Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752922AbdLDQER (ORCPT ); Mon, 4 Dec 2017 11:04:17 -0500 Received: from mail.linuxfoundation.org ([140.211.169.12]:33866 "EHLO mail.linuxfoundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753398AbdLDQEN (ORCPT ); Mon, 4 Dec 2017 11:04:13 -0500 From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman , stable@vger.kernel.org, "Kirill A. Shutemov" , Michal Hocko , Hugh Dickins , Linus Torvalds Subject: [PATCH 4.14 04/95] mm, thp: Do not make page table dirty unconditionally in touch_p[mu]d() Date: Mon, 4 Dec 2017 16:59:28 +0100 Message-Id: <20171204160046.374315077@linuxfoundation.org> X-Mailer: git-send-email 2.15.1 In-Reply-To: <20171204160046.206920966@linuxfoundation.org> References: <20171204160046.206920966@linuxfoundation.org> User-Agent: quilt/0.65 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3473 Lines: 108 4.14-stable review patch. If anyone has any objections, please let me know. ------------------ From: Kirill A. Shutemov commit a8f97366452ed491d13cf1e44241bc0b5740b1f0 upstream. Currently, we unconditionally make page table dirty in touch_pmd(). It may result in false-positive can_follow_write_pmd(). We may avoid the situation, if we would only make the page table entry dirty if caller asks for write access -- FOLL_WRITE. The patch also changes touch_pud() in the same way. Signed-off-by: Kirill A. Shutemov Cc: Michal Hocko Cc: Hugh Dickins Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- mm/huge_memory.c | 36 +++++++++++++----------------------- 1 file changed, 13 insertions(+), 23 deletions(-) --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -842,20 +842,15 @@ EXPORT_SYMBOL_GPL(vmf_insert_pfn_pud); #endif /* CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD */ static void touch_pmd(struct vm_area_struct *vma, unsigned long addr, - pmd_t *pmd) + pmd_t *pmd, int flags) { pmd_t _pmd; - /* - * We should set the dirty bit only for FOLL_WRITE but for now - * the dirty bit in the pmd is meaningless. And if the dirty - * bit will become meaningful and we'll only set it with - * FOLL_WRITE, an atomic set_bit will be required on the pmd to - * set the young bit, instead of the current set_pmd_at. - */ - _pmd = pmd_mkyoung(pmd_mkdirty(*pmd)); + _pmd = pmd_mkyoung(*pmd); + if (flags & FOLL_WRITE) + _pmd = pmd_mkdirty(_pmd); if (pmdp_set_access_flags(vma, addr & HPAGE_PMD_MASK, - pmd, _pmd, 1)) + pmd, _pmd, flags & FOLL_WRITE)) update_mmu_cache_pmd(vma, addr, pmd); } @@ -884,7 +879,7 @@ struct page *follow_devmap_pmd(struct vm return NULL; if (flags & FOLL_TOUCH) - touch_pmd(vma, addr, pmd); + touch_pmd(vma, addr, pmd, flags); /* * device mapped pages can only be returned if the @@ -995,20 +990,15 @@ out: #ifdef CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD static void touch_pud(struct vm_area_struct *vma, unsigned long addr, - pud_t *pud) + pud_t *pud, int flags) { pud_t _pud; - /* - * We should set the dirty bit only for FOLL_WRITE but for now - * the dirty bit in the pud is meaningless. And if the dirty - * bit will become meaningful and we'll only set it with - * FOLL_WRITE, an atomic set_bit will be required on the pud to - * set the young bit, instead of the current set_pud_at. - */ - _pud = pud_mkyoung(pud_mkdirty(*pud)); + _pud = pud_mkyoung(*pud); + if (flags & FOLL_WRITE) + _pud = pud_mkdirty(_pud); if (pudp_set_access_flags(vma, addr & HPAGE_PUD_MASK, - pud, _pud, 1)) + pud, _pud, flags & FOLL_WRITE)) update_mmu_cache_pud(vma, addr, pud); } @@ -1031,7 +1021,7 @@ struct page *follow_devmap_pud(struct vm return NULL; if (flags & FOLL_TOUCH) - touch_pud(vma, addr, pud); + touch_pud(vma, addr, pud, flags); /* * device mapped pages can only be returned if the @@ -1407,7 +1397,7 @@ struct page *follow_trans_huge_pmd(struc page = pmd_page(*pmd); VM_BUG_ON_PAGE(!PageHead(page) && !is_zone_device_page(page), page); if (flags & FOLL_TOUCH) - touch_pmd(vma, addr, pmd); + touch_pmd(vma, addr, pmd, flags); if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) { /* * We don't mlock() pte-mapped THPs. This way we can avoid