Song noticed switch_mm_irqs_off taking a lot of CPU time in recent
kernels,using 2.4% of a 48 CPU system during a netperf to localhost run.
Digging into the profile, we noticed that cpumask_clear_cpu and
cpumask_set_cpu together take about half of the CPU time taken by
switch_mm_irqs_off.
However, the CPUs running netperf end up switching back and forth
between netperf and the idle task, which does not require changes
to the mm_cpumask. Furthermore, the init_mm cpumask ends up being
the most heavily contended one in the system.`
Skipping cpumask_clear_cpu and cpumask_set_cpu for init_mm
(mostly the idle task) reduced CPU use of switch_mm_irqs_off
from 2.4% of the CPU to 1.9% of the CPU, with the following
netperf commandline:
./super_netperf 300 -P 0 -t TCP_RR -p 8888 -H kerneltest008.09.atn1 -l 30 \
-- -r 300,300 -o -s 1M,1M -S 1M,1M
perf output w/o this patch:
1.26% netserver [kernel.vmlinux] [k] switch_mm_irqs_off
1.17% swapper [kernel.vmlinux] [k] switch_mm_irqs_off
perf output w/ this patch:
1.01% swapper [kernel.vmlinux] [k] switch_mm_irqs_off
0.88% netserver [kernel.vmlinux] [k] switch_mm_irqs_off
Netperf throughput is about the same before and after.
Signed-off-by: Rik van Riel <r...@surriel.com>
Reported-and-tested-by: Song Liu <songliubrav...@fb.com>
---
arch/x86/mm/tlb.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index e055d1a06699..c8f9c550f7ec 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -288,12 +288,14 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct
mm_struct *next,
/* Stop remote flushes for the previous mm */
VM_WARN_ON_ONCE(!cpumask_test_cpu(cpu, mm_cpumask(real_prev)) &&
real_prev != &init_mm);
- cpumask_clear_cpu(cpu, mm_cpumask(real_prev));
+ if (real_prev != &init_mm)
+ cpumask_clear_cpu(cpu, mm_cpumask(real_prev));
/*
* Start remote flushes and then read tlb_gen.
*/
- cpumask_set_cpu(cpu, mm_cpumask(next));
+ if (next != &init_mm)
+ cpumask_set_cpu(cpu, mm_cpumask(next));
next_tlb_gen = atomic64_read(&next->context.tlb_gen);
choose_new_asid(next, next_tlb_gen, &new_asid, &need_flush);
