From patchwork Wed Mar 30 01:59:50 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Verma, Vishal L" X-Patchwork-Id: 8692371 Return-Path: X-Original-To: patchwork-linux-fsdevel@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 3C7E1C0553 for ; Wed, 30 Mar 2016 02:00:53 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 22A5620155 for ; Wed, 30 Mar 2016 02:00:52 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id EFD9820148 for ; Wed, 30 Mar 2016 02:00:50 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754392AbcC3CA0 (ORCPT ); Tue, 29 Mar 2016 22:00:26 -0400 Received: from mga14.intel.com ([192.55.52.115]:35358 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754191AbcC3CAZ (ORCPT ); Tue, 29 Mar 2016 22:00:25 -0400 Received: from fmsmga004.fm.intel.com ([10.253.24.48]) by fmsmga103.fm.intel.com with ESMTP; 29 Mar 2016 19:00:24 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.24,414,1455004800"; d="scan'208";a="75601566" Received: from omniknight.lm.intel.com ([10.232.112.171]) by fmsmga004.fm.intel.com with ESMTP; 29 Mar 2016 19:00:21 -0700 From: Vishal Verma To: linux-nvdimm@lists.01.org Cc: Vishal Verma , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, xfs@oss.sgi.com, linux-ext4@vger.kernel.org, linux-mm@kvack.org, Matthew Wilcox , Ross Zwisler , Dan Williams , Dave Chinner , Jan Kara , Jens Axboe , Al Viro , Andrew Morton , linux-kernel@vger.kernel.org, Christoph Hellwig Subject: [PATCH v2 5/5] dax: handle media errors in dax_do_io Date: Tue, 29 Mar 2016 19:59:50 -0600 Message-Id: <1459303190-20072-6-git-send-email-vishal.l.verma@intel.com> X-Mailer: git-send-email 2.5.5 In-Reply-To: <1459303190-20072-1-git-send-email-vishal.l.verma@intel.com> References: <1459303190-20072-1-git-send-email-vishal.l.verma@intel.com> Sender: linux-fsdevel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-fsdevel@vger.kernel.org X-Spam-Status: No, score=-7.9 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_HI, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP dax_do_io (called for read() or write() for a dax file system) may fail in the presence of bad blocks or media errors. Since we expect that a write should clear media errors on nvdimms, make dax_do_io fall back to the direct_IO path, which will send down a bio to the driver, which can then attempt to clear the error. Cc: Matthew Wilcox Cc: Dan Williams Cc: Ross Zwisler Cc: Dave Chinner Cc: Jan Kara Cc: Jens Axboe Cc: Al Viro Cc: Christoph Hellwig Signed-off-by: Vishal Verma --- fs/block_dev.c | 17 ++++++++++++++--- fs/ext2/inode.c | 22 +++++++++++++++------- fs/ext4/indirect.c | 18 +++++++++++++----- fs/ext4/inode.c | 21 ++++++++++++++------- fs/xfs/xfs_aops.c | 14 ++++++++++++-- 5 files changed, 68 insertions(+), 24 deletions(-) diff --git a/fs/block_dev.c b/fs/block_dev.c index c5837fa..d6113b9 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -166,13 +166,24 @@ blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) { struct file *file = iocb->ki_filp; struct inode *inode = bdev_file_inode(file); + ssize_t ret, ret_saved = 0; - if (IS_DAX(inode)) - return dax_do_io(iocb, inode, iter, offset, blkdev_get_block, + if (IS_DAX(inode)) { + ret = dax_do_io(iocb, inode, iter, offset, blkdev_get_block, NULL, DIO_SKIP_DIO_COUNT); - return __blockdev_direct_IO(iocb, inode, I_BDEV(inode), iter, offset, + if (ret == -EIO && (iov_iter_rw(iter) == WRITE)) + ret_saved = ret; + else + return ret; + } + + ret = __blockdev_direct_IO(iocb, inode, I_BDEV(inode), iter, offset, blkdev_get_block, NULL, NULL, DIO_SKIP_DIO_COUNT); + if (ret < 0 && ret_saved) + return ret_saved; + + return ret; } int __sync_blockdev(struct block_device *bdev, int wait) diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index 824f249..64792c6 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -859,14 +859,22 @@ ext2_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; size_t count = iov_iter_count(iter); - ssize_t ret; + ssize_t ret, ret_saved = 0; - if (IS_DAX(inode)) - ret = dax_do_io(iocb, inode, iter, offset, ext2_get_block, NULL, - DIO_LOCKING); - else - ret = blockdev_direct_IO(iocb, inode, iter, offset, - ext2_get_block); + if (IS_DAX(inode)) { + ret = dax_do_io(iocb, inode, iter, offset, ext2_get_block, + NULL, DIO_LOCKING | DIO_SKIP_HOLES); + if (ret == -EIO && iov_iter_rw(iter) == WRITE) + ret_saved = ret; + else + goto out; + } + + ret = blockdev_direct_IO(iocb, inode, iter, offset, ext2_get_block); + if (ret < 0 && ret_saved) + ret = ret_saved; + + out: if (ret < 0 && iov_iter_rw(iter) == WRITE) ext2_write_failed(mapping, offset + count); return ret; diff --git a/fs/ext4/indirect.c b/fs/ext4/indirect.c index 3027fa6..798f341 100644 --- a/fs/ext4/indirect.c +++ b/fs/ext4/indirect.c @@ -716,14 +716,22 @@ retry: NULL, NULL, 0); inode_dio_end(inode); } else { + ssize_t ret_saved = 0; + locked: - if (IS_DAX(inode)) + if (IS_DAX(inode)) { ret = dax_do_io(iocb, inode, iter, offset, ext4_dio_get_block, NULL, DIO_LOCKING); - else - ret = blockdev_direct_IO(iocb, inode, iter, offset, - ext4_dio_get_block); - + if (ret == -EIO && iov_iter_rw(iter) == WRITE) + ret_saved = ret; + else + goto skip_dio; + } + ret = blockdev_direct_IO(iocb, inode, iter, offset, + ext4_get_block); + if (ret < 0 && ret_saved) + ret = ret_saved; +skip_dio: if (unlikely(iov_iter_rw(iter) == WRITE && ret < 0)) { loff_t isize = i_size_read(inode); loff_t end = offset + count; diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index dab84a2..27f07c2 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -3341,7 +3341,7 @@ static ssize_t ext4_ext_direct_IO(struct kiocb *iocb, struct iov_iter *iter, { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; - ssize_t ret; + ssize_t ret, ret_saved = 0; size_t count = iov_iter_count(iter); int overwrite = 0; get_block_t *get_block_func = NULL; @@ -3401,15 +3401,22 @@ static ssize_t ext4_ext_direct_IO(struct kiocb *iocb, struct iov_iter *iter, #ifdef CONFIG_EXT4_FS_ENCRYPTION BUG_ON(ext4_encrypted_inode(inode) && S_ISREG(inode->i_mode)); #endif - if (IS_DAX(inode)) + if (IS_DAX(inode)) { ret = dax_do_io(iocb, inode, iter, offset, get_block_func, ext4_end_io_dio, dio_flags); - else - ret = __blockdev_direct_IO(iocb, inode, - inode->i_sb->s_bdev, iter, offset, - get_block_func, - ext4_end_io_dio, NULL, dio_flags); + if (ret == -EIO && iov_iter_rw(iter) == WRITE) + ret_saved = ret; + else + goto skip_dio; + } + ret = __blockdev_direct_IO(iocb, inode, + inode->i_sb->s_bdev, iter, offset, + get_block_func, + ext4_end_io_dio, NULL, dio_flags); + if (ret < 0 && ret_saved) + ret = ret_saved; + skip_dio: if (ret > 0 && !overwrite && ext4_test_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN)) { int err; diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index d445a64..7cfcf86 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -1413,6 +1413,7 @@ xfs_vm_direct_IO( dio_iodone_t *endio = NULL; int flags = 0; struct block_device *bdev; + ssize_t ret, ret_saved = 0; if (iov_iter_rw(iter) == WRITE) { endio = xfs_end_io_direct_write; @@ -1420,13 +1421,22 @@ xfs_vm_direct_IO( } if (IS_DAX(inode)) { - return dax_do_io(iocb, inode, iter, offset, + ret = dax_do_io(iocb, inode, iter, offset, xfs_get_blocks_direct, endio, 0); + if (ret == -EIO && iov_iter_rw(iter) == WRITE) + ret_saved = ret; + else + return ret; } bdev = xfs_find_bdev_for_inode(inode); - return __blockdev_direct_IO(iocb, inode, bdev, iter, offset, + ret = __blockdev_direct_IO(iocb, inode, bdev, iter, offset, xfs_get_blocks_direct, endio, NULL, flags); + + if (ret < 0 && ret_saved) + ret = ret_saved; + + return ret; } /*