From mboxrd@z Thu Jan 1 00:00:00 1970 From: Zachary Amsden Subject: Re: [PATCH] paravirt.h Date: Tue, 22 Aug 2006 19:12:11 -0700 Message-ID: <44EBB97B.9030707@vmware.com> References: <1155202505.18420.5.camel@localhost.localdomain> <44DB7596.6010503@goop.org> <1156254965.27114.17.camel@localhost.localdomain> <200608221544.26989.ak@muc.de> <44EB3BF0.3040805@vmware.com> <1156271386.2976.102.camel@laptopd505.fenrus.org> <1156275004.27114.34.camel@localhost.localdomain> <44EB584A.5070505@vmware.com> <44EB5A76.9060402@vmware.com> <44EB7F0C.60402@vmware.com> <1156298131.12015.42.camel@localhost.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1156298131.12015.42.camel@localhost.localdomain> Sender: linux-kernel-owner@vger.kernel.org To: Rusty Russell Cc: Andi Kleen , Andrew Morton , virtualization@lists.osdl.org, Chris Wright , Linux Kernel Mailing List , Arjan van de Ven List-Id: virtualization@lists.linuxfoundation.org Rusty Russell wrote: > On Tue, 2006-08-22 at 15:02 -0700, Zachary Amsden wrote: > >> Well, I don't think anything is sufficient for a preemptible kernel. I >> think that's just plain not going to work. You could have a kernel >> thread that got preempted in a paravirt-op patch point >> > > Patching over the 6 native cases is actually not that bad: they're > listed below (each one has trailing noops). > > cli > sti > push %eax; popf > pushf; pop %eax > pushf; pop %eax; cli > iret > sti; sysexit > > If you're at the first insn you don't have to do anything, since you're > about to replace that code. If you're in the noops, you can just > advance EIP to the end. You can't be preempted between sti and sysexit, > since we only use that when interrupts are already disabled. And > reversing either "push %eax" or "pushf; pop %eax" is fairly easy. > > Depending on your hypervisor, you might need to catch those threads who > are currently doing the paravirt_ops function calls, as well. This > introduces more (and more complex) cases. > Yes, but the problem gets far worse. You don't need to worry about just those. You need to worry about all that C code that runs in the native paravirt-ops as well, because you could have preempted it in the middle of a callout. And the paravirt_ops code isn't isolated in a separate section (though it well could be). Zach