diff options
| author | Rik van Riel <riel@redhat.com> | 2013-07-31 22:14:21 -0400 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2013-09-26 17:21:43 -0700 |
| commit | 3bf030f755a2be92e6bd8b74743f01a57643db74 (patch) | |
| tree | 3a223585a07624207c8745c3b72109fc513161d7 /arch | |
| parent | 04111416b44ffb1140ff7801de5dd31d2b7fe207 (diff) | |
sched/x86: Optimize switch_mm() for multi-threaded workloads
commit 8f898fbbe5ee5e20a77c4074472a1fd088dc47d1 upstream.
Dick Fowles, Don Zickus and Joe Mario have been working on
improvements to perf, and noticed heavy cache line contention
on the mm_cpumask, running linpack on a 60 core / 120 thread
system.
The cause turned out to be unnecessary atomic accesses to the
mm_cpumask. When in lazy TLB mode, the CPU is only removed from
the mm_cpumask if there is a TLB flush event.
Most of the time, no such TLB flush happens, and the kernel
skips the TLB reload. It can also skip the atomic memory
set & test.
Here is a summary of Joe's test results:
* The __schedule function dropped from 24% of all program cycles down
to 5.5%.
* The cacheline contention/hotness for accesses to that bitmask went
from being the 1st/2nd hottest - down to the 84th hottest (0.3% of
all shared misses which is now quite cold)
* The average load latency for the bit-test-n-set instruction in
__schedule dropped from 10k-15k cycles down to an average of 600 cycles.
* The linpack program results improved from 133 GFlops to 144 GFlops.
Peak GFlops rose from 133 to 153.
Reported-by: Don Zickus <dzickus@redhat.com>
Reported-by: Joe Mario <jmario@redhat.com>
Tested-by: Joe Mario <jmario@redhat.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Reviewed-by: Paul Turner <pjt@google.com>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/20130731221421.616d3d20@annuminas.surriel.com
[ Made the comments consistent around the modified code. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'arch')
| -rw-r--r-- | arch/x86/include/asm/mmu_context.h | 20 |
1 files changed, 13 insertions, 7 deletions
diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h index cdbf36776106..be12c534fd59 100644 --- a/arch/x86/include/asm/mmu_context.h +++ b/arch/x86/include/asm/mmu_context.h @@ -45,22 +45,28 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, /* Re-load page tables */ load_cr3(next->pgd); - /* stop flush ipis for the previous mm */ + /* Stop flush ipis for the previous mm */ cpumask_clear_cpu(cpu, mm_cpumask(prev)); - /* - * load the LDT, if the LDT is different: - */ + /* Load the LDT, if the LDT is different: */ if (unlikely(prev->context.ldt != next->context.ldt)) load_LDT_nolock(&next->context); } #ifdef CONFIG_SMP - else { + else { this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK); BUG_ON(this_cpu_read(cpu_tlbstate.active_mm) != next); - if (!cpumask_test_and_set_cpu(cpu, mm_cpumask(next))) { - /* We were in lazy tlb mode and leave_mm disabled + if (!cpumask_test_cpu(cpu, mm_cpumask(next))) { + /* + * On established mms, the mm_cpumask is only changed + * from irq context, from ptep_clear_flush() while in + * lazy tlb mode, and here. Irqs are blocked during + * schedule, protecting us from simultaneous changes. + */ + cpumask_set_cpu(cpu, mm_cpumask(next)); + /* + * We were in lazy tlb mode and leave_mm disabled * tlb flush IPI delivery. We must reload CR3 * to make sure to use no freed page tables. */ |