From 267fb27352b6fc9fdbad753127a239f75618ecbc Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 30 Oct 2020 15:50:32 +0100 Subject: perf: Reduce stack usage of perf_output_begin() __perf_output_begin() has an on-stack struct perf_sample_data in the unlikely case it needs to generate a LOST record. However, every call to perf_output_begin() must already have a perf_sample_data on-stack. Reported-by: Thomas Gleixner Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20201030151954.985416146@infradead.org --- include/linux/perf_event.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 0c19d279b97f..b775ae0a8c87 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1400,11 +1400,14 @@ perf_event_addr_filters(struct perf_event *event) extern void perf_event_addr_filters_sync(struct perf_event *event); extern int perf_output_begin(struct perf_output_handle *handle, + struct perf_sample_data *data, struct perf_event *event, unsigned int size); extern int perf_output_begin_forward(struct perf_output_handle *handle, - struct perf_event *event, - unsigned int size); + struct perf_sample_data *data, + struct perf_event *event, + unsigned int size); extern int perf_output_begin_backward(struct perf_output_handle *handle, + struct perf_sample_data *data, struct perf_event *event, unsigned int size); -- cgit v1.2.3 From 76a4efa80900fc40e0fdf243b42aec9fb8c35d24 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 30 Oct 2020 12:14:21 +0100 Subject: perf/arch: Remove perf_sample_data::regs_user_copy struct perf_sample_data lives on-stack, we should be careful about it's size. Furthermore, the pt_regs copy in there is only because x86_64 is a trainwreck, solve it differently. Reported-by: Thomas Gleixner Signed-off-by: Peter Zijlstra (Intel) Tested-by: Steven Rostedt Link: https://lkml.kernel.org/r/20201030151955.258178461@infradead.org --- include/linux/perf_event.h | 6 ------ include/linux/perf_regs.h | 6 ++---- 2 files changed, 2 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index b775ae0a8c87..96450f6fb1de 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1022,13 +1022,7 @@ struct perf_sample_data { struct perf_callchain_entry *callchain; u64 aux_size; - /* - * regs_user may point to task_pt_regs or to regs_user_copy, depending - * on arch details. - */ struct perf_regs regs_user; - struct pt_regs regs_user_copy; - struct perf_regs regs_intr; u64 stack_user_size; diff --git a/include/linux/perf_regs.h b/include/linux/perf_regs.h index 2d12e97d5e7b..f632c5725f16 100644 --- a/include/linux/perf_regs.h +++ b/include/linux/perf_regs.h @@ -20,8 +20,7 @@ u64 perf_reg_value(struct pt_regs *regs, int idx); int perf_reg_validate(u64 mask); u64 perf_reg_abi(struct task_struct *task); void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy); + struct pt_regs *regs); #else #define PERF_REG_EXTENDED_MASK 0 @@ -42,8 +41,7 @@ static inline u64 perf_reg_abi(struct task_struct *task) } static inline void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { regs_user->regs = task_pt_regs(current); regs_user->abi = perf_reg_abi(current); -- cgit v1.2.3 From 8a3c84b649b033024d2349f96234b26cbd6083a6 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Tue, 10 Nov 2020 16:50:21 -0800 Subject: vfs: separate __sb_start_write into blocking and non-blocking helpers Break this function into two helpers so that it's obvious that the trylock versions return a value that must be checked, and the blocking versions don't require that. While we're at it, clean up the return type mismatch. Signed-off-by: Darrick J. Wong