On 2014-04-25 at 15:11:05 +0200, Dan Carpenter <dan.carpenter@xxxxxxxxxx> wrote: > On Fri, Apr 25, 2014 at 01:54:51PM +0200, Tobias Klauser wrote: > > pFifoHead is dereferenced before it is checked for NULL which implies it > > can potentially be NULL. Thus move the check before the dereference. > > > > Found by the coverity scanner, CID 1127221. > > > > pFifoHead is this: > > struct vnt_tx_fifo_head *pFifoHead = &tx_buffer->fifo_head; > > Even if tx_buffer were NULL, the address of tx_buffer->fifo_head would > be non-NULL. The NULL test serves no purpose. If tx_buffer were NULL > we would still get an oops though. According to smatch, tx_buffer is > never NULL but I haven't verified manually. Hm, silly me... Thanks for pointing out Dan. I manually checked now and it seems that tx_buffer can never be NULL. So it should be fine to remove the test altogether. I'll send an updated patch. Thanks _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
