From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: Reset problem vs. MMIO emulation, hypercalls, etc... Date: Wed, 08 Aug 2012 15:42:02 +0300 Message-ID: <50225E9A.4070808@redhat.com> References: <20120803174113.GA13174@amt.cnet> <1344033008.24037.67.camel@pasglop> <20120806031344.GG16664@truffala.fritz.box> <1344286677.24037.100.camel@pasglop> <20120807013228.GL16664@truffala.fritz.box> <5020D5EB.9060104@redhat.com> <20120807121442.GN16664@truffala.fritz.box> <5021148D.4000107@redhat.com> <20120808004948.GO16664@truffala.fritz.box> <50222A52.6010201@redhat.com> <20120808115943.GU16664@truffala.fritz.box> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Benjamin Herrenschmidt , Marcelo Tosatti , kvm@vger.kernel.org, Alexander Graf , Paul Mackerras , kvm-ppc@vger.kernel.org To: David Gibson Return-path: In-Reply-To: <20120808115943.GU16664@truffala.fritz.box> Sender: kvm-ppc-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On 08/08/2012 02:59 PM, David Gibson wrote: >> >> No, you're correct. HLT could have been emulated in userspace, it just >> wasn't. The correct statement is that HLT was arbitrarily chosen to be >> emulated in userspace with the synchronous model, but the asynchronous >> model forced it into the kernel. > > Aha! Ok, understood. Uh, assuming you meant kernelspace, not > userspace in the first line there, anyway. I did. > Ok, so I am now reassured > that our current handling of CEDE in kernelspace does not cause > problems. Great. It's a real pity the original local APIC implementation was in userspace, it causes no end of confusion, and in fact was broken for smp until recently even though it is 7 years old. -- error compiling committee.c: too many arguments to function