Skip to contentSkip to conversation

ext4: Reject on-disk mount options with missing NUL-terminator

PATCH In review Kees Cook 4 participants 9 messages Feb 6 – Feb 11 ext4 file system

lore
Share & tools
lore

Kees Cook

author
#0 lore raw

PATCH 1/1 ext4: Reject on-disk mount options with missing NUL-terminator

When mounting an ext4 filesystem, the on-disk superblock's s_mount_opts field (which stores default mount options set via tune2fs) is read and parsed. Unlike userspace-provided mount options which are validated by the VFS layer before reaching the filesystem, the on-disk s_mount_opts is read directly from the disk buffer without NUL-termination validation.

The two option paths use the same parser but arrive differently:

Userspace mount options:

    VFS -> ext4_parse_param()

On-disk default options:

    parse_apply_sb_mount_options() -> parse_options() -> ext4_parse_param()

When s_mount_opts lacks NUL-termination, strscpy_pad()'s internal fortified strnlen() detects reading beyond the 64-byte field, triggering an Oops:

strnlen: detected buffer overflow: 65 byte read of buffer size 64

  WARNING: CPU: 0 PID: 179 at lib/string_helpers.c:1035 __fortify_report+0x5a/0x100
  ...
  Call Trace:
   strnlen+0x71/0xa0 lib/string.c:155
   sized_strscpy+0x48/0x2a0 lib/string.c:298
   parse_apply_sb_mount_options+0x94/0x4a0 fs/ext4/super.c:2486
   __ext4_fill_super+0x31d6/0x51b0 fs/ext4/super.c:5306
   ext4_fill_super+0x3972/0xaf70 fs/ext4/super.c:5736
   get_tree_bdev_flags+0x38c/0x620 fs/super.c:1698
   vfs_get_tree+0x8e/0x340 fs/super.c:1758
   fc_mount fs/namespace.c:1199
   do_new_mount fs/namespace.c:3718
   path_mount+0x7b9/0x23a0 fs/namespace.c:4028
   ...

Reject the mount with an error instead. While there is an existing similar check in the ioctl path (which validates userspace input before writing TO disk), this check validates data read FROM disk before parsing.

The painful history here is:

8b67f04ab9de ("ext4: Add mount options in superblock")
s_mount_opts is created and treated as __nonstring: kstrndup is used to make sure all 64 potential characters are available for use (i.e. up to 65 bytes may be allocated).

04a91570ac67 ("ext4: implemet new ioctls to set and get superblock parameters")
Created ext4_tune_sb_params::mount_opts as 64 bytes in size but incorrectly treated it and s_mount_opts as a C strings (it used strscpy_pad() to copy between them).

8ecb790ea8c3 ("ext4: avoid potential buffer over-read in parse_apply_sb_mount_options()")
As a prerequisite to the ioctl treating s_mount_opts as a C string, this attempted to switch to using strscpy_pad() with a 65 byte destination
for the case of an unterminated s_mount_opts. (But strscpy_pad() will fail due to the over-read of s_mount_opts by strnlen().)

3db63d2c2d1d ("ext4: check if mount_opts is NUL-terminated in ext4_ioctl_set_tune_sb()")
As a continuation of trying to solve the 64/65 mismatch, this started enforcing a 63 character limit (i.e. 64 bytes total) to incoming values from userspace to the ioctl API. (But did not check s_mount_opts coming from disk.)

ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")
Notices the loud failures of strscpy_pad() introduced by 8ecb790ea8c3, and attempted to silence them by making the destination 64 and rejecting too-long strings from the on-disk copy of s_mount_opts, but didn't actually solve it at all, since the problem was always the over-read of the source seen by strnlen(). (Note that the report quoted in this commit exactly matches the report today.)

The other option is to go back in time and mark both s_mount_opts and mount_opts as __nonstring and switch to using memcpy_and_pad() to copy them around between userspace and kernel and disk instead of making them C strings.

What do the ext4 regression tests expect for s_mount_opts? Is there a test for a non-terminated s_mount_opts in an image?

