LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: "Eric D. Mudama" <edmudama@gmail.com>
To: "Jeff Garzik" <jeff@garzik.org>
Cc: "Jens Axboe" <jens.axboe@oracle.com>,
	"Robert Hancock" <hancockr@shaw.ca>,
	"Björn Steinbrink" <B.Steinbrink@gmx.de>,
	linux-kernel@vger.kernel.org, htejun@gmail.com
Subject: Re: SATA exceptions with 2.6.20-rc5
Date: Mon, 22 Jan 2007 11:17:51 -0700	[thread overview]
Message-ID: <311601c90701221017v65866b6dm565d09bc756bdf3b@mail.gmail.com> (raw)
In-Reply-To: <45AC2F99.3040209@garzik.org>

On 1/15/07, Jeff Garzik <jeff@garzik.org> wrote:
> Jens Axboe wrote:
> > On Mon, Jan 15 2007, Jeff Garzik wrote:
> >> Jens Axboe wrote:
> >>> I'd be surprised if the device would not obey the 7 second timeout rule
> >>> that seems to be set in stone and not allow more dirty in-drive cache
> >>> than it could flush out in approximately that time.
> >> AFAIK Windows flush-cache timeout is 30 seconds, not 7 as with other
> >> commands...
> >
> > Ok, 7 seconds for FLUSH_CACHE would have been nice for us too though, as
> > it would pretty much guarentee lower latencies for random writes and
> > write back caching. The concern is the barrier code, of course. I guess
> > I should do some timings on potential worst case patterns some day. Alan
> > may have done that sometime in the past, iirc.
>
> FWIW:  According to the drive guys (Eric M, among others), FLUSH CACHE
> will "probably" be under 30 seconds, but pathological cases might even
> extend beyond that.
>
> Definitely more than 7 seconds in less-than-pathological cases,
> unfortunately...

The mentioned Maxtor model (6Yxxx) isn't susceptible to the
large-buffer long completion times, due to architectural differences
and availability of only small buffers.  Any "real" long-completion
flush on this device would, I believe, involve damage to the disk that
hinders the ability to seek, settle, or write.  (e.g. 30-second
flushes are easy to hit if you mount the disk on a shaker-table with
sufficient amplitude)

Later in the thread I think people have pretty much isolated it as not
the disk's problem, but just wanted to point this out.

I assume that large enough customers can buy enterprise-type command
completion ("all commands within X seconds") from most any disk
vendor.  However, these firmwares require much smarter or more active
drivers or block layers, to handle the higher error rate when the data
on the device is valid, but it will take longer than allowed by the
arbitrary enterprise rules.  Most customers who are buying this many
devices have software engineers customizing the drivers or disk
management applications to handle this differing behavior.

