Re: [LTP] [PATCH] ioctl_ficlone03: fix capabilities on immutable files

Re: [LTP] [PATCH] ioctl_ficlone03: fix capabilities on immutable files

[Petr Vorel]

Hi all,

[ Cc Kent and other filesystem folks to be sure we don't hide a bug ]

> From: Andrea Cervesato <andrea.cervesato@suse.com>

> Make sure that capabilities requirements are satisfied when accessing
> immutable files. On OpenSUSE Tumbleweed 32bit bcachefs fails with the
> following error due to missing capabilities:

> tst_test.c:1833: TINFO: === Testing on bcachefs ===
> ..
> ioctl_ficlone03.c:74: TBROK: ioctl .. failed: ENOTTY (25)
> ioctl_ficlone03.c:89: TWARN: ioctl ..  failed: ENOTTY (25)
> ioctl_ficlone03.c:91: TWARN: ioctl ..  failed: ENOTTY (25)
> ioctl_ficlone03.c:98: TWARN: close(-1) failed: EBADF (9)

> Introduce CAP_LINUX_IMMUTABLE capability to make sure that test won't
> fail on other systems as well.

> Signed-off-by: Andrea Cervesato <andrea.cervesato@suse.com>
> ---
>  testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)

> diff --git a/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c b/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c
> index 6a9d270d9fb56f3a263f0aed976f62c4576e6a5f..6716423d9c96d9fc1d433f28e0ae1511b912ae2c 100644
> --- a/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c
> +++ b/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c
> @@ -122,5 +122,9 @@ static struct tst_test test = {
>  	.bufs = (struct tst_buffers []) {
>  		{&clone_range, .size = sizeof(struct file_clone_range)},
>  		{},
> -	}
> +	},
> +	.caps = (struct tst_cap []) {
> +		TST_CAP(TST_CAP_REQ, CAP_LINUX_IMMUTABLE),
> +		{}
> +	},

Reviewed-by: Petr Vorel <pvorel@suse.cz>

LGTM, obviously it is CAP_LINUX_IMMUTABLE related.

This looks like fs/bcachefs/fs-ioctl.c

static int bch2_inode_flags_set(struct btree_trans *trans,
				struct bch_inode_info *inode,
				struct bch_inode_unpacked *bi,
				void *p)
{
	...
	if (((newflags ^ oldflags) & (BCH_INODE_append|BCH_INODE_immutable)) &&
	    !capable(CAP_LINUX_IMMUTABLE))
		return -EPERM;


We don't test on vfat and exfat. If I try to do it fail the same way (bcachefs,
fat, exfat and ocfs2 use custom handler for FAT_IOCTL_SET_ATTRIBUTES).

I wonder why it does not fail for generic VFS fs/ioctl.c used by Btrfs and XFS:

/*
 * Generic function to check FS_IOC_FSSETXATTR/FS_IOC_SETFLAGS values and reject
 * any invalid configurations.
 *
 * Note: must be called with inode lock held.
 */
static int fileattr_set_prepare(struct inode *inode,
			      const struct fileattr *old_ma,
			      struct fileattr *fa)
{
	int err;

	/*
	 * The IMMUTABLE and APPEND_ONLY flags can only be changed by
	 * the relevant capability.
	 */
	if ((fa->flags ^ old_ma->flags) & (FS_APPEND_FL | FS_IMMUTABLE_FL) &&
	    !capable(CAP_LINUX_IMMUTABLE))
		return -EPERM;


Kind regards,
Petr

>  };

> ---
> base-commit: 753bd13472d4be44eb70ff183b007fe9c5fffa07
> change-id: 20250326-fix_ioctl_ficlone03_32bit_bcachefs-2ec15bd6c0c6

> Best regards,

Re: [LTP] [PATCH] ioctl_ficlone03: fix capabilities on immutable files

[Kent Overstreet]

On Wed, Mar 26, 2025 at 02:47:49PM +0100, Petr Vorel wrote:
> Hi all,
> 
> [ Cc Kent and other filesystem folks to be sure we don't hide a bug ]

I'm missing context here, and the original thread doesn't seem to be on
lore?

