| Message ID | 20190416202013.4034148-9-arnd@arndb.de |
|---|---|
| State | New |
| Headers | show |
| Series | compat_ioctl: cleanups | expand |
On Tue, Apr 16, 2019 at 10:19:46PM +0200, Arnd Bergmann wrote: > Many drivers have ioctl() handlers that are completely compatible between > 32-bit and 64-bit architectures, except for the argument that is passed > down from user space and may have to be passed through compat_ptr() > in order to become a valid 64-bit pointer. > > Using ".compat_ptr = compat_ptr_ioctl" in file operations should let > us simplify a lot of those drivers to avoid #ifdef checks, and convert > additional drivers that don't have proper compat handling yet. You forgot to mention that it even gets the previous commit to build ;-) IOW, that needs to be reordered.
On Wed, Apr 17, 2019 at 11:19 PM Al Viro <viro@zeniv.linux.org.uk> wrote: > > On Tue, Apr 16, 2019 at 10:19:46PM +0200, Arnd Bergmann wrote: > > Many drivers have ioctl() handlers that are completely compatible between > > 32-bit and 64-bit architectures, except for the argument that is passed > > down from user space and may have to be passed through compat_ptr() > > in order to become a valid 64-bit pointer. > > > > Using ".compat_ptr = compat_ptr_ioctl" in file operations should let > > us simplify a lot of those drivers to avoid #ifdef checks, and convert > > additional drivers that don't have proper compat handling yet. > > You forgot to mention that it even gets the previous commit to build ;-) > IOW, that needs to be reordered. Right. I had reordered the series multiple times and got this part wrong in the end. Arnd
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index fee116e822d8..ab2ff530313b 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -78,6 +78,16 @@ get_user(val, srcptr) || put_user(val, dstptr); \ }) +/* helper function to avoid trivial compat_ioctl() implementations */ +long compat_ptr_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + if (!file->f_op->unlocked_ioctl) + return -ENOIOCTLCMD; + + return file->f_op->unlocked_ioctl(file, cmd, (unsigned long)compat_ptr(arg)); +} +EXPORT_SYMBOL(compat_ptr_ioctl); + static inline int do_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { int err; diff --git a/include/linux/fs.h b/include/linux/fs.h index dd28e7679089..dc4138314635 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1698,6 +1698,13 @@ int vfs_mkobj(struct dentry *, umode_t, extern long vfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg); +#ifdef CONFIG_COMPAT +extern long compat_ptr_ioctl(struct file *file, unsigned int cmd, + unsigned long arg); +#else +#define compat_ptr_ioctl NULL +#endif + /* * VFS file helper functions. */
Many drivers have ioctl() handlers that are completely compatible between 32-bit and 64-bit architectures, except for the argument that is passed down from user space and may have to be passed through compat_ptr() in order to become a valid 64-bit pointer. Using ".compat_ptr = compat_ptr_ioctl" in file operations should let us simplify a lot of those drivers to avoid #ifdef checks, and convert additional drivers that don't have proper compat handling yet. Signed-off-by: Arnd Bergmann <arnd@arndb.de> --- v2: use compat_ptr_ioctl instead of generic_compat_ioctl_ptrarg, as suggested by Al Viro --- fs/compat_ioctl.c | 10 ++++++++++ include/linux/fs.h | 7 +++++++ 2 files changed, 17 insertions(+) -- 2.20.0