public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* [patch 8/19] go back to 256 requests per queue
@ 2002-06-17  6:52 Andrew Morton
  0 siblings, 0 replies; only message in thread
From: Andrew Morton @ 2002-06-17  6:52 UTC (permalink / raw)
  To: Linus Torvalds; +Cc: lkml, Jens Axboe



The request queue was increased from 256 slots to 512 in 2.5.20.  The
throughput of `dbench 128' on Randy's 384 megabyte machine fell 40%.

We do need to understand why that happened, and what we can learn from
it.  But in the meanwhile I'd suggest that we go back to 256 slots so
that this known problem doesn't impact people's evaluation and tuning
of 2.5 performance.



--- 2.5.22/drivers/block/ll_rw_blk.c~256-requests	Sun Jun 16 22:50:18 2002
+++ 2.5.22-akpm/drivers/block/ll_rw_blk.c	Sun Jun 16 23:22:46 2002
@@ -2002,8 +2002,8 @@ int __init blk_dev_init(void)
 	queue_nr_requests = (total_ram >> 8) & ~15;	/* One per quarter-megabyte */
 	if (queue_nr_requests < 32)
 		queue_nr_requests = 32;
-	if (queue_nr_requests > 512)
-		queue_nr_requests = 512;
+	if (queue_nr_requests > 256)
+		queue_nr_requests = 256;
 
 	/*
 	 * Batch frees according to queue length

-

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2002-06-17  6:52 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2002-06-17  6:52 [patch 8/19] go back to 256 requests per queue Andrew Morton

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox