public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@kernel.dk>
To: Vivek Goyal <vgoyal@redhat.com>
Cc: Tao Ma <tm@tao.ma>, Shaohua Li <shaohua.li@intel.com>,
	linux-kernel@vger.kernel.org
Subject: Re: CFQ: async queue blocks the whole system
Date: Fri, 10 Jun 2011 11:21:57 +0200	[thread overview]
Message-ID: <4DF1E235.8010104@kernel.dk> (raw)
In-Reply-To: <20110610092031.GD4183@redhat.com>

On 2011-06-10 11:20, Vivek Goyal wrote:
> On Fri, Jun 10, 2011 at 11:02:11AM +0800, Tao Ma wrote:
> 
> [..]
>>> I don't think we can give a deadline for async request, because we
>>> still want to give sync high priority. We can give async some slices,
>>> so for a workload of small number of async requests and large number
>>> sync requests, we don't starve async too much. But for a workload with
>>> large number of sync/async requests, async will be starved for sure
>>> and we can't solve this in cfq.
>> OK, so if you guys thinks a 500 seconds wait is good for an async write
>> to complete, fine, then we have to switch to deadline.
> 
> I don't think that starving WRITES completely is a good idea. Especially
> given the fact that you were not able to dispatch WRITES for 500 seconds.
> This needs fixing.
> 
> Its not about giving hard deadline to WRITES, but making sure we don't
> starve them completely and they also make some progress.

Agree, we need to have some sort of forward progress guarantee at least.
Starving buffered writes indefinitely is surely a BUG. And it must be a
regression from not that long ago.

Trying to catch up with this thread...

-- 
Jens Axboe


  reply	other threads:[~2011-06-10  9:22 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-09 10:49 CFQ: async queue blocks the whole system Tao Ma
2011-06-09 14:14 ` Vivek Goyal
2011-06-09 14:34   ` Jens Axboe
2011-06-09 14:47   ` Tao Ma
2011-06-09 15:37     ` Vivek Goyal
2011-06-09 15:44       ` Tao Ma
2011-06-09 18:27         ` Vivek Goyal
2011-06-10  5:48           ` Tao Ma
2011-06-10  9:14             ` Vivek Goyal
2011-06-10 10:00               ` Tao Ma
2011-06-10 15:44                 ` Vivek Goyal
2011-06-11  7:24                   ` Tao Ma
2011-06-13 10:08                   ` Tao Ma
2011-06-13 21:41                     ` Vivek Goyal
2011-06-14  7:03                       ` Tao Ma
2011-06-14 13:30                         ` Vivek Goyal
2011-06-14 15:42                           ` Tao Ma
2011-06-14 21:14                             ` Vivek Goyal
2011-06-17  3:04                   ` Tao Ma
2011-06-17 12:50                     ` Vivek Goyal
2011-06-17 14:34                       ` Tao Ma
2011-06-10  1:19       ` Shaohua Li
2011-06-10  1:34         ` Shaohua Li
2011-06-10  2:06           ` Tao Ma
2011-06-10  2:35             ` Shaohua Li
2011-06-10  3:02               ` Tao Ma
2011-06-10  9:20                 ` Vivek Goyal
2011-06-10  9:21                   ` Jens Axboe [this message]
2011-06-13  1:03                   ` Shaohua Li
2011-06-10  9:17         ` Vivek Goyal
2011-06-10  9:20           ` Jens Axboe
2011-06-10  9:29             ` Vivek Goyal
2011-06-10  9:31               ` 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=4DF1E235.8010104@kernel.dk \
    --to=axboe@kernel.dk \
    --cc=linux-kernel@vger.kernel.org \
    --cc=shaohua.li@intel.com \
    --cc=tm@tao.ma \
    --cc=vgoyal@redhat.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