Reviewed-by: Jan Kara Reviewed-by: Christoph Hellwig --- include/linux/fs.h | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 0bd126418bb6..305989afd49c 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1581,7 +1581,8 @@ extern struct timespec64 current_time(struct inode *inode); */ void __sb_end_write(struct super_block *sb, int level); -int __sb_start_write(struct super_block *sb, int level, bool wait); +void __sb_start_write(struct super_block *sb, int level); +bool __sb_start_write_trylock(struct super_block *sb, int level); #define __sb_writers_acquired(sb, lev) \ percpu_rwsem_acquire(&(sb)->s_writers.rw_sem[(lev)-1], 1, _THIS_IP_) @@ -1645,12 +1646,12 @@ static inline void sb_end_intwrite(struct super_block *sb) */ static inline void sb_start_write(struct super_block *sb) { - __sb_start_write(sb, SB_FREEZE_WRITE, true); + __sb_start_write(sb, SB_FREEZE_WRITE); } -static inline int sb_start_write_trylock(struct super_block *sb) +static inline bool sb_start_write_trylock(struct super_block *sb) { - return __sb_start_write(sb, SB_FREEZE_WRITE, false); + return __sb_start_write_trylock(sb, SB_FREEZE_WRITE); } /** @@ -1674,7 +1675,7 @@ static inline int sb_start_write_trylock(struct super_block *sb) */ static inline void sb_start_pagefault(struct super_block *sb) { - __sb_start_write(sb, SB_FREEZE_PAGEFAULT, true); + __sb_start_write(sb, SB_FREEZE_PAGEFAULT); } /* @@ -1692,12 +1693,12 @@ static inline void sb_start_pagefault(struct super_block *sb) */ static inline void sb_start_intwrite(struct super_block *sb) { - __sb_start_write(sb, SB_FREEZE_FS, true); + __sb_start_write(sb, SB_FREEZE_FS); } -static inline int sb_start_intwrite_trylock(struct super_block *sb) +static inline bool sb_start_intwrite_trylock(struct super_block *sb) { - return __sb_start_write(sb, SB_FREEZE_FS, false); + return __sb_start_write_trylock(sb, SB_FREEZE_FS); } @@ -2756,14 +2757,14 @@ static inline void file_start_write(struct file *file) { if (!S_ISREG(file_inode(file)->i_mode)) return; - __sb_start_write(file_inode(file)->i_sb, SB_FREEZE_WRITE, true); + sb_start_write(file_inode(file)->i_sb); } static inline bool file_start_write_trylock(struct file *file) { if (!S_ISREG(file_inode(file)->i_mode)) return true; - return __sb_start_write(file_inode(file)->i_sb, SB_FREEZE_WRITE, false); + return sb_start_write_trylock(file_inode(file)->i_sb); } static inline void file_end_write(struct file *file) -- cgit v1.2.3 From 9b8523423b23ee3dfd88e32f5b7207be56a4e782 Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Tue, 10 Nov 2020 16:50:21 -0800 Subject: vfs: move __sb_{start,end}_write* to fs.h Now that we've straightened out the callers, move these three functions to fs.h since they're fairly trivial. Signed-off-by: Darrick J. Wong Reviewed-by: Christoph Hellwig Reviewed-by: Jan Kara --- include/linux/fs.h | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 305989afd49c..6dabd019cab0 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1580,9 +1580,24 @@ extern struct timespec64 current_time(struct inode *inode); * Snapshotting support. */ -void __sb_end_write(struct super_block *sb, int level); -void __sb_start_write(struct super_block *sb, int level); -bool __sb_start_write_trylock(struct super_block *sb, int level); +/* + * These are internal functions, please use sb_start_{write,pagefault,intwrite} + * instead. + */ +static inline void __sb_end_write(struct super_block *sb, int level) +{ + percpu_up_read(sb->s_writers.rw_sem + level-1); +} + +static inline void __sb_start_write(struct super_block *sb, int