Be very generous with timeouts for synchronize cache. We may wait a very long
authorPeter Avalos <pavalos@dragonflybsd.org>
Sat, 24 Nov 2007 01:31:39 +0000 (01:31 +0000)
committerPeter Avalos <pavalos@dragonflybsd.org>
Sat, 24 Nov 2007 01:31:39 +0000 (01:31 +0000)
time in the cases where it really sends the drive out to lunch, but it also
allows us to catch very weird edge cases of strange drives that might take
a very long time.

Obtained-from: FreeBSD

sys/bus/cam/scsi/scsi_da.c

index 96a6721..a2c529a 100644 (file)
@@ -26,7 +26,7 @@
  * SUCH DAMAGE.
  *
  * $FreeBSD: src/sys/cam/scsi/scsi_da.c,v 1.42.2.46 2003/10/21 22:18:19 thomas Exp $
- * $DragonFly: src/sys/bus/cam/scsi/scsi_da.c,v 1.45 2007/11/21 21:28:41 pavalos Exp $
+ * $DragonFly: src/sys/bus/cam/scsi/scsi_da.c,v 1.46 2007/11/24 01:31:39 pavalos Exp $
  */
 
 #ifdef _KERNEL
@@ -2124,7 +2124,7 @@ dashutdown(void * arg, int howto)
                                       /*begin_lba*/0, /* whole disk */
                                       /*lb_count*/0,
                                       SSD_FULL_SIZE,
-                                      5 * 60 * 1000);
+                                      60 * 60 * 1000);
 
                xpt_polled_action(&ccb);