From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932705AbcBJFHS (ORCPT ); Wed, 10 Feb 2016 00:07:18 -0500 Received: from e34.co.us.ibm.com ([32.97.110.152]:41742 "EHLO e34.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932614AbcBJFHP (ORCPT ); Wed, 10 Feb 2016 00:07:15 -0500 X-IBM-Helo: d03dlp03.boulder.ibm.com X-IBM-MailFrom: aneesh.kumar@linux.vnet.ibm.com X-IBM-RcptTo: linux-kernel@vger.kernel.org;stable@vger.kernel.org From: "Aneesh Kumar K.V" To: Vineet Gupta , "Kirill A. Shutemov" Cc: Vineet Gupta , Andrea Arcangeli , Andrew Morton , stable@vger.kernel.org, linux-snps-arc@lists.infradead.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: Re: [PATCH] mm,thp: khugepaged: call pte flush at the time of collapse In-Reply-To: <1455080175-10987-1-git-send-email-vgupta@synopsys.com> References: <1455080175-10987-1-git-send-email-vgupta@synopsys.com> User-Agent: Notmuch/0.20.2 (http://notmuchmail.org) Emacs/24.5.1 (x86_64-pc-linux-gnu) Date: Wed, 10 Feb 2016 10:37:02 +0530 Message-ID: <87fux1xifd.fsf@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain X-TM-AS-MML: disable X-Content-Scanned: Fidelis XPS MAILER x-cbid: 16021005-0017-0000-0000-000011F9719F Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Vineet Gupta writes: > This showed up on ARC when running LMBench bw_mem tests as > Overlapping TLB Machine Check Exception triggered due to STLB entry > (2M pages) overlapping some NTLB entry (regular 8K page). > > bw_mem 2m touches a large chunk of vaddr creating NTLB entries. > In the interim khugepaged kicks in, collapsing the contiguous ptes into > a single pmd. pmdp_collapse_flush()->flush_pmd_tlb_range() is called to > flush out NTLB entries for the ptes. This for ARC (by design) can only > shootdown STLB entries (for pmd). The stray NTLB entries cause the overlap > with the subsequent STLB entry for collapsed page. > So make pmdp_collapse_flush() call pte flush interface not pmd flush. > > Note that originally all thp flush call sites in generic code called > flush_tlb_range() leaving it to architecture to implement the flush for > pte and/or pmd. Commit 12ebc1581ad11454 changed this by calling a new > opt-in API flush_pmd_tlb_range() which made the semantics more explicit > but failed to distinguish the pte vs pmd flush in generic code, which is > what this patch fixes. > > Note that ARC can fixed w/o touching the generic pmdp_collapse_flush() > by defining a ARC version, but that defeats the purpose of generic > version, plus sementically this is the right thing to do. > > Fixes STAR 9000961194: LMBench on AXS103 triggering duplicate TLB > exceptions with super pages > > Cc: Kirill A. Shutemov > Cc: Aneesh Kumar K.V > Cc: Andrea Arcangeli > Cc: Andrew Morton > Cc: #4.4 > Cc: > Cc: linux-kernel@vger.kernel.org > Cc: linux-mm@kvack.org > Fixes: 12ebc1581ad11454 ("mm,thp: introduce flush_pmd_tlb_range") > Signed-off-by: Vineet Gupta Reviewed-by: Aneesh Kumar K.V We do have reverse usage in migration code path, which I have as a patch here. https://github.com/kvaneesh/linux/commit/b8a78933fea93cb0b2978868e59a0a4b12eb92eb > --- > mm/pgtable-generic.c | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) > > diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c > index 7d3db0247983..1ba58213ad65 100644 > --- a/mm/pgtable-generic.c > +++ b/mm/pgtable-generic.c > @@ -210,7 +210,9 @@ pmd_t pmdp_collapse_flush(struct vm_area_struct *vma, unsigned long address, > VM_BUG_ON(address & ~HPAGE_PMD_MASK); > VM_BUG_ON(pmd_trans_huge(*pmdp)); > pmd = pmdp_huge_get_and_clear(vma->vm_mm, address, pmdp); > - flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE); > + > + /* collapse entails shooting down ptes not pmd */ > + flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE); > return pmd; > } > #endif > -- > 2.5.0 -aneesh