Hi, When one request is dispatched to LLD via dm-rq, if the result is BLK_STS_*RESOURCE, dm-rq will free the request. However, LLD may allocate private data for this request, so this way will cause memory leak.
Add .cleanup_rq() callback and implement it in SCSI for fixing the issue, since SCSI is the only driver which allocates private requst data in .queue_rq() path. Another use case of this callback is to free the request and re-submit bios during cpu hotplug when the hctx is dead, see the following link: https://lore.kernel.org/linux-block/[email protected]/T/#t V4: - add more commit log on the new .cleanup_rq callback, as suggested by Mike V3: - run .cleanup_rq() from dm-rq because this issue is dm-rq specific, and even in future it should be still very unusual to free request in this way. If we call .cleanup_rq() in generic rq free code(fast path), cost will be introduced unnecessarily, also we have to consider related race. V2: - run .cleanup_rq() in blk_mq_free_request(), as suggested by Mike Ming Lei (2): blk-mq: add callback of .cleanup_rq scsi: implement .cleanup_rq callback drivers/md/dm-rq.c | 1 + drivers/scsi/scsi_lib.c | 13 +++++++++++++ include/linux/blk-mq.h | 13 +++++++++++++ 3 files changed, 27 insertions(+) Cc: Ewan D. Milne <[email protected]> Cc: Bart Van Assche <[email protected]> Cc: Hannes Reinecke <[email protected]> Cc: Christoph Hellwig <[email protected]> Cc: Mike Snitzer <[email protected]> Cc: [email protected] Cc: <[email protected]> Fixes: 396eaf21ee17 ("blk-mq: improve DM's blk-mq IO merging via blk_insert_cloned_request feedback") -- 2.20.1
