diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index d17518ca19b8..8b66a555d2f0 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -1825,11 +1825,18 @@ static inline temp_mm_state_t use_temporary_mm(struct mm_struct *mm) return temp_state; } +__ro_after_init struct mm_struct *poking_mm; +__ro_after_init unsigned long poking_addr; + static inline void unuse_temporary_mm(temp_mm_state_t prev_state) { lockdep_assert_irqs_disabled(); + switch_mm_irqs_off(NULL, prev_state.mm, current); + /* Clear the cpumask, to indicate no TLB flushing is needed anywhere */ + cpumask_clear_cpu(raw_smp_processor_id(), mm_cpumask(poking_mm)); + /* * Restore the breakpoints if they were disabled before the temporary mm * was loaded. @@ -1838,9 +1845,6 @@ static inline void unuse_temporary_mm(temp_mm_state_t prev_state) hw_breakpoint_restore(); } -__ro_after_init struct mm_struct *poking_mm; -__ro_after_init unsigned long poking_addr; - static void text_poke_memcpy(void *dst, const void *src, size_t len) { memcpy(dst, src, len); diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index b0d5a644fc84..cc4e57ae690f 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -606,18 +606,15 @@ void switch_mm_irqs_off(struct mm_struct *unused, struct mm_struct *next, cond_mitigation(tsk); /* - * Stop remote flushes for the previous mm. - * Skip kernel threads; we never send init_mm TLB flushing IPIs, - * but the bitmap manipulation can cause cache line contention. + * Leave this CPU in prev's mm_cpumask. Atomic writes to + * mm_cpumask can be expensive under contention. The CPU + * will be removed lazily at TLB flush time. */ - if (prev != &init_mm) { - VM_WARN_ON_ONCE(!cpumask_test_cpu(cpu, - mm_cpumask(prev))); - cpumask_clear_cpu(cpu, mm_cpumask(prev)); - } + VM_WARN_ON_ONCE(prev != &init_mm && !cpumask_test_cpu(cpu, + mm_cpumask(prev))); /* Start receiving IPIs and then read tlb_gen (and LAM below) */ - if (next != &init_mm) + if (next != &init_mm && !cpumask_test_cpu(cpu, mm_cpumask(next))) cpumask_set_cpu(cpu, mm_cpumask(next)); next_tlb_gen = atomic64_read(&next->context.tlb_gen); @@ -761,8 +758,10 @@ static void flush_tlb_func(void *info) count_vm_tlb_event(NR_TLB_REMOTE_FLUSH_RECEIVED); /* Can only happen on remote CPUs */ - if (f->mm && f->mm != loaded_mm) + if (f->mm && f->mm != loaded_mm) { + cpumask_clear_cpu(raw_smp_processor_id(), mm_cpumask(f->mm)); return; + } } if (unlikely(loaded_mm == &init_mm)) -- knotmirror 0.0.1 From 2815a56e4b7252a836969f5674ee356ea1ce482c Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Thu, 14 Nov 2024 10:26:17 -0500 Subject: [PATCH 2/7] x86/mm/tlb: Add tracepoint for TLB flush IPI to stale CPU Add a tracepoint when we send a TLB flush IPI to a CPU that used to be in the mm_cpumask, but isn't any more. Suggested-by: Dave Hansen Signed-off-by: Rik van Riel Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20241114152723.1294686-3-riel@surriel.com --- arch/x86/mm/tlb.c | 1 + include/linux/mm_types.h | 1 + 2 files changed, 2 insertions(+) diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index cc4e57ae690f..1aac4fa90d3d 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -760,6 +760,7 @@ static void flush_tlb_func(void *info) /* Can only happen on remote CPUs */ if (f->mm && f->mm != loaded_mm) { cpumask_clear_cpu(raw_smp_processor_id(), mm_cpumask(f->mm)); + trace_tlb_flush(TLB_REMOTE_WRONG_CPU, 0); return; } } diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 