Fixes ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")

---
Cc: "Theodore Ts'o" <[email protected]>
Cc: Andreas Dilger <[email protected]>
Cc: Andy Shevchenko <[email protected]>
Cc: <[email protected]>

---
 fs/ext4/super.c | 7 +++++++
 1 file changed, 7 insertions(+)
fs/ext4/super.c +7 -0
index 87205660c5d0..9ad6005615d8 100644
@@ -2485,6 +2485,13 @@ static int parse_apply_sb_mount_options(struct super_block *sb,
 	if (!sbi->s_es->s_mount_opts[0])
 		return 0;
 
+	if (strnlen(sbi->s_es->s_mount_opts, sizeof(sbi->s_es->s_mount_opts)) ==
+	    sizeof(sbi->s_es->s_mount_opts)) {
+		ext4_msg(sb, KERN_ERR,
+			 "Mount options in superblock are not NUL-terminated");
+		return -EINVAL;
+	}
+
 	if (strscpy_pad(s_mount_opts, sbi->s_es->s_mount_opts) < 0)
 		return -E2BIG;
 
Signature · 1 line
2.34.1
#1 lore raw
Quoted 62 lines · Kees Cook

When mounting an ext4 filesystem, the on-disk superblock's s_mount_opts field (which stores default mount options set via tune2fs) is read and parsed. Unlike userspace-provided mount options which are validated by the VFS layer before reaching the filesystem, the on-disk s_mount_opts is read directly from the disk buffer without NUL-termination validation.

The two option paths use the same parser but arrive differently:

Userspace mount options:

    VFS -> ext4_parse_param()

On-disk default options:

    parse_apply_sb_mount_options() -> parse_options() -> ext4_parse_param()

When s_mount_opts lacks NUL-termination, strscpy_pad()'s internal fortified strnlen() detects reading beyond the 64-byte field, triggering an Oops:

strnlen: detected buffer overflow: 65 byte read of buffer size 64

  WARNING: CPU: 0 PID: 179 at lib/string_helpers.c:1035 __fortify_report+0x5a/0x100
  ...
  Call Trace:
   strnlen+0x71/0xa0 lib/string.c:155
   sized_strscpy+0x48/0x2a0 lib/string.c:298
   parse_apply_sb_mount_options+0x94/0x4a0 fs/ext4/super.c:2486
   __ext4_fill_super+0x31d6/0x51b0 fs/ext4/super.c:5306
   ext4_fill_super+0x3972/0xaf70 fs/ext4/super.c:5736
   get_tree_bdev_flags+0x38c/0x620 fs/super.c:1698
   vfs_get_tree+0x8e/0x340 fs/super.c:1758
   fc_mount fs/namespace.c:1199
   do_new_mount fs/namespace.c:3718
   path_mount+0x7b9/0x23a0 fs/namespace.c:4028
   ...

Reject the mount with an error instead. While there is an existing similar check in the ioctl path (which validates userspace input before writing TO disk), this check validates data read FROM disk before parsing.

The painful history here is:

8b67f04ab9de ("ext4: Add mount options in superblock")
s_mount_opts is created and treated as __nonstring: kstrndup is used to make sure all 64 potential characters are available for use (i.e. up to 65 bytes may be allocated).

04a91570ac67 ("ext4: implemet new ioctls to set and get superblock parameters")
Created ext4_tune_sb_params::mount_opts as 64 bytes in size but incorrectly treated it and s_mount_opts as a C strings (it used strscpy_pad() to copy between them).

8ecb790ea8c3 ("ext4: avoid potential buffer over-read in parse_apply_sb_mount_options()")
As a prerequisite to the ioctl treating s_mount_opts as a C string, this attempted to switch to using strscpy_pad() with a 65 byte destination
for the case of an unterminated s_mount_opts. (But strscpy_pad() will fail due to the over-read of s_mount_opts by strnlen().)

3db63d2c2d1d ("ext4: check if mount_opts is NUL-terminated in ext4_ioctl_set_tune_sb()")
As a continuation of trying to solve the 64/65 mismatch, this started enforcing a 63 character limit (i.e. 64 bytes total) to incoming values from userspace to the ioctl API. (But did not check s_mount_opts coming from disk.)

Quoted 7 lines

ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")
Notices the loud failures of strscpy_pad() introduced by 8ecb790ea8c3, and attempted to silence them by making the destination 64 and rejecting too-long strings from the on-disk copy of s_mount_opts, but didn't actually solve it at all, since the problem was always the over-read of the source seen by strnlen(). (Note that the report quoted in this commit exactly matches the report today.)

