From: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
To: Jeff Garzik <jgarzik@pobox.com>, Jens Axboe <axboe@suse.de>
Cc: Linux Kernel <linux-kernel@vger.kernel.org>,
Chris Mason <mason@suse.com>
Subject: Re: [PATCH] barrier patch set
Date: Sat, 20 Mar 2004 00:59:22 +0100 [thread overview]
Message-ID: <200403200059.22234.bzolnier@elka.pw.edu.pl> (raw)
In-Reply-To: <405B2127.8090705@pobox.com>
On Friday 19 of March 2004 17:34, Jeff Garzik wrote:
> Jens Axboe wrote:
> > Cosmetic stuff that will get ironed out. You can find the patches here:
> >
> > ftp://ftp.kernel.org/pub/linux/kernel/people/axboe/patches/v2.6/2.6.5-rc1
> >-mm2/
> >
> > ide-barrier-2.6.5-rc1-mm2-1
> > ide/core part
Jens, am I right that you didn't do any changes/cleanups I asked you to do?
Here they are once again (probably some new items added as a bonus). ;-)
- do not use hwgroup->wrq (die!) and do not add drive->special_buf,
just do what PM code does and other special commands do - use taskfile
(yes, dirty stack allocation)
- SCSI -> IDE transform should die, please use something like REQ_FLUSH
and let subsystems deal with it
- ide_get_error_location() is cool but clean other places doing same thing
as you are duplicating existing code
(please use u64 not sector_t - you are getting raw info from the disk)
- why does blkdev_issue_flush() add REQ_BLOCK_PC to rq->flags?
- why are we doing pre-flush?
> WRT ATA and flush-cache... before the IDE pieces of this patch are
> merged, IMO it is a requirement that the entire flush-cache stuff gets a
> review. ide_get_error_location() is one of the important pieces
> (great!). Another important piece is to make sure that a drive's
> flush-cache capability is correctly deduced and set up from the
> identify-device. The steps look like
Yep.
> - check identify-device word 83, bits 12 (flush cache) and 13 (flush
> cache ext)
> - issue set-features command to get flush-cache into proper state
> (enabled or disabled, as the user desires), if identify-device word 86
> indicates it is not already in the state you seek.
> - re-read identify [packet] device page from device, make sure
> flush-cache[-ext] is enabled. A slacker could just make sure the
> set-features command completed successfully, but to be 100% correct you
> need to re-read the identify-device page. :/
The fact that spec says "supported" not "enabled" in description of word86
makes me wonder - can they be disabled? (FLUSH CACHE is mandatory for General
feature set and FLUSH CACHE EXT is mandatory if 48-bit LBA is supported)
Jeff, please note that these bits were introduced by ATA-6 spec
and take a look at ATA-5 spec:
...
FLUSH CACHE
General feature set
- Mandatory for all devices
...
and ATA-4 spec:
...
FLUSH CACHE
General feature set
- Optional for all devices
...
IMO to test if FLUSH CACHE works we should just issue it during disk setup
and check result. This way we can use FLUSH CACHE also on < ATA-6 devices
(there is a lot of them).
Cheers,
Bartlomiej
next prev parent reply other threads:[~2004-03-19 23:51 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-03-19 15:35 Jens Axboe
2004-03-19 16:30 ` Mika Penttilä
2004-03-19 18:16 ` Jens Axboe
2004-03-19 18:44 ` Mika Penttilä
2004-03-20 9:55 ` Jens Axboe
2004-03-19 16:34 ` Jeff Garzik
2004-03-19 18:19 ` Jens Axboe
2004-03-19 23:01 ` Matthias Andree
2004-03-20 0:02 ` Bartlomiej Zolnierkiewicz
2004-03-20 1:48 ` Johannes Stezenbach
2004-03-20 2:13 ` Bartlomiej Zolnierkiewicz
2004-03-20 2:53 ` Johannes Stezenbach
2004-03-20 16:03 ` Bartlomiej Zolnierkiewicz
2004-03-20 11:36 ` Matthias Andree
2004-03-20 16:00 ` Bartlomiej Zolnierkiewicz
2004-03-20 23:36 ` Johannes Stezenbach
2004-03-21 1:33 ` Bartlomiej Zolnierkiewicz
2004-03-20 18:52 ` Helge Hafting
2004-03-22 11:15 ` Matthias Andree
2004-03-19 23:59 ` Bartlomiej Zolnierkiewicz [this message]
2004-03-20 0:14 ` Jeff Garzik
2004-03-20 0:40 ` Bartlomiej Zolnierkiewicz
2004-03-20 0:42 ` Jeff Garzik
2004-03-20 1:24 ` Bartlomiej Zolnierkiewicz
2004-03-20 9:58 ` Jens Axboe
2004-03-20 10:12 ` Jeff Garzik
2004-03-20 10:19 ` Jens Axboe
2004-03-20 10:37 ` Jeff Garzik
2004-03-20 16:30 ` Bartlomiej Zolnierkiewicz
2004-03-21 18:12 ` Jeff Garzik
2004-03-20 10:21 ` Jeff Garzik
2004-03-20 15:54 ` Bartlomiej Zolnierkiewicz
2004-03-20 0:17 ` Jeff Garzik
2004-03-20 9:53 ` Jens Axboe
2004-03-20 16:23 ` Bartlomiej Zolnierkiewicz
2004-03-20 16:27 ` Jens Axboe
2004-03-20 16:32 ` Chris Mason
2004-03-20 17:05 ` Bartlomiej Zolnierkiewicz
2004-03-20 17:10 ` Chris Mason
2004-03-20 20:16 ` Bartlomiej Zolnierkiewicz
2004-03-21 9:43 ` Jens Axboe
2004-03-30 16:04 ` Stephen C. Tweedie
2004-03-30 19:19 ` Chris Mason
2004-03-30 21:50 ` Stephen C. Tweedie
2004-03-30 22:13 ` Chris Mason
2004-03-31 14:03 ` Stephen C. Tweedie
2004-03-31 14:27 ` Chris Mason
2004-03-31 18:28 ` Ric Wheeler
2004-03-30 22:21 ` Jeff Garzik
2004-03-30 22:36 ` Chris Wedgwood
2004-03-30 22:39 ` Jeff Garzik
2004-03-30 22:41 ` Chris Wedgwood
2004-03-30 22:40 ` Bartlomiej Zolnierkiewicz
2004-03-30 22:38 ` Jeff Garzik
2004-03-31 14:08 ` Stephen C. Tweedie
2004-03-31 14:21 ` Chris Mason
2004-03-31 21:26 ` Jeff Garzik
2004-03-31 22:09 ` Chris Mason
2004-03-31 21:27 ` Jeff Garzik
2004-03-19 16:48 ` Marc-Christian Petersen
2004-03-19 18:19 ` Jens Axboe
2004-03-22 11:09 ` Andrew Morton
2004-03-22 11:10 ` Jens Axboe
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=200403200059.22234.bzolnier@elka.pw.edu.pl \
--to=b.zolnierkiewicz@elka.pw.edu.pl \
--cc=axboe@suse.de \
--cc=jgarzik@pobox.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mason@suse.com \
/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
all inboxes | Powered by JetHome®