> 
> > From: Andrea Cervesato <andrea.cervesato@suse.com>
> 
> > Make sure that capabilities requirements are satisfied when accessing
> > immutable files. On OpenSUSE Tumbleweed 32bit bcachefs fails with the
> > following error due to missing capabilities:
> 
> > tst_test.c:1833: TINFO: === Testing on bcachefs ===
> > ..
> > ioctl_ficlone03.c:74: TBROK: ioctl .. failed: ENOTTY (25)
> > ioctl_ficlone03.c:89: TWARN: ioctl ..  failed: ENOTTY (25)
> > ioctl_ficlone03.c:91: TWARN: ioctl ..  failed: ENOTTY (25)
> > ioctl_ficlone03.c:98: TWARN: close(-1) failed: EBADF (9)
> 
> > Introduce CAP_LINUX_IMMUTABLE capability to make sure that test won't
> > fail on other systems as well.
> 
> > Signed-off-by: Andrea Cervesato <andrea.cervesato@suse.com>
> > ---
> >  testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c | 6 +++++-
> >  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> > diff --git a/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c b/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c
> > index 6a9d270d9fb56f3a263f0aed976f62c4576e6a5f..6716423d9c96d9fc1d433f28e0ae1511b912ae2c 100644
> > --- a/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c
> > +++ b/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c
> > @@ -122,5 +122,9 @@ static struct tst_test test = {
> >  	.bufs = (struct tst_buffers []) {
> >  		{&clone_range, .size = sizeof(struct file_clone_range)},
> >  		{},
> > -	}
> > +	},
> > +	.caps = (struct tst_cap []) {
> > +		TST_CAP(TST_CAP_REQ, CAP_LINUX_IMMUTABLE),
> > +		{}
> > +	},
> 
> Reviewed-by: Petr Vorel <pvorel@suse.cz>
> 
> LGTM, obviously it is CAP_LINUX_IMMUTABLE related.
> 
> This looks like fs/bcachefs/fs-ioctl.c
> 
> static int bch2_inode_flags_set(struct btree_trans *trans,
> 				struct bch_inode_info *inode,
> 				struct bch_inode_unpacked *bi,
> 				void *p)
> {
> 	...
> 	if (((newflags ^ oldflags) & (BCH_INODE_append|BCH_INODE_immutable)) &&
> 	    !capable(CAP_LINUX_IMMUTABLE))
> 		return -EPERM;
> 
> 
> We don't test on vfat and exfat. If I try to do it fail the same way (bcachefs,
> fat, exfat and ocfs2 use custom handler for FAT_IOCTL_SET_ATTRIBUTES).
> 
> I wonder why it does not fail for generic VFS fs/ioctl.c used by Btrfs and XFS:
> 
> /*
>  * Generic function to check FS_IOC_FSSETXATTR/FS_IOC_SETFLAGS values and reject
>  * any invalid configurations.
>  *
>  * Note: must be called with inode lock held.
>  */
> static int fileattr_set_prepare(struct inode *inode,
> 			      const struct fileattr *old_ma,
> 			      struct fileattr *fa)
> {
> 	int err;
> 
> 	/*
> 	 * The IMMUTABLE and APPEND_ONLY flags can only be changed by
> 	 * the relevant capability.
> 	 */
> 	if ((fa->flags ^ old_ma->flags) & (FS_APPEND_FL | FS_IMMUTABLE_FL) &&
> 	    !capable(CAP_LINUX_IMMUTABLE))
> 		return -EPERM;
> 
> 
> Kind regards,
> Petr
> 
> >  };
> 
> > ---
> > base-commit: 753bd13472d4be44eb70ff183b007fe9c5fffa07
> > change-id: 20250326-fix_ioctl_ficlone03_32bit_bcachefs-2ec15bd6c0c6
> 
> > Best regards,

Re: [LTP] [PATCH] ioctl_ficlone03: fix capabilities on immutable files

[Petr Vorel]

Hi Kent,

> On Wed, Mar 26, 2025 at 02:47:49PM +0100, Petr Vorel wrote:
> > Hi all,

> > [ Cc Kent and other filesystem folks to be sure we don't hide a bug ]

> I'm missing context here, and the original thread doesn't seem to be on
> lore?

I'm sorry, to put more info: this is an attempt to fix of LTP test
ioctl_ficlone03.c [1], which is failing on 32 bit compatibility mode:

# ./ioctl_ficlone03
tst_test.c:1833: TINFO: === Testing on bcachefs ===
..
ioctl_ficlone03.c:74: TBROK: ioctl .. failed: ENOTTY (25)
ioctl_ficlone03.c:89: TWARN: ioctl ..  failed: ENOTTY (25)
ioctl_ficlone03.c:91: TWARN: ioctl ..  failed: ENOTTY (25)
ioctl_ficlone03.c:98: TWARN: close(-1) failed: EBADF (9)

Original thread of this is on LTP ML [2].

Kind regards,
Petr

[1] https://github.com/linux-test-project/ltp/tree/master/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c
[2] https://lore.kernel.org/ltp/20250326-fix_ioctl_ficlone03_32bit_bcachefs-v1-1-554a0315ebf5@suse.com/

> > > From: Andrea Cervesato <andrea.cervesato@suse.com>

> > > Make sure that capabilities requirements are satisfied when accessing
> > > immutable files. On OpenSUSE Tumbleweed 32bit bcachefs fails with the
> > > following error due to missing capabilities:

> > > tst_test.c:1833: TINFO: === Testing on bcachefs ===
> > > ..
> > > ioctl_ficlone03.c:74: TBROK: ioctl .. failed: ENOTTY (25)
> > > ioctl_ficlone03.c:89: TWARN: ioctl ..  failed: ENOTTY (25)
> > > ioctl_ficlone03.c:91: TWARN: ioctl ..  failed: ENOTTY (25)
> > > ioctl_ficlone03.c:98: TWARN: close(-1) failed: EBADF (9)

> > > Introduce CAP_LINUX_IMMUTABLE capability to make sure that test won't
> > > fail on other systems as well.

