From: Murthy Bhat <Murthy.Bhat@xxxxxxxxxxxxx> Correct request leakage during reset operations. * While failing queued IOs in TMF path, there was a request leak and hence stale entries in request pool with ref count being nonzero. In shutdown path, we have a BUG_ON to catch stuck IO either in f/w or in the driver. The stale requests caused a system crash. The IO request pool leakage also lead to a significant performance drop. 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> --- 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 c154e4578e55..2886884ad584 100644 --- a/drivers/scsi/smartpqi/smartpqi_init.c +++ b/drivers/scsi/smartpqi/smartpqi_init.c @@ -5489,6 +5489,8 @@ 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); } @@ -5525,6 +5527,8 @@ 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); }
