[tip: x86/mm] x86/mm/tlb: Only trim the mm_cpumask once a second
From: tip-bot2 for Rik van Riel
Date: Fri Dec 06 2024 - 04:40:37 EST
The following commit has been merged into the x86/mm branch of tip:
Commit-ID: 6db2526c1d694c91c6e05e2f186c085e9460f202
Gitweb: https://git.kernel.org/tip/6db2526c1d694c91c6e05e2f186c085e9460f202
Author: Rik van Riel <riel@xxxxxx>
AuthorDate: Wed, 04 Dec 2024 21:03:16 -05:00
Committer: Ingo Molnar <mingo@xxxxxxxxxx>
CommitterDate: Fri, 06 Dec 2024 10:26:20 +01:00
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 <oliver.sang@xxxxxxxxx>
Signed-off-by: Rik van Riel <riel@xxxxxxxxxxx>
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
Cc: Andy Lutomirski <luto@xxxxxxxxxx>
Cc: Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Link: https://lore.kernel.org/r/20241204210316.612ee573@fangorn
Closes: https://lore.kernel.org/oe-lkp/202411282207.6bd28eae-lkp@xxxxxxxxx/
---
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 ce4677b..3b496cd 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 2886cb6..795fdd5 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 69e79ff..02fc2aa 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 3c30817..458a5d5 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();