- Kernel - mailweb.openeuler.org

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

23 Dec '20

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

1 1

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

23 Dec '20

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

1 9

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

23 Dec '20

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

1 2

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

23 Dec '20

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

3 28

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

22 Dec '20

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

1 0

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

22 Dec '20

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

2 25

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

22 Dec '20

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

1 24

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

22 Dec '20

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

4 30

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

22 Dec '20

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

1 4

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

22 Dec '20

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

2 5

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

22 Dec '20

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

2 4

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

22 Dec '20

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

1 4

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

22 Dec '20

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

1 0

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

22 Dec '20

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

1 2

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

22 Dec '20

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

1 0

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

21 Dec '20

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

1 2

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

21 Dec '20

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

2 1

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

21 Dec '20

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

1 0

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

21 Dec '20

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

1 2

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

21 Dec '20

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

1 0

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

21 Dec '20

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

2 2

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

19 Dec '20

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

2 27

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

17 Dec '20

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

1 3

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

17 Dec '20

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

1 1

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

17 Dec '20

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

1 14

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

17 Dec '20

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

1 0

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

17 Dec '20

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

1 0

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

17 Dec '20

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

2 6

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

16 Dec '20

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

2 6

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

16 Dec '20

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

1 0

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

16 Dec '20

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

1 1

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

15 Dec '20

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

1 0

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

15 Dec '20

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

1 0

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

14 Dec '20

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

1 38

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

14 Dec '20

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

1 4

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

11 Dec '20

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

1 0

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

11 Dec '20

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

1 0

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

10 Dec '20

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

2 1

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

10 Dec '20

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

1 32

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

10 Dec '20

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

1 0

openEuler kernel sig meeting
by openEuler conference 09 Dec '20

09 Dec '20

3 3

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

08 Dec '20

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

1 16

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

08 Dec '20

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

1 6

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

08 Dec '20

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

2 2

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

07 Dec '20

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

1 0

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

03 Dec '20

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

1 49

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

02 Dec '20

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

1 87

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

02 Dec '20

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

1 0

openeuler patch tag咨询
by Zhengyuan Liu 02 Dec '20

02 Dec '20

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

3 3

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

02 Dec '20

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

1 0

2024

2023

2022

2021

2020

2019

Kernel