xfs
[Top] [All Lists]

Re: [PATCH] xfs: improve metadata I/O merging in the elevator

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: [PATCH] xfs: improve metadata I/O merging in the elevator
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Mon, 16 Nov 2009 06:05:23 -0500
Cc: Christoph Hellwig <hch@xxxxxxxxxxxxx>, xfs@xxxxxxxxxxx
In-reply-to: <20091116035019.GI9467@xxxxxxxxxxxxxxxx>
References: <20091112190930.GB32110@xxxxxxxxxxxxx> <20091116035019.GI9467@xxxxxxxxxxxxxxxx>
User-agent: Mutt/1.5.19 (2009-01-05)
On Mon, Nov 16, 2009 at 02:50:19PM +1100, Dave Chinner wrote:
> Definitely should be done, but....
> 
> It looks like the patch you posted isn't quite doing what was
> intended - async write buffers are being classified as WRITE, not
> WRITE_META. That means we get more write combining in the elevator
> (performance increase) like with WRITE_META, but don't get the
> faster dispatch (latency reduction) by using the META queue to keep
> the metadata writeback separate from the bulk data writeback.
> That may be why deadline is not showing any improvement...
> 
> FWIW, the original patch here:
> 
> http://oss.sgi.com/archives/xfs/2008-01/msg00630.html
> 
> uses WRITE_META, but it looks like you've taken bits of this
> patch:

Indeed.  I'l re-add the write-side markings and will re-bench and
re-submit.

<Prev in Thread] Current Thread [Next in Thread>