Messages in this thread | | | Date | Fri, 2 Jul 2010 12:52:41 +0200 | From | Christoph Hellwig <> | Subject | Re: scsi: address leak in the error path of discard page allocation |
| |
On Fri, Jul 02, 2010 at 01:53:14PM +0900, FUJITA Tomonori wrote: > Seems that scsi-ml calls scsi_unprep_request() for not-prepped > requests in scsi_init_io error path. So we could move that > scsi_unprep_request() to the error path in scsi_prep_return(). Then we > can free discard page in the single place. > > Applying the rule strictly is fine by me too; we remove > scsi_unprep_request() in scsi_init_io error path and clean up things > in each prep function's error path.
That would be my preference. Making sure a function cleans up all allocations / state changes on errors means code is a lot fragile and easier to understand.
> Btw, blk_clear_request_payload() is necessary? > > Making sure that a request is clean is not a bad idea but if we hit > BLKPREP_KILL or BLKPREP_DEFER, we call > blk_end_request(). blk_end_request() can free a request properly even > if we don't do something like blk_clear_request_payload?
For BLKPREP_KILL we do call __blk_end_request_all, but for BLKPREP_DEFER we don't. In that case we just leave it on the queue for a later retry. So we either have to clean it up, or leave the detect the case of a partially constructed command in ->prep_fn. I think cleaning up properly and having defined state when entering ->prep_fn is the better variant.
| |