--eric

  reply	other threads:[~2007-01-22 18:17 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <fa.hif5u4ZXua+b0mVNaWEcItWv9i0@ifi.uio.no>
2007-01-14 23:43 ` Robert Hancock
2007-01-15  0:22   ` Jeff Garzik
2007-01-15  0:34     ` Björn Steinbrink
2007-01-15  2:47       ` Björn Steinbrink
2007-01-15  2:25     ` Robert Hancock
2007-01-15  2:53       ` Jens Axboe
2007-01-15 13:42         ` Jeff Garzik
2007-01-16  0:23           ` Jens Axboe
2007-01-16  0:36             ` Robert Hancock
2007-01-16  1:51               ` Jeff Garzik
2007-01-16  1:51             ` Jeff Garzik
2007-01-22 18:17               ` Eric D. Mudama [this message]
2007-01-15 21:17   ` Björn Steinbrink
2007-01-15 23:46     ` Björn Steinbrink
2007-01-16  0:34       ` Robert Hancock
2007-01-16  1:35         ` Björn Steinbrink
2007-01-16  3:00           ` Robert Hancock
2007-01-16  1:53         ` Jeff Garzik
2007-01-19 15:05           ` Alistair John Strachan
2007-01-19 19:51             ` chunkeey
2007-01-20  2:41             ` Robert Hancock
2007-01-20  2:47               ` Alistair John Strachan
2007-01-20  4:15               ` Björn Steinbrink
     [not found]               ` <20070120072755.GA4652@atjola.homenet>
2007-01-20  7:50                 ` Björn Steinbrink
2007-01-20 18:50               ` Chr
2007-01-20 22:32               ` Chr
2007-01-21  1:50                 ` Robert Hancock
2007-01-21  3:34                   ` Jeff Garzik
2007-01-21  4:54                     ` Björn Steinbrink
2007-01-21  6:39                       ` Robert Hancock
2007-01-21  8:36                         ` Björn Steinbrink
2007-01-21 17:34                           ` Chr
2007-01-21 18:01                             ` Björn Steinbrink
2007-01-21 20:13                               ` Chr
2007-01-22  2:39                                 ` Tejun Heo
2007-01-22 12:32                                   ` Chr
2007-01-21 18:40                           ` Björn Steinbrink
2007-01-21 19:58                             ` Robert Hancock
2007-01-21 22:08                               ` Björn Steinbrink
2007-01-21 22:11                                 ` Björn Steinbrink
2007-01-21 22:26                                   ` Robert Hancock
2007-01-21 22:27                               ` Björn Steinbrink
2007-01-22  0:17                                 ` Robert Hancock
2007-01-22 16:12                                   ` Björn Steinbrink
2007-01-22 16:57                                     ` Björn Steinbrink
2007-01-22 17:53                                       ` Björn Steinbrink
2007-01-19 14:53         ` Alistair John Strachan
     [not found] <fa.1kBz5luWz8nR0lLqm1VD4hZZYdw@ifi.uio.no>
     [not found] ` <fa.QZxgjxcwtENaZNY24NMTlKBSgIM@ifi.uio.no>
     [not found]   ` <fa.fkPTbUGmKc/1pt0eD6TE4d02n+Q@ifi.uio.no>
     [not found]     ` <fa.6iQt5OtHZ3x5w8eYbLxwULhLTJ0@ifi.uio.no>
     [not found]       ` <fa.1aqo3IxNGJClHcBVZNTagX6bL9o@ifi.uio.no>
     [not found]         ` <fa.rI60BGlFbSyfLyumqmgiOfDqCI4@ifi.uio.no>
2007-01-23 23:18           ` Robert Hancock
2007-01-24  0:39             ` Björn Steinbrink
2007-02-03  1:42               ` Björn Steinbrink
2007-02-03  5:48                 ` Robert Hancock
2007-02-04  1:13                   ` Björn Steinbrink
2007-02-09 12:03                     ` Björn Steinbrink
2007-01-24  8:24             ` Ian Kumlien
2007-01-24 14:41               ` Björn Steinbrink
     [not found] <fa.ow4pXUncgdZmfLf3oyfrn1W+Bk0@ifi.uio.no>
     [not found] ` <fa.SoSeidhDuEr/K0kN+L4vW61Vpnc@ifi.uio.no>
     [not found]   ` <fa.m8QbVQMhuOshKzTdlSjNjOhaNcc@ifi.uio.no>
     [not found]     ` <fa.hDS02YCM8Tv1/STeTpJGEQD/49s@ifi.uio.no>
     [not found]       ` <fa.eqBbU9XvtTizNMpuUjctnk8vuOI@ifi.uio.no>
     [not found]         ` <fa.4QxeKMcmkoyhlL26AivZV6BFQJQ@ifi.uio.no>
2007-01-23  1:24           ` Robert Hancock
2007-01-23  1:34             ` Alistair John Strachan
2007-01-23  1:41               ` Robert Hancock
2007-01-23 15:29                 ` Larry Walton
2007-01-23  2:44             ` Björn Steinbrink
2007-01-23  5:03               ` Robert Hancock
2007-01-20 15:03 Ian Kumlien
2007-01-20 19:59 ` Robert Hancock
2007-01-20 21:43   ` Alistair John Strachan
2007-01-20 22:11     ` Ian Kumlien
     [not found] <fa.U/G88R1fWKOeQK3EBPHKK4MeRsQ@ifi.uio.no>
     [not found] ` <fa.2D0TIXbVTOgZmGg9ZJU+R7te70k@ifi.uio.no>
     [not found]   ` <fa.hMhdefkReYJ4idUyqqEWJFnWUBE@ifi.uio.no>
     [not found]     ` <fa.8TPWeOrcwkkHutPX5NOcJsTBO8Y@ifi.uio.no>
     [not found]       ` <fa.b92BqwV090pDj7q0iBG6BChksbI@ifi.uio.no>
     [not found]         ` <fa.O3RzvckSjB73Y0uL8P1nTXDRd6U@ifi.uio.no>
2007-01-19  0:09           ` Robert Hancock
2007-01-19  0:52             ` Björn Steinbrink
  -- strict thread matches above, loose matches on Subject: below --
2007-01-14 22:44 Björn Steinbrink
2007-01-15  6:48 ` Mikael Pettersson
2007-01-15 13:43   ` Jeff Garzik
2007-01-15 13:47   ` Björn Steinbrink

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=311601c90701221017v65866b6dm565d09bc756bdf3b@mail.gmail.com \
    --to=edmudama@gmail.com \
    --cc=B.Steinbrink@gmx.de \
    --cc=hancockr@shaw.ca \
    --cc=htejun@gmail.com \
    --cc=jeff@garzik.org \
    --cc=jens.axboe@oracle.com \
    --cc=linux-kernel@vger.kernel.org \
    --subject='Re: SATA exceptions with 2.6.20-rc5' \
    /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

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).