From: "Darrick J. Wong" <djwong@kernel.org>
To: tytso@mit.edu
Cc: joannelkoong@gmail.com, miklos@szeredi.hu, John@groves.net,
linux-fsdevel@vger.kernel.org, bernd@bsbernd.com,
linux-ext4@vger.kernel.org, neal@gompa.dev
Subject: [PATCH 18/22] fuse2fs: don't allow hardlinks for now
Date: Thu, 17 Jul 2025 16:44:04 -0700 [thread overview]
Message-ID: <175279461359.715479.1220608915548013053.stgit@frogsfrogsfrogs> (raw)
In-Reply-To: <175279460935.715479.15460687085573767955.stgit@frogsfrogsfrogs>
From: Darrick J. Wong <djwong@kernel.org>
XXX see the comment for why we have to do this bellicosely stupid thing.
Signed-off-by: "Darrick J. Wong" <djwong@kernel.org>
---
misc/fuse2fs.c | 30 +++++++++++++++++++++++++++++-
1 file changed, 29 insertions(+), 1 deletion(-)
diff --git a/misc/fuse2fs.c b/misc/fuse2fs.c
index 82b59c1ac89774..e281b5fc589d82 100644
--- a/misc/fuse2fs.c
+++ b/misc/fuse2fs.c
@@ -261,6 +261,7 @@ struct fuse2fs {
uint8_t dirsync;
uint8_t unmount_in_destroy;
uint8_t noblkdev;
+ uint8_t can_hardlink;
enum fuse2fs_opstate opstate;
int blocklog;
@@ -1382,9 +1383,31 @@ static void *op_init(struct fuse_conn_info *conn
/*
* If we're mounting in iomap mode, we need to unmount in op_destroy
* so that the block device will be released before umount(2) returns.
+ *
+ * XXX: It turns out that fuse2fs creates internal node ids that have
+ * nothing to do with the ext2_ino_t that we give it. These internal
+ * node ids are what actually gets igetted in the kernel, which means
+ * that there can be multiple fuse_inode objects for the same fuse2fs
+ * inode.
+ *
+ * What this means, horrifyingly, is that on a fuse filesystem that
+ * supports hard links, the in-kernel i_rwsem does not protect against
+ * concurrent writes between files that point to the same inode. That
+ * in turn means that the file mode and size can get desynchronized
+ * between the multiple fuse_inode objects. This also means that we
+ * cannot cache iomaps in the kernel AT ALL because the caches will
+ * get out of sync, leading to WARN_ONs from the iomap zeroing code and
+ * probably data corruption after that.
+ *
+ * So for now we just disable hardlinking on iomap to see if the weird
+ * fstests failures (particularly g/476) go away. Long term it means
+ * we probably have to find a way around this, like porting fuse2fs
+ * to be a low level fuse driver.
*/
- if (fuse2fs_iomap_enabled(ff))
+ if (fuse2fs_iomap_enabled(ff)) {
ff->unmount_in_destroy = 1;
+ ff->can_hardlink = 0;
+ }
/* Clear the valid flag so that an unclean shutdown forces a fsck */
if (ff->opstate == F2OP_WRITABLE) {
@@ -2751,6 +2774,10 @@ static int op_link(const char *src, const char *dest)
int ret = 0;
FUSE2FS_CHECK_CONTEXT(ff);
+
+ if (!ff->can_hardlink)
+ return -ENOSYS;
+
dbg_printf(ff, "%s: src=%s dest=%s\n", __func__, src, dest);
temp_path = strdup(dest);
if (!temp_path) {
@@ -6380,6 +6407,7 @@ int main(int argc, char *argv[])
.iomap_state = IOMAP_UNKNOWN,
.iomap_dev = FUSE_IOMAP_DEV_NULL,
#endif
+ .can_hardlink = 1,
};
errcode_t err;
FILE *orig_stderr = stderr;
next prev parent reply other threads:[~2025-07-17 23:44 UTC|newest]
