Date: Wed, 10 Mar 2004 22:02:07 +0100
From: Jens Axboe <axboe@suse.de>
To: Andrew Morton <akpm@osdl.org>
Cc: linux-kernel@vger.kernel.org, kenneth.w.chen@intel.com,
       thornber@redhat.com
Subject: Re: [PATCH] backing dev unplugging
Message-ID: <20040310210207.GL15087@suse.de>
References: <20040310124507.GU4949@suse.de> <20040310130046.2df24f0e.akpm@osdl.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20040310130046.2df24f0e.akpm@osdl.org>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 944
Lines: 26

On Wed, Mar 10 2004, Andrew Morton wrote:
> Jens Axboe <axboe@suse.de> wrote:
> >
> > Here's a first cut at killing global plugging of block devices to reduce
> > the nasty contention blk_plug_lock caused.
> 
> Shouldn't we take read_lock(&md->map_lock) in dm_table_unplug_all()?

Ugh yes, we certainly should.

> If so, what are the lock ranking issues here?  The queue lock is not
> held yet, so it seems pretty simple?

As far as I can tell, it's pretty straight forward. The unplug_fn() will
grab the queue lock for 'ordinary' devices, for dm on dm you'd nest the
maplock inside each other (which should be quite alright, as far as I
can tell, without pulling any nasty tricks).

-- 
Jens Axboe

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/