Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1460309

Re: [PATCH] x86/irq: do not substract irq_tlb_count from irq_call_count

Show all headers | View raw

* Aaron Lu <aaron.lu@intel.com> wrote:

> This is found by LKP's cyclic performance regression tracking recently
> with the vm-scalability test suite. I have bisected to commit
> 0a7ce4b5a632 ("mm/rmap: share the i_mmap_rwsem"). This commit didn't do
> anything wrong but revealed the irq_call_count problem. IIUC, the commit
> makes rwc->remap_one in rmap_walk_file concurrent with multiple threads.
> When remap_one is try_to_unmap_one, then multiple threads could queue
> flush tlb to the same CPU but only one IPI will be sent.

Note, for some reason the commit ID you used is wrong, the real one is:

  3dec0ba0be6a ("mm/rmap: share the i_mmap_rwsem")

I have fixed this in the changelog.

Thanks,

	Ingo

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH] x86/irq: do not substract irq_tlb_count from irq_call_count Aaron Lu <aaron.lu@intel.com> - 2016-08-11 09:50 +0200
  Re: [PATCH] x86/irq: do not substract irq_tlb_count from  irq_call_count Aaron Lu <aaron.lu@intel.com> - 2016-08-11 11:20 +0200
  Re: [PATCH] x86/irq: do not substract irq_tlb_count from  irq_call_count Ingo Molnar <mingo@kernel.org> - 2016-08-11 11:20 +0200
  [tip:x86/urgent] x86/irq: Do not substract irq_tlb_count from  irq_call_count tip-bot for Aaron Lu <tipbot@zytor.com> - 2016-08-11 14:00 +0200
  Re: [PATCH] x86/irq: do not substract irq_tlb_count from irq_call_count "Huang\, Ying" <ying.huang@intel.com> - 2016-08-11 17:20 +0200
    Re: [PATCH] x86/irq: do not substract irq_tlb_count from  irq_call_count Aaron Lu <aaron.lu@intel.com> - 2016-08-12 04:20 +0200

csiph-web