level) +{ + percpu_down_read(sb->s_writers.rw_sem + level - 1); +} + +static inline bool __sb_start_write_trylock(struct super_block *sb, int level) +{ + return percpu_down_read_trylock(sb->s_writers.rw_sem + level - 1); +} #define __sb_writers_acquired(sb, lev) \ percpu_rwsem_acquire(&(sb)->s_writers.rw_sem[(lev)-1], 1, _THIS_IP_) -- cgit v1.2.3 From 7e890c37c25c7cbca37ff0ab292873d8146e713b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 12 Nov 2020 17:50:04 +0100 Subject: block: add a return value to set_capacity_revalidate_and_notify Return if the function ended up sending an uevent or not. Cc: stable@vger.kernel.org # v5.9 Signed-off-by: Christoph Hellwig Reviewed-by: Petr Vorel Signed-off-by: Jens Axboe --- include/linux/genhd.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 38f23d757013..03da3f603d30 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -315,7 +315,7 @@ static inline int get_disk_ro(struct gendisk *disk) extern void disk_block_events(struct gendisk *disk); extern void disk_unblock_events(struct gendisk *disk); extern void disk_flush_events(struct gendisk *disk, unsigned int mask); -void set_capacity_revalidate_and_notify(struct gendisk *disk, sector_t size, +bool set_capacity_revalidate_and_notify(struct gendisk *disk, sector_t size, bool update_bdev); /* drivers/char/random.c */ -- cgit v1.2.3 From 3347acc6fcd4ee71ad18a9ff9d9dac176b517329 Mon Sep 17 00:00:00 2001 From: Arvind Sankar Date: Fri, 13 Nov 2020 22:51:59 -0800 Subject: compiler.h: fix barrier_data() on clang Commit 815f0ddb346c ("include/linux/compiler*.h: make compiler-*.h mutually exclusive") neglected to copy barrier_data() from compiler-gcc.h into compiler-clang.h. The definition in compiler-gcc.h was really to work around clang's more aggressive optimization, so this broke barrier_data() on clang, and consequently memzero_explicit() as well. For example, this results in at least the memzero_explicit() call in lib/crypto/sha256.c:sha256_transform() being optimized away by clang. Fix this by moving the definition of barrier_data() into compiler.h. Also move the gcc/clang definition of barrier() into compiler.h, __memory_barrier() is icc-specific (and barrier() is already defined using it in compiler-intel.h) and doesn't belong in compiler.h. [rdunlap@infradead.org: fix ALPHA builds when SMP is not enabled] Link: https://lkml.kernel.org/r/20201101231835.4589-1-rdunlap@infradead.org Fixes: 815f0ddb346c ("include/linux/compiler*.h: make compiler-*.h mutually exclusive") Signed-off-by: Arvind Sankar Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Tested-by: Nick Desaulniers Reviewed-by: Nick Desaulniers Reviewed-by: Kees Cook Cc: Link: https://lkml.kernel.org/r/20201014212631.207844-1-nivedita@alum.mit.edu Signed-off-by: Linus Torvalds --- include/linux/compiler-clang.h | 6 ------ include/linux/compiler-gcc.h | 19 ------------------- include/linux/compiler.h | 18 ++++++++++++++++-- 3 files changed, 16 insertions(+), 27 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compiler-clang.h b/include/linux/compiler-clang.h index 230604e7f057..dd7233c48bf3 100644 --- a/include/linux/compiler-clang.h +++ b/include/linux/compiler-clang.h @@ -60,12 +60,6 @@ #define COMPILER_HAS_GENERIC_BUILTIN_OVERFLOW 1 #endif -/* The following are for compatibility with GCC, from compiler-gcc.h, - * and may be redefined here because they should not be shared with other - * compilers, like ICC. - */ -#define barrier() __asm__ __volatile__("" : : : "memory") - #if __has_feature(shadow_call_stack) # define __noscs __attribute__((__no_sanitize__("shadow-call-stack"))) #endif diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h index 5deb37024574..74c6c0486eed 100644 --- a/include/linux/compiler-gcc.h +++ b/include/linux/compiler-gcc.h @@ -15,25 +15,6 @@ # error Sorry, your version of GCC is too old - please use 4.9 or newer. #endif -/* Optimization barrier */ - -/* The "volatile" is due to gcc bugs */ -#define barrier() __asm__ __volatile__("": : :"memory") -/* - * This version is i.e. to prevent dead stores elimination on @ptr - * where gcc and llvm may behave differently when otherwise using - * normal barrier(): while gcc behavior gets along with a normal - * barrier(), llvm needs an explicit input variable to be assumed - * clobbered. The issue is as follows: while the inline asm might - * access any memory it wants, the compiler could have fit all of - * @ptr into memory registers instead, and since @ptr never escaped - * from that, it proved that the inline asm wasn't touching any of - * it. This version works well with both compilers, i.e. we're telling - * the compiler that the inline asm absolutely may see the contents - * of @ptr. See also: https://llvm.org/bugs/show_bug.cgi?id=15495 - */ -#define barrier_data(ptr) __asm__ __volatile__("": :"r"(ptr) :"memory") - /* * This macro obfuscates arithmetic on a variable address so that gcc * shouldn't recognize the original var, and make assumptions about it. diff --git a/include/linux/compiler.h b/include/linux/compiler.h index e512f5505dad..b8fe0c23cfff 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -80,11 +80,25 @@ void ftrace_likely_update(struct ftrace_likely_data *f, int val, /* Optimization barrier */ #ifndef barrier -# define barrier() __memory_barrier() +/* The "volatile" is due to gcc bugs */ +# define barrier() __asm__ __volatile__("": : :"memory") #endif #ifndef barrier_data -# define barrier_data(ptr) barrier() +/* + * This version is i.e. to prevent dead stores elimination on @ptr + * where gcc and llvm may behave differently when otherwise using + * normal barrier(): while gcc behavior gets along with a normal + * barrier(), llvm needs an explicit input variable to be assumed + * clobbered. The issue is as follows: while the inline asm might + * access any memory it wants, the compiler could have fit all of + * @ptr into memory registers instead, and since @ptr never escaped + * from that, it proved that the inline asm wasn't touching any of + * it. This version works well with both compilers, i.e. we're telling + * the compiler that the inline asm absolutely may see the contents + * of @ptr. See also: https://llvm.org/bugs/show_bug.cgi?id=15495 + */ +# define barrier_data(ptr) __asm__ __volatile__("": :"r"(ptr) :"memory") #endif /* workaround for GCC PR82365 if needed */ -- cgit v1.2.3 From 8b21ca0218d29cc6bb7028125c7e5a10dfb4730c Mon Sep 17 00:00:00 2001 From: Muchun Song Date: Fri, 13 Nov 2020 22:52:13 -0800 Subject: mm: memcontrol: fix missing wakeup polling thread When we poll the swap.events, we can miss being woken up when the swap event occurs. Because we didn't notify. Fixes: f3a53a3a1e5b ("mm, memcontrol: implement memory.swap.events") Signed-off-by: Muchun Song Signed-off-by: Andrew Morton Reviewed-by: Shakeel Butt Acked-by: Johannes Weiner Cc: Roman Gushchin Cc: Michal Hocko Cc: Yafang Shao Cc: Chris Down Cc: Tejun Heo Link: https://lkml.kernel.org/r/20201105161936.98312-1-songmuchun@bytedance.com Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index e391e3c56de5..a80c59af2c60 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -900,12 +900,19 @@ static inline void count_memcg_event_mm(struct mm_struct *mm, static inline void memcg_memory_event(struct mem_cgroup *memcg, enum memcg_memory_event event) { + bool swap_event = event == MEMCG_SWAP_HIGH || event == MEMCG_SWAP_MAX || + event == MEMCG_SWAP_FAIL; + atomic_long_inc(&memcg->memory_events_local[event]); - cgroup_file_notify(&memcg->events_local_file); + if (!swap_event) + cgroup_file_notify(&memcg->events_local_file); do { atomic_long_inc(&memcg->memory_events[event]); - cgroup_file_notify(&memcg->events_file); + if (swap_event) + cgroup_file_notify(&memcg->swap_events_file); + else + cgroup_file_notify(&memcg->events_file); if (!cgroup_subsys_on_dfl(memory_cgrp_subsys)) break; -- cgit v1.2.3 From 57e3cebd022fbc035dcf190ac789fd2ffc747f5b Mon Sep 17 00:00:00 2001 From: Shenming Lu Date: Sat, 28 Nov 2020 22:18:57 +0800 Subject: KVM: arm64: Delay the polling of the GICR_VPENDBASER.Dirty bit In order to reduce the impact of the VPT parsing happening on the GIC, we can split the vcpu reseidency in two phases: - programming GICR_VPENDBASER: this still happens in vcpu_load() - checking for the VPT parsing to be complete: this can happen on vcpu entry (in kvm_vgic_flush_hwstate()) This allows the GIC and the CPU to work in parallel, rewmoving some of the entry overhead. Suggested-by: Marc Zyngier Signed-off-by: Shenming Lu Signed-off-by: Marc Zyngier Link: https://lore.kernel.org/r/20201128141857.983-3-lushenming@huawei.com --- include/linux/irqchip/arm-gic-v4.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/irqchip/arm-gic-v4.h b/include/linux/irqchip/arm-gic-v4.h index 6976b8331b60..943c3411ca10 100644 --- a/include/linux/irqchip/arm-gic-v4.h +++ b/include/linux/irqchip/arm-gic-v4.h @@ -39,6 +39,8 @@ struct its_vpe { irq_hw_number_t vpe_db_lpi; /* VPE resident */ bool resident; + /* VPT parse complete */ + bool ready; union { /* GICv4.0 implementations */ struct { @@ -104,6 +106,7 @@ enum its_vcpu_info_cmd_type { PROP_UPDATE_AND_INV_VLPI, SCHEDULE_VPE, DESCHEDULE_VPE, + COMMIT_VPE, INVALL_VPE, PROP_UPDATE_VSGI, }; @@ -129,6 +132,7 @@ int its_alloc_vcpu_irqs(struct its_vm *vm); void its_free_vcpu_irqs(struct its_vm *vm); int its_make_vpe_resident(struct its_vpe *vpe, bool g0en, bool g1en); int its_make_vpe_non_resident(struct its_vpe *vpe, bool db); +int its_commit_vpe(struct its_vpe *vpe); int its_invall_vpe(struct its_vpe *vpe); int its_map_vlpi(int irq, struct its_vlpi_map *map); int its_get_vlpi(int irq, struct its_vlpi_map *map); -- cgit v1.2.3 From 6df3e14436f6ee254b1a4952d90ee8988be59c89 Mon Sep 17 00:00:00 2001 From: David Brazdil Date: Wed, 2 Dec 2020 18:41:02 +0000 Subject: psci: Add accessor for psci_0_1_function_ids Make it possible to retrieve a copy of the psci_0_1_function_ids struct. This is useful for KVM if it is configured to intercept host's PSCI SMCs. Signed-off-by: David Brazdil Signed-off-by: Marc Zyngier Acked-by: Mark Rutland Link: https://lore.kernel.org/r/20201202184122.26046-7-dbrazdil@google.com --- include/linux/psci.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/psci.h b/include/linux/psci.h index 2a1bfb890e58..4ca0060a3fc4 100644 --- a/include/linux/psci.h +++ b/include/linux/psci.h @@ -34,6 +34,15 @@ struct psci_operations { extern struct psci_operations psci_ops; +struct psci_0_1_function_ids { + u32 cpu_suspend; + u32 cpu_on; + u32 cpu_off; + u32 migrate; +}; + +struct psci_0_1_function_ids get_psci_0_1_function_ids(void); + #if defined(CONFIG_ARM_PSCI_FW) int __init psci_dt_init(void); #else -- cgit v1.2.3