From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753558Ab2GTTW5 (ORCPT ); Fri, 20 Jul 2012 15:22:57 -0400 Received: from mx1.redhat.com ([209.132.183.28]:54918 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753337Ab2GTTWy (ORCPT ); Fri, 20 Jul 2012 15:22:54 -0400 Date: Fri, 20 Jul 2012 16:04:54 -0300 From: Marcelo Tosatti To: "Michael S. Tsirkin" Cc: Avi Kivity , gleb@redhat.com, kvm@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCHv2] kvm: fix race with level interrupts Message-ID: <20120720190454.GA24024@amt.cnet> References: <20120719104520.GA14748@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20120719104520.GA14748@redhat.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Jul 19, 2012 at 01:45:20PM +0300, Michael S. Tsirkin wrote: > When more than 1 source id is in use for the same GSI, we have the > following race related to handling irq_states race: > > CPU 0 clears bit 0. CPU 0 read irq_state as 0. CPU 1 sets level to 1. > CPU 1 calls kvm_ioapic_set_irq(1). CPU 0 calls kvm_ioapic_set_irq(0). > Now ioapic thinks the level is 0 but irq_state is not 0. > > Fix by performing all irq_states bitmap handling under pic/ioapic lock. > This also removes the need for atomics with irq_states handling. > > Reported-by: Gleb Natapov > Signed-off-by: Michael S. Tsirkin > --- Applied, thanks. > Changes from v1: > Address comments by Gleb and Alex: > renamed some variables for clarify > renamed kvm_irq_line_state -> __kvm_irq_line_state > > Any chance we can put this in 3.5? I know level IRQs are not widely > used, which is likely why this went unnoticed for so long, but still ... http://yarchive.net/comp/linux/merge_window.html "From: Linus Torvalds The thing is, I don't take bug fixes late in the -rc just because they are bug fixes. And I really shouldn't. If it's an old bug, and doesn't cause an oops or a security issue, it had damn well better wait for the next merge window. There is absolutely _no_ reason to just blindly "fix bugs" at the end of the rc stage, because quite frankly, the risks coming from fixing a bug is often bigger than the advantage. "