Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965223AbbEMOpN (ORCPT ); Wed, 13 May 2015 10:45:13 -0400 Received: from prod-mail-xrelay02.akamai.com ([72.246.2.14]:49048 "EHLO prod-mail-xrelay02.akamai.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934274AbbEMOpH (ORCPT ); Wed, 13 May 2015 10:45:07 -0400 Date: Wed, 13 May 2015 10:45:06 -0400 From: Eric B Munson To: Michal Hocko Cc: Michael Kerrisk , Andrew Morton , Linus Torvalds , David Rientjes , LKML , Linux API , linux-mm@kvack.org, Michal Hocko Subject: Re: [PATCH 1/2] mmap.2: clarify MAP_LOCKED semantic Message-ID: <20150513144506.GD1227@akamai.com> References: <1431527892-2996-1-git-send-email-miso@dhcp22.suse.cz> <1431527892-2996-2-git-send-email-miso@dhcp22.suse.cz> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="rz+pwK2yUstbofK6" Content-Disposition: inline In-Reply-To: <1431527892-2996-2-git-send-email-miso@dhcp22.suse.cz> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2834 Lines: 70 --rz+pwK2yUstbofK6 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, 13 May 2015, Michal Hocko wrote: > From: Michal Hocko >=20 > MAP_LOCKED had a subtly different semantic from mmap(2)+mlock(2) since > it has been introduced. > mlock(2) fails if the memory range cannot get populated to guarantee > that no future major faults will happen on the range. mmap(MAP_LOCKED) on > the other hand silently succeeds even if the range was populated only > partially. >=20 > Fixing this subtle difference in the kernel is rather awkward because > the memory population happens after mm locks have been dropped and so > the cleanup before returning failure (munlock) could operate on something > else than the originally mapped area. >=20 > E.g. speculative userspace page fault handler catching SEGV and doing > mmap(fault_addr, MAP_FIXED|MAP_LOCKED) might discard portion of a racing > mmap and lead to lost data. Although it is not clear whether such a > usage would be valid, mmap page doesn't explicitly describe requirements > for threaded applications so we cannot exclude this possibility. >=20 > This patch makes the semantic of MAP_LOCKED explicit and suggest using > mmap + mlock as the only way to guarantee no later major page faults. >=20 > Signed-off-by: Michal Hocko Does the problem still happend when MAP_POPULATE | MAP_LOCKED is used (AFAICT MAP_POPULATE will cause the mmap to fail if all the pages cannot be made present). Either way this is a good catch. Acked-by: Eric B Munson --rz+pwK2yUstbofK6 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJVU2NyAAoJELbVsDOpoOa9qrcQAL7qkobw8zBNPCW62SwMI8Gt HnilnO3Y42MDYOVSNzcPvmwDbETpqrS5mzK4cEpM5KQFLIz6XMvmZr2D/16BIOED 3ieX+d2Eg7kw5O87/rq2MVrmLYfAXoDnni25d6SQEQBQihuAGEk7M45KBRaa8zcb aKElv4ov8xjMMcRwRzTdpdzaWogFjNz5TuRsm4Zu9Xj/MrPhmhzaCwoBpOqxGKp5 kt9utHzVh4h6MDC0ZCZ4dfw9UCAMfA4wtEncS4JcEQ+a/LPN1n/ZG7LqqqwIL50x 1miEgtCxIOVlaLXUUCHeCa0tfvSBPFznDfjpnnnGeowFKdG9HfoakbgrzeGTWFX+ +6MLsCzba/ODKwK4sALB0U9KUCLg+qjBFuTAX16mpaRJLjiaPvK1MHOWoD70yC+z fKAKjFlXRmebAM9nmpx0Q0uKUzUT8wK7WN0IFB9sJrqM7HoR+nzAweGtbXQrNKwm O5jA/jPChP7zshrllyfgp56rRiYF3ztsT1kGDA7147c8S17NIzW4oFGYCUG7s8ua XPsjaMQHgrPn33LVMq9dfX01kLMft1GKVYLil6xoGW+AfqBrNwbOLvVQfy0WrxM2 XKNwGCa5G/V4RYVzQEKdzfvwKYslxbpr3iqmjR0PQNhjdZWFAgIHonrboQmc59PC fkrgPry07zHAVqVQVIDX =+Q1g -----END PGP SIGNATURE----- --rz+pwK2yUstbofK6-- -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/