All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chris Mason <mason@kernel.org>
To: mason@kernel.org, tytso@mit.edu, cem@kernel.org,
	david@fromorbit.com, jack@suse.cz, djwong@kernel.org,
	josef@toxicpanda.com, hch@lst.de, linux-xfs@vger.kernel.org,
	linux-fsdevel@vger.kernel.org, linux-ext4@vger.kernel.org,
	tz2294@columbia.edu, brauner@kernel.org
Subject: [PATCH v9 13/14] xfs: don't enable DAX on zoned inodes
Date: Thu, 12 Nov 2026 16:46:17 -0500	[thread overview]
Message-ID: <20261112214604.67875-14-mason@kernel.org> (raw)
In-Reply-To: <20261112214604.67875-1-mason@kernel.org>

A store through a shared mapping can dereference a NULL pointer if the
file is zoned and has S_DAX.

The zoned code does not support DAX, but XFS doesn't check for this:
with dax=always or FS_XFLAG_DAX, an inode gets S_DAX if its device does
DAX, and an internal zoned section on pmem does. The write fault then
takes the zoned path:

    xfs_filemap_fault()
      xfs_write_fault()
        xfs_write_fault_zoned()
          page_folio(vmf->page)     /* NULL in a DAX fault */

ftruncate() crashes in a similar way, in xfs_reflink_end_cow().

The fix used here is to never set S_DAX on a zoned inode, and to refuse
to change the realtime flag of an S_DAX inode on a zoned file system.

Zoned files now use the page cache with dax=always or FS_XFLAG_DAX. A
file whose realtime flag is cleared stays without DAX until its inode
is evicted.

Fixes: be458049ffe3 ("xfs: enable the zoned RT device feature")
Signed-off-by: Chris Mason <mason@kernel.org>
---

Notes:
    v8->v9:
    - The commit log and the comments say that the zoned code does not
      support DAX, no longer that it cannot (Christoph Hellwig).
    - The commit log no longer says that ftruncate() only crashes on kernels
      without an earlier patch of the series. That patch has lost the test
      that stopped it.
    - Was 15/15.
    
    v7->v8:
    - New patch.

 fs/xfs/xfs_ioctl.c | 8 +++++---
 fs/xfs/xfs_iops.c  | 4 ++++
 2 files changed, 9 insertions(+), 3 deletions(-)

diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
index af13ad6784b7..49ba762ea969 100644
--- a/fs/xfs/xfs_ioctl.c
+++ b/fs/xfs/xfs_ioctl.c
@@ -564,11 +564,13 @@ xfs_ioctl_setattr_xflags(
 
 		/*
 		 * If S_DAX is enabled on this file, we can only switch the
-		 * device if both support fsdax.  We can't update S_DAX because
-		 * there might be other threads walking down the access paths.
+		 * device if both support fsdax, and the zoned code does not.
+		 * We can't update S_DAX because there might be other threads
+		 * walking down the access paths.
 		 */
 		if (IS_DAX(VFS_I(ip)) &&
-		    (mp->m_ddev_targp->bt_daxdev == NULL ||
+		    (xfs_has_zoned(mp) ||
+		     mp->m_ddev_targp->bt_daxdev == NULL ||
 		     (mp->m_rtdev_targp &&
 		      mp->m_rtdev_targp->bt_daxdev == NULL)))
 			return -EINVAL;
diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
index d307ec97fac8..a84a4c1c3fc2 100644
--- a/fs/xfs/xfs_iops.c
+++ b/fs/xfs/xfs_iops.c
@@ -1373,6 +1373,10 @@ xfs_inode_supports_dax(
 	if (mp->m_sb.sb_blocksize != PAGE_SIZE)
 		return false;
 
+	/* The zoned code does not support DAX. */
+	if (xfs_is_zoned_inode(ip))
+		return false;
+
 	/* Device has to support DAX too. */
 	return xfs_inode_buftarg(ip)->bt_daxdev != NULL;
 }
-- 
2.55.0


  parent reply	other threads:[~2026-11-12 21:46 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-12 21:46 [PATCH v9 00/14] reflink and atomic write bugs Chris Mason
2026-11-12 21:46 ` [PATCH v9 01/14] xfs: don't skip the flush before shifting extents Chris Mason
2026-11-12 21:55   ` sashiko-bot
2026-11-12 21:46 ` [PATCH v9 02/14] xfs: flush the shifted range before changing the file Chris Mason
2026-11-12 21:54   ` sashiko-bot
2026-11-12 21:46 ` [PATCH v9 03/14] xfs: don't change the block map under a dirty folio Chris Mason
2026-11-12 22:03   ` sashiko-bot
2026-11-12 21:46 ` [PATCH v9 04/14] iomap: don't punch delalloc under a dirty EOF block Chris Mason
2026-11-12 21:54   ` sashiko-bot
2026-11-12 21:46 ` [PATCH v9 05/14] fs: validate atomic writes after O_APPEND and limits Chris Mason
2026-11-12 22:16   ` sashiko-bot
2026-11-12 21:46 ` [PATCH v9 06/14] xfs: recheck direct write alignment after write checks Chris Mason
2026-11-12 22:22   ` sashiko-bot
2026-11-12 21:46 ` [PATCH v9 07/14] xfs: test for reflink after the dio " Chris Mason
2026-11-12 21:54   ` sashiko-bot
2026-11-12 21:46 ` [PATCH v9 08/14] xfs: refuse writes to files racing with realtime inode flag changes Chris Mason
2026-11-12 22:11   ` sashiko-bot
2026-11-12 21:46 ` [PATCH v9 09/14] xfs: test for always COW after the dio write checks Chris Mason
2026-11-12 21:53   ` sashiko-bot
2026-11-12 21:46 ` [PATCH v9 10/14] xfs: refuse fallocate racing with realtime flag changes Chris Mason
2026-11-12 22:11   ` sashiko-bot
2026-11-12 21:46 ` [PATCH v9 11/14] xfs: take the MMAPLOCK in xfs_fileattr_set() Chris Mason
2026-11-12 22:13   ` sashiko-bot
2026-11-12 23:49   ` Christoph Hellwig
2026-11-12 21:46 ` [PATCH v9 12/14] xfs: restart faults racing with realtime flag changes Chris Mason
2026-11-12 22:08   ` sashiko-bot
2026-11-12 23:50   ` Christoph Hellwig
2026-11-12 21:46 ` Chris Mason [this message]
2026-11-12 22:03   ` [PATCH v9 13/14] xfs: don't enable DAX on zoned inodes sashiko-bot
2026-11-12 23:50   ` Christoph Hellwig
2026-11-12 21:46 ` [PATCH v9 14/14] xfs: test for realtime and reflink under the pNFS locks Chris Mason
2026-11-12 22:02   ` sashiko-bot
2026-11-12 23:51   ` Christoph Hellwig

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20261112214604.67875-14-mason@kernel.org \
    --to=mason@kernel.org \
    --cc=brauner@kernel.org \
    --cc=cem@kernel.org \
    --cc=david@fromorbit.com \
    --cc=djwong@kernel.org \
    --cc=hch@lst.de \
    --cc=jack@suse.cz \
    --cc=josef@toxicpanda.com \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-xfs@vger.kernel.org \
    --cc=tytso@mit.edu \
    --cc=tz2294@columbia.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.