> > > Signed-off-by: Andrea Cervesato <andrea.cervesato@suse.com>
> > > ---
> > >  testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c | 6 +++++-
> > >  1 file changed, 5 insertions(+), 1 deletion(-)

> > > diff --git a/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c b/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c
> > > index 6a9d270d9fb56f3a263f0aed976f62c4576e6a5f..6716423d9c96d9fc1d433f28e0ae1511b912ae2c 100644
> > > --- a/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c
> > > +++ b/testcases/kernel/syscalls/ioctl/ioctl_ficlone03.c
> > > @@ -122,5 +122,9 @@ static struct tst_test test = {
> > >  	.bufs = (struct tst_buffers []) {
> > >  		{&clone_range, .size = sizeof(struct file_clone_range)},
> > >  		{},
> > > -	}
> > > +	},
> > > +	.caps = (struct tst_cap []) {
> > > +		TST_CAP(TST_CAP_REQ, CAP_LINUX_IMMUTABLE),
> > > +		{}
> > > +	},

> > Reviewed-by: Petr Vorel <pvorel@suse.cz>

> > LGTM, obviously it is CAP_LINUX_IMMUTABLE related.

> > This looks like fs/bcachefs/fs-ioctl.c

> > static int bch2_inode_flags_set(struct btree_trans *trans,
> > 				struct bch_inode_info *inode,
> > 				struct bch_inode_unpacked *bi,
> > 				void *p)
> > {
> > 	...
> > 	if (((newflags ^ oldflags) & (BCH_INODE_append|BCH_INODE_immutable)) &&
> > 	    !capable(CAP_LINUX_IMMUTABLE))
> > 		return -EPERM;


> > We don't test on vfat and exfat. If I try to do it fail the same way (bcachefs,
> > fat, exfat and ocfs2 use custom handler for FAT_IOCTL_SET_ATTRIBUTES).

> > I wonder why it does not fail for generic VFS fs/ioctl.c used by Btrfs and XFS:

> > /*
> >  * Generic function to check FS_IOC_FSSETXATTR/FS_IOC_SETFLAGS values and reject
> >  * any invalid configurations.
> >  *
> >  * Note: must be called with inode lock held.
> >  */
> > static int fileattr_set_prepare(struct inode *inode,
> > 			      const struct fileattr *old_ma,
> > 			      struct fileattr *fa)
> > {
> > 	int err;

> > 	/*
> > 	 * The IMMUTABLE and APPEND_ONLY flags can only be changed by
> > 	 * the relevant capability.
> > 	 */
> > 	if ((fa->flags ^ old_ma->flags) & (FS_APPEND_FL | FS_IMMUTABLE_FL) &&
> > 	    !capable(CAP_LINUX_IMMUTABLE))
> > 		return -EPERM;


> > Kind regards,
> > Petr

> > >  };

> > > ---
> > > base-commit: 753bd13472d4be44eb70ff183b007fe9c5fffa07
> > > change-id: 20250326-fix_ioctl_ficlone03_32bit_bcachefs-2ec15bd6c0c6

> > > Best regards,

Re: [LTP] [PATCH] ioctl_ficlone03: fix capabilities on immutable files

[Dave Chinner]

On Wed, Mar 26, 2025 at 02:47:49PM +0100, Petr Vorel wrote:
> Hi all,
> 
> [ Cc Kent and other filesystem folks to be sure we don't hide a bug ]
> 
> > From: Andrea Cervesato <andrea.cervesato@suse.com>
> 
> > Make sure that capabilities requirements are satisfied when accessing
> > immutable files. On OpenSUSE Tumbleweed 32bit bcachefs fails with the
> > following error due to missing capabilities:
> 
> > tst_test.c:1833: TINFO: === Testing on bcachefs ===
> > ..
> > ioctl_ficlone03.c:74: TBROK: ioctl .. failed: ENOTTY (25)
> > ioctl_ficlone03.c:89: TWARN: ioctl ..  failed: ENOTTY (25)
> > ioctl_ficlone03.c:91: TWARN: ioctl ..  failed: ENOTTY (25)
> > ioctl_ficlone03.c:98: TWARN: close(-1) failed: EBADF (9)

None of these are -EPERM, so how is a missing capability that
results in -EPERM being returned cause -ENOTTY or -EBADF failures?

ohhhhh. ENOTTY is a result of a kernel side compat ioctl handling bug
w/ bcachefs.

bcachefs doesn't implement ->fileattr_set().

sys_compat_ioctl() does:

        case FS_IOC32_GETFLAGS:
        case FS_IOC32_SETFLAGS:
                cmd = (cmd == FS_IOC32_GETFLAGS) ?
                        FS_IOC_GETFLAGS : FS_IOC_SETFLAGS;

and then calls do_vfs_ioctl().

