public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@suse.de>
To: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Cc: martin@dalecki.de, linux-kernel@vger.kernel.org
Subject: Re: please DON'T run 2.5.27 with IDE!
Date: Wed, 24 Jul 2002 15:36:52 +0200	[thread overview]
Message-ID: <20020724133652.GH15201@suse.de> (raw)
In-Reply-To: <Pine.SOL.4.30.0207241531200.15605-100000@mion.elka.pw.edu.pl>

On Wed, Jul 24 2002, Bartlomiej Zolnierkiewicz wrote:
> 
> On Wed, 24 Jul 2002, Jens Axboe wrote:
> 
> > On Wed, Jul 24 2002, Marcin Dalecki wrote:
> > > Jens Axboe wrote:
> > >
> > > >>>>2.5.27:drivers/block/ll_rw_blk.c
> > > >>>>void blk_start_queue(request_queue_t *q)
> > > >>>>{
> > > >>>>       if (test_and_clear_bit(QUEUE_FLAG_STOPPED, &q->queue_flags)) {
> > > >>>>               unsigned long flags;
> > > >>>>
> > > >>>>               spin_lock_irqsave(q->queue_lock, flags);
> > > >>>>               if (!elv_queue_empty(q))
> > > >>>>                       q->request_fn(q);
> > > >>>>               spin_unlock_irqrestore(q->queue_lock, flags);
> > > >>>>       }
> > > >>>>}
> > >
> > > >There were buggy versions at one point, however they may not have made it
> > > >into a full release. In that case it was just bk version of 2.5.19-pre
> > > >effectively. I forget the details :-)
> > >
> > > Naj - it's far more trivial I just looked at wrong tree at hand...
> > > But anyway. What happens if somone does set QUEUE_FLAG_STOPPED
> > > between the test_and_claer_bit and taking the spin_lock? Setting
> > > the QUEUE_FLAG_STOPPED isn't maintaining the spin_lock protection!
> >
> > It doesn't matter. If QUEUE_FLAG_STOPPED was set when entering
> > blk_start_queue(), it will call into the request_fn. If blk_stop_queue()
> > is called between clearing QUEUE_FLAG_STOPPED in blk_start_queue() and
> > grabbing the spin_lock, the worst that can happen is a spurios extra
> > request_fn call.
> >
> > > My goal is to make sure that the QUEUE_FLAG_STOPPED has a valid value
> > > *inside* the q->request_fn call.
> >
> > So you want the queue_lock to protect the flags as well... I don't
> > really see the point of this.
> 
> If driver corectly uses blk_start/stop_queue() it is not needed.
> Whole point of introducing this flag was not to take lock to check
> status of queue.

Right. The way it was meant to be used it how cpqarray and cciss does
it -- stopping queue in request_fn, restarting it from isr.

-- 
Jens Axboe


  reply	other threads:[~2002-07-24 13:33 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-07-22 19:37 please DON'T run 2.5.27 with IDE! Bartlomiej Zolnierkiewicz
2002-07-22 20:39 ` Andries Brouwer
2002-07-22 23:25   ` Thunder from the hill
2002-07-23  0:39 ` A Guy Called Tyketto
2002-07-23  0:58   ` Thunder from the hill
2002-07-23  1:10     ` Bartlomiej Zolnierkiewicz
2002-07-23  8:03 ` Morten Helgesen
2002-07-23 12:47   ` Bartlomiej Zolnierkiewicz
2002-07-23 13:00     ` Marcin Dalecki
2002-07-23 13:42       ` Bartlomiej Zolnierkiewicz
2002-07-23 13:58         ` Marcin Dalecki
2002-07-23 19:52           ` Jan Harkes
2002-07-23 20:08             ` Andre Hedrick
2002-07-24 10:24             ` Marcin Dalecki
2002-07-23 20:24           ` Bartlomiej Zolnierkiewicz
2002-07-24 10:30             ` Marcin Dalecki
2002-07-24 10:54               ` Bartlomiej Zolnierkiewicz
2002-07-24 11:35                 ` Marcin Dalecki
2002-07-24 11:53                   ` Jens Axboe
2002-07-24 12:08                     ` Marcin Dalecki
2002-07-24 12:39                   ` Bartlomiej Zolnierkiewicz
2002-07-24 12:41                     ` Jens Axboe
2002-07-24 12:49                       ` Bartlomiej Zolnierkiewicz
2002-07-24 12:50                         ` Jens Axboe
2002-07-24 13:08                           ` Marcin Dalecki
2002-07-24 13:25                             ` Jens Axboe
2002-07-24 13:35                               ` Bartlomiej Zolnierkiewicz
2002-07-24 13:36                                 ` Jens Axboe [this message]
2002-07-24 13:38                                   ` Marcin Dalecki
2002-07-24 13:35                               ` Marcin Dalecki
2002-07-24 12:43                   ` Bartlomiej Zolnierkiewicz
2002-07-24 13:10                     ` Marcin Dalecki
2002-07-24 13:21                       ` Bartlomiej Zolnierkiewicz
  -- strict thread matches above, loose matches on Subject: below --
2002-07-22 19:43 Petr Vandrovec
2002-07-22 19:46 ` Bartlomiej Zolnierkiewicz

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=20020724133652.GH15201@suse.de \
    --to=axboe@suse.de \
    --cc=B.Zolnierkiewicz@elka.pw.edu.pl \
    --cc=linux-kernel@vger.kernel.org \
    --cc=martin@dalecki.de \
    /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