On Wed, Apr 17, 2019 at 11:23 PM Al Viro viro@zeniv.linux.org.uk wrote:
On Tue, Apr 16, 2019 at 10:19:49PM +0200, Arnd Bergmann wrote:
The ceph_ioctl function is used both for files and directories, but only the files support doing that in 32-bit compat mode.
For consistency, add the same compat handler to the dir operations as well.
Reviewed-by: "Yan, Zheng" zyan@redhat.com Cc: stable@vger.kernel.org Signed-off-by: Arnd Bergmann arnd@arndb.de
fs/ceph/dir.c | 1 + 1 file changed, 1 insertion(+)
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index a8f429882249..7c060cb22aa3 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -1785,6 +1785,7 @@ const struct file_operations ceph_dir_fops = { .open = ceph_open, .release = ceph_release, .unlocked_ioctl = ceph_ioctl,
.compat_ioctl = ceph_ioctl,Again, broken on s390 (and so's the ceph_file_ops, of course). It wants compat_ptr() applied to argument...
This gets changed in a later patch, I intentionally left this one doing the same as ceph_file_fops so the patch can be backported if necessary, without also backporting the patch that adds compat_ptr_ioctl.
Arnd