Looking through the history of tune2fs, I see v1.42 has commit 9a976ac732a7e9 ("tune2fs: Fix mount_opts handling"), which always set byte 63 to a null, which implies that s_mount_opts is supposed to be a 64-byte field that contains a null-terminated string. It also doesn't appear that slack space after the null terminator is set to any particular value.

In e2fsprogs v1.41.13, commit 9345f02671fd39 ("tune2fs, debugfs,
libext2fs: Add support for ext4 default mount options") added the ability to set s_mount_opts, and it rejects any string whose length is greater than or equal to 64 bytes. This validation is still in place.

My guess is that the ext4 code should go back to requiring a null terminated string, and the __nonstring annotations dropped.

I guess the hard part is, were there any filesystems written out with a 64-byte unterminated s_mount_opts? It looks like 04a91570ac67 used strscpy_pad to copy the string to the superblock and that function always zero-terminates the destination, so perhaps we can stick with "the ondisk superblock s_mount_opts is always a null terminated string"?

--D

Quoted 43 lines

The other option is to go back in time and mark both s_mount_opts and mount_opts as __nonstring and switch to using memcpy_and_pad() to copy them around between userspace and kernel and disk instead of making them C strings.

What do the ext4 regression tests expect for s_mount_opts? Is there a test for a non-terminated s_mount_opts in an image?

Fixes ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")

---
Cc: "Theodore Ts'o" <[email protected]>
Cc: Andreas Dilger <[email protected]>
Cc: Andy Shevchenko <[email protected]>
Cc: <[email protected]>

---
 fs/ext4/super.c | 7 +++++++
 1 file changed, 7 insertions(+)
fs/ext4/super.c +7 -0
index 87205660c5d0..9ad6005615d8 100644
@@ -2485,6 +2485,13 @@ static int parse_apply_sb_mount_options(struct super_block *sb,
 	if (!sbi->s_es->s_mount_opts[0])
 		return 0;
 
+	if (strnlen(sbi->s_es->s_mount_opts, sizeof(sbi->s_es->s_mount_opts)) ==
+	    sizeof(sbi->s_es->s_mount_opts)) {
+		ext4_msg(sb, KERN_ERR,
+			 "Mount options in superblock are not NUL-terminated");
+		return -EINVAL;
+	}
+
 	if (strscpy_pad(s_mount_opts, sbi->s_es->s_mount_opts) < 0)
 		return -E2BIG;
 
Signature · 1 line
2.34.1
author
#2 lore raw
Quoted 71 lines · Darrick J. Wong
Quoted 62 lines · Kees Cook

When mounting an ext4 filesystem, the on-disk superblock's s_mount_opts field (which stores default mount options set via tune2fs) is read and parsed. Unlike userspace-provided mount options which are validated by the VFS layer before reaching the filesystem, the on-disk s_mount_opts is read directly from the disk buffer without NUL-termination validation.

The two option paths use the same parser but arrive differently:

Userspace mount options:

    VFS -> ext4_parse_param()

On-disk default options:

    parse_apply_sb_mount_options() -> parse_options() -> ext4_parse_param()

When s_mount_opts lacks NUL-termination, strscpy_pad()'s internal fortified strnlen() detects reading beyond the 64-byte field, triggering an Oops:

strnlen: detected buffer overflow: 65 byte read of buffer size 64

  WARNING: CPU: 0 PID: 179 at lib/string_helpers.c:1035 __fortify_report+0x5a/0x100
  ...
  Call Trace:
   strnlen+0x71/0xa0 lib/string.c:155
   sized_strscpy+0x48/0x2a0 lib/string.c:298
   parse_apply_sb_mount_options+0x94/0x4a0 fs/ext4/super.c:2486
   __ext4_fill_super+0x31d6/0x51b0 fs/ext4/super.c:5306
   ext4_fill_super+0x3972/0xaf70 fs/ext4/super.c:5736
   get_tree_bdev_flags+0x38c/0x620 fs/super.c:1698
   vfs_get_tree+0x8e/0x340 fs/super.c:1758
   fc_mount fs/namespace.c:1199
   do_new_mount fs/namespace.c:3718
   path_mount+0x7b9/0x23a0 fs/namespace.c:4028
   ...

Reject the mount with an error instead. While there is an existing similar check in the ioctl path (which validates userspace input before writing TO disk), this check validates data read FROM disk before parsing.

