pagg
[Top] [All Lists]

Re: CpuSet on PAGG (Re: PAGG in Open Source projects?)

To: Kaigai Kohei <kaigai@xxxxxxxxxxxxx>
Subject: Re: CpuSet on PAGG (Re: PAGG in Open Source projects?)
From: Guillaume Thouvenin <guillaume.thouvenin@xxxxxxxx>
Date: Fri, 28 Jan 2005 14:27:44 +0100
Cc: Paul Jackson <pj@xxxxxxx>, erikj@xxxxxxxxxxxxxxxxxxxxxxx, pagg@xxxxxxxxxxx, limin@xxxxxxxxxxxxxxxxxx, LSE-Tech <lse-tech@xxxxxxxxxxxxxxxxxxxxx>, guillaume.thouvenin@xxxxxxxx
In-reply-to: <41FA330A.2030303@ak.jp.nec.com>
References: <Pine.SGI.4.53.0501181437280.627920@subway.americas.sgi.com> <41F8E117.5030501@ak.jp.nec.com> <20050127081753.5a9d16af.pj@sgi.com> <41FA330A.2030303@ak.jp.nec.com>
Sender: pagg-bounce@xxxxxxxxxxx
On Fri, 2005-01-28 at 21:41 +0900, Kaigai Kohei wrote:
> But my main subject is not this point. The purpose of those patches is
> to restrain incrementation of hook functions in fork() or exit().
> I used PAGG for this, as a common event handling framework.

I agree with this point. It seems that several applications need hook
functions in fork() or/and exit(). I can give example like CSA, ELSA,
CKRM, CpuSet, LSM or Dprobes. Thus, if I need a hook in fork() for my
accounting application, ELSA for example , and if I don't want to add my
own hook, PAGG is a solution. AFAIU, I can't use LSM hooks because it's
a security framework, I can't use Dprobes because it's a debugging
framework and the hooks used by CpuSet and CKRM don't allow any
registration. There was also another project called kernelhooks (the
former GKHI I think) but I don't know if it's still maintained...

Best,
Guillaume


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