From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-2.8 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,PDS_BAD_THREAD_QP_64, SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 39FBDC2B9F4 for ; Tue, 22 Jun 2021 22:31:43 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id D720861026 for ; Tue, 22 Jun 2021 22:31:42 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org D720861026 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=ACULAB.COM Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id BD78D6B0089; Tue, 22 Jun 2021 18:31:41 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id B86816B008A; Tue, 22 Jun 2021 18:31:41 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 9B2206B008C; Tue, 22 Jun 2021 18:31:41 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0030.hostedemail.com [216.40.44.30]) by kanga.kvack.org (Postfix) with ESMTP id 3FBFF6B0089 for ; Tue, 22 Jun 2021 18:31:41 -0400 (EDT) Received: from smtpin08.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay05.hostedemail.com (Postfix) with ESMTP id 6CD50181CC1B4 for ; Tue, 22 Jun 2021 22:31:41 +0000 (UTC) X-FDA: 78282808002.08.C28C310 Received: from eu-smtp-delivery-151.mimecast.com (eu-smtp-delivery-151.mimecast.com [185.58.86.151]) by imf03.hostedemail.com (Postfix) with ESMTP id A919DC0201C0 for ; Tue, 22 Jun 2021 22:31:40 +0000 (UTC) Received: from AcuMS.aculab.com (156.67.243.121 [156.67.243.121]) (Using TLS) by relay.mimecast.com with ESMTP id uk-mta-35-rM-Uk-sYOYOVhFr34Mu4Kw-1; Tue, 22 Jun 2021 23:31:37 +0100 X-MC-Unique: rM-Uk-sYOYOVhFr34Mu4Kw-1 Received: from AcuMS.Aculab.com (10.202.163.6) by AcuMS.aculab.com (10.202.163.6) with Microsoft SMTP Server (TLS) id 15.0.1497.18; Tue, 22 Jun 2021 23:31:36 +0100 Received: from AcuMS.Aculab.com ([fe80::994c:f5c2:35d6:9b65]) by AcuMS.aculab.com ([fe80::994c:f5c2:35d6:9b65%12]) with mapi id 15.00.1497.018; Tue, 22 Jun 2021 23:31:36 +0100 From: David Laight To: 'Matthew Wilcox' CC: 'David Howells' , Al Viro , "torvalds@linux-foundation.org" , Ted Ts'o , Dave Hansen , Andrew Morton , "linux-mm@kvack.org" , "linux-ext4@vger.kernel.org" , "linux-fsdevel@vger.kernel.org" , "linux-kernel@vger.kernel.org" Subject: RE: Do we need to unrevert "fs: do not prefault sys_write() user buffer pages"? Thread-Topic: Do we need to unrevert "fs: do not prefault sys_write() user buffer pages"? Thread-Index: AQHXZ4N03eCp9KNCtEagRX54mD94w6sgkJXg///0tICAABcJIA== Date: Tue, 22 Jun 2021 22:31:36 +0000 Message-ID: <0aac203b65b24776a27468e63114a24b@AcuMS.aculab.com> References: <3221175.1624375240@warthog.procyon.org.uk> <3225322.1624379221@warthog.procyon.org.uk> <7a6d8c55749d46d09f6f6e27a99fde36@AcuMS.aculab.com> In-Reply-To: Accept-Language: en-GB, en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.202.205.107] MIME-Version: 1.0 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: aculab.com Content-Language: en-US Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Authentication-Results: imf03.hostedemail.com; dkim=none; spf=pass (imf03.hostedemail.com: domain of david.laight@aculab.com designates 185.58.86.151 as permitted sender) smtp.mailfrom=david.laight@aculab.com; dmarc=pass (policy=none) header.from=ACULAB.COM X-Stat-Signature: kssgi6ge8hrfmhi77gy74uxrzb53ir5i X-Rspamd-Server: rspam04 X-Rspamd-Queue-Id: A919DC0201C0 X-HE-Tag: 1624401100-589173 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: From: Matthew Wilcox > Sent: 22 June 2021 23:04 >=20 > On Tue, Jun 22, 2021 at 09:55:09PM +0000, David Laight wrote: > > From: David Howells > > > Sent: 22 June 2021 17:27 > > > > > > Al Viro wrote: > > > > > > > On Tue, Jun 22, 2021 at 04:20:40PM +0100, David Howells wrote: > > > > > > > > > and wondering if the iov_iter_fault_in_readable() is actually eff= ective. > > > > > Yes, it can make sure that the page we're intending to modify is = dragged > > > > > into the pagecache and marked uptodate so that it can be read fro= m, but is > > > > > it possible for the page to then get reclaimed before we get to > > > > > iov_iter_copy_from_user_atomic()? a_ops->write_begin() could pot= entially > > > > > take a long time, say if it has to go and get a lock/lease from a= server. > > > > > > > > Yes, it is. So what? We'll just retry. You *can't* take faults w= hile > > > > holding some pages locked; not without shitloads of deadlocks. > > > > > > In that case, can we amend the comment immediately above > > > iov_iter_fault_in_readable()? > > > > > > =09/* > > > =09 * Bring in the user page that we will copy from _first_. > > > =09 * Otherwise there's a nasty deadlock on copying from the > > > =09 * same page as we're writing to, without it being marked > > > =09 * up-to-date. > > > =09 * > > > =09 * Not only is this an optimisation, but it is also required > > > =09 * to check that the address is actually valid, when atomic > > > =09 * usercopies are used, below. > > > =09 */ > > > =09if (unlikely(iov_iter_fault_in_readable(i, bytes))) { > > > > > > The first part suggests this is for deadlock avoidance. If that's no= t true, > > > then this should perhaps be changed. > > > > I'd say something like: > > =09/* > > =09 * The actual copy_from_user() is done with a lock held > > =09 * so cannot fault in missing pages. > > =09 * So fault in the pages first. > > =09 * If they get paged out the inatomic usercopy will fail > > =09 * and the whole operation is retried. > > =09 * > > =09 * Hopefully there are enough memory pages available to > > =09 * stop this looping forever. > > =09 */ > > > > It is perfectly possible for another application thread to > > invalidate one of the buffer fragments after iov_iter_fault_in_readable= () > > return success - so it will then fail on the second pass. > > > > The maximum number of pages required is twice the maximum number > > of iov fragments. > > If the system is crawling along with no available memory pages > > the same physical page could get used for two user pages. >=20 > I would suggest reading the function before you suggest modifications > to it. >=20 > offset =3D (pos & (PAGE_SIZE - 1)); > bytes =3D min_t(unsigned long, PAGE_SIZE - offset, > iov_iter_count(i)); Right the transfer size is limited to PAGE_SIZE. But the user supplied iov[] could be a lot of 2 byte buffers all with base addresses (PAGE_SIZE * n - 1). So you might need two pages for each iov[] fragment. And you need not to recycle the earlier ones. =09David - Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1= PT, UK Registration No: 1397386 (Wales)