The painful history here is:

8b67f04ab9de ("ext4: Add mount options in superblock")
s_mount_opts is created and treated as __nonstring: kstrndup is used to make sure all 64 potential characters are available for use (i.e. up to 65 bytes may be allocated).

04a91570ac67 ("ext4: implemet new ioctls to set and get superblock parameters")
Created ext4_tune_sb_params::mount_opts as 64 bytes in size but incorrectly treated it and s_mount_opts as a C strings (it used strscpy_pad() to copy between them).

8ecb790ea8c3 ("ext4: avoid potential buffer over-read in parse_apply_sb_mount_options()")
As a prerequisite to the ioctl treating s_mount_opts as a C string, this attempted to switch to using strscpy_pad() with a 65 byte destination
for the case of an unterminated s_mount_opts. (But strscpy_pad() will fail due to the over-read of s_mount_opts by strnlen().)

3db63d2c2d1d ("ext4: check if mount_opts is NUL-terminated in ext4_ioctl_set_tune_sb()")
As a continuation of trying to solve the 64/65 mismatch, this started enforcing a 63 character limit (i.e. 64 bytes total) to incoming values from userspace to the ioctl API. (But did not check s_mount_opts coming from disk.)

Quoted 7 lines

ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")
Notices the loud failures of strscpy_pad() introduced by 8ecb790ea8c3, and attempted to silence them by making the destination 64 and rejecting too-long strings from the on-disk copy of s_mount_opts, but didn't actually solve it at all, since the problem was always the over-read of the source seen by strnlen(). (Note that the report quoted in this commit exactly matches the report today.)

Quoted 6 lines

Looking through the history of tune2fs, I see v1.42 has commit 9a976ac732a7e9 ("tune2fs: Fix mount_opts handling"), which always set byte 63 to a null, which implies that s_mount_opts is supposed to be a 64-byte field that contains a null-terminated string. It also doesn't appear that slack space after the null terminator is set to any particular value.

Okay, so the userspace side is treating it as a C string. Then ignore my "PATCH ALTERNATIVE" I just sent. ;)

Quoted 5 lines

In e2fsprogs v1.41.13, commit 9345f02671fd39 ("tune2fs, debugfs,
libext2fs: Add support for ext4 default mount options") added the ability to set s_mount_opts, and it rejects any string whose length is greater than or equal to 64 bytes. This validation is still in place.

Great, thanks for tracking that down. That looks to be Dec 2010.

Quoted 2 lines

My guess is that the ext4 code should go back to requiring a null terminated string, and the __nonstring annotations dropped.

It does already; no __nonstring annotations currently exist for s_mount_opts. The trouble was that it was originally treated as effectively __nonstring between 8b67f04ab9de (v2.6.36, Oct 2010) and 04a91570ac67 (v6.18, Nov 2025).

Quoted 5 lines

I guess the hard part is, were there any filesystems written out with a 64-byte unterminated s_mount_opts? It looks like 04a91570ac67 used strscpy_pad to copy the string to the superblock and that function always zero-terminates the destination, so perhaps we can stick with "the ondisk superblock s_mount_opts is always a null terminated string"?

Yeah, if the userspace tools always wrote images with a NUL-terminated s_mount_opts, then it should be perfectly sane to reject mounts that don't follow that. (i.e. this patch.)

Signature · 1 line
Kees Cook
#3 lore raw
Quoted 104 lines · Kees Cook
Quoted 71 lines · Darrick J. Wong
Quoted 62 lines · Kees Cook

When mounting an ext4 filesystem, the on-disk superblock's s_mount_opts field (which stores default mount options set via tune2fs) is read and parsed. Unlike userspace-provided mount options which are validated by the VFS layer before reaching the filesystem, the on-disk s_mount_opts is read directly from the disk buffer without NUL-termination validation.

