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=-6.8 required=3.0 tests=DKIMWL_WL_HIGH,DKIM_SIGNED, DKIM_VALID,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,SIGNED_OFF_BY, SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED,USER_AGENT_GIT autolearn=unavailable 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 AE08FECE599 for ; Wed, 16 Oct 2019 21:57:06 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 8791821A4C for ; Wed, 16 Oct 2019 21:57:06 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=default; t=1571263026; bh=igHqsfJ/KTYG9kzGSxI1nBlYQVC8EhHXmU66VJt5M2Q=; h=From:To:Cc:Subject:Date:In-Reply-To:References:List-ID:From; b=YBvszdAxgCaikrZ46jUolhH9tTO2i2KOPTXhl62j9FxHUC8M+VqRAuBE6Bf1Dibb5 Z2XFyH/eZr42cfgDb4Z1Ao8Se4ebSaRb5aqyaK9lVm8CjWsA5SUVMtVrWor82ESu/T JgWG/+CvILuzfDOhyFaa2Or0TSeinfvIl2+7udSA= Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2395451AbfJPV5F (ORCPT ); Wed, 16 Oct 2019 17:57:05 -0400 Received: from mail.kernel.org ([198.145.29.99]:48150 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2406653AbfJPV4a (ORCPT ); Wed, 16 Oct 2019 17:56:30 -0400 Received: from localhost (unknown [192.55.54.58]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPSA id BF2BF21925; Wed, 16 Oct 2019 21:56:28 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=default; t=1571262988; bh=igHqsfJ/KTYG9kzGSxI1nBlYQVC8EhHXmU66VJt5M2Q=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=UwZAkgfxrKIp58P/ET3sZOrFrX6VoZ26TP1jaw01AonA8hI1PBPDWGe0P3xZDhyuv pM5Jbyquyg4m7LxnHTOISDYHccciSPNSIO29Dovf7wPW9NjpVqXJNmk74VVd4LPfQy VrCG6yqLCDRz96/kjjbb5ApfarvR8wL0sHgR3074= From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman , stable@vger.kernel.org, Su Yanjun , Trond Myklebust , Anna Schumaker Subject: [PATCH 4.14 56/65] NFS: Fix O_DIRECT accounting of number of bytes read/written Date: Wed, 16 Oct 2019 14:51:10 -0700 Message-Id: <20191016214838.027140327@linuxfoundation.org> X-Mailer: git-send-email 2.23.0 In-Reply-To: <20191016214756.457746573@linuxfoundation.org> References: <20191016214756.457746573@linuxfoundation.org> User-Agent: quilt/0.66 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Trond Myklebust commit 031d73ed768a40684f3ca21992265ffdb6a270bf upstream. When a series of O_DIRECT reads or writes are truncated, either due to eof or due to an error, then we should return the number of contiguous bytes that were received/sent starting at the offset specified by the application. Currently, we are failing to correctly check contiguity, and so we're failing the generic/465 in xfstests when the race between the read and write RPCs causes the file to get extended while the 2 reads are outstanding. If the first read RPC call wins the race and returns with eof set, we should treat the second read RPC as being truncated. Reported-by: Su Yanjun Fixes: 1ccbad9f9f9bd ("nfs: fix DIO good bytes calculation") Cc: stable@vger.kernel.org # 4.1+ Signed-off-by: Trond Myklebust Signed-off-by: Anna Schumaker Signed-off-by: Greg Kroah-Hartman --- fs/nfs/direct.c | 78 ++++++++++++++++++++++++++++++-------------------------- 1 file changed, 43 insertions(+), 35 deletions(-) --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -122,32 +122,49 @@ static inline int put_dreq(struct nfs_di } static void -nfs_direct_good_bytes(struct nfs_direct_req *dreq, struct nfs_pgio_header *hdr) +nfs_direct_handle_truncated(struct nfs_direct_req *dreq, + const struct nfs_pgio_header *hdr, + ssize_t dreq_len) +{ + struct nfs_direct_mirror *mirror = &dreq->mirrors[hdr->pgio_mirror_idx]; + + if (!