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@xxxxxxxx>
>
> > 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@xxxxxxxx>
> > ---
> > 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@xxxxxxx>
>
> 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,
