xfs
[Top] [All Lists]

Re: [PATCH 5/5] xfs: kick inode writeback when low on memory

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: [PATCH 5/5] xfs: kick inode writeback when low on memory
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Wed, 2 Mar 2011 09:12:20 -0500
Cc: xfs@xxxxxxxxxxx, chris.mason@xxxxxxxxxx
In-reply-to: <20110302030602.GD4905@dastard>
References: <1298412969-14389-1-git-send-email-david@xxxxxxxxxxxxx> <1298412969-14389-6-git-send-email-david@xxxxxxxxxxxxx> <20110302030602.GD4905@dastard>
User-agent: Mutt/1.5.21 (2010-09-15)
On Wed, Mar 02, 2011 at 02:06:02PM +1100, Dave Chinner wrote:
> I'm open to ideas here - I could convert the bdi flusher
> infrastructure to cmwqs rather than using worker threads, or move
> all dirty inode tracking and writeback into XFS, or ???

Tejun posted patches to convert the writeback threads to workqueues.
But I think sooner or later we should stop using VFS dirty state for
metadata.  By allowing the dirty_inode operation to return a value
and say it shouldn't be marked dirty that could be done relatively
easily.

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