linux-kernel - [PATCH] mm/gup: fix gup_pmd

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-Id: <20190111034033.601-1-yuzhao@google.com>
Date:   Thu, 10 Jan 2019 20:40:33 -0700
From:   Yu Zhao <yuzhao@...gle.com>
To:     Andrew Morton <akpm@...ux-foundation.org>,
        Dan Williams <dan.j.williams@...el.com>,
        Huang Ying <ying.huang@...el.com>
Cc:     Matthew Wilcox <willy@...radead.org>,
        Keith Busch <keith.busch@...el.com>,
        "Michael S . Tsirkin" <mst@...hat.com>, Jan Kara <jack@...e.cz>,
        John Hubbard <jhubbard@...dia.com>,
        Wei Yang <richard.weiyang@...il.com>,
        Mike Rapoport <rppt@...ux.ibm.com>,
        Andrea Arcangeli <aarcange@...hat.com>, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org,
        "Kirill A . Shutemov" <kirill.shutemov@...ux.intel.com>,
        Yu Zhao <yuzhao@...gle.com>
Subject: [PATCH] mm/gup: fix gup_pmd_range() for dax

For dax pmd, pmd_trans_huge() returns false but pmd_huge() returns
true on x86. So the function works as long as hugetlb is configured.
However, dax doesn't depend on hugetlb.

Signed-off-by: Yu Zhao <yuzhao@...gle.com>
---
 mm/gup.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/mm/gup.c b/mm/gup.c
index 05acd7e2eb22..75029649baca 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -1674,7 +1674,8 @@ static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end,
 		if (!pmd_present(pmd))
 			return 0;
 
-		if (unlikely(pmd_trans_huge(pmd) || pmd_huge(pmd))) {
+		if (unlikely(pmd_trans_huge(pmd) || pmd_huge(pmd) ||
+			     pmd_devmap(pmd))) {
 			/*
 			 * NUMA hinting faults need to be handled in the GUP
 			 * slowpath for accounting purposes and so that they
-- 
2.20.1.97.g81188d93c3-goog