On 11/23/18 1:24 PM, Dan Carpenter wrote: > On Fri, Nov 23, 2018 at 12:54:32PM -0700, Jens Axboe wrote: >> On 11/23/18 12:18 AM, Dan Carpenter wrote: >>> If blk_mq_alloc_request() fails then it leads to an error pointer >>> dereference. >> >> Thanks, this should not error, as we're getting the one and only >> reserved tag. That's why it doesn't check. So I think if we're >> going to add code to a case that can't happen, it should at least >> be accompanied by a error print of some sort. > > Ah. Fine. Let's not add code just to make the static checker happy. > > We should leave it as-is. Agree, but I think it deserves a comment :-) -- Jens Axboe
