) - ret = fixup_pi_state_owner(uaddr, q, current); - goto out; + return fixup_pi_state_owner(uaddr, q, current); + return 1; } /* @@ -2576,10 +2574,8 @@ static int fixup_owner(u32 __user *uaddr, struct futex_q *q, int locked) * Another speculative read; pi_state->owner == current is unstable * but needs our attention. */ - if (q->pi_state->owner == current) { - ret = fixup_pi_state_owner(uaddr, q, NULL); - goto out; - } + if (q->pi_state->owner == current) + return fixup_pi_state_owner(uaddr, q, NULL); /* * Paranoia check. If we did not take the lock, then we should not be @@ -2592,8 +2588,7 @@ static int fixup_owner(u32 __user *uaddr, struct futex_q *q, int locked) q->pi_state->owner); } -out: - return ret ? ret : locked; + return 0; } /** @@ -3315,7 +3310,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, if (q.pi_state && (q.pi_state->owner != current)) { spin_lock(q.lock_ptr); ret = fixup_pi_state_owner(uaddr2, &q, current); - if (ret && rt_mutex_owner(&q.pi_state->pi_mutex) == current) { + if (ret < 0 && rt_mutex_owner(&q.pi_state->pi_mutex) == current) { pi_state = q.pi_state; get_pi_state(pi_state); } @@ -3325,6 +3320,11 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, */ put_pi_state(q.pi_state); spin_unlock(q.lock_ptr); + /* + * Adjust the return value. It's either -EFAULT or + * success (1) but the caller expects 0 for success. + */ + ret = ret < 0 ? ret : 0; } } else { struct rt_mutex *pi_mutex; -- 2.25.1

1 6

[PATCH openEuler-1.0-LTS 1/3] netfilter: clear skb->next in NF_HOOK_LIST()
by Yang Yingliang 04 Feb '21

04 Feb '21

From: Cong Wang <cong.wang(a)bytedance.com> stable inclusion from linux-4.19.161 commit 5460d62d661c0fc53bfe83493821b1dc3dc969f4 -------------------------------- NF_HOOK_LIST() uses list_del() to remove skb from the linked list, however, it is not sufficient as skb->next still points to other skb. We should just call skb_list_del_init() to clear skb->next, like the rest places which using skb list. This has been fixed in upstream by commit ca58fbe06c54 ("netfilter: add and use nf_hook_slow_list()"). Fixes: 9f17dbf04ddf ("netfilter: fix use-after-free in NF_HOOK_LIST") Reported-by: liuzx(a)knownsec.com Tested-by: liuzx(a)knownsec.com Cc: Florian Westphal <fw(a)strlen.de> Cc: Edward Cree <ecree(a)solarflare.com> Cc: stable(a)vger.kernel.org # between 4.19 and 5.4 Signed-off-by: Cong Wang <cong.wang(a)bytedance.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/linux/netfilter.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 72cb19c3db6a..9460a5635c90 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -300,7 +300,7 @@ NF_HOOK_LIST(uint8_t pf, unsigned int hook, struct net *net, struct sock *sk, INIT_LIST_HEAD(&sublist); list_for_each_entry_safe(skb, next, head, list) { - list_del(&skb->list); + skb_list_del_init(skb); if (nf_hook(pf, hook, net, sk, skb, in, out, okfn) == 1) list_add_tail(&skb->list, &sublist); } -- 2.25.1

1 2

[PATCH kernel-4.19 1/4] arm64: module: rework special section handling
by Yang Yingliang 04 Feb '21

