CVE-2025-39723

Source
https://nvd.nist.gov/vuln/detail/CVE-2025-39723
Import Source
https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2025-39723.json
JSON Data
https://api.osv.dev/v1/vulns/CVE-2025-39723
Downstream
Published
2025-09-05T17:21:31Z
Modified
2025-10-16T04:30:36.911790Z
Summary
netfs: Fix unbuffered write error handling
Details

In the Linux kernel, the following vulnerability has been resolved:

netfs: Fix unbuffered write error handling

If all the subrequests in an unbuffered write stream fail, the subrequest collector doesn't update the stream->transferred value and it retains its initial LONGMAX value. Unfortunately, if all active streams fail, then we take the smallest value of { LONGMAX, LONGMAX, ... } as the value to set in wreq->transferred - which is then returned from ->writeiter().

LONG_MAX was chosen as the initial value so that all the streams can be quickly assessed by taking the smallest value of all stream->transferred - but this only works if we've set any of them.

Fix this by adding a flag to indicate whether the value in stream->transferred is valid and checking that when we integrate the values. stream->transferred can then be initialised to zero.

This was found by running the generic/750 xfstest against cifs with cache=none. It splices data to the target file. Once (if) it has used up all the available scratch space, the writes start failing with ENOSPC. This causes ->writeiter() to fail. However, it was returning wreq->transferred, i.e. LONGMAX, rather than an error (because it thought the amount transferred was non-zero) and iterfilesplice_write() would then try to clean up that amount of pipe bufferage - leading to an oops when it overran. The kernel log showed:

CIFS: VFS: Send error in write = -28

followed by:

BUG: kernel NULL pointer dereference, address: 0000000000000008

with:

RIP: 0010:iter_file_splice_write+0x3a4/0x520
do_splice+0x197/0x4e0

or:

RIP: 0010:pipe_buf_release (include/linux/pipe_fs_i.h:282)
iter_file_splice_write (fs/splice.c:755)

Also put a warning check into splice to announce if ->write_iter() returned that it had written more than it was asked to.

References

Affected packages

Git / git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git

Affected ranges

Type
GIT
Repo
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
Events
Introduced
288ace2f57c9d06dd2e42bd80d03747d879a4068
Fixed
f08c80af3c9a9849cd178b4843b7c01d103506a1
Type
GIT
Repo
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
Events
Introduced
288ace2f57c9d06dd2e42bd80d03747d879a4068
Fixed
387164a2b97e1f5404c6d0049a7409bac7d2bc5b
Type
GIT
Repo
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
Events
Introduced
288ace2f57c9d06dd2e42bd80d03747d879a4068
Fixed
a3de58b12ce074ec05b8741fa28d62ccb1070468

Affected versions

v6.*

v6.10
v6.10-rc1
v6.10-rc2
v6.10-rc3
v6.10-rc4
v6.10-rc5
v6.10-rc6
v6.10-rc7
v6.11
v6.11-rc1
v6.11-rc2
v6.11-rc3
v6.11-rc4
v6.11-rc5
v6.11-rc6
v6.11-rc7
v6.12
v6.12-rc1
v6.12-rc2
v6.12-rc3
v6.12-rc4
v6.12-rc5
v6.12-rc6
v6.12-rc7
v6.12.1
v6.12.10
v6.12.11
v6.12.12
v6.12.13
v6.12.14
v6.12.15
v6.12.16
v6.12.17
v6.12.18
v6.12.19
v6.12.2
v6.12.20
v6.12.21
v6.12.22
v6.12.23
v6.12.24
v6.12.25
v6.12.26
v6.12.27
v6.12.28
v6.12.29
v6.12.3
v6.12.30
v6.12.31
v6.12.32
v6.12.33
v6.12.34
v6.12.35
v6.12.36
v6.12.37
v6.12.38
v6.12.39
v6.12.4
v6.12.40
v6.12.41
v6.12.42
v6.12.43
v6.12.5
v6.12.6
v6.12.7
v6.12.8
v6.12.9
v6.13
v6.13-rc1
v6.13-rc2
v6.13-rc3
v6.13-rc4
v6.13-rc5
v6.13-rc6
v6.13-rc7
v6.14
v6.14-rc1
v6.14-rc2
v6.14-rc3
v6.14-rc4
v6.14-rc5
v6.14-rc6
v6.14-rc7
v6.15
v6.15-rc1
v6.15-rc2
v6.15-rc3
v6.15-rc4
v6.15-rc5
v6.15-rc6
v6.15-rc7
v6.16
v6.16-rc1
v6.16-rc2
v6.16-rc3
v6.16-rc4
v6.16-rc5
v6.16-rc6
v6.16-rc7
v6.16.1
v6.16.2
v6.16.3
v6.17-rc1
v6.9
v6.9-rc7

Linux / Kernel

Package

Name
Kernel

Affected ranges

Type
ECOSYSTEM
Events
Introduced
6.10.0
Fixed
6.12.44
Type
ECOSYSTEM
Events
Introduced
6.13.0
Fixed
6.16.4