On Thu, Jun 06, 2013 at 11:39:46AM -0700, Greg Kroah-Hartman wrote:
> On Thu, Jun 06, 2013 at 11:04:07PM +0800, Peng Tao wrote:
> > No need to cast count since it is already ssize_t. No need
> > to cast payload to const, but need __force instead to avoid
> > Sparse complaining.
>
> That usually means something is really wrong here.
>
> > Reported-and-Suggested-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
> > Signed-off-by: Peng Tao <tao.peng@xxxxxxx>
> > Signed-off-by: Andreas Dilger <andreas.dilger@xxxxxxxxx>
> > ---
> > .../lustre/lustre/libcfs/kernel_user_comm.c | 4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/staging/lustre/lustre/libcfs/kernel_user_comm.c b/drivers/staging/lustre/lustre/libcfs/kernel_user_comm.c
> > index 8084ada..d6d3b2e 100644
> > --- a/drivers/staging/lustre/lustre/libcfs/kernel_user_comm.c
> > +++ b/drivers/staging/lustre/lustre/libcfs/kernel_user_comm.c
> > @@ -170,8 +170,8 @@ int libcfs_kkuc_msg_put(struct file *filp, void *payload)
>
> payload should be __user, right? That should solve this problem
> instead. Ah, but you need to unwind some other functions to get this to
> work properly, but that looks like the correct thing to do, instead of
> this:
>
> >
> > fs = get_fs();
> > set_fs(KERNEL_DS);
> > - while ((ssize_t)count > 0) {
> > - rc = vfs_write(filp, (const void __user *)payload,
> > + while (count > 0) {
> > + rc = vfs_write(filp, (void __force __user *)payload,
> > count, &offset);
>
> I don't want to take this, sorry.
Ugh, it's vfs_write(), nevermind, this is valid as you are doing the
proper set_fs() "magic" thing here. I'll go queue this up, but ick,
ick, ick...
greg k-h
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