(test_bit(NFS_IOHDR_ERROR, &hdr->flags) || + test_bit(NFS_IOHDR_EOF, &hdr->flags))) + return; + if (dreq->max_count >= dreq_len) { + dreq->max_count = dreq_len; + if (dreq->count > dreq_len) + dreq->count = dreq_len; + + if (test_bit(NFS_IOHDR_ERROR, &hdr->flags)) + dreq->error = hdr->error; + else /* Clear outstanding error if this is EOF */ + dreq->error = 0; + } + if (mirror->count > dreq_len) + mirror->count = dreq_len; +} + +static void +nfs_direct_count_bytes(struct nfs_direct_req *dreq, + const struct nfs_pgio_header *hdr) { - int i; - ssize_t count; + struct nfs_direct_mirror *mirror = &dreq->mirrors[hdr->pgio_mirror_idx]; + loff_t hdr_end = hdr->io_start + hdr->good_bytes; + ssize_t dreq_len = 0; - WARN_ON_ONCE(dreq->count >= dreq->max_count); + if (hdr_end > dreq->io_start) + dreq_len = hdr_end - dreq->io_start; - if (dreq->mirror_count == 1) { - dreq->mirrors[hdr->pgio_mirror_idx].count += hdr->good_bytes; - dreq->count += hdr->good_bytes; - } else { - /* mirrored writes */ - count = dreq->mirrors[hdr->pgio_mirror_idx].count; - if (count + dreq->io_start < hdr->io_start + hdr->good_bytes) { - count = hdr->io_start + hdr->good_bytes - dreq->io_start; - dreq->mirrors[hdr->pgio_mirror_idx].count = count; - } - /* update the dreq->count by finding the minimum agreed count from all - * mirrors */ - count = dreq->mirrors[0].count; + nfs_direct_handle_truncated(dreq, hdr, dreq_len); - for (i = 1; i < dreq->mirror_count; i++) - count = min(count, dreq->mirrors[i].count); + if (dreq_len > dreq->max_count) + dreq_len = dreq->max_count; - dreq->count = count; - } + if (mirror->count < dreq_len) + mirror->count = dreq_len; + if (dreq->count < dreq_len) + dreq->count = dreq_len; } /* @@ -401,20 +418,12 @@ static void nfs_direct_read_completion(s struct nfs_direct_req *dreq = hdr->dreq; spin_lock(&dreq->lock); - if (test_bit(NFS_IOHDR_ERROR, &hdr->flags)) - dreq->error = hdr->error; - if (test_bit(NFS_IOHDR_REDO, &hdr->flags)) { spin_unlock(&dreq->lock); goto out_put; } - if (hdr->good_bytes != 0) - nfs_direct_good_bytes(dreq, hdr); - - if (test_bit(NFS_IOHDR_EOF, &hdr->flags)) - dreq->error = 0; - + nfs_direct_count_bytes(dreq, hdr); spin_unlock(&dreq->lock); while (!list_empty(&hdr->pages)) { @@ -651,6 +660,9 @@ static void nfs_direct_write_reschedule( nfs_direct_write_scan_commit_list(dreq->inode, &reqs, &cinfo); dreq->count = 0; + dreq->max_count = 0; + list_for_each_entry(req, &reqs, wb_list) + dreq->max_count += req->wb_bytes; dreq->verf.committed = NFS_INVALID_STABLE_HOW; nfs_clear_pnfs_ds_commit_verifiers(&dreq->ds_cinfo); for (i = 0; i < dreq->mirror_count; i++) @@ -783,17 +795,13 @@ static void nfs_direct_write_completion( nfs_init_cinfo_from_dreq(&cinfo, dreq); spin_lock(&dreq->lock); - - if (test_bit(NFS_IOHDR_ERROR, &hdr->flags)) - dreq->error = hdr->error; - if (test_bit(NFS_IOHDR_REDO, &hdr->flags)) { spin_unlock(&dreq->lock); goto out_put; } + nfs_direct_count_bytes(dreq, hdr); if (hdr->good_bytes != 0) { - nfs_direct_good_bytes(dreq, hdr); if (nfs_write_need_commit(hdr)) { if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES) request_commit = true;