From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752105AbXCUAOV (ORCPT ); Tue, 20 Mar 2007 20:14:21 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752093AbXCUAOV (ORCPT ); Tue, 20 Mar 2007 20:14:21 -0400 Received: from smtp-outbound-1.vmware.com ([65.113.40.141]:42948 "EHLO smtp-outbound-1.vmware.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752034AbXCUAOU (ORCPT ); Tue, 20 Mar 2007 20:14:20 -0400 Message-ID: <46008705.6010007@vmware.com> Date: Tue, 20 Mar 2007 17:14:45 -0800 From: Zachary Amsden User-Agent: Thunderbird 1.5.0.10 (X11/20070221) MIME-Version: 1.0 To: Jeremy Fitzhardinge CC: Matt Mackall , Linus Torvalds , "Eric W. Biederman" , Rusty Russell , Andi Kleen , David Miller , mingo@elte.hu, akpm@linux-foundation.org, linux-kernel@vger.kernel.org, virtualization@lists.osdl.org, xen-devel@lists.xensource.com, chrisw@sous-sol.org, anthony@codemonkey.ws, netdev@vger.kernel.org Subject: Re: [patch 13/26] Xen-paravirt_ops: Consistently wrap paravirt ops callsites to make them patchable References: <1174272469.11680.23.camel@localhost.localdomain> <1174348905.11680.54.camel@localhost.localdomain> <45FF4043.4000805@vmware.com> <45FF770C.7050301@goop.org> <46000C7E.4070001@goop.org> <20070320224324.GL10459@waste.org> <46006963.1020401@vmware.com> <46006F66.4030901@goop.org> In-Reply-To: <46006F66.4030901@goop.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Jeremy Fitzhardinge wrote: > Zachary Amsden wrote: > >> I think Jeremy's idea was to have interrupt handlers leave interrupts >> disabled on exit if pda.intr_mask was set. In which case, they would >> bypass all work and we could never get preempted. >> > > Yes, I was worried that if we left the isr without actually handling the > interrupt, it would still be asserted and we'd just get interrupted > again. The idea is that we avoid touching cli/sti for the common case > of no interrupts while interrupts are disabled, but we'd still need to > fall back to using them if an interrupt becomes pending. > > >> I don't think leaving hardware interrupts disabled for such a long >> time is good though. >> > > How long? It would be no longer than now, and possibly less, wouldn't it? > Hmm. Perhaps. Something about the asymmetry bothers me alot though. Zach