Thread overview: 174+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-07-17 23:10 [RFC v3] fuse: use fs-iomap for better performance so we can containerize ext4 Darrick J. Wong
2025-07-17 23:23 ` [PATCHSET RFC v3 1/4] fuse: fixes and cleanups ahead of iomap support Darrick J. Wong
2025-07-17 23:26 ` [PATCH 1/7] fuse: fix livelock in synchronous file put from fuseblk workers Darrick J. Wong
2025-07-17 23:26 ` [PATCH 2/7] fuse: flush pending fuse events before aborting the connection Darrick J. Wong
2025-07-18 16:37 ` Bernd Schubert
2025-07-18 17:50 ` Joanne Koong
2025-07-18 17:57 ` Bernd Schubert
2025-07-18 18:38 ` Darrick J. Wong
2025-07-18 18:07 ` Bernd Schubert
2025-07-18 18:13 ` Bernd Schubert
2025-07-18 19:34 ` Darrick J. Wong
2025-07-18 21:03 ` Bernd Schubert
2025-07-18 22:23 ` Joanne Koong
2025-07-19 0:32 ` Darrick J. Wong
2025-07-21 20:32 ` Joanne Koong
2025-07-23 17:34 ` Darrick J. Wong
2025-07-23 21:02 ` Joanne Koong
2025-07-23 21:11 ` Joanne Koong
2025-07-24 22:28 ` Darrick J. Wong
2025-07-22 12:30 ` Jeff Layton
2025-07-22 12:38 ` Jeff Layton
2025-07-23 15:37 ` Darrick J. Wong
2025-07-23 16:24 ` Jeff Layton
2025-07-31 9:45 ` Christian Brauner
2025-07-31 17:52 ` Darrick J. Wong
2025-07-19 7:18 ` Amir Goldstein
2025-07-21 20:05 ` Joanne Koong
2025-07-23 17:06 ` Darrick J. Wong
2025-07-23 20:27 ` Joanne Koong
2025-07-24 22:34 ` Darrick J. Wong
2025-07-17 23:27 ` [PATCH 3/7] fuse: capture the unique id of fuse commands being sent Darrick J. Wong
2025-07-18 17:10 ` Bernd Schubert
2025-07-18 18:13 ` Darrick J. Wong
2025-07-22 22:20 ` Bernd Schubert
2025-07-17 23:27 ` [PATCH 4/7] fuse: implement file attributes mask for statx Darrick J. Wong
2025-08-18 15:11 ` Miklos Szeredi
2025-08-18 20:01 ` Darrick J. Wong
2025-08-18 20:04 ` Darrick J. Wong
2025-08-19 15:01 ` Miklos Szeredi
2025-08-19 22:51 ` Darrick J. Wong
2025-08-20 9:16 ` Miklos Szeredi
2025-08-20 9:40 ` Miklos Szeredi
2025-08-20 15:16 ` Darrick J. Wong
2025-08-20 15:31 ` Miklos Szeredi
2025-08-20 15:09 ` Darrick J. Wong
2025-08-20 15:23 ` Miklos Szeredi
2025-08-20 15:29 ` Darrick J. Wong
2025-07-17 23:27 ` [PATCH 5/7] iomap: exit early when iomap_iter is called with zero length Darrick J. Wong
2025-07-17 23:27 ` [PATCH 6/7] iomap: trace iomap_zero_iter zeroing activities Darrick J. Wong
2025-07-17 23:28 ` [PATCH 7/7] iomap: error out on file IO when there is no inline_data buffer Darrick J. Wong
2025-07-17 23:24 ` [PATCHSET RFC v3 2/4] fuse: allow servers to use iomap for better file IO performance Darrick J. Wong
2025-07-17 23:28 ` [PATCH 01/13] fuse: implement the basic iomap mechanisms Darrick J. Wong
2025-07-17 23:28 ` [PATCH 02/13] fuse: add an ioctl to add new iomap devices Darrick J. Wong
2025-07-17 23:28 ` [PATCH 03/13] fuse: flush events and send FUSE_SYNCFS and FUSE_DESTROY on unmount Darrick J. Wong
2025-07-17 23:29 ` [PATCH 04/13] fuse: implement basic iomap reporting such as FIEMAP and SEEK_{DATA,HOLE} Darrick J. Wong
2025-07-17 23:29 ` [PATCH 05/13] fuse: implement direct IO with iomap Darrick J. Wong
2025-07-17 23:29 ` [PATCH 06/13] fuse: implement buffered " Darrick J. Wong
2025-07-18 15:10 ` Amir Goldstein
