From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752511AbcBPFU1 (ORCPT ); Tue, 16 Feb 2016 00:20:27 -0500 Received: from mail-pa0-f44.google.com ([209.85.220.44]:33027 "EHLO mail-pa0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750952AbcBPFUZ (ORCPT ); Tue, 16 Feb 2016 00:20:25 -0500 Message-ID: <1455600014.3308.9.camel@gmail.com> Subject: Re: [PATCH V3] powerpc/mm: Fix Multi hit ERAT cause by recent THP update From: Balbir Singh To: "Aneesh Kumar K.V" , benh@kernel.crashing.org, paulus@samba.org, mpe@ellerman.id.au, akpm@linux-foundation.org, Mel Gorman , "Kirill A. Shutemov" Cc: linuxppc-dev@lists.ozlabs.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org Date: Tue, 16 Feb 2016 16:20:14 +1100 In-Reply-To: <87d1ryfd94.fsf@linux.vnet.ibm.com> References: <1454980831-16631-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com> <1455504278.16012.18.camel@gmail.com> <87lh6mfv2j.fsf@linux.vnet.ibm.com> <1455512997.16012.24.camel@gmail.com> <87d1ryfd94.fsf@linux.vnet.ibm.com> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.18.4 (3.18.4-1.fc23) Mime-Version: 1.0 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 2016-02-15 at 16:31 +0530, Aneesh Kumar K.V wrote: > Balbir Singh writes: > > > > Now we can't depend for mm_cpumask, a parallel find_linux_pte_hugepte > > > can happen outside that. Now i had a variant for kick_all_cpus_sync that > > > ignored idle cpus. But then that needs more verification. > > > > > > http://article.gmane.org/gmane.linux.ports.ppc.embedded/81105 > > Can be racy as a CPU moves from non-idle to idle > > > > In > > > > > +     pmd_hugepage_update(vma->vm_mm, address, pmdp, ~0UL, 0); > > > +     /* > > > +      * This ensures that generic code that rely on IRQ disabling > > > +      * to prevent a parallel THP split work as expected. > > > +      */ > > > +     kick_all_cpus_sync(); > > > > pmdp_invalidate()->pmd_hugepage_update() can still run in parallel with  > > find_linux_pte_or_hugepte() and race.. Am I missing something? > > > > Yes. But then we make sure that the pte_t returned by > find_linux_pte_or_hugepte doesn't change to a regular pmd entry by using > that kick. Now callers of find_lnux_pte_or_hugepte will check for > _PAGE_PRESENT. So if it called before > pmd_hugepage_update(_PAGE_PRESENT), we wait for the caller to finish the > usage (via kick()). Or they bail out after finding _PAGE_PRESENT cleared. Makes sense, but I would still check the assumption about checking for _PAGE_PRESENT Balbir Singh