From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id F3A471FECD9; Fri, 21 Feb 2025 15:23:40 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1740151421; cv=none; b=H71s+QmmXFRDnuJmZo+qbYxrHTc1BJH2sAdUn83zGvzC2MSHKzBw4Fbs4DZ0qpEWvylFxT5IWcuyphIHs9uLoTKRQsSndgkQkuMVlvAjFVeamAD2a1w7U7WEL/R4S9RFqGfHQFpezdvRMvoUofdAKX4/UT5h+Rt3moUfxkioGvA= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1740151421; c=relaxed/simple; bh=Mp5kAwKKJQ2x9SogxNJaqftBzNGl2rKtXwcDdtS8gZ0=; h=Subject:To:Cc:From:Date:In-Reply-To:Message-ID:MIME-Version: Content-Type; b=cef4ZL5Vk71Kfq+V0NbLnjAB23vns5WPKwvJxiH+E2vxwlCqm55czsmFUMh7aMBIjPG+0WXRJOfkttYUo8nzHcYCBD2fAW36ZSVK6iUgm17M37yOe7YK9UdbQJoqYiamO5bbG7E1Qv0vYQs1mYssGNtYTu2X7LE8DBvpUXZJB2Q= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b=jp6y8INX; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="jp6y8INX" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 7EEC2C4CEE4; Fri, 21 Feb 2025 15:23:40 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1740151420; bh=Mp5kAwKKJQ2x9SogxNJaqftBzNGl2rKtXwcDdtS8gZ0=; h=Subject:To:Cc:From:Date:In-Reply-To:From; b=jp6y8INX4CQL5eL4WrsApL+lLQ36wz4lkqq9NqYujSUPKzgjRmj8TxrpSAW+HtToS gyONj56X87OCcD7GRd12D2ag/K+OCIZs9BM89LY3KWXHa2VY0BxdW2k80KGcKxlIta bzTHDC7SKAnoKOqKJ+1PsnDiTRQqcMZyZPKUtd38= Subject: Patch "xfs: don't free cowblocks from under dirty pagecache on unshare" has been added to the 6.6-stable tree To: bfoster@redhat.com,catherine.hoang@oracle.com,cem@kernel.org,djwong@kernel.org,gregkh@linuxfoundation.org,xfs-stable@lists.linux.dev Cc: From: Date: Fri, 21 Feb 2025 16:23:27 +0100 In-Reply-To: <20250205214025.72516-7-catherine.hoang@oracle.com> Message-ID: <2025022127-wisplike-quartet-1f98@gregkh> Precedence: bulk X-Mailing-List: xfs-stable@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=ANSI_X3.4-1968 Content-Transfer-Encoding: 8bit X-stable: commit X-Patchwork-Hint: ignore This is a note to let you know that I've just added the patch titled xfs: don't free cowblocks from under dirty pagecache on unshare to the 6.6-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary The filename of the patch is: xfs-don-t-free-cowblocks-from-under-dirty-pagecache-on-unshare.patch and it can be found in the queue-6.6 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let know about it. >From stable+bounces-113972-greg=kroah.com@vger.kernel.org Wed Feb 5 22:40:54 2025 From: Catherine Hoang Date: Wed, 5 Feb 2025 13:40:07 -0800 Subject: xfs: don't free cowblocks from under dirty pagecache on unshare To: stable@vger.kernel.org Cc: xfs-stable@lists.linux.dev Message-ID: <20250205214025.72516-7-catherine.hoang@oracle.com> From: Brian Foster commit 4390f019ad7866c3791c3d768d2ff185d89e8ebe upstream. fallocate unshare mode explicitly breaks extent sharing. When a command completes, it checks the data fork for any remaining shared extents to determine whether the reflink inode flag and COW fork preallocation can be removed. This logic doesn't consider in-core pagecache and I/O state, however, which means we can unsafely remove COW fork blocks that are still needed under certain conditions. For example, consider the following command sequence: xfs_io -fc "pwrite 0 1k" -c "reflink 0 256k 1k" \ -c "pwrite 0 32k" -c "funshare 0 1k" This allocates a data block at offset 0, shares it, and then overwrites it with a larger buffered write. The overwrite triggers COW fork preallocation, 32 blocks by default, which maps the entire 32k write to delalloc in the COW fork. All but the shared block at offset 0 remains hole mapped in the data fork. The unshare command redirties and flushes the folio at offset 0, removing the only shared extent from the inode. Since the inode no longer maps shared extents, unshare purges the COW fork before the remaining 28k may have written back. This leaves dirty pagecache backed by holes, which writeback quietly skips, thus leaving clean, non-zeroed pagecache over holes in the file. To verify, fiemap shows holes in the first 32k of the file and reads return different data across a remount: $ xfs_io -c "fiemap -v" : EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS ... 1: [8..511]: hole 504 ... $ xfs_io -c "pread -v 4k 8" 00001000: cd cd cd cd cd cd cd cd ........ $ umount ; mount $ xfs_io -c "pread -v 4k 8" 00001000: 00 00 00 00 00 00 00 00 ........ To avoid this problem, make unshare follow the same rules used for background cowblock scanning and never purge the COW fork for inodes with dirty pagecache or in-flight I/O. Fixes: 46afb0628b86347 ("xfs: only flush the unshared range in xfs_reflink_unshare") Signed-off-by: Brian Foster Reviewed-by: Darrick J. Wong Signed-off-by: Carlos Maiolino Signed-off-by: Catherine Hoang Acked-by: Darrick J. Wong Signed-off-by: Greg Kroah-Hartman --- fs/xfs/xfs_icache.c | 8 +------- fs/xfs/xfs_reflink.c | 3 +++ fs/xfs/xfs_reflink.h | 19 +++++++++++++++++++ 3 files changed, 23 insertions(+), 7 deletions(-) --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -1271,13 +1271,7 @@ xfs_prep_free_cowblocks( */ if (!sync && inode_is_open_for_write(VFS_I(ip))) return false; - if ((VFS_I(ip)->i_state & I_DIRTY_PAGES) || - mapping_tagged(VFS_I(ip)->i_mapping, PAGECACHE_TAG_DIRTY) || - mapping_tagged(VFS_I(ip)->i_mapping, PAGECACHE_TAG_WRITEBACK) || - atomic_read(&VFS_I(ip)->i_dio_count)) - return false; - - return true; + return xfs_can_free_cowblocks(ip); } /* --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -1600,6 +1600,9 @@ xfs_reflink_clear_inode_flag( ASSERT(xfs_is_reflink_inode(ip)); + if (!xfs_can_free_cowblocks(ip)) + return 0; + error = xfs_reflink_inode_has_shared_extents(*tpp, ip, &needs_flag); if (error || needs_flag) return error; --- a/fs/xfs/xfs_reflink.h +++ b/fs/xfs/xfs_reflink.h @@ -16,6 +16,25 @@ static inline bool xfs_is_cow_inode(stru return xfs_is_reflink_inode(ip) || xfs_is_always_cow_inode(ip); } +/* + * Check whether it is safe to free COW fork blocks from an inode. It is unsafe + * to do so when an inode has dirty cache or I/O in-flight, even if no shared + * extents exist in the data fork, because outstanding I/O may target blocks + * that were speculatively allocated to the COW fork. + */ +static inline bool +xfs_can_free_cowblocks(struct xfs_inode *ip) +{ + struct inode *inode = VFS_I(ip); + + if ((inode->i_state & I_DIRTY_PAGES) || + mapping_tagged(inode->i_mapping, PAGECACHE_TAG_DIRTY) || + mapping_tagged(inode->i_mapping, PAGECACHE_TAG_WRITEBACK) || + atomic_read(&inode->i_dio_count)) + return false; + return true; +} + extern int xfs_reflink_trim_around_shared(struct xfs_inode *ip, struct xfs_bmbt_irec *irec, bool *shared); int xfs_bmap_trim_cow(struct xfs_inode *ip, struct xfs_bmbt_irec *imap, Patches currently in stable-queue which might be from catherine.hoang@oracle.com are queue-6.6/xfs-return-bool-from-xfs_attr3_leaf_add.patch queue-6.6/xfs-fix-a-sloppy-memory-handling-bug-in-xfs_iroot_realloc.patch queue-6.6/xfs-streamline-xfs_filestream_pick_ag.patch queue-6.6/xfs-merge-xfs_attr_leaf_try_add-into-xfs_attr_leaf_addname.patch queue-6.6/xfs-don-t-free-cowblocks-from-under-dirty-pagecache-on-unshare.patch queue-6.6/xfs-pass-the-exact-range-to-initialize-to-xfs_initialize_perag.patch queue-6.6/xfs-assert-a-valid-limit-in-xfs_rtfind_forw.patch queue-6.6/xfs-don-t-use-__gfp_retry_mayfail-in-xfs_initialize_perag.patch queue-6.6/xfs-use-try_cmpxchg-in-xlog_cil_insert_pcp_aggregate.patch queue-6.6/xfs-don-t-ifdef-around-the-exact-minlen-allocations.patch queue-6.6/xfs-reduce-unnecessary-searches-when-searching-for-the-best-extents.patch queue-6.6/xfs-validate-inumber-in-xfs_iget.patch queue-6.6/xfs-support-lowmode-allocations-in-xfs_bmap_exact_minlen_extent_alloc.patch queue-6.6/xfs-skip-background-cowblock-trims-on-inodes-open-for-write.patch queue-6.6/xfs-remove-empty-declartion-in-header-file.patch queue-6.6/xfs-fold-xfs_bmap_alloc_userdata-into-xfs_bmapi_allocate.patch queue-6.6/xfs-update-the-file-system-geometry-after-recoverying-superblock-buffers.patch queue-6.6/xfs-call-xfs_bmap_exact_minlen_extent_alloc-from-xfs_bmap_btalloc.patch queue-6.6/xfs-distinguish-extra-split-from-real-enospc-from-xfs_attr_node_try_addname.patch queue-6.6/xfs-error-out-when-a-superblock-buffer-update-reduces-the-agcount.patch queue-6.6/xfs-update-the-pag-for-the-last-ag-at-recovery-time.patch queue-6.6/xfs-check-for-delayed-allocations-before-setting-extsize.patch queue-6.6/xfs-fix-a-typo.patch queue-6.6/xfs-distinguish-extra-split-from-real-enospc-from-xfs_attr3_leaf_split.patch