From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([140.186.70.92]:48833) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1QPF7a-0004CH-M4 for qemu-devel@nongnu.org; Wed, 25 May 2011 10:31:39 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1QPF7W-0000eS-BH for qemu-devel@nongnu.org; Wed, 25 May 2011 10:31:38 -0400 Received: from aaar.vm.bytemark.co.uk ([80.68.92.230]:56720) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1QPF7W-0000e2-3I for qemu-devel@nongnu.org; Wed, 25 May 2011 10:31:34 -0400 From: Ian Campbell In-Reply-To: <4DDCBC6A.1060107@siemens.com> References: <4DD64E1F.8020603@siemens.com> <4DD8ECB0.9050803@web.de> <1306253200.20576.186.camel@zakaz.uk.xensource.com> <4DDBDA51.7010104@siemens.com> <1306311214.20576.201.camel@zakaz.uk.xensource.com> <4DDCBC6A.1060107@siemens.com> Content-Type: text/plain; charset="UTF-8" Date: Wed, 25 May 2011 09:34:57 +0100 Message-ID: <1306312497.20576.209.camel@zakaz.uk.xensource.com> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH v2 1/2] Generalize -machine command line option List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Jan Kiszka Cc: Anthony PERARD , Anthony Liguori , qemu-devel On Wed, 2011-05-25 at 10:23 +0200, Jan Kiszka wrote: > On 2011-05-25 10:13, Ian Campbell wrote: > > On Tue, 2011-05-24 at 18:18 +0200, Jan Kiszka wrote: > >> On 2011-05-24 18:06, Ian Campbell wrote: > >> Well, first of all I think this revealed a Xen bug because it crashes > >> when you try to run xenfv with an inappropriate accelerator, no? What is > >> the result of -machine xenfv,accel=tcg or, without my patch, -M xenfv > >> -machine accel=tcg? > > > > Unsurprisingly it crashed... > > > > I'm not sure if this is a bug in the xen side of simply a case of > > providing enough rope. I didn't really follow the threads which resulted > > in the accel stuff all that closely so I'm not sure what the intention > > was, it seems to me that the accel option is invalid with certain > > machine types though. > > > > I suppose in theory -machine xenpv,accel=kvm might result in xenner or > > something, accel=xenpv,tcg I'm less sure about (perhaps xenner too?). > > For -machine xenfv I don't expect anything other than accel=xen makes > > much sense. > > The point is that crashing is always a very poor way of reporting a > misconfiguration to the user. I bet you are able to detect and report > that more gracefully, e.g. during xenfv machine init. ACK. How about: 8<------------------------------------ >>From b23cadda8da72da480fe7cc8e31ffccd449f4b47 Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Wed, 25 May 2011 09:32:07 +0100 Subject: [PATCH] Fail to initialise Xen HVM support gracefully if Xen acceleration not enabled. This is preferable to the existing segmentation fault due to use of uninitialised xen_xc context, this is normally inialisted in xen_init which is not called if xen acceleration is not enabled. Signed-off-by: Ian Campbell Cc: Jan Kiszka Cc: Anthony PERARD --- xen-all.c | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/xen-all.c b/xen-all.c index 377aff7..c04bcbf 100644 --- a/xen-all.c +++ b/xen-all.c @@ -783,6 +783,12 @@ int xen_hvm_init(void) unsigned long ioreq_pfn; XenIOState *state; + if (!xen_xc) { + errno = ENOTSUP; + perror("xen: xen hvm requires accel=xen"); + return -errno; + } + state = qemu_mallocz(sizeof (XenIOState)); state->xce_handle = xen_xc_evtchn_open(NULL, 0); -- 1.7.2.5 -- Ian Campbell Current Noise: Blind Melon - No Rain Who are you?