04 Feb '21

From: Mark Rutland <mark.rutland(a)arm.com> mainline inclusion from mainline-v5.5-rc1 commit bd8b21d3dd661658addc1cd4cc869bab11d28596 category: bugfix bugzilla: 25285 CVE: NA -------------------------------- When we load a module, we have to perform some special work for a couple of named sections. To do this, we iterate over all of the module's sections, and perform work for each section we recognize. To make it easier to handle the unexpected absence of a section, and to make the section-specific logic easer to read, let's factor the section search into a helper. Similar is already done in the core module loader, and other architectures (and ideally we'd unify these in future). If we expect a module to have an ftrace trampoline section, but it doesn't have one, we'll now reject loading the module. When ARM64_MODULE_PLTS is selected, any correctly built module should have one (and this is assumed by arm64's ftrace PLT code) and the absence of such a section implies something has gone wrong at build time. Subsequent patches will make use of the new helper. Signed-off-by: Mark Rutland <mark.rutland(a)arm.com> Reviewed-by: Ard Biesheuvel <ard.biesheuvel(a)linaro.org> Reviewed-by: Torsten Duwe <duwe(a)suse.de> Tested-by: Amit Daniel Kachhap <amit.kachhap(a)arm.com> Tested-by: Torsten Duwe <duwe(a)suse.de> Cc: Catalin Marinas <catalin.marinas(a)arm.com> Cc: James Morse <james.morse(a)arm.com> Cc: Will Deacon <will(a)kernel.org> Signed-off-by: Pu Lehui <pulehui(a)huawei.com> Reviewed-by: Jian Cheng <cj.chengjian(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/arm64/kernel/module.c | 35 ++++++++++++++++++++++++++--------- 1 file changed, 26 insertions(+), 9 deletions(-) diff --git a/arch/arm64/kernel/module.c b/arch/arm64/kernel/module.c index 834ecf108b53..55b00bfb704f 100644 --- a/arch/arm64/kernel/module.c +++ b/arch/arm64/kernel/module.c @@ -484,22 +484,39 @@ int apply_relocate_add(Elf64_Shdr *sechdrs, return -ENOEXEC; } -int module_finalize(const Elf_Ehdr *hdr, - const Elf_Shdr *sechdrs, - struct module *me) +static const Elf_Shdr *find_section(const Elf_Ehdr *hdr, + const Elf_Shdr *sechdrs, + const char *name) { const Elf_Shdr *s, *se; const char *secstrs = (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset; for (s = sechdrs, se = sechdrs + hdr->e_shnum; s < se; s++) { - if (strcmp(".altinstructions", secstrs + s->sh_name) == 0) - apply_alternatives_module((void *)s->sh_addr, s->sh_size); + if (strcmp(name, secstrs + s->sh_name) == 0) + return s; + } + + return NULL; +} + +int module_finalize(const Elf_Ehdr *hdr, + const Elf_Shdr *sechdrs, + struct module *me) +{ + const Elf_Shdr *s; + + s = find_section(hdr, sechdrs, ".altinstructions"); + if (s) + apply_alternatives_module((void *)s->sh_addr, s->sh_size); + #ifdef CONFIG_ARM64_MODULE_PLTS - if (IS_ENABLED(CONFIG_DYNAMIC_FTRACE) && - !strcmp(".text.ftrace_trampoline", secstrs + s->sh_name)) - me->arch.ftrace_trampoline = (void *)s->sh_addr; -#endif + if (IS_ENABLED(CONFIG_DYNAMIC_FTRACE)) { + s = find_section(hdr, sechdrs, ".text.ftrace_trampoline"); + if (!s) + return -ENOEXEC; + me->arch.ftrace_trampoline = (void *)s->sh_addr; } +#endif return 0; } -- 2.25.1

1 3

[PATCH kernel-4.19 00/27] backport stable-4.19.172
by Yang Yingliang 04 Feb '21

04 Feb '21

Baruch Siach (1): gpio: mvebu: fix pwm .get_state period calculation Eric Biggers (1): fs: fix lazytime expiration handling in __writeback_single_inode() Gaurav Kohli (1): tracing: Fix race in trace_open and buffer resize call Greg Kroah-Hartman (1): Linux 4.19.172 Jan Kara (1): writeback: Drop I_DIRTY_TIME_EXPIRE Jason Gerecke (1): HID: wacom: Correct NULL dereference on AES pen proximity Jean-Philippe Brucker (1): tools: Factor HOSTCC, HOSTLD, HOSTAR definitions Mikulas Patocka (1): dm integrity: conditionally disable "recalculate" feature Thomas Gleixner (18): futex: Move futex exit handling into futex code futex: Replace PF_EXITPIDONE with a state exit/exec: Seperate mm_release() futex: Split futex_mm_release() for exit/exec futex: Set task::futex_state to DEAD right after handling futex exit futex: Mark the begin of futex exit explicitly futex: Sanitize exit state handling futex: Provide state handling for exec() as well futex: Add mutex around futex exit futex: Provide distinct return value when owner is exiting futex: Prevent exit livelock futex: Ensure the correct return value from futex_lock_pi() futex: Replace pointless printk in fixup_owner() futex: Provide and use pi_state_update_owner() rtmutex: Remove unused argument from rt_mutex_proxy_unlock() futex: Use pi_state_update_owner() in put_pi_state() futex: Simplify fixup_pi_state_owner() futex: Handle faults correctly for PI futexes Wang Hai (1): Revert "mm/slub: fix a memory leak in sysfs_slab_add()" Documentation/device-mapper/dm-integrity.txt | 7 + Makefile | 2 +- drivers/gpio/gpio-mvebu.c | 25 +- drivers/hid/wacom_sys.c | 7 +- drivers/hid/wacom_wac.h | 2 +- drivers/md/dm-integrity.c | 24 +- fs/exec.c | 2 +- fs/ext4/inode.c | 2 +- fs/fs-writeback.c | 36 +- fs/xfs/xfs_trans_inode.c | 4 +- include/linux/compat.h | 2 - include/linux/fs.h | 1 - include/linux/futex.h | 40 +- include/linux/sched.h | 3 +- include/linux/sched/mm.h | 6 +- include/trace/events/writeback.h | 1 - kernel/exit.c | 30 +- kernel/fork.c | 40 +- kernel/futex.c | 485 +++++++++++++------ kernel/locking/rtmutex.c | 3 +- kernel/locking/rtmutex_common.h | 3 +- kernel/trace/ring_buffer.c | 4 + mm/slub.c | 4 +- tools/build/Makefile | 4 - tools/objtool/Makefile | 9 - tools/perf/Makefile.perf | 4 - tools/power/acpi/Makefile.config | 1 - tools/scripts/Makefile.include | 10 + 28 files changed, 459 insertions(+), 302 deletions(-) -- 2.25.1