2025-07-18 18:01 ` Darrick J. Wong
2025-07-18 18:39 ` Bernd Schubert
2025-07-18 18:46 ` Darrick J. Wong
2025-07-18 19:45 ` Amir Goldstein
2025-07-18 20:20 ` Darrick J. Wong
2025-07-17 23:29 ` [PATCH 07/13] fuse: enable caching of timestamps Darrick J. Wong
2025-07-17 23:30 ` [PATCH 08/13] fuse: implement large folios for iomap pagecache files Darrick J. Wong
2025-07-17 23:30 ` [PATCH 09/13] fuse: use an unrestricted backing device with iomap pagecache io Darrick J. Wong
2025-07-17 23:30 ` [PATCH 10/13] fuse: advertise support for iomap Darrick J. Wong
2025-07-17 23:31 ` [PATCH 11/13] fuse: query filesystem geometry when using iomap Darrick J. Wong
2025-07-17 23:31 ` [PATCH 12/13] fuse: implement fadvise for iomap files Darrick J. Wong
2025-07-17 23:31 ` [PATCH 13/13] fuse: implement inline data file IO via iomap Darrick J. Wong
2025-07-17 23:24 ` [PATCHSET RFC v3 3/4] fuse: cache iomap mappings for even better file IO performance Darrick J. Wong
2025-07-17 23:31 ` [PATCH 1/4] fuse: cache iomaps Darrick J. Wong
2025-07-17 23:32 ` [PATCH 2/4] fuse: use the iomap cache for iomap_begin Darrick J. Wong
2025-07-17 23:32 ` [PATCH 3/4] fuse: invalidate iomap cache after file updates Darrick J. Wong
2025-07-17 23:32 ` [PATCH 4/4] fuse: enable iomap cache management Darrick J. Wong
2025-07-17 23:24 ` [PATCHSET RFC v3 4/4] fuse: handle timestamps and ACLs correctly when iomap is enabled Darrick J. Wong
2025-07-17 23:32 ` [PATCH 1/7] fuse: force a ctime update after a fileattr_set call when in iomap mode Darrick J. Wong
2025-07-17 23:33 ` [PATCH 2/7] fuse: synchronize inode->i_flags after fileattr_[gs]et Darrick J. Wong
2025-07-17 23:33 ` [PATCH 3/7] fuse: cache atime when in iomap mode Darrick J. Wong
2025-07-17 23:33 ` [PATCH 4/7] fuse: update file mode when updating acls Darrick J. Wong
2025-07-17 23:33 ` [PATCH 5/7] fuse: propagate default and file acls on creation Darrick J. Wong
2025-07-17 23:34 ` [PATCH 6/7] fuse: let the kernel handle KILL_SUID/KILL_SGID for iomap filesystems Darrick J. Wong
2025-07-17 23:34 ` [PATCH 7/7] fuse: update ctime when updating acls on an iomap inode Darrick J. Wong
2025-07-17 23:25 ` [PATCHSET RFC v3 1/3] libfuse: allow servers to use iomap for better file IO performance Darrick J. Wong
2025-07-17 23:34 ` [PATCH 01/14] libfuse: add kernel gates for FUSE_IOMAP and bump libfuse api version Darrick J. Wong
2025-07-17 23:34 ` [PATCH 02/14] libfuse: add fuse commands for iomap_begin and end Darrick J. Wong
2025-07-17 23:35 ` [PATCH 03/14] libfuse: add upper level iomap commands Darrick J. Wong
2025-07-17 23:35 ` [PATCH 04/14] libfuse: add a notification to add a new device to iomap Darrick J. Wong
2025-07-17 23:35 ` [PATCH 05/14] libfuse: add iomap ioend low level handler Darrick J. Wong
2025-07-17 23:35 ` [PATCH 06/14] libfuse: add upper level iomap ioend commands Darrick J. Wong
2025-07-17 23:36 ` [PATCH 07/14] libfuse: add a reply function to send FUSE_ATTR_* to the kernel Darrick J. Wong
2025-07-18 14:10 ` Amir Goldstein
2025-07-18 15:48 ` Darrick J. Wong
2025-07-19 7:34 ` Amir Goldstein
2025-07-17 23:36 ` [PATCH 08/14] libfuse: connect high level fuse library to fuse_reply_attr_iflags Darrick J. Wong