The two option paths use the same parser but arrive differently:

Userspace mount options:

    VFS -> ext4_parse_param()

On-disk default options:

    parse_apply_sb_mount_options() -> parse_options() -> ext4_parse_param()

When s_mount_opts lacks NUL-termination, strscpy_pad()'s internal fortified strnlen() detects reading beyond the 64-byte field, triggering an Oops:

strnlen: detected buffer overflow: 65 byte read of buffer size 64

  WARNING: CPU: 0 PID: 179 at lib/string_helpers.c:1035 __fortify_report+0x5a/0x100
  ...
  Call Trace:
   strnlen+0x71/0xa0 lib/string.c:155
   sized_strscpy+0x48/0x2a0 lib/string.c:298
   parse_apply_sb_mount_options+0x94/0x4a0 fs/ext4/super.c:2486
   __ext4_fill_super+0x31d6/0x51b0 fs/ext4/super.c:5306
   ext4_fill_super+0x3972/0xaf70 fs/ext4/super.c:5736
   get_tree_bdev_flags+0x38c/0x620 fs/super.c:1698
   vfs_get_tree+0x8e/0x340 fs/super.c:1758
   fc_mount fs/namespace.c:1199
   do_new_mount fs/namespace.c:3718
   path_mount+0x7b9/0x23a0 fs/namespace.c:4028
   ...

Reject the mount with an error instead. While there is an existing similar check in the ioctl path (which validates userspace input before writing TO disk), this check validates data read FROM disk before parsing.

The painful history here is:

8b67f04ab9de ("ext4: Add mount options in superblock")
s_mount_opts is created and treated as __nonstring: kstrndup is used to make sure all 64 potential characters are available for use (i.e. up to 65 bytes may be allocated).

04a91570ac67 ("ext4: implemet new ioctls to set and get superblock parameters")
Created ext4_tune_sb_params::mount_opts as 64 bytes in size but incorrectly treated it and s_mount_opts as a C strings (it used strscpy_pad() to copy between them).

8ecb790ea8c3 ("ext4: avoid potential buffer over-read in parse_apply_sb_mount_options()")
As a prerequisite to the ioctl treating s_mount_opts as a C string, this attempted to switch to using strscpy_pad() with a 65 byte destination
for the case of an unterminated s_mount_opts. (But strscpy_pad() will fail due to the over-read of s_mount_opts by strnlen().)

3db63d2c2d1d ("ext4: check if mount_opts is NUL-terminated in ext4_ioctl_set_tune_sb()")
As a continuation of trying to solve the 64/65 mismatch, this started enforcing a 63 character limit (i.e. 64 bytes total) to incoming values from userspace to the ioctl API. (But did not check s_mount_opts coming from disk.)

Quoted 7 lines

ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")
Notices the loud failures of strscpy_pad() introduced by 8ecb790ea8c3, and attempted to silence them by making the destination 64 and rejecting too-long strings from the on-disk copy of s_mount_opts, but didn't actually solve it at all, since the problem was always the over-read of the source seen by strnlen(). (Note that the report quoted in this commit exactly matches the report today.)

Quoted 6 lines

Looking through the history of tune2fs, I see v1.42 has commit 9a976ac732a7e9 ("tune2fs: Fix mount_opts handling"), which always set byte 63 to a null, which implies that s_mount_opts is supposed to be a 64-byte field that contains a null-terminated string. It also doesn't appear that slack space after the null terminator is set to any particular value.

Okay, so the userspace side is treating it as a C string. Then ignore my "PATCH ALTERNATIVE" I just sent. ;)

Quoted 5 lines

In e2fsprogs v1.41.13, commit 9345f02671fd39 ("tune2fs, debugfs,
libext2fs: Add support for ext4 default mount options") added the ability to set s_mount_opts, and it rejects any string whose length is greater than or equal to 64 bytes. This validation is still in place.

Great, thanks for tracking that down. That looks to be Dec 2010.

Quoted 2 lines

My guess is that the ext4 code should go back to requiring a null terminated string, and the __nonstring annotations dropped.

It does already; no __nonstring annotations currently exist for s_mount_opts. The trouble was that it was originally treated as effectively __nonstring between 8b67f04ab9de (v2.6.36, Oct 2010) and 04a91570ac67 (v6.18, Nov 2025).

Quoted 5 lines

I guess the hard part is, were there any filesystems written out with a 64-byte unterminated s_mount_opts? It looks like 04a91570ac67 used strscpy_pad to copy the string to the superblock and that function always zero-terminates the destination, so perhaps we can stick with "the ondisk superblock s_mount_opts is always a null terminated string"?

Quoted 3 lines

Yeah, if the userspace tools always wrote images with a NUL-terminated s_mount_opts, then it should be perfectly sane to reject mounts that don't follow that. (i.e. this patch.)

