From: Murthy Bhat <Murthy.Bhat@xxxxxxxxxxxxx> [ Upstream commit b622a601a13ae5974c5b0aeecb990c224b8db0d9 ] While failing queued I/Os in TMF path, there was a request leak and hence stale entries in request pool with ref count being non-zero. In shutdown path we have a BUG_ON to catch stuck I/O either in firmware or in the driver. The stale requests caused a system crash. The I/O request pool leakage also lead to a significant performance drop. Link: https://lore.kernel.org/r/161549370379.25025.12793264112620796062.stgit@brunhilda Reviewed-by: Scott Teel <scott.teel@xxxxxxxxxxxxx> Reviewed-by: Scott Benesh <scott.benesh@xxxxxxxxxxxxx> Reviewed-by: Kevin Barnett <kevin.barnett@xxxxxxxxxxxxx> Signed-off-by: Murthy Bhat <Murthy.Bhat@xxxxxxxxxxxxx> Signed-off-by: Don Brace <don.brace@xxxxxxxxxxxxx> Signed-off-by: Martin K. Petersen <martin.petersen@xxxxxxxxxx> Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx> --- drivers/scsi/smartpqi/smartpqi_init.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/scsi/smartpqi/smartpqi_init.c b/drivers/scsi/smartpqi/smartpqi_init.c index 093ed5d1eef2..3480a0a66771 100644 --- a/drivers/scsi/smartpqi/smartpqi_init.c +++ b/drivers/scsi/smartpqi/smartpqi_init.c @@ -5513,6 +5513,8 @@ static void pqi_fail_io_queued_for_device(struct pqi_ctrl_info *ctrl_info, list_del(&io_request->request_list_entry); set_host_byte(scmd, DID_RESET); + pqi_free_io_request(io_request); + scsi_dma_unmap(scmd); pqi_scsi_done(scmd); } @@ -5549,6 +5551,8 @@ static void pqi_fail_io_queued_for_all_devices(struct pqi_ctrl_info *ctrl_info) list_del(&io_request->request_list_entry); set_host_byte(scmd, DID_RESET); + pqi_free_io_request(io_request); + scsi_dma_unmap(scmd); pqi_scsi_done(scmd); } -- 2.30.2
