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 Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 4CBEAE7C4EB for ; Wed, 4 Oct 2023 18:22:38 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S244072AbjJDSWk (ORCPT ); Wed, 4 Oct 2023 14:22:40 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:37454 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S243770AbjJDSWj (ORCPT ); Wed, 4 Oct 2023 14:22:39 -0400 Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 629669E for ; Wed, 4 Oct 2023 11:22:35 -0700 (PDT) Received: by smtp.kernel.org (Postfix) with ESMTPSA id A33A1C433C7; Wed, 4 Oct 2023 18:22:34 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1696443755; bh=jcxNQ1plkWsy9x84sXn858tyghzck7/RWCbO365Cf2Q=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=V+TtX+8ggy7d1jJHZVtKax1HXvTfUdISKLZ1iQGn2CxTL/MeZRL6G8gdqG21JD7nl Ly1BVHtrk1bEzeYEZd96SVD+JC5Yej4Ng+1Igc6pqvqXDjlL9utkQ/qXHAXq8JTRd8 LTafDNqrTsiMDNNH/hw9gnhBDZO1mOvlUANgdP1k= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Trond Myklebust , Anna Schumaker , Sasha Levin Subject: [PATCH 6.5 001/321] NFS: Fix error handling for O_DIRECT write scheduling Date: Wed, 4 Oct 2023 19:52:26 +0200 Message-ID: <20231004175229.288366501@linuxfoundation.org> X-Mailer: git-send-email 2.42.0 In-Reply-To: <20231004175229.211487444@linuxfoundation.org> References: <20231004175229.211487444@linuxfoundation.org> User-Agent: quilt/0.67 X-stable: review X-Patchwork-Hint: ignore MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: stable@vger.kernel.org 6.5-stable review patch. If anyone has any objections, please let me know. ------------------ From: Trond Myklebust [ Upstream commit 954998b60caa8f2a3bf3abe490de6f08d283687a ] If we fail to schedule a request for transmission, there are 2 possibilities: 1) Either we hit a fatal error, and we just want to drop the remaining requests on the floor. 2) We were asked to try again, in which case we should allow the outstanding RPC calls to complete, so that we can recoalesce requests and try again. Fixes: d600ad1f2bdb ("NFS41: pop some layoutget errors to application") Signed-off-by: Trond Myklebust Signed-off-by: Anna Schumaker Signed-off-by: Sasha Levin --- fs/nfs/direct.c | 62 ++++++++++++++++++++++++++++++++++++------------- 1 file changed, 46 insertions(+), 16 deletions(-) diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 47d892a1d363d..ee88f0a6e7b81 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -528,10 +528,9 @@ nfs_direct_write_scan_commit_list(struct inode *inode, static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) { struct nfs_pageio_descriptor desc; - struct nfs_page *req, *tmp; + struct nfs_page *req; LIST_HEAD(reqs); struct nfs_commit_info cinfo; - LIST_HEAD(failed); nfs_init_cinfo_from_dreq(&cinfo, dreq); nfs_direct_write_scan_commit_list(dreq->inode, &reqs, &cinfo); @@ -549,27 +548,36 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) &nfs_direct_write_completion_ops); desc.pg_dreq = dreq; - list_for_each_entry_safe(req, tmp, &reqs, wb_list) { + while (!list_empty(&reqs)) { + req = nfs_list_entry(reqs.next); /* Bump the transmission count */ req->wb_nio++; if (!nfs_pageio_add_request(&desc, req)) { - nfs_list_move_request(req, &failed); spin_lock(&cinfo.inode->i_lock); - dreq->flags = 0; - if (desc.pg_error < 0) + if (dreq->error < 0) { + desc.pg_error = dreq->error; + } else if (desc.pg_error != -EAGAIN) { + dreq->flags = 0; + if (!desc.pg_error) + desc.pg_error = -EIO; dreq->error = desc.pg_error; - else - dreq->error = -EIO; + } else + dreq->flags = NFS_ODIRECT_RESCHED_WRITES; spin_unlock(&cinfo.inode->i_lock); + break; } nfs_release_request(req); } nfs_pageio_complete(&desc); - while (!list_empty(&failed)) { - req = nfs_list_entry(failed.next); + while (!list_empty(&reqs)) { + req = nfs_list_entry(reqs.next); nfs_list_remove_request(req); nfs_unlock_and_release_request(req); + if (desc.pg_error == -EAGAIN) + nfs_mark_request_commit(req, NULL, &cinfo, 0); + else + nfs_release_request(req); } if (put_dreq(dreq)) @@ -794,9 +802,11 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq, { struct nfs_pageio_descriptor desc; struct inode *inode = dreq->inode; + struct nfs_commit_info cinfo; ssize_t result = 0; size_t requested_bytes = 0; size_t wsize = max_t(size_t, NFS_SERVER(inode)->wsize, PAGE_SIZE); + bool defer = false; trace_nfs_direct_write_schedule_iovec(dreq); @@ -837,17 +847,37 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq, break; } - nfs_lock_request(req); - if (!nfs_pageio_add_request(&desc, req)) { - result = desc.pg_error; - nfs_unlock_and_release_request(req); - break; - } pgbase = 0; bytes -= req_len; requested_bytes += req_len; pos += req_len; dreq->bytes_left -= req_len; + + if (defer) { + nfs_mark_request_commit(req, NULL, &cinfo, 0); + continue; + } + + nfs_lock_request(req); + if (nfs_pageio_add_request(&desc, req)) + continue; + + /* Exit on hard errors */ + if (desc.pg_error < 0 && desc.pg_error != -EAGAIN) { + result = desc.pg_error; + nfs_unlock_and_release_request(req); + break; + } + + /* If the error is soft, defer remaining requests */ + nfs_init_cinfo_from_dreq(&cinfo, dreq); + spin_lock(&cinfo.inode->i_lock); + dreq->flags = NFS_ODIRECT_RESCHED_WRITES; + spin_unlock(&cinfo.inode->i_lock); + nfs_unlock_request(req); + nfs_mark_request_commit(req, NULL, &cinfo, 0); + desc.pg_error = 0; + defer = true; } nfs_direct_release_pages(pagevec, npages); kvfree(pagevec); -- 2.40.1