6e3bdf8e38bc..6b6f05404304 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -1335,6 +1335,7 @@ enum tlb_flush_reason { TLB_LOCAL_SHOOTDOWN, TLB_LOCAL_MM_SHOOTDOWN, TLB_REMOTE_SEND_IPI, + TLB_REMOTE_WRONG_CPU, NR_TLB_FLUSH_REASONS, }; -- knotmirror 0.0.1 From 953753db887f9d70f70f61d6ecbe5cf209107672 Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Thu, 5 Dec 2024 10:46:30 -0500 Subject: [PATCH 3/7] x86/mm/tlb: Also remove local CPU from mm_cpumask if stale The code in flush_tlb_func() that removes a remote CPU from the cpumask if it is no longer running the target mm is also needed on the originating CPU of a TLB flush, now that CPUs are no longer cleared from the mm_cpumask at context switch time. Flushing the TLB when we are not running the target mm is harmless, because the CPU's tlb_gen only gets updated to match the mm_tlb_gen, but it does hit this warning: WARN_ON_ONCE(local_tlb_gen > mm_tlb_gen); [ 210.343902][ T4668] WARNING: CPU: 38 PID: 4668 at arch/x86/mm/tlb.c:815 flush_tlb_func (arch/x86/mm/tlb.c:815) Removing both local and remote CPUs from the mm_cpumask when doing a flush for a not currently loaded mm avoids that warning. Reported-by: kernel test robot Tested-by: kernel test robot Signed-off-by: Rik van Riel Signed-off-by: Ingo Molnar Cc: Dave Hansen Cc: Mathieu Desnoyers Cc: Andy Lutomirski Cc: Peter Zijlstra Cc: Linus Torvalds Link: https://lore.kernel.org/r/20241205104630.755706ca@fangorn Closes: https://lore.kernel.org/oe-lkp/202412051551.690e9656-lkp@intel.com --- arch/x86/mm/tlb.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index 1aac4fa90d3d..3c30817ec6a2 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -756,13 +756,13 @@ static void flush_tlb_func(void *info) if (!local) { inc_irq_stat(irq_tlb_count); count_vm_tlb_event(NR_TLB_REMOTE_FLUSH_RECEIVED); + } - /* Can only happen on remote CPUs */ - if (f->mm && f->mm != loaded_mm) { - cpumask_clear_cpu(raw_smp_processor_id(), mm_cpumask(f->mm)); - trace_tlb_flush(TLB_REMOTE_WRONG_CPU, 0); - return; - } + /* The CPU was left in the mm_cpumask of the target mm. Clear it. */ + if (f->mm && f->mm != loaded_mm) { + cpumask_clear_cpu(raw_smp_processor_id(), mm_cpumask(f->mm)); + trace_tlb_flush(TLB_REMOTE_WRONG_CPU, 0); + return; } if (unlikely(loaded_mm == &init_mm)) -- knotmirror 0.0.1 From 6db2526c1d694c91c6e05e2f186c085e9460f202 Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Wed, 4 Dec 2024 21:03:16 -0500 Subject: [PATCH 4/7] x86/mm/tlb: Only trim the mm_cpumask once a second Setting and clearing CPU bits in the mm_cpumask is only ever done by the CPU itself, from the context switch code or the TLB flush code. Synchronization is handled by switch_mm_irqs_off() blocking interrupts. Sending TLB flush IPIs to CPUs that are in the mm_cpumask, but no longer running the program causes a regression in the will-it-scale tlbflush2 test. This test is contrived, but a large regression here might cause a small regression in some real world workload. Instead of always sending IPIs to CPUs that are in the mm_cpumask, but no longer running the program, send these IPIs only once a second. The rest of the time we can skip over CPUs where the loaded_mm is different from the target mm. Reported-by: kernel test roboto Signed-off-by: Rik van Riel Signed-off-by: Ingo Molnar Cc: Dave Hansen Cc: Andy Lutomirski Cc: Mathieu Desnoyers Cc: Peter Zijlstra Cc: Linus