From 209954cbc7d0ce1a190fc725d20ce303d74d2680 Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Thu, 14 Nov 2024 10:26:16 -0500 Subject: x86/mm/tlb: Update mm_cpumask lazily On busy multi-threaded workloads, there can be significant contention on the mm_cpumask at context switch time. Reduce that contention by updating mm_cpumask lazily, setting the CPU bit at context switch time (if not already set), and clearing the CPU bit at the first TLB flush sent to a CPU where the process isn't running. When a flurry of TLB flushes for a process happen, only the first one will be sent to CPUs where the process isn't running. The others will be sent to CPUs where the process is currently running. On an AMD Milan system with 36 cores, there is a noticeable difference: $ hackbench --groups 20 --loops 10000 Before: ~4.5s +/- 0.1s After: ~4.2s +/- 0.1s Signed-off-by: Rik van Riel Signed-off-by: Ingo Molnar Cc: Dave Hansen Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Mel Gorman Link: https://lore.kernel.org/r/20241114152723.1294686-2-riel@surriel.com --- arch/x86/kernel/alternative.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'arch/x86/kernel') 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); -- cgit v1.2.3 From dd4059634dab548c904eeae2660ba3c8f7ce843c Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Mon, 2 Dec 2024 09:31:39 +0200 Subject: 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/kernel/cpu/mtrr/generic.c | 6 +++--- arch/x86/kernel/cpu/mtrr/mtrr.c | 2 +- arch/x86/kernel/kvm.c | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/x86/kernel') 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) -- cgit v1.2.3