SCSI: Retry ADD_TO_MLQUEUE return value for EH commands

commit 6e883b0e42739aa560133cfaf41be1138c51a500 upstream.

A target reset when I/O is ongoing might result
an eventual device offline, as scsi_eh_completed_normally()
might return ADD_TO_MLQUEUE in addition to the
advertised SUCCESS, FAILED, and NEEDS_RETRY.

Which is unfortunate as scsi_send_eh_cmnd() will
therefore map ADD_TO_MLQUEUE to FAILED instead of
the more appropriate NEEDS_RETRY.

Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c
index a1689353d7fd715cf91776e9b290381795876477..c253e9c051b2067de673f5cb68c4b8d44e0e551d 100644 (file)
--- a/drivers/scsi/scsi_error.c
+++ b/drivers/scsi/scsi_error.c
@@ -721,6 +721,9 @@ static int scsi_send_eh_cmnd(struct scsi_cmnd *scmd, unsigned char *cmnd,
                case NEEDS_RETRY:
                case FAILED:
                        break;
+               case ADD_TO_MLQUEUE:
+                       rtn = NEEDS_RETRY;
+                       break;
                default:
                        rtn = FAILED;
                        break;