Re: [PATCH] various IDE patches/cleanups

[Davin McCall]

4th patch notes
---------------

choose_drive() selects which drive in a hwgroup will have a request serviced.
It tries not to return a drive whose queue is plugged, but there's a small
mistake which means it might do this sometimes.

Once we've guarenteed that the returned drive is never plugged, it's possible
to remove a check in ide_do_request() (or actually, modify it, seeing as the
drive can still get plugged between requests in the tcq case).

- (ide-io.c)
  - ide_do_request()
    - Move check of blk_queue_plugged() just after the check for the service
      routine returning ide_released. So we only check it between tcq
      requests.
  - choose_drive()
    - A plugged drive can't benefit from "nice" behaviour. Neither can
      a drive with no requests in its queue.

--- linux-2.6.0-patch3/drivers/ide/ide-io.c	Thu Jan 29 12:43:26 2004
+++ linux-2.6.0/drivers/ide/ide-io.c	Thu Jan 29 12:48:09 2004
@@ -776,7 +776,9 @@
 				if (!drive->sleep
 				/* FIXME: use time_before */
 				 && 0 < (signed long)(WAKEUP(drive) - (jiffies - best->service_time))
-				 && 0 < (signed long)((jiffies + t) - WAKEUP(drive)))
+				 && 0 < (signed long)((jiffies + t) - WAKEUP(drive))
+				 && !blk_queue_plugged(drive->queue)
+				 && !elv_queue_empty(drive->queue))
 				{
 					ide_stall_queue(best, IDE_MIN(t, 10 * WAIT_MIN_SLEEP));
 					goto repeat;
@@ -908,14 +910,6 @@
 			break;
 		}
 
-		if (blk_queue_plugged(drive->queue)) {
-			if (drive->using_tcq)
-				break;
-
-			printk(KERN_ERR "ide: huh? queue was plugged!\n");
-			break;
-		}
-
 		/*
 		 * we know that the queue isn't empty, but this can happen
 		 * if the q->prep_rq_fn() decides to kill a request
@@ -967,8 +961,11 @@
 		spin_lock_irq(&ide_lock);
 		if (hwif->irq != masked_irq)
 			enable_irq(hwif->irq);
-		if (startstop == ide_released)
+		if (startstop == ide_released) {
+			if (blk_queue_plugged(drive->queue))
+				break;		
 			goto queue_next;
+		}
 		if (startstop == ide_stopped)
 			hwgroup->busy = 0;
 	}
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/