From mboxrd@z Thu Jan  1 00:00:00 1970
From: Erik Mouw <mouw@nl.linux.org>
Subject: Re: interrupt processing
Date: Mon, 2 Jul 2007 17:55:41 +0200
Message-ID: <20070702155541.GG6356@gateway.home>
References: <c565abbb0706150746w4bf83459n864347729823b191@mail.gmail.com>
Mime-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature"; boundary="8S1fMsFYqgBC+BN/"
Return-path: <linux-smp-owner@vger.kernel.org>
Content-Disposition: inline
In-Reply-To: <c565abbb0706150746w4bf83459n864347729823b191@mail.gmail.com>
Sender: linux-smp-owner@vger.kernel.org
List-Id: <linux-smp.vger.kernel.org>
To: Matthew Faulkner <matthew.faulkner@gmail.com>
Cc: linux-smp@vger.kernel.org


--8S1fMsFYqgBC+BN/
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Fri, Jun 15, 2007 at 03:46:39PM +0100, Matthew Faulkner wrote:
> Probably a very simple question, but can anyone tell me how the kernel
> decides which CPU (or core) a hardware interrupt gets handled by?

It programs the APIC. IRQs can be handled by a random CPU, by a set of
CPUs, or by a single CPU.

You can have some influence on the APIC IRQ routing by writing
/proc/irq/<irqno>/smp_affinity .


Erik

--=20
They're all fools. Don't worry. Darwin may be slow, but he'll
eventually get them. -- Matthew Lammers in alt.sysadmin.recovery

--8S1fMsFYqgBC+BN/
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFGiR/9/PlVHJtIto0RAh1iAJ43Ypuls603z/uTZ9YtYyCCSJ7hXwCeM2hu
Jxf4RsAwwQftVtJHMefEgmU=
=FoB/
-----END PGP SIGNATURE-----

--8S1fMsFYqgBC+BN/--