I agree with your patch, so here a couple of remarks:

- the questions you asked in the commit messages probably need to be replaced with the summary of what Darrick said

- should the not NUL-terminated string be a fatal error? Maybe we can tell the user that it's truncated and go on with it? Dunno (on the first glance this sounds better from user perspective, but it may lead to a weird results if the cut string will be parsed perfectly and give something dangerous to the data on the volume.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#4 lore raw
In reply to Kees Cook
Quoted 8 lines · Kees Cook

ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")
Notices the loud failures of strscpy_pad() introduced by 8ecb790ea8c3, and attempted to silence them by making the destination 64 and rejecting too-long strings from the on-disk copy of s_mount_opts, but didn't actually solve it at all, since the problem was always the over-read of the source seen by strnlen(). (Note that the report quoted in this commit exactly matches the report today.)

[...]

Quoted 3 lines
Fixes ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")

Hi there,

[ I'd better be Cc'ed as the author of the commit in Fixes ]

The mentioned reports are for v6.18.2 kernel while ee5a977b4e77 ("ext4:
fix string copying in parse_apply_sb_mount_options()") landed in v6.18.3. Back at the time I've tested the patch with different bogus s_mount_opts values and the fortify warnings should have been gone.

I don't think there is an error in ee5a977b4e77 unless these warnings actually appear on the latest kernels with ee5a977b4e77 applied.

fs/ext4/super.c static int parse_apply_sb_mount_options(struct super_block *sb, · line 2493

Quoted 10 lines
fs/ext4/super.c +6 -0
@@ -2485,6 +2485,13 @@ static int parse_apply_sb_mount_options(struct super_block *sb,
 	if (!sbi->s_es->s_mount_opts[0])
 		return 0;
 
+	if (strnlen(sbi->s_es->s_mount_opts, sizeof(sbi->s_es->s_mount_opts)) ==
+	    sizeof(sbi->s_es->s_mount_opts)) {
+		ext4_msg(sb, KERN_ERR,
+			 "Mount options in superblock are not NUL-terminated");
+		return -EINVAL;
+	}

strscpy_pad() returns -E2BIG if the source string was truncated. This happens for the above condition as well - the last byte is truncated and replaced with a NUL-terminator.

The check at 3db63d2c2d1d ("ext4: check if mount_opts is NUL-terminated in ext4_ioctl_set_tune_sb()") was done in that manner as there is currently no way to propagate strscpy_pad() return value up from ext4_sb_setparams(). So the string is independently checked inside ext4_ioctl_set_tune_sb() directly.

As for the 64/65 byte length part, now the rationale of the checks works as Darrick Wong described at the other part of this thread and corresponds to how relevant userspace stuff treats the s_mount_opts field: the buffer is at most 63 payload characters long + NUL-terminator. Jan Kara also shared similar thoughts during the discussion of ee5a977b4e77 [1].

[1]: https://lore.kernel.org/linux-ext4/yq6rbx54jt4btntsh37urd6u63wwcd3lyhovbrm6w7occaveea@riljfkx5jmhi/

Quoted 6 lines
fs/ext4/super.c +1 -0
+
 	if (strscpy_pad(s_mount_opts, sbi->s_es->s_mount_opts) < 0)
 		return -E2BIG;
 
Signature · 1 line
2.34.1
author
#5 lore raw
Quoted 12 lines · Fedor Pchelkin
Quoted 8 lines · Kees Cook

ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")
Notices the loud failures of strscpy_pad() introduced by 8ecb790ea8c3, and attempted to silence them by making the destination 64 and rejecting too-long strings from the on-disk copy of s_mount_opts, but didn't actually solve it at all, since the problem was always the over-read of the source seen by strnlen(). (Note that the report quoted in this commit exactly matches the report today.)

[...]

Quoted 7 lines
Quoted 3 lines
Fixes ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")

Hi there,

[ I'd better be Cc'ed as the author of the commit in Fixes ]

Agreed! Sorry I missed adding you to Cc.

Quoted 4 lines

The mentioned reports are for v6.18.2 kernel while ee5a977b4e77 ("ext4:
fix string copying in parse_apply_sb_mount_options()") landed in v6.18.3. Back at the time I've tested the patch with different bogus s_mount_opts values and the fortify warnings should have been gone.

Ah-ha! Okay, thank you for catching this versioning issue. I had been scratching my head over how it could have been the same warning. This report is effectively a duplicate of the report you fixed with ee5a977b4e77.