2025-07-18 14:27 ` Amir Goldstein
2025-07-18 15:55 ` Darrick J. Wong
2025-07-21 18:51 ` Bernd Schubert
2025-07-23 17:50 ` Darrick J. Wong
2025-07-24 19:56 ` Amir Goldstein
2025-07-29 5:35 ` Darrick J. Wong
2025-07-29 7:50 ` Amir Goldstein
2025-07-29 14:22 ` Darrick J. Wong
2025-07-17 23:36 ` [PATCH 09/14] libfuse: add FUSE_IOMAP_DIRECTIO Darrick J. Wong
2025-07-17 23:37 ` [PATCH 10/14] libfuse: add FUSE_IOMAP_FILEIO Darrick J. Wong
2025-07-17 23:37 ` [PATCH 11/14] libfuse: allow discovery of the kernel's iomap capabilities Darrick J. Wong
2025-07-17 23:37 ` [PATCH 12/14] libfuse: add lower level iomap_config implementation Darrick J. Wong
2025-07-17 23:37 ` [PATCH 13/14] libfuse: add upper " Darrick J. Wong
2025-07-17 23:38 ` [PATCH 14/14] libfuse: add strictatime/lazytime mount options Darrick J. Wong
2025-07-17 23:25 ` [PATCHSET RFC v3 2/3] libfuse: cache iomap mappings for even better file IO performance Darrick J. Wong
2025-07-17 23:38 ` [PATCH 1/1] libfuse: enable iomap cache management Darrick J. Wong
2025-07-18 16:16 ` Bernd Schubert
2025-07-18 18:22 ` Darrick J. Wong
2025-07-18 18:35 ` Bernd Schubert
2025-07-18 18:40 ` Darrick J. Wong
2025-07-18 18:51 ` Bernd Schubert
2025-07-17 23:25 ` [PATCHSET RFC v3 3/3] libfuse: implement statx and syncfs Darrick J. Wong
2025-07-17 23:38 ` [PATCH 1/4] libfuse: wire up FUSE_SYNCFS to the low level library Darrick J. Wong
2025-07-17 23:38 ` [PATCH 2/4] libfuse: add syncfs support to the upper library Darrick J. Wong
2025-07-17 23:39 ` [PATCH 3/4] libfuse: add statx support to the lower level library Darrick J. Wong
2025-07-18 13:28 ` Amir Goldstein
2025-07-18 15:58 ` Darrick J. Wong
2025-07-18 16:27 ` Darrick J. Wong
2025-07-18 16:54 ` Bernd Schubert
2025-07-18 18:42 ` Darrick J. Wong
2025-07-17 23:39 ` [PATCH 4/4] libfuse: add upper level statx hooks Darrick J. Wong
2025-07-17 23:25 ` [PATCHSET RFC v3 1/3] fuse2fs: use fuse iomap data paths for better file I/O performance Darrick J. Wong
2025-07-17 23:39 ` [PATCH 01/22] fuse2fs: implement bare minimum iomap for file mapping reporting Darrick J. Wong
2025-07-17 23:39 ` [PATCH 02/22] fuse2fs: add iomap= mount option Darrick J. Wong
2025-07-17 23:40 ` [PATCH 03/22] fuse2fs: implement iomap configuration Darrick J. Wong
2025-07-17 23:40 ` [PATCH 04/22] fuse2fs: register block devices for use with iomap Darrick J. Wong
2025-07-17 23:40 ` [PATCH 05/22] fuse2fs: always use directio disk reads with fuse2fs Darrick J. Wong
2025-07-17 23:40 ` [PATCH 06/22] fuse2fs: implement directio file reads Darrick J. Wong
2025-07-17 23:41 ` [PATCH 07/22] fuse2fs: use tagged block IO for zeroing sub-block regions Darrick J. Wong
2025-07-17 23:41 ` [PATCH 08/22] fuse2fs: only flush the cache for the file under directio read Darrick J. Wong
2025-07-17 23:41 ` [PATCH 09/22] fuse2fs: add extent dump function for debugging Darrick J. Wong
2025-07-17 23:41 ` [PATCH 10/22] fuse2fs: implement direct write support Darrick J. Wong
2025-07-17 23:42 ` [PATCH 11/22] fuse2fs: turn on iomap for pagecache IO Darrick J. Wong
2025-07-17 23:42 ` [PATCH 12/22] fuse2fs: improve tracing for fallocate Darrick J. Wong
