From: Tejun Heo <htejun@gmail.com>
To: Mark Lord <liml@rtr.ca>
Cc: Jeff Garzik <jgarzik@pobox.com>,
IDE/ATA development list <linux-ide@vger.kernel.org>
Subject: Re: [PATCH 09/13] sata_mv ncq Use DMA memory pools for hardware memory tables
Date: Thu, 31 Jan 2008 12:31:14 +0900 [thread overview]
Message-ID: <47A14102.9080605@gmail.com> (raw)
In-Reply-To: <47A13F28.7050001@gmail.com>
Tejun Heo wrote:
> I'm skeptical about the benefit of IRQ coalescing on storage
> controllers. Coalescing improves performance when there are many small
> requests to complete and if you put a lot of small non-consecutive
> requests to a disk, it gets really really really slow and IRQ coalescing
> just doesn't matter at all. The only way to achieve high number of
> completions is to issue small commands to consecutive addresses which is
> just silly. In storage, high volume transfer is achieved through
> request coalescing not completion coalescing and this is true for even SDDs.
To add a bit, I was thinking about writes regarding SDDs. Completion
coalescing makes more sense for reads from SDDs if log based filesystem
is used on it because then read operations are scattered all over the
disk and the device doesn't have to seek. Limited queue size compared
to network interfaces will make completion coalescing less effective but
yeah, I think it will be worth playing with it on SDD + lfs.
Thanks.
--
tejun
next prev parent reply other threads:[~2008-01-31 3:31 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-26 23:28 [PATCH 01/13] sata_mv ncq EH fixes Mark Lord
2008-01-26 23:30 ` [PATCH 02/13] sata_mv ncq Mask transient IRQs Mark Lord
2008-01-26 23:31 ` [PATCH 03/13] sata_mv ncq Rename base to port mmio Mark Lord
2008-01-26 23:31 ` [PATCH 04/13] sata_mv ncq Fix EDMA configuration Mark Lord
2008-01-26 23:31 ` [PATCH 05/13] sata_mv ncq Add want ncq parameter for " Mark Lord
2008-01-26 23:31 ` [PATCH 06/13] sata_mv ncq Use hqtag instead of ioid Mark Lord
2008-01-26 23:32 ` [PATCH 07/13] sata_mv ncq Ignore response status LSB on NCQ Mark Lord
2008-01-26 23:32 ` [PATCH 08/13] sata_mv ncq Restrict max sectors to 8-bits on GenII NCQ Mark Lord
2008-01-26 23:32 ` [PATCH 09/13] sata_mv ncq Use DMA memory pools for hardware memory tables Mark Lord
2008-01-29 17:10 ` Jeff Garzik
2008-01-29 18:24 ` Mark Lord
2008-01-30 9:54 ` Jeff Garzik
2008-01-30 16:40 ` Mark Lord
2008-01-30 17:08 ` Jeff Garzik
2008-01-30 17:19 ` Mark Lord
2008-01-30 17:45 ` Jeff Garzik
2008-01-30 18:57 ` Mark Lord
2008-01-31 3:23 ` Tejun Heo
2008-01-31 3:31 ` Tejun Heo [this message]
2008-01-31 3:59 ` Mark Lord
2008-01-31 9:00 ` Mikael Pettersson
2008-01-26 23:32 ` [PATCH 10/13] sata_mv ncq Introduce per-tag SG tables Mark Lord
2008-01-30 9:50 ` Jeff Garzik
2008-01-26 23:33 ` [PATCH 11/13] sata_mv ncq Enable NCQ operation Mark Lord
2008-01-26 23:33 ` [PATCH 12/13] sata_mv ncq Remove post internal cmd op Mark Lord
2008-01-26 23:33 ` [PATCH 13/13] sata_mv ncq Comments and version bump Mark Lord
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=47A14102.9080605@gmail.com \
--to=htejun@gmail.com \
--cc=jgarzik@pobox.com \
--cc=liml@rtr.ca \
--cc=linux-ide@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).