xfs
[Top] [All Lists]

Re: [PATCH 2/9]: Reduce Log I/O latency

To: Stewart Smith <stewart@xxxxxxxxx>
Subject: Re: [PATCH 2/9]: Reduce Log I/O latency
From: Andi Kleen <andi@xxxxxxxxxxxxxx>
Date: Thu, 22 Nov 2007 13:06:11 +0100
Cc: David Chinner <dgc@xxxxxxx>, Andi Kleen <andi@xxxxxxxxxxxxxx>, xfs-oss <xfs@xxxxxxxxxxx>, lkml <linux-kernel@xxxxxxxxxxxxxxx>
In-reply-to: <1195702123.8369.78.camel@xxxxxxxxxxxxxxxxxxxxx>
References: <20071122003339.GH114266761__34694.2978365861$1195691722$gmane$org@xxxxxxx> <p73oddnhzoq.fsf@xxxxxxxxxxxxxx> <20071122011214.GR114266761@xxxxxxx> <1195702123.8369.78.camel@xxxxxxxxxxxxxxxxxxxxx>
Sender: xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.4.2.1i
> FWIW from a "real time" database POV this seems to make sense to me...
> in fact, we probably rely on filesystem metadata way too much
> (historically it's just "worked".... although we do seem to get issues
> on ext3).

For that case you really would need priority inheritance: any metadata
IO on behalf or blocking a process needs to use the process' block IO 
priority.

David's change just fixes a limited set of cases, but breaks others.

-Andi


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