2025-07-17 23:42 ` [PATCH 13/22] fuse2fs: don't zero bytes in punch hole Darrick J. Wong
2025-07-17 23:43 ` [PATCH 14/22] fuse2fs: don't do file data block IO when iomap is enabled Darrick J. Wong
2025-07-17 23:43 ` [PATCH 15/22] fuse2fs: disable most io channel flush/invalidate in iomap pagecache mode Darrick J. Wong
2025-07-17 23:43 ` [PATCH 16/22] fuse2fs: re-enable the block device pagecache for metadata IO Darrick J. Wong
2025-07-17 23:43 ` [PATCH 17/22] fuse2fs: avoid fuseblk mode if fuse-iomap support is likely Darrick J. Wong
2025-07-17 23:44 ` Darrick J. Wong [this message]
2025-07-17 23:44 ` [PATCH 19/22] fuse2fs: enable file IO to inline data files Darrick J. Wong
2025-07-17 23:44 ` [PATCH 20/22] fuse2fs: set iomap-related inode flags Darrick J. Wong
2025-07-17 23:44 ` [PATCH 21/22] fuse2fs: add strictatime/lazytime mount options Darrick J. Wong
2025-07-17 23:45 ` [PATCH 22/22] fuse2fs: configure block device block size Darrick J. Wong
2025-07-17 23:26 ` [PATCHSET RFC v3 2/3] fuse2fs: use fuse iomap data paths for better file I/O performance Darrick J. Wong
2025-07-17 23:45 ` [PATCH 1/1] fuse2fs: enable caching of iomaps Darrick J. Wong
2025-07-17 23:26 ` [PATCHSET RFC v3 3/3] fuse2fs: handle timestamps and ACLs correctly when iomap is enabled Darrick J. Wong
2025-07-17 23:45 ` [PATCH 01/10] fuse2fs: allow O_APPEND and O_TRUNC opens Darrick J. Wong
2025-07-17 23:45 ` [PATCH 02/10] fuse2fs: skip permission checking on utimens when iomap is enabled Darrick J. Wong
2025-07-17 23:46 ` [PATCH 03/10] fuse2fs: let the kernel tell us about acl/mode updates Darrick J. Wong
2025-07-17 23:46 ` [PATCH 04/10] fuse2fs: better debugging for file mode updates Darrick J. Wong
2025-07-17 23:46 ` [PATCH 05/10] fuse2fs: debug timestamp updates Darrick J. Wong
2025-07-17 23:46 ` [PATCH 06/10] fuse2fs: use coarse timestamps for iomap mode Darrick J. Wong
2025-07-17 23:47 ` [PATCH 07/10] fuse2fs: add tracing for retrieving timestamps Darrick J. Wong
2025-07-17 23:47 ` [PATCH 08/10] fuse2fs: enable syncfs Darrick J. Wong
2025-07-17 23:47 ` [PATCH 09/10] fuse2fs: skip the gdt write in op_destroy if syncfs is working Darrick J. Wong
2025-07-17 23:47 ` [PATCH 10/10] fuse2fs: implement statx Darrick J. Wong
2025-07-18 8:54 ` [RFC v3] fuse: use fs-iomap for better performance so we can containerize ext4 Christian Brauner
2025-07-18 11:55 ` Amir Goldstein
2025-07-18 19:31 ` Darrick J. Wong
2025-07-18 19:56 ` Amir Goldstein
2025-07-18 20:21 ` Darrick J. Wong
2025-07-23 13:05 ` Christian Brauner
2025-07-23 18:04 ` Darrick J. Wong
2025-07-31 10:13 ` Christian Brauner
2025-07-31 17:22 ` Darrick J. Wong
2025-08-04 10:12 ` Christian Brauner
2025-08-12 20:20 ` Darrick J. Wong
2025-08-15 14:20 ` Christian Brauner
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=175279461359.715479.1220608915548013053.stgit@frogsfrogsfrogs \
--to=djwong@kernel.org \
--cc=John@groves.net \
--cc=bernd@bsbernd.com \
--cc=joannelkoong@gmail.com \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=miklos@szeredi.hu \
--cc=neal@gompa.dev \
--cc=tytso@mit.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox