Zhang, Yang Z <yang.z.zhang <at> intel.com> writes: > > Marcelo Tosatti wrote on 2013-01-24: > > On Wed, Jan 23, 2013 at 10:47:23PM +0800, Yang Zhang wrote: > >> From: Yang Zhang <yang.z.zhang <at> Intel.com> > >> > >> APIC virtualization is a new feature which can eliminate most of VM exit > >> when vcpu handle a interrupt: > >> > >> APIC register virtualization: > >> APIC read access doesn't cause APIC-access VM exits. > >> APIC write becomes trap-like. > >> Virtual interrupt delivery: > >> Virtual interrupt delivery avoids KVM to inject vAPIC interrupts > >> manually, which is fully taken care of by the hardware. > > > > Can you provide numbers, and also what tests have been performed. > This first patch is adding APIC register virtualization supporting. > The second patch is adding enable virtual x2apic mode supporting since it is required by APICv when guest > uses msr based way to access APIC. > Then third patch is is add virtual interrupt delivery supporting. > > The test cover booting different guests: include windows 2k3, win7, winxp, rhel6u3, rhel 5u5 and upstream > Linux. And also did some scp inside guest and migration. Didn't see any issues. > > Also, we did some performance measure with previous patch and it shows about 3% ~5% improvement. > Can you pl. elaborate on the performance setup and the specific tests that were done ? Also any improvements observed on the latency front when using virtio with these patches ? Any improvements observed with VFIO (i.e. in the absence of the posted interrupt related changes?) Thanks! Vinod > Best regards, > Yang > > -- > To unsubscribe from this list: send the line "unsubscribe kvm" in > the body of a message to majordomo <at> vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > > -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
