Date: Sun, 14 Nov 2004 13:49:32 +0100
From: Ingo Molnar <mingo@elte.hu>
To: Gunther Persoons <gunther_persoons@spymac.com>
Cc: linux-kernel@vger.kernel.org
Subject: Re: [patch] Real-Time Preemption, -RT-2.6.10-rc1-mm3-V0.7.25-1
Message-ID: <20041114124932.GB11042@elte.hu>
References: <20041103105840.GA3992@elte.hu> <20041106155720.GA14950@elte.hu> <20041108091619.GA9897@elte.hu> <20041108165718.GA7741@elte.hu> <20041109160544.GA28242@elte.hu> <20041111144414.GA8881@elte.hu> <20041111215122.GA5885@elte.hu> <41951380.2080801@spymac.com> <20041112201936.GA15133@elte.hu> <41961C03.10607@spymac.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <41961C03.10607@spymac.com>
User-Agent: Mutt/1.4.1i
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1008
Lines: 22


* Gunther Persoons <gunther_persoons@spymac.com> wrote:

> As i thought the init scripts were my problem. But i have an other
> question. I recently started to use NFS. But with the mainline kernel
> cpu usage is 100%, and when i look in top si shows bewteen 40 and 60%
> cpu usage. With your kernel si is 0%, but ksoftriqd/0 shows around 38%
> cpu usage and total cpu usage is around 52%. Is this normal? on my
> server cpu usage is 2% but it uses a intel network card. My laptop is
> using a wireless pcmcia card (cisco).

normally the RT kernel has higher system overhead (all IRQ traffic goes
to separate thread contexts, involving context-switching, etc.) so a
_reduction_ in system overhead looks a bit strange. Is there a
difference in performance?

	Ingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/