From patchwork Wed Jul 18 00:07:28 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Martin Wilck X-Patchwork-Id: 10531081 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 2BCB0602C2 for ; Wed, 18 Jul 2018 00:07:38 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 1AD9A292B5 for ; Wed, 18 Jul 2018 00:07:38 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 0F1BB292BA; Wed, 18 Jul 2018 00:07:38 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.9 required=2.0 tests=BAYES_00, MAILING_LIST_MULTI, RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id A24FF292B5 for ; Wed, 18 Jul 2018 00:07:37 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1731412AbeGRAmk (ORCPT ); Tue, 17 Jul 2018 20:42:40 -0400 Received: from smtp2.provo.novell.com ([137.65.250.81]:56117 "EHLO smtp2.provo.novell.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1729805AbeGRAmk (ORCPT ); Tue, 17 Jul 2018 20:42:40 -0400 Received: from [192.168.1.40] (prv-ext-foundry1int.gns.novell.com [137.65.251.240]) by smtp2.provo.novell.com with ESMTP (TLS encrypted); Tue, 17 Jul 2018 18:07:32 -0600 Message-ID: Subject: Re: Silent data corruption in blkdev_direct_IO() From: Martin Wilck To: Ming Lei Cc: Jens Axboe , Hannes Reinecke , Christoph Hellwig , "linux-block@vger.kernel.org" , Ming Lei , jack@suse.com, kent.overstreet@gmail.com Date: Wed, 18 Jul 2018 02:07:28 +0200 In-Reply-To: References: X-Mailer: Evolution 3.28.2 Mime-Version: 1.0 Sender: linux-block-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-block@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP On Mon, 2018-07-16 at 19:45 +0800, Ming Lei wrote: > On Sat, Jul 14, 2018 at 6:29 AM, Martin Wilck > wrote: > > Hi Ming & Jens, > > > > On Fri, 2018-07-13 at 12:54 -0600, Jens Axboe wrote: > > > On 7/12/18 5:29 PM, Ming Lei wrote: > > > > > > > > Maybe you can try the following patch from Christoph to see if > > > > it > > > > makes a > > > > difference: > > > > > > > > https://marc.info/?l=linux-kernel&m=153013977816825&w=2 > > > > > > That's not a bad idea. > > > > Are you saying that the previous "nasty hack" in > > bio_iov_iter_get_pages() was broken, and the new one is not? > > I've scratched my head over that (old) code lately, but I couldn't > > spot > > an actual error in it. > > Yeah, I think the new patch in above link is better, and looks its > correctness > is easy to prove. > > https://marc.info/?t=152812081400001&r=1&w=2 > > So please test the patch if possible. I haven't tested yet, but upon further inspection, I can tell that the current code (without Christoph's patch) is actually broken if the function is called with bio->bi_vcnt > 0. The following patch explains the problem. AFAICS, Christoph's new code is correct. From b75adc856119346e02126cf8975755300f2d9b7f Mon Sep 17 00:00:00 2001 From: Martin Wilck Date: Wed, 18 Jul 2018 01:56:37 +0200 Subject: [PATCH] block: bio_iov_iter_get_pages: fix size of last iovec If the last page of the bio is not "full", the length of the last vector bin needs to be corrected. This bin has the index (bio->bi_vcnt - 1), but in bio->bi_io_vec, not in the "bv" helper array which is shifted by the value of bio->bi_vcnt at function invocation. Signed-off-by: Martin Wilck --- block/bio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/block/bio.c b/block/bio.c index 53e0f0a..c22e76f 100644 --- a/block/bio.c +++ b/block/bio.c @@ -913,7 +913,7 @@ int bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) bv[0].bv_offset += offset; bv[0].bv_len -= offset; if (diff) - bv[bio->bi_vcnt - 1].bv_len -= diff; + bio->bi_io_vec[bio->bi_vcnt - 1].bv_len -= diff; iov_iter_advance(iter, size); return 0;