Received: from eggs.gnu.org ([209.51.188.92]:51858) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1hMd8Z-00070a-Fg for qemu-devel@nongnu.org; Fri, 03 May 2019 14:42:24 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1hMd8X-00024l-6T for qemu-devel@nongnu.org; Fri, 03 May 2019 14:42:23 -0400 Received: from mx1.redhat.com ([209.132.183.28]:54532) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1hMd8W-00023P-09 for qemu-devel@nongnu.org; Fri, 03 May 2019 14:42:20 -0400 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.phx2.redhat.com [10.5.11.14]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 1757681F11 for ; Fri, 3 May 2019 18:42:19 +0000 (UTC) Date: Fri, 3 May 2019 15:42:06 -0300 From: Eduardo Habkost Message-ID: <20190503184206.GU28722@habkost.net> References: <20181220054037.24320-1-peterx@redhat.com> <20181220054037.24320-2-peterx@redhat.com> <20190426132744.2b594bf5@x1.home> <20190426150236.1af2ff08@x1.home> <94415012.15677076.1556342950794.JavaMail.zimbra@redhat.com> <20190429082106.4fd59e77@x1.home> <20190429145556.GA28722@habkost.net> <20190429092212.3b03e4bb@x1.home> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20190429092212.3b03e4bb@x1.home> Subject: Re: [Qemu-devel] [PATCH v2 1/3] q35: set split kernel irqchip as default List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alex Williamson Cc: Paolo Bonzini , Igor Mammedov , "Michael S . Tsirkin" , qemu-devel@nongnu.org, Peter Xu On Mon, Apr 29, 2019 at 09:22:12AM -0600, Alex Williamson wrote: [...] > > > What's a good 4.0.1 strategy to resolve this? Re-instate KVM irqchip > > > as the Q35 default? I can't see that simply switching to current QEMU > > > handling is a viable option for performance? What about 4.1? We could > > > certainly improve EOI support in QEMU, there's essentially no support > > > currently, but it seems like an uphill battle for an iothread based > > > userspace ioapic to ever compare to KVM handling? Thanks, > > > > irqchip=split and irqchip=kernel aren't guest ABI compatible, are > > they? That would make it impossible to fix this in pc-q35-4.0 > > for a 4.0.1 update. > > I suppose it would require a pc-q35-4.0.1 machine type :-\ Thanks, I wonder if it's possible to untangle this and make the irqchip option stop affecting guest ABI on 4.1+ machine-types? This way QEMU could choose smarter defaults in the future without the compatibility code hassle. -- Eduardo