[Top] [All Lists]

Re: [MMTests] IO metadata on XFS

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: [MMTests] IO metadata on XFS
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Mon, 2 Jul 2012 02:32:26 -0400
Cc: Mel Gorman <mgorman@xxxxxxx>, linux-mm@xxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, linux-fsdevel@xxxxxxxxxxxxxxx, xfs@xxxxxxxxxxx
In-reply-to: <20120701235458.GM19223@dastard>
References: <20120620113252.GE4011@xxxxxxx> <20120629111932.GA14154@xxxxxxx> <20120629112505.GF14154@xxxxxxx> <20120701235458.GM19223@dastard>
User-agent: Mutt/1.5.21 (2010-09-15)
On Mon, Jul 02, 2012 at 09:54:58AM +1000, Dave Chinner wrote:
> That will be caused by the fact we changed all the metadata updates
> to be logged, which means a transaction every time .dirty_inode is
> called.
> This should mostly go away when XFS is converted to use .update_time
> rather than .dirty_inode to only issue transactions when the VFS
> updates the atime rather than every .dirty_inode call...

I think the patch to do that conversion still needs review..

> It increases the CPU overhead (dirty_inode can be called up to 4
> times per write(2) call, IIRC), so with limited numbers of
> threads/limited CPU power it will result in lower performance. Where
> you have lots of CPU power, there will be little difference in
> performance...

When I checked it it could only be called twice, and we'd already
optimize away the second call.  I'd defintively like to track down where
the performance changes happend, at least to a major version but even
better to a -rc or git commit.

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