Quoted 7 lines

I don't think there is an error in ee5a977b4e77 unless these warnings actually appear on the latest kernels with ee5a977b4e77 applied.

Quoted 4 lines
fs/ext4/super.c +0 -0
@@ -2485,6 +2485,13 @@ static int parse_apply_sb_mount_options(struct super_block *sb,
 	if (!sbi->s_es->s_mount_opts[0])
 		return 0;
 
Quoted 10 lines

fs/ext4/super.c static int parse_apply_sb_mount_options(struct super_block *sb, · line 2493

Quoted 6 lines
fs/ext4/super.c +6 -0
+	if (strnlen(sbi->s_es->s_mount_opts, sizeof(sbi->s_es->s_mount_opts)) ==
+	    sizeof(sbi->s_es->s_mount_opts)) {
+		ext4_msg(sb, KERN_ERR,
+			 "Mount options in superblock are not NUL-terminated");
+		return -EINVAL;
+	}

strscpy_pad() returns -E2BIG if the source string was truncated. This happens for the above condition as well - the last byte is truncated and replaced with a NUL-terminator.

Yeah, I've double-checked this now. The second half of the overflow check in the fortified strnlen eluded by eyes when I went through this originally. Thanks for sanity checking this!

Quoted 7 lines

The check at 3db63d2c2d1d ("ext4: check if mount_opts is NUL-terminated in ext4_ioctl_set_tune_sb()") was done in that manner as there is currently no way to propagate strscpy_pad() return value up from ext4_sb_setparams(). So the string is independently checked inside ext4_ioctl_set_tune_sb() directly.

Quoted 7 lines

As for the 64/65 byte length part, now the rationale of the checks works as Darrick Wong described at the other part of this thread and corresponds to how relevant userspace stuff treats the s_mount_opts field: the buffer is at most 63 payload characters long + NUL-terminator. Jan Kara also shared similar thoughts during the discussion of ee5a977b4e77 [1].

[1]: https://lore.kernel.org/linux-ext4/yq6rbx54jt4btntsh37urd6u63wwcd3lyhovbrm6w7occaveea@riljfkx5jmhi/

Okay, great. I figure I can do two things:

1) rework this patch with adjusted commit log to reflect the notes raised so far, so that we reject mounts that lack a NUL-terminated s_mount_opts (as silent truncation may induce an unintended option string, e.g. "...,journal_path=/dev/sda2" into "...,journal_path=/dev/sda" or something weird like that).

2) Leave everything as-is, live with above corner case since it should be unreachable with userspace tooling as they have always existed.

I'm fine either way! :)

-Kees

Signature · 1 line
Kees Cook
#6 lore raw
Quoted 78 lines · Kees Cook
Quoted 12 lines · Fedor Pchelkin
Quoted 8 lines · Kees Cook

ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")
Notices the loud failures of strscpy_pad() introduced by 8ecb790ea8c3, and attempted to silence them by making the destination 64 and rejecting too-long strings from the on-disk copy of s_mount_opts, but didn't actually solve it at all, since the problem was always the over-read of the source seen by strnlen(). (Note that the report quoted in this commit exactly matches the report today.)

[...]

Quoted 7 lines
Quoted 3 lines
Fixes ee5a977b4e77 ("ext4: fix string copying in parse_apply_sb_mount_options()")

Hi there,

[ I'd better be Cc'ed as the author of the commit in Fixes ]

Agreed! Sorry I missed adding you to Cc.

Quoted 4 lines

The mentioned reports are for v6.18.2 kernel while ee5a977b4e77 ("ext4:
fix string copying in parse_apply_sb_mount_options()") landed in v6.18.3. Back at the time I've tested the patch with different bogus s_mount_opts values and the fortify warnings should have been gone.

Ah-ha! Okay, thank you for catching this versioning issue. I had been scratching my head over how it could have been the same warning. This report is effectively a duplicate of the report you fixed with ee5a977b4e77.

Quoted 7 lines

I don't think there is an error in ee5a977b4e77 unless these warnings actually appear on the latest kernels with ee5a977b4e77 applied.