Torvalds Link: https://lore.kernel.org/r/20241204210316.612ee573@fangorn Closes: https://lore.kernel.org/oe-lkp/202411282207.6bd28eae-lkp@intel.com/ --- arch/x86/include/asm/mmu.h | 2 ++ arch/x86/include/asm/mmu_context.h | 1 + arch/x86/include/asm/tlbflush.h | 1 + arch/x86/mm/tlb.c | 35 +++++++++++++++++++++++++++--- 4 files changed, 36 insertions(+), 3 deletions(-) diff --git a/arch/x86/include/asm/mmu.h b/arch/x86/include/asm/mmu.h index ce4677b8b735..3b496cdcb74b 100644 --- a/arch/x86/include/asm/mmu.h +++ b/arch/x86/include/asm/mmu.h @@ -37,6 +37,8 @@ typedef struct { */ atomic64_t tlb_gen; + unsigned long next_trim_cpumask; + #ifdef CONFIG_MODIFY_LDT_SYSCALL struct rw_semaphore ldt_usr_sem; struct ldt_struct *ldt; diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h index 2886cb668d7f..795fdd53bd0a 100644 --- a/arch/x86/include/asm/mmu_context.h +++ b/arch/x86/include/asm/mmu_context.h @@ -151,6 +151,7 @@ static inline int init_new_context(struct task_struct *tsk, mm->context.ctx_id = atomic64_inc_return(&last_mm_ctx_id); atomic64_set(&mm->context.tlb_gen, 0); + mm->context.next_trim_cpumask = jiffies + HZ; #ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS if (cpu_feature_enabled(X86_FEATURE_OSPKE)) { diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h index 69e79fff41b8..02fc2aa06e9e 100644 --- a/arch/x86/include/asm/tlbflush.h +++ b/arch/x86/include/asm/tlbflush.h @@ -222,6 +222,7 @@ struct flush_tlb_info { unsigned int initiating_cpu; u8 stride_shift; u8 freed_tables; + u8 trim_cpumask; }; void flush_tlb_local(void); diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index 3c30817ec6a2..458a5d5be594 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -892,9 +892,36 @@ done: nr_invalidate); } -static bool tlb_is_not_lazy(int cpu, void *data) +static bool should_flush_tlb(int cpu, void *data) { - return !per_cpu(cpu_tlbstate_shared.is_lazy, cpu); + struct flush_tlb_info *info = data; + + /* Lazy TLB will get flushed at the next context switch. */ + if (per_cpu(cpu_tlbstate_shared.is_lazy, cpu)) + return false; + + /* No mm means kernel memory flush. */ + if (!info->mm) + return true; + + /* The target mm is loaded, and the CPU is not lazy. */ + if (per_cpu(cpu_tlbstate.loaded_mm, cpu) == info->mm) + return true; + + /* In cpumask, but not the loaded mm? Periodically remove by flushing. */ + if (info->trim_cpumask) + return true; + + return false; +} + +static bool should_trim_cpumask(struct mm_struct *mm) +{ + if (time_after(jiffies, READ_ONCE(mm->context.next_trim_cpumask))) { + WRITE_ONCE(mm->context.next_trim_cpumask, jiffies + HZ); + return true; + } + return false; } DEFINE_PER_CPU_SHARED_ALIGNED(struct tlb_state_shared, cpu_tlbstate_shared); @@ -928,7 +955,7 @@ STATIC_NOPV void native_flush_tlb_multi(const struct cpumask *cpumask, if (info->freed_tables) on_each_cpu_mask(cpumask, flush_tlb_func, (void *)info, true); else - on_each_cpu_cond_mask(tlb_is_not_lazy, flush_tlb_func, + on_each_cpu_cond_mask(should_flush_tlb, flush_tlb_func, (void *)info, 1, cpumask); } @@ -979,6 +1006,7 @@ static struct flush_tlb_info *get_flush_tlb_info(struct mm_struct *mm, info->freed_tables = freed_tables; info->new_tlb_gen = new_tlb_gen; info->initiating_cpu = smp_processor_id(); + info->trim_cpumask = 0; return info; } @@ -1021,6 +1049,7 @@ void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, * flush_tlb_func_local() directly in this case. */ if (cpumask_any_but(mm_cpumask(mm), cpu) < nr_cpu_ids) { + info->trim_cpumask = should_trim_cpumask(mm); flush_tlb_multi(mm_cpumask(mm), info); } else if (mm == this_cpu_read(cpu_tlbstate.loaded_mm)) { lockdep_assert_irqs_enabled(); -- knotmirror 0.0.1 From 9d93db0d1881c9e37e1528cd796e20ff13b7692c Mon Sep 17 00:00:00 2001 From: Gautam Somani Date: Sun, 1 Dec 2024 03:41:02 +0900 Subject: [PATCH 5/7] x86/mm/selftests: Fix typo in lam.c Change the spelling from metadate -> metadata Signed-off-by: Gautam Somani Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20241130184102.2182-1-gautamsomani@gmail.com --- tools/testing/selftests/x86/lam.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/x86/lam.c b/tools/testing/selftests/x86/lam.c index 0ea4f6813930..4d4a76532dc9 100644 --- a/tools/testing/selftests/x86/lam.c +++ b/tools/testing/selftests/x86/lam.c @@ -237,7 +237,7 @@ static uint64_t set_metadata(uint64_t src, unsigned long lam) * both pointers should point to the same address. * * @return: - * 0: value on the pointer with metadate and value on original are same + * 0: value on the pointer with metadata and value on original are same * 1: not same. */ static int handle_lam_test(void *src, unsigned int lam) -- knotmirror 0.0.1 From dd4059634dab548c904eeae2660ba3c8f7ce843c Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Mon, 2 Dec 2024 09:31:39 +0200 Subject: [PATCH 6/7] x86/mtrr: Rename mtrr_overwrite_state() to guest_force_mtrr_state() Rename the helper to better reflect its function. Suggested-by: Dave Hansen Signed-off-by: Kirill A. Shutemov Signed-off-by: Ingo Molnar Acked-by: Dave Hansen Reviewed-by: Michael Kelley Link: https://lore.kernel.org/r/20241202073139.448208-1-kirill.shutemov@linux.intel.com --- arch/x86/hyperv/ivm.c | 2 +- arch/x86/include/asm/mtrr.h | 10 +++++----- arch/x86/kernel/cpu/mtrr/generic.c | 6 +++--- arch/x86/kernel/cpu/mtrr/mtrr.c | 2 +- arch/x86/kernel/kvm.c | 2 +- arch/x86/xen/enlighten_pv.c | 4 ++-- 6 files changed, 13 insertions(+), 13 deletions(-) diff --git a/arch/x86/hyperv/ivm.c b/arch/x86/hyperv/ivm.c index 60fc3ed72830..90aabe1fd3b6 100644 --- a/arch/x86/hyperv/ivm.c +++ b/arch/x86/hyperv/ivm.c @@ -664,7 +664,7 @@ void __init hv_vtom_init(void) x86_platform.guest.enc_status_change_finish = hv_vtom_set_host_visibility; /* Set WB as the default cache mode. */ - mtrr_overwrite_state(NULL, 0, MTRR_TYPE_WRBACK); + guest_force_mtrr_state(NULL, 0, MTRR_TYPE_WRBACK); } #endif /* defined(CONFIG_AMD_MEM_ENCRYPT) || defined(CONFIG_INTEL_TDX_GUEST) */ diff --git a/arch/x86/include/asm/mtrr.h b/arch/x86/include/asm/mtrr.h index 4218248083d9..c69e269937c5 100644 --- a/arch/x86/include/asm/mtrr.h +++ b/arch/x86/include/asm/mtrr.h @@ -58,8 +58,8 @@ struct mtrr_state_type { */ # ifdef CONFIG_MTRR void mtrr_bp_init(void); -void mtrr_overwrite_state(struct mtrr_var_range *var, unsigned int num_var, - mtrr_type def_type); +void guest_force_mtrr_state(struct mtrr_var_range *var, unsigned int num_var, + mtrr_type def_type); extern u8 mtrr_type_lookup(u64 addr, u64 end, u8 *uniform); extern void mtrr_save_fixed_ranges(void *); extern void mtrr_save_state(void); @@ -75,9 +75,9 @@ void mtrr_disable(void); void mtrr_enable(void); void mtrr_generic_set_state(void); # else -static inline void mtrr_overwrite_state(struct mtrr_var_range *var, - unsigned int num_var, - mtrr_type def_type) +static inline void guest_force_mtrr_state(struct mtrr_var_range *var, + unsigned int num_var, + mtrr_type def_type) { } diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index 7b29ebda024f..2fdfda2b60e4 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -423,7 +423,7 @@ void __init mtrr_copy_map(void) } /** - * mtrr_overwrite_state - set static MTRR state + * guest_force_mtrr_state - set static MTRR state for a guest * * Used to set MTRR state via different means (e.g. with data obtained from * a hypervisor). @@ -436,8 +436,8 @@ void __init mtrr_copy_map(void) * @num_var: length of the @var array * @def_type: default caching type */ -void mtrr_overwrite_state(struct mtrr_var_range *var, unsigned int num_var, - mtrr_type def_type) +void guest_force_mtrr_state(struct mtrr_var_range *var, unsigned int num_var, + mtrr_type def_type) { unsigned int i; diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.c b/arch/x86/kernel/cpu/mtrr/mtrr.c index 989d368be04f..ecbda0341a8a 100644 --- a/arch/x86/kernel/cpu/mtrr/mtrr.c +++ b/arch/x86/kernel/cpu/mtrr/mtrr.c @@ -625,7 +625,7 @@ void mtrr_save_state(void) static int __init mtrr_init_finalize(void) { /* - * Map might exist if mtrr_overwrite_state() has been called or if + * Map might exist if guest_force_mtrr_state() has been called or if * mtrr_enabled() returns true. */ mtrr_copy_map(); diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c index 21e9e4845354..7a422a6c5983 100644 --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -983,7 +983,7 @@ static void __init kvm_init_platform(void) x86_platform.apic_post_init = kvm_apic_init; /* Set WB as the default cache mode for SEV-SNP and TDX */ - mtrr_overwrite_state(NULL, 0, MTRR_TYPE_WRBACK); + guest_force_mtrr_state(NULL, 0, MTRR_TYPE_WRBACK); } #if defined(CONFIG_AMD_MEM_ENCRYPT) diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c index d6818c6cafda..633469fab536 100644 --- a/arch/x86/xen/enlighten_pv.c +++ b/arch/x86/xen/enlighten_pv.c @@ -171,7 +171,7 @@ static void __init xen_set_mtrr_data(void) /* Only overwrite MTRR state if any MTRR could be got from Xen. */ if (reg) - mtrr_overwrite_state(var, reg, MTRR_TYPE_UNCACHABLE); + guest_force_mtrr_state(var, reg, MTRR_TYPE_UNCACHABLE); #endif } @@ -195,7 +195,7 @@ static void __init xen_pv_init_platform(void) if (xen_initial_domain()) xen_set_mtrr_data(); else - mtrr_overwrite_state(NULL, 0, MTRR_TYPE_WRBACK); + guest_force_mtrr_state(NULL, 0, MTRR_TYPE_WRBACK); /* Adjust nr_cpu_ids before "enumeration" happens */ xen_smp_count_cpus(); -- knotmirror 0.0.1 From aa135d1d0902c49ed45bec98c61c1b4022652b7e Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Fri, 20 Dec 2024 09:40:29 +0100 Subject: [PATCH 7/7] x86/mm: Remove unnecessary include of The header file linux/extable.h is included for search_exception_tables(). That function is no longer used since commit: c2508ec5a58db ("mm: introduce new 'lock_mm_and_find_vma()' page fault helper") Remove it. Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20241220084029.473617-1-bigeasy@linutronix.de --- arch/x86/mm/fault.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index e6c469b323cc..ef12ff3db903 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -7,7 +7,6 @@ #include /* test_thread_flag(), ... */ #include /* task_stack_*(), ... */ #include /* oops_begin/end, ... */ -#include /* search_exception_tables */ #include /* max_low_pfn */ #include /* kfence_handle_page_fault */ #include /* NOKPROBE_SYMBOL, ... */