From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Hounschell Subject: processor isolation inquiry Date: Sun, 21 Dec 2008 09:53:32 -0500 Message-ID: <494E586C.7070107@cfl.rr.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Max Krasnyanskiy To: linux-kernel , linux-kernel-rt Return-path: Received: from cdptpa-omtalb.mail.rr.com ([75.180.132.121]:54788 "EHLO cdptpa-omtalb.mail.rr.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753260AbYLUQhS (ORCPT ); Sun, 21 Dec 2008 11:37:18 -0500 Sender: linux-rt-users-owner@vger.kernel.org List-ID: It's been some time now since Max K. brought up the complete processor isolation discussion. I know his original patch was NAK'd but some work on this was put into cpusets. I'm unclear as to what actually has been or will be implemented. The 2 things I'm most interested in are work queues, and local timer interrupts. Is it yet possible to isolate a processor from either of these 2 things with recent kernels? The kernel doc concerning cpusets doesn't seem to clarify this for me. The Process shielding instructions at http://rt.wiki.kernel.org/index.php/CPU_shielding_using_/proc_and_/dev/cpuset don't seem to help with work queues or local timer interrupts. Should they? Thanks in advance for any pointers Mark