xfs
[Top] [All Lists]

Re: [PATCH 2/3] xfs: use per-filesystem I/O completion workqueues

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: [PATCH 2/3] xfs: use per-filesystem I/O completion workqueues
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Thu, 25 Aug 2011 01:20:04 -0400
Cc: Christoph Hellwig <hch@xxxxxxxxxxxxx>, xfs@xxxxxxxxxxx
In-reply-to: <20110825004811.GK3162@dastard>
References: <20110824055924.139283426@xxxxxxxxxxxxxxxxxxxxxx> <20110824060150.001321834@xxxxxxxxxxxxxxxxxxxxxx> <20110825004811.GK3162@dastard>
User-agent: Mutt/1.5.21 (2010-09-15)
On Thu, Aug 25, 2011 at 10:48:11AM +1000, Dave Chinner wrote:
> The only issue I see with this is that it brings back per-filesystem
> workqueue threads. Because all the workqueues are defined with
> MEM_RECLAIM, there is a rescuer thread per workqueue that is used
> when the CWMQ cannot allocate memory to queue the work to the
> appropriate per-cpu queue.

Not much we can do about it.

> If we are going to block here, then we probably should increase the
> per-cpu concurrency of the work queue so that we can continue to
> process other ioends while this one is blocked.

True.

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