From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kiszka Subject: Re: [PATCH] KVM: nVMX: Fix IRQs inject to L2 which belong to L1 since race Date: Fri, 04 Jul 2014 12:52:49 +0200 Message-ID: <53B68781.9080200@siemens.com> References: <1404284054-51863-1-git-send-email-wanpeng.li@linux.intel.com> <53B3CA6A.4050902@siemens.com> <20140703065955.GA4236@kernel> <20140704025250.GA2849@kernel> <53B63EF2.6000800@siemens.com> <53B674CD.2090906@siemens.com> <53B67602.4040604@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Gleb Natapov , Hu Robert , kvm@vger.kernel.org, linux-kernel@vger.kernel.org To: Paolo Bonzini , Wanpeng Li , Bandan Das Return-path: In-Reply-To: <53B67602.4040604@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On 2014-07-04 11:38, Paolo Bonzini wrote: > Il 04/07/2014 11:33, Jan Kiszka ha scritto: >> >> The compiler is not aware of the fact that push/pop exists in this >> function and, thus, places the vmcs parameter on the stack without >> reserving the space. So the pushfq will overwrite the vmcs pointer and >> let the function fail. > > Is that just a missing "memory" clobber? push/pop clobbers memory. Nope, we would needs some clobber like "stack". I wonder what is required to use push in inline assembly safely? Jan -- Siemens AG, Corporate Technology, CT RTC ITP SES-DE Corporate Competence Center Embedded Linux