xfs
[Top] [All Lists]

Re: 2.4.20pre5aa2

To: Dave Hansen <haveblue@xxxxxxxxxx>
Subject: Re: 2.4.20pre5aa2
From: Andrea Arcangeli <andrea@xxxxxxx>
Date: Mon, 16 Sep 2002 18:20:01 +0200
Cc: Samuel Flory <sflory@xxxxxxxxxxxx>, Stephen Lord <lord@xxxxxxx>, Austin Gonyou <austin@xxxxxxxxxxxxxxx>, Christian Guggenberger <christian.guggenberger@xxxxxxxxxxxxxxxxxxxxxxxx>, Linux Kernel <linux-kernel@xxxxxxxxxxxxxxx>, linux-xfs@xxxxxxxxxxx
In-reply-to: <3D8600DD.1010707@us.ibm.com>
References: <20020911201602.A13655@pc9391.uni-regensburg.de> <1031768655.24629.23.camel@UberGeek.coremetrics.com> <20020911184111.GY17868@dualathlon.random> <3D81235B.6080809@rackable.com> <20020913002316.GG11605@dualathlon.random> <1031878070.1236.29.camel@snafu> <20020913005440.GJ11605@dualathlon.random> <3D8149F6.9060702@rackable.com> <20020913125345.GO11605@dualathlon.random> <3D8600DD.1010707@us.ibm.com>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.3.27i
On Mon, Sep 16, 2002 at 09:03:41AM -0700, Dave Hansen wrote:
> +     vmi = get_vmalloc_info();

hmm, not sure if it's better to slowdown vmalloc instead of
/proc/meminfo and to keep meminfo o1. In theory vmalloc should be used
only for persistent infrequent allocations, so meminfo has a chance to
be recalled more frequently with monitors like xosview during workloads.
Admittedly in final production with no monitoring meminfo is going to
never be recalled, however I like the idea to keep meminfo very quick.

Andrea


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