| Message ID | 20231215121008.2881653-1-alexander.atanasov@virtuozzo.com (mailing list archive) |
|---|---|
| State | Accepted |
| Headers | show |
| Series | [v2,RESEND] scsi: code: always send batch on reset or error handling command | expand |
On Fri, Dec 15, 2023 at 8:10 PM Alexander Atanasov <alexander.atanasov@virtuozzo.com> wrote: > > In commit 8930a6c20791 ("scsi: core: add support for request batching") > blk-mq last flags was mapped to SCMD_LAST and used as an indicator to > send the batch for the drivers that implement it but the error handling > code was not updated. > > scsi_send_eh_cmnd(...) is used to send error handling commands and > request sense. The problem is that request sense comes as a single > command that gets into the batch queue and times out. As result > device goes offline after several failed resets. This was observed > on virtio_scsi device resize operation. > > [ 496.316946] sd 0:0:4:0: [sdd] tag#117 scsi_eh_0: requesting sense > [ 506.786356] sd 0:0:4:0: [sdd] tag#117 scsi_send_eh_cmnd timeleft: 0 > [ 506.787981] sd 0:0:4:0: [sdd] tag#117 abort > > To fix this always set SCMD_LAST flag in scsi_send_eh_cmnd and > scsi_reset_ioctl(...). > > Fixes: 8930a6c20791 ("scsi: core: add support for request batching") > Signed-off-by: Alexander Atanasov <alexander.atanasov@virtuozzo.com> Reviewed-by: Ming Lei <ming.lei@redhat.com>
On Fri, 15 Dec 2023 14:10:08 +0200, Alexander Atanasov wrote: > In commit 8930a6c20791 ("scsi: core: add support for request batching") > blk-mq last flags was mapped to SCMD_LAST and used as an indicator to > send the batch for the drivers that implement it but the error handling > code was not updated. > > scsi_send_eh_cmnd(...) is used to send error handling commands and > request sense. The problem is that request sense comes as a single > command that gets into the batch queue and times out. As result > device goes offline after several failed resets. This was observed > on virtio_scsi device resize operation. > > [...] Applied to 6.7/scsi-fixes, thanks! [1/1] scsi: code: always send batch on reset or error handling command https://git.kernel.org/mkp/scsi/c/066c5b46b6ea
diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index c67cdcdc3ba8..1223d34c04da 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -1152,6 +1152,7 @@ static enum scsi_disposition scsi_send_eh_cmnd(struct scsi_cmnd *scmd, scsi_log_send(scmd); scmd->submitter = SUBMITTED_BY_SCSI_ERROR_HANDLER; + scmd->flags |= SCMD_LAST; /* * Lock sdev->state_mutex to avoid that scsi_device_quiesce() can @@ -2459,6 +2460,7 @@ scsi_ioctl_reset(struct scsi_device *dev, int __user *arg) scsi_init_command(dev, scmd); scmd->submitter = SUBMITTED_BY_SCSI_RESET_IOCTL; + scmd->flags |= SCMD_LAST; memset(&scmd->sdb, 0, sizeof(scmd->sdb)); scmd->cmd_len = 0;
In commit 8930a6c20791 ("scsi: core: add support for request batching") blk-mq last flags was mapped to SCMD_LAST and used as an indicator to send the batch for the drivers that implement it but the error handling code was not updated. scsi_send_eh_cmnd(...) is used to send error handling commands and request sense. The problem is that request sense comes as a single command that gets into the batch queue and times out. As result device goes offline after several failed resets. This was observed on virtio_scsi device resize operation. [ 496.316946] sd 0:0:4:0: [sdd] tag#117 scsi_eh_0: requesting sense [ 506.786356] sd 0:0:4:0: [sdd] tag#117 scsi_send_eh_cmnd timeleft: 0 [ 506.787981] sd 0:0:4:0: [sdd] tag#117 abort To fix this always set SCMD_LAST flag in scsi_send_eh_cmnd and scsi_reset_ioctl(...). Fixes: 8930a6c20791 ("scsi: core: add support for request batching") Signed-off-by: Alexander Atanasov <alexander.atanasov@virtuozzo.com> --- drivers/scsi/scsi_error.c | 2 ++ 1 file changed, 2 insertions(+) v1->v2: fix it globally not only for virtio_scsi, as suggested by Paolo Bonzini, to avoid reintroducing the same bug. RESEND -> add linux-scsi, remove stable