From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1NGB0h-0006Rp-0Y for qemu-devel@nongnu.org; Thu, 03 Dec 2009 07:42:15 -0500 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1NGB0b-0006Q3-P7 for qemu-devel@nongnu.org; Thu, 03 Dec 2009 07:42:14 -0500 Received: from [199.232.76.173] (port=46555 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1NGB0a-0006Ps-V9 for qemu-devel@nongnu.org; Thu, 03 Dec 2009 07:42:09 -0500 Received: from mx1.redhat.com ([209.132.183.28]:13256) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1NGB0a-0001VL-Gu for qemu-devel@nongnu.org; Thu, 03 Dec 2009 07:42:08 -0500 Date: Thu, 3 Dec 2009 14:39:24 +0200 From: "Michael S. Tsirkin" Subject: Re: [Qemu-devel] Re: [PATCH 0/9] in-kernel irqchip, new spin Message-ID: <20091203123924.GF7352@redhat.com> References: <1259761575-3953-1-git-send-email-glommer@redhat.com> <20091203102701.GC6752@redhat.com> <5d6222a80912030342m499eb913lff5fc7a21b142718@mail.gmail.com> <20091203115729.GC7352@redhat.com> <5d6222a80912030404m2cc09946wf3391f10227efcee@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <5d6222a80912030404m2cc09946wf3391f10227efcee@mail.gmail.com> List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Glauber Costa Cc: Glauber Costa , avi@redhat.com, aliguori@us.ibm.com, qemu-devel@nongnu.org On Thu, Dec 03, 2009 at 10:04:21AM -0200, Glauber Costa wrote: > >> Haven't tested. But since I see that msix need some special code in qemu-kvm, > >> it probably won't. But I assume we can just add a patch ontop of this to add > >> that code and make it work, right? > > > > Sure. However - are you making in-kernel irqchip the default? > > If so, I think the best way to do it would be: > > 1. add in-kernel irqchip, off by default > > 2. fix msix with in-kernel > > 3. make in-kernel on by default > > > Right now there are no knobs to disable it, since last time I checked, > people were inclined to > solve that by adding a machine type that does not do irqchip in > kernel, if wanted. Can't everything doable by machine type also doable from command line switch? Disabling irqchip used to be a valuable debugging tool. > However, since it is probably not going to reach 0.12 anyway, we can > come up with a patch > that fixes msix, and bundle in this series before I actually enable > the irqchip (which is one > of the last patches) Fine. > -- > Glauber Costa. > "Free as in Freedom" > http://glommer.net > > "The less confident you are, the more serious you have to act."