From: Keir Fraser Subject: Re: [PATCH] x86: allow NMI injection Date: Wed, 28 Feb 2007 11:59:58 +0000 Message-ID: References: <45E57803.76E4.0078.0@novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <45E57803.76E4.0078.0@novell.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Jan Beulich , xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org On 28/2/07 11:39, "Jan Beulich" wrote: > NetWare's internal debugger needs the ability to send NMI IPIs, and > there is no reason to not allow domUs or dom0's vCPUs other than vCPU 0 > to handle NMIs (they just will never see hardware generated ones). > > While currently not having a frontend, the added hypercall also allows > for being used to inject NMIs into foreign VMs. > > Along the lines, this fixes a potential race condition caused by > previously accessing the VCPU flags field non-atomically in entry.S. > > Signed-off-by: Jan Beulich Notably this patch changes the way that a NMI handler is registered to use the native-like vector 2. This changes the guest interface though. Do you really need to be able to specify a custom CS? Can you not vector to the flat CS and then far jump? I'm not sure about making the IPI function a physdev_op(), since this is still a virtual NMI (it has nothing to do with real hardware NMIs). It might be better to make it a vcpu_op. Then it would not be a great fit to allow send-to-all send-to-all-butself overrides, but I'm not sure how important that optimisation is (or is to make the NMI deliveries as simultaneous as possible)? -- Keir