[Top] [All Lists]

Re: buffered writeback torture program

To: Andreas Dilger <adilger@xxxxxxxxx>
Subject: Re: buffered writeback torture program
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Thu, 21 Apr 2011 14:02:13 -0400
Cc: Christoph Hellwig <hch@xxxxxxxxxxxxx>, Chris Mason <chris.mason@xxxxxxxxxx>, linux-fsdevel <linux-fsdevel@xxxxxxxxxxxxxxx>, linux-ext4 <linux-ext4@xxxxxxxxxxxxxxx>, xfs <xfs@xxxxxxxxxxx>, jack <jack@xxxxxxx>, axboe <axboe@xxxxxxxxx>, dchinner <dchinner@xxxxxxxxxx>
In-reply-to: <93CB867E-B908-4B38-A146-A9DC958ACF64@xxxxxxxxx>
References: <1303322378-sup-1722@think> <20110421083258.GA26784@xxxxxxxxxxxxx> <1303407205-sup-6141@think> <20110421174120.GA7267@xxxxxxxxxxxxx> <93CB867E-B908-4B38-A146-A9DC958ACF64@xxxxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Thu, Apr 21, 2011 at 11:59:37AM -0600, Andreas Dilger wrote:
> But doesn't XFS have potentially very large extents, especially in the case 
> of files that were fallocate()'d or linearly written?  If there is a single 
> 8GB extent, and then random writes within that extent (seems very database 
> like) grouping the all of the writes in the extent doesn't seem so great.

It doesn't cluster any writes in an extent.  It only writes out
additional dirty pages directly following that one we were asked to
write out.  As soon as we hit a non-dirty page we give up.

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