This then ends up in vfs_fileattr_set(), which does:

	if (!inode->i_op->fileattr_set)
                return -ENOIOCTLCMD;

which means sys_compat_ioctl() then falls back to calling
->compat_ioctl().

However, cmd has been overwritten to FS_IOC_SETFLAGS and
bch2_compat_fs_ioctl() looks for FS_IOC32_SETFLAGS, so it returns
-ENOIOCTLCMD as it doesn't handle the ioctl command passed in.

sys_compat_ioctl() then converts the -ENOIOCTLCMD to -ENOTTY, and
there's the error being reported.

OK, this is a bcachefs compat ioctl handling bug, triggered by not
implementing ->fileattr_set and implementing FS_IOC_SETFLAGS
directly itself via ->unlocked_ioctl.

Yeah, fixes to bcachefs needed here, not LTP.

> I wonder why it does not fail for generic VFS fs/ioctl.c used by Btrfs and XFS:

Because they implement ->fileattr_set() and so all the compat ioctl
FS_IOC32_SETFLAGS handling works as expected.

-Dave.
-- 
Dave Chinner
david@fromorbit.com

Re: [LTP] [PATCH] ioctl_ficlone03: fix capabilities on immutable files

[Kent Overstreet]

On Thu, Mar 27, 2025 at 11:49:19AM +1100, Dave Chinner wrote:
> On Wed, Mar 26, 2025 at 02:47:49PM +0100, Petr Vorel wrote:
> > Hi all,
> > 
> > [ Cc Kent and other filesystem folks to be sure we don't hide a bug ]
> > 
> > > From: Andrea Cervesato <andrea.cervesato@suse.com>
> > 
> > > Make sure that capabilities requirements are satisfied when accessing
> > > immutable files. On OpenSUSE Tumbleweed 32bit bcachefs fails with the
> > > following error due to missing capabilities:
> > 
> > > tst_test.c:1833: TINFO: === Testing on bcachefs ===
> > > ..
> > > ioctl_ficlone03.c:74: TBROK: ioctl .. failed: ENOTTY (25)
> > > ioctl_ficlone03.c:89: TWARN: ioctl ..  failed: ENOTTY (25)
> > > ioctl_ficlone03.c:91: TWARN: ioctl ..  failed: ENOTTY (25)
> > > ioctl_ficlone03.c:98: TWARN: close(-1) failed: EBADF (9)
> 
> None of these are -EPERM, so how is a missing capability that
> results in -EPERM being returned cause -ENOTTY or -EBADF failures?
> 
> ohhhhh. ENOTTY is a result of a kernel side compat ioctl handling bug
> w/ bcachefs.
> 
> bcachefs doesn't implement ->fileattr_set().
> 
> sys_compat_ioctl() does:
> 
>         case FS_IOC32_GETFLAGS:
>         case FS_IOC32_SETFLAGS:
>                 cmd = (cmd == FS_IOC32_GETFLAGS) ?
>                         FS_IOC_GETFLAGS : FS_IOC_SETFLAGS;
> 
> and then calls do_vfs_ioctl().
> 
> This then ends up in vfs_fileattr_set(), which does:
> 
> 	if (!inode->i_op->fileattr_set)
>                 return -ENOIOCTLCMD;
> 
> which means sys_compat_ioctl() then falls back to calling
> ->compat_ioctl().
> 
> However, cmd has been overwritten to FS_IOC_SETFLAGS and
> bch2_compat_fs_ioctl() looks for FS_IOC32_SETFLAGS, so it returns
> -ENOIOCTLCMD as it doesn't handle the ioctl command passed in.
> 
> sys_compat_ioctl() then converts the -ENOIOCTLCMD to -ENOTTY, and
> there's the error being reported.
> 
> OK, this is a bcachefs compat ioctl handling bug, triggered by not
> implementing ->fileattr_set and implementing FS_IOC_SETFLAGS
> directly itself via ->unlocked_ioctl.
> 
> Yeah, fixes to bcachefs needed here, not LTP.
> 
> > I wonder why it does not fail for generic VFS fs/ioctl.c used by Btrfs and XFS:
> 
> Because they implement ->fileattr_set() and so all the compat ioctl
> FS_IOC32_SETFLAGS handling works as expected.

thanks for the analysis, I'll try to get this fixed soon