From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751427AbaLQQxV (ORCPT ); Wed, 17 Dec 2014 11:53:21 -0500 Received: from foss-mx-na.foss.arm.com ([217.140.108.86]:51340 "EHLO foss-mx-na.foss.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751023AbaLQQxU (ORCPT ); Wed, 17 Dec 2014 11:53:20 -0500 Date: Wed, 17 Dec 2014 16:53:10 +0000 From: Will Deacon To: Linus Torvalds Cc: Dave Hansen , Benjamin Herrenschmidt , Peter Zijlstra , Russell King - ARM Linux , Michal Simek , LKML , Linux-MM Subject: Re: post-3.18 performance regression in TLB flushing code Message-ID: <20141217165310.GJ870@arm.com> References: <5490A5F8.6050504@sr71.net> <20141217100810.GA3461@arm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Dec 17, 2014 at 04:28:23PM +0000, Linus Torvalds wrote: > On Wed, Dec 17, 2014 at 2:08 AM, Will Deacon wrote: > > > > I think there are a couple of things you could try to see if that 2% comes > > back: > > > > * Revert the patch and try the one here [1] instead (which only does part > > (1) of the above). > > > > -- or -- > > > > * Instead of adding the tlb->end check to tlb_flush_mmu, add it to > > tlb_flush_mmu_free > > or just move the check back to tlb_flush_mmu() where it belongs. > > I don't see why you moved it to "tlb_flush_mmu_tlbonly()" in the first > place, or why you'd now want to add it to tlb_flush_mmu_free(). > > Both of those helper functions have two callers: > > - tlb_flush_mmu(). Doing it here (instead of in the helper functions) > is the right thing to do > > - the "force_flush" case: we know we have added at least one page to > the TLB state so checking for it is pointless. > > So I'm not seeing why you wanted to do it in tlb_flush_mmu_tlbonly(), > and now add it to tlb_flush_mmu_free(). That seems bogus. I guess I was being overly cautious in case tlb_flush_mmu_tlbonly grows additional users, but you're right. > So why not just this trivial patch, to make the logic be the same it > used to be (just using "end > 0" instead of the old "need_flush")? Looks fine to me... Dave? Will > diff --git a/mm/memory.c b/mm/memory.c > index c3b9097251c5..6efe36a998ba 100644 > --- a/mm/memory.c > +++ b/mm/memory.c > @@ -235,9 +235,6 @@ void tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm, unsigned long > > static void tlb_flush_mmu_tlbonly(struct mmu_gather *tlb) > { > - if (!tlb->end) > - return; > - > tlb_flush(tlb); > mmu_notifier_invalidate_range(tlb->mm, tlb->start, tlb->end); > #ifdef CONFIG_HAVE_RCU_TABLE_FREE > @@ -259,6 +256,9 @@ static void tlb_flush_mmu_free(struct mmu_gather *tlb) > > void tlb_flush_mmu(struct mmu_gather *tlb) > { > + if (!tlb->end) > + return; > + > tlb_flush_mmu_tlbonly(tlb); > tlb_flush_mmu_free(tlb); > }