On Sat, 11 Jul 2009, Jiri Slaby wrote:
> On 07/11/2009 12:17 PM, Jaswinder Singh Rajput wrote:
> > On Sat, 2009-07-11 at 10:17 +0200, Jiri Slaby wrote:
> >> On 07/11/2009 09:51 AM, Julia Lawall wrote:
> >>> --- a/drivers/staging/otus/wwrap.c
> >>> +++ b/drivers/staging/otus/wwrap.c
> >>> @@ -1016,12 +1016,13 @@ void kevent(struct work_struct *work)
> >>> {
> >>> struct usbdrv_private *macp =
> >>> container_of(work, struct usbdrv_private, kevent);
> >>> - zdev_t *dev = macp->device;
> >>> + zdev_t *dev;
> >>>
> >>> if (macp == NULL)
> >>> {
> >>> return;
> >>> }
> >>
> >> The test is rather useless here.
> >>
> >
> > Why useless, it should be non-null before setting to dev.
>
> Even if work was NULL, macp won't be NULL.
work could be a small negative number such that calling container_of would
return 0.
julia
