Message-ID: <4AB25A20.3040103@kernel.org>
Date: Fri, 18 Sep 2009 00:47:44 +0900
From: Tejun Heo <tj@kernel.org>
User-Agent: Thunderbird 2.0.0.22 (X11/20090605)
MIME-Version: 1.0
To: Mark Lord <liml@rtr.ca>
CC: Chris Webb <chris@arachsys.com>, Ric Wheeler <rwheeler@redhat.com>,
       Andrei Tanas <andrei@tanas.ca>, NeilBrown <neilb@suse.de>,
       linux-kernel@vger.kernel.org,
       IDE/ATA development list <linux-ide@vger.kernel.org>,
       linux-scsi@vger.kernel.org, Jeff Garzik <jgarzik@redhat.com>,
       Mark Lord <mlord@pobox.com>
Subject: Re: MD/RAID time out writing superblock
References: <20090916222842.GB16053@arachsys.com> <4AB17905.90606@kernel.org> <4AB23B17.2040204@rtr.ca>
In-Reply-To: <4AB23B17.2040204@rtr.ca>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1019
Lines: 30

Hello,

Mark Lord wrote:
>> That's interesting.  One important side effect of issuing IDENTIFY is
>> that they will serialize command streams as they are not NCQ commands
>> and thus could change command patterns significantly.
> ..
> 
> SMART is the opcode that is most frequently implicated here, not IDENTIFY.

Yeap, any non-NCQ commands would do it.

> Note that even a barrier FLUSH CACHE is non NCQ and will serialize
> the stream.

Yeah, I was just thinking that issuing non-NCQ commands mixed with NCQ
commands would make the command stream fluctuate more.  Modern drives
are pretty good at lowering power consumption while idle so being more
fluctuative (is it a word?) might have something to do with the
problem.  Just a wild speculation tho.

Thanks.

-- 
tejun
--
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/