From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751470AbdFHKwK (ORCPT ); Thu, 8 Jun 2017 06:52:10 -0400 Received: from mx2.suse.de ([195.135.220.15]:53030 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1750752AbdFHKwJ (ORCPT ); Thu, 8 Jun 2017 06:52:09 -0400 Subject: Re: [PATCH 3/3] mm: migrate: Stabilise page count when migrating transparent hugepages To: Will Deacon , linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: mark.rutland@arm.com, akpm@linux-foundation.org, kirill.shutemov@linux.intel.com, Punit.Agrawal@arm.com, mgorman@suse.de, steve.capper@arm.com References: <1496771916-28203-1-git-send-email-will.deacon@arm.com> <1496771916-28203-4-git-send-email-will.deacon@arm.com> From: Vlastimil Babka Message-ID: <33165470-21b8-15d2-b952-a7bbf74ee83d@suse.cz> Date: Thu, 8 Jun 2017 12:52:07 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.1.1 MIME-Version: 1.0 In-Reply-To: <1496771916-28203-4-git-send-email-will.deacon@arm.com> Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 06/06/2017 07:58 PM, Will Deacon wrote: > When migrating a transparent hugepage, migrate_misplaced_transhuge_page > guards itself against a concurrent fastgup of the page by checking that > the page count is equal to 2 before and after installing the new pmd. > > If the page count changes, then the pmd is reverted back to the original > entry, however there is a small window where the new (possibly writable) > pmd is installed and the underlying page could be written by userspace. > Restoring the old pmd could therefore result in loss of data. > > This patch fixes the problem by freezing the page count whilst updating > the page tables, which protects against a concurrent fastgup without the > need to restore the old pmd in the failure case (since the page count can > no longer change under our feet). > > Cc: Mel Gorman > Signed-off-by: Will Deacon > --- > mm/migrate.c | 15 ++------------- > 1 file changed, 2 insertions(+), 13 deletions(-) > > diff --git a/mm/migrate.c b/mm/migrate.c > index 89a0a1707f4c..8b21f1b1ec6e 100644 > --- a/mm/migrate.c > +++ b/mm/migrate.c > @@ -1913,7 +1913,6 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, > int page_lru = page_is_file_cache(page); > unsigned long mmun_start = address & HPAGE_PMD_MASK; > unsigned long mmun_end = mmun_start + HPAGE_PMD_SIZE; > - pmd_t orig_entry; > > /* > * Rate-limit the amount of data that is being migrated to a node. > @@ -1956,8 +1955,7 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, > /* Recheck the target PMD */ > mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); > ptl = pmd_lock(mm, pmd); > - if (unlikely(!pmd_same(*pmd, entry) || page_count(page) != 2)) { > -fail_putback: > + if (unlikely(!pmd_same(*pmd, entry) || !page_ref_freeze(page, 2))) { > spin_unlock(ptl); > mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end); > > @@ -1979,7 +1977,6 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, > goto out_unlock; > } > > - orig_entry = *pmd; > entry = mk_huge_pmd(new_page, vma->vm_page_prot); > entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma); > > @@ -1996,15 +1993,7 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, There's a comment above this: /* * Clear the old entry under pagetable lock and establish the new PTE. * Any parallel GUP will either observe the old page blocking on the * page lock, block on the page table lock or observe the new page. * The SetPageUptodate on the new page and page_add_new_anon_rmap * guarantee the copy is visible before the pagetable update. */ Is it still correct? Didn't the freezing prevent some of the cases above? > set_pmd_at(mm, mmun_start, pmd, entry); > update_mmu_cache_pmd(vma, address, &entry); > > - if (page_count(page) != 2) { BTW, how did the old code recognize that page count would increase and then decrease back? > - set_pmd_at(mm, mmun_start, pmd, orig_entry); > - flush_pmd_tlb_range(vma, mmun_start, mmun_end); > - mmu_notifier_invalidate_range(mm, mmun_start, mmun_end); > - update_mmu_cache_pmd(vma, address, &entry); > - page_remove_rmap(new_page, true); > - goto fail_putback; > - } > - > + page_ref_unfreeze(page, 2); > mlock_migrate_page(new_page, page); > page_remove_rmap(page, true); > set_page_owner_migrate_reason(new_page, MR_NUMA_MISPLACED); >