1 27

[PATCH kernel-4.19 1/2] netfilter: add and use nf_hook_slow_list()
by Yang Yingliang 04 Feb '21

04 Feb '21

From: Florian Westphal <fw(a)strlen.de> mainline inclusion from mainline-v5.5-rc1 commit ca58fbe06c54795f00db79e447f94c2028d30124 category: bugfix bugzilla: NA CVE: CVE-2021-20177 -------------------------------- At this time, NF_HOOK_LIST() macro will iterate the list and then calls nf_hook() for each individual skb. This makes it so the entire list is passed into the netfilter core. The advantage is that we only need to fetch the rule blob once per list instead of per-skb. NF_HOOK_LIST now only works for ipv4 and ipv6, as those are the only callers. v2: use skb_list_del_init() instead of list_del (Edward Cree) Signed-off-by: Florian Westphal <fw(a)strlen.de> Acked-by: Edward Cree <ecree(a)solarflare.com> Signed-off-by: Pablo Neira Ayuso <pablo(a)netfilter.org> Conflicts: include/linux/netfilter.h net/netfilter/core.c [yyl: adjust context] Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Yue Haibing <yuehaibing(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- include/linux/netfilter.h | 41 +++++++++++++++++++++++++++++---------- net/netfilter/core.c | 20 +++++++++++++++++++ 2 files changed, 51 insertions(+), 10 deletions(-) diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 9460a5635c90..8bf57abee9e4 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -183,6 +183,8 @@ extern struct static_key nf_hooks_needed[NFPROTO_NUMPROTO][NF_MAX_HOOKS]; int nf_hook_slow(struct sk_buff *skb, struct nf_hook_state *state, const struct nf_hook_entries *e, unsigned int i); +void nf_hook_slow_list(struct list_head *head, struct nf_hook_state *state, + const struct nf_hook_entries *e); /** * nf_hook - call a netfilter hook * @@ -295,17 +297,36 @@ NF_HOOK_LIST(uint8_t pf, unsigned int hook, struct net *net, struct sock *sk, struct list_head *head, struct net_device *in, struct net_device *out, int (*okfn)(struct net *, struct sock *, struct sk_buff *)) { - struct sk_buff *skb, *next; - struct list_head sublist; - - INIT_LIST_HEAD(&sublist); - list_for_each_entry_safe(skb, next, head, list) { - skb_list_del_init(skb); - if (nf_hook(pf, hook, net, sk, skb, in, out, okfn) == 1) - list_add_tail(&skb->list, &sublist); + struct nf_hook_entries *hook_head = NULL; + +#ifdef CONFIG_JUMP_LABEL + if (__builtin_constant_p(pf) && + __builtin_constant_p(hook) && + !static_key_false(&nf_hooks_needed[pf][hook])) + return; +#endif + + rcu_read_lock(); + switch (pf) { + case NFPROTO_IPV4: + hook_head = rcu_dereference(net->nf.hooks_ipv4[hook]); + break; + case NFPROTO_IPV6: + hook_head = rcu_dereference(net->nf.hooks_ipv6[hook]); + break; + default: + WARN_ON_ONCE(1); + break; } - /* Put passed packets back on main list */ - list_splice(&sublist, head); + + if (hook_head) { + struct nf_hook_state state; + + nf_hook_state_init(&state, hook, pf, in, out, sk, net, okfn); + + nf_hook_slow_list(head, &state, hook_head); + } + rcu_read_unlock(); } /* Call setsockopt() */ diff --git a/net/netfilter/core.c b/net/netfilter/core.c index 93aaec3a54ec..3f0bdc728f89 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -536,6 +536,26 @@ int nf_hook_slow(struct sk_buff *skb, struct nf_hook_state *state, EXPORT_SYMBOL(nf_hook_slow); +void nf_hook_slow_list(struct list_head *head, struct nf_hook_state *state, + const struct nf_hook_entries *e) +{ + struct sk_buff *skb, *next; + struct list_head sublist; + int ret; + + INIT_LIST_HEAD(&sublist); + + list_for_each_entry_safe(skb, next, head, list) { + skb_list_del_init(skb); + ret = nf_hook_slow(skb, state, e, 0); + if (ret == 1) + list_add_tail(&skb->list, &sublist); + } + /* Put passed packets back on main list */ + list_splice(&sublist, head); +} +EXPORT_SYMBOL(nf_hook_slow_list); + int skb_make_writable(struct sk_buff *skb, unsigned int writable_len) { if (writable_len > skb->len) -- 2.25.1

1 1

2024

2023

2022

2021

2020

2019

Kernel