Quoted 4 lines
fs/ext4/super.c +0 -0
@@ -2485,6 +2485,13 @@ static int parse_apply_sb_mount_options(struct super_block *sb,
 	if (!sbi->s_es->s_mount_opts[0])
 		return 0;
 
Quoted 10 lines

fs/ext4/super.c static int parse_apply_sb_mount_options(struct super_block *sb, · line 2493

Quoted 6 lines
fs/ext4/super.c +6 -0
+	if (strnlen(sbi->s_es->s_mount_opts, sizeof(sbi->s_es->s_mount_opts)) ==
+	    sizeof(sbi->s_es->s_mount_opts)) {
+		ext4_msg(sb, KERN_ERR,
+			 "Mount options in superblock are not NUL-terminated");
+		return -EINVAL;
+	}

strscpy_pad() returns -E2BIG if the source string was truncated. This happens for the above condition as well - the last byte is truncated and replaced with a NUL-terminator.

Yeah, I've double-checked this now. The second half of the overflow check in the fortified strnlen eluded by eyes when I went through this originally. Thanks for sanity checking this!

Quoted 7 lines

The check at 3db63d2c2d1d ("ext4: check if mount_opts is NUL-terminated in ext4_ioctl_set_tune_sb()") was done in that manner as there is currently no way to propagate strscpy_pad() return value up from ext4_sb_setparams(). So the string is independently checked inside ext4_ioctl_set_tune_sb() directly.

Quoted 7 lines

As for the 64/65 byte length part, now the rationale of the checks works as Darrick Wong described at the other part of this thread and corresponds to how relevant userspace stuff treats the s_mount_opts field: the buffer is at most 63 payload characters long + NUL-terminator. Jan Kara also shared similar thoughts during the discussion of ee5a977b4e77 [1].

[1]: https://lore.kernel.org/linux-ext4/yq6rbx54jt4btntsh37urd6u63wwcd3lyhovbrm6w7occaveea@riljfkx5jmhi/

Okay, great. I figure I can do two things:

1) rework this patch with adjusted commit log to reflect the notes raised so far, so that we reject mounts that lack a NUL-terminated s_mount_opts (as silent truncation may induce an unintended option string, e.g. "...,journal_path=/dev/sda2" into "...,journal_path=/dev/sda" or something weird like that).

Quoted 4 lines

2) Leave everything as-is, live with above corner case since it should be unreachable with userspace tooling as they have always existed.

I'm fine either way! :)

I'd pick #1, unless someone knows of a userspace program that could have set a 64-byte s_mount_ops string with no null terminator. I didn't find any, but there are many implementations of ext4 out there. :/

(and yes, it's better to reject an unterminated s_mount_opts than accidentally point the kernel at the wrong block device)

--D

Quoted 5 lines

-Kees

Signature · 1 line
Kees Cook
#7 lore raw
Quoted 12 lines · Darrick J. Wong
Quoted 11 lines · Kees Cook

Okay, great. I figure I can do two things:

1) rework this patch with adjusted commit log to reflect the notes raised so far, so that we reject mounts that lack a NUL-terminated s_mount_opts (as silent truncation may induce an unintended option string, e.g. "...,journal_path=/dev/sda2" into "...,journal_path=/dev/sda" or something weird like that).

2) Leave everything as-is, live with above corner case since it should be unreachable with userspace tooling as they have always existed.

Quoted 8 lines
Quoted 1 line

I'm fine either way! :)

I'd pick #1, unless someone knows of a userspace program that could have set a 64-byte s_mount_ops string with no null terminator. I didn't find any, but there are many implementations of ext4 out there. :/

(and yes, it's better to reject an unterminated s_mount_opts than accidentally point the kernel at the wrong block device)

If I understand the issue correctly, it's already being rejected with the existing check:

	if (strscpy_pad(s_mount_opts, sbi->s_es->s_mount_opts) < 0)
		return -E2BIG;

If the source string is truncated at least by one symbol (which is the case for unterminated string), strscpy_pad() returns -E2BIG and the mount fails.

author
#8 lore raw
Quoted 9 lines · Fedor Pchelkin

If I understand the issue correctly, it's already being rejected with the existing check:

	if (strscpy_pad(s_mount_opts, sbi->s_es->s_mount_opts) < 0)
		return -E2BIG;

If the source string is truncated at least by one symbol (which is the case for unterminated string), strscpy_pad() returns -E2BIG and the mount fails.

Agreed. Sorry for the noise! Things appear to be fine as-is. I had missed the version tested in the original report.

Signature · 1 line
Kees Cook