* [Qemu-devel] -cpu check, rdtscp and AMD KVM @ 2015-11-12 10:53 Paolo Bonzini 2015-11-12 15:25 ` Eduardo Habkost 0 siblings, 1 reply; 6+ messages in thread From: Paolo Bonzini @ 2015-11-12 10:53 UTC (permalink / raw) To: Eduardo Habkost, qemu-devel Eduardo, running QEMU 2.5 on SVM causes a warning because rdtscp is not supported by KVM. While I can fix it, that leaves a warning on all existing kernels. What do you think is the best way forward? Paolo ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Qemu-devel] -cpu check, rdtscp and AMD KVM 2015-11-12 10:53 [Qemu-devel] -cpu check, rdtscp and AMD KVM Paolo Bonzini @ 2015-11-12 15:25 ` Eduardo Habkost 2015-11-12 15:58 ` Paolo Bonzini 0 siblings, 1 reply; 6+ messages in thread From: Eduardo Habkost @ 2015-11-12 15:25 UTC (permalink / raw) To: Paolo Bonzini; +Cc: qemu-devel On Thu, Nov 12, 2015 at 11:53:44AM +0100, Paolo Bonzini wrote: > Eduardo, > > running QEMU 2.5 on SVM causes a warning because rdtscp is not supported > by KVM. While I can fix it, that leaves a warning on all existing kernels. You mean when using Opteron_G2 or newer, right? qemu64 and kvm64 don't have rdtscp enabled. > > What do you think is the best way forward? To simply remove rdtscp from all Opteron_G* models? -- Eduardo ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Qemu-devel] -cpu check, rdtscp and AMD KVM 2015-11-12 15:25 ` Eduardo Habkost @ 2015-11-12 15:58 ` Paolo Bonzini 2015-11-12 16:27 ` Eduardo Habkost 0 siblings, 1 reply; 6+ messages in thread From: Paolo Bonzini @ 2015-11-12 15:58 UTC (permalink / raw) To: Eduardo Habkost; +Cc: qemu-devel On 12/11/2015 16:25, Eduardo Habkost wrote: > On Thu, Nov 12, 2015 at 11:53:44AM +0100, Paolo Bonzini wrote: >> Eduardo, >> >> running QEMU 2.5 on SVM causes a warning because rdtscp is not supported >> by KVM. While I can fix it, that leaves a warning on all existing kernels. > > You mean when using Opteron_G2 or newer, right? qemu64 and kvm64 > don't have rdtscp enabled. > >> What do you think is the best way forward? > > To simply remove rdtscp from all Opteron_G* models? Not sure this is the right thing to do... Real hardware has it, and going forward KVM will provide it. Paolo ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Qemu-devel] -cpu check, rdtscp and AMD KVM 2015-11-12 15:58 ` Paolo Bonzini @ 2015-11-12 16:27 ` Eduardo Habkost 2015-11-12 16:35 ` Paolo Bonzini 0 siblings, 1 reply; 6+ messages in thread From: Eduardo Habkost @ 2015-11-12 16:27 UTC (permalink / raw) To: Paolo Bonzini; +Cc: qemu-devel On Thu, Nov 12, 2015 at 04:58:37PM +0100, Paolo Bonzini wrote: > On 12/11/2015 16:25, Eduardo Habkost wrote: > > On Thu, Nov 12, 2015 at 11:53:44AM +0100, Paolo Bonzini wrote: > >> Eduardo, > >> > >> running QEMU 2.5 on SVM causes a warning because rdtscp is not supported > >> by KVM. While I can fix it, that leaves a warning on all existing kernels. > > > > You mean when using Opteron_G2 or newer, right? qemu64 and kvm64 > > don't have rdtscp enabled. > > > >> What do you think is the best way forward? > > > > To simply remove rdtscp from all Opteron_G* models? > > Not sure this is the right thing to do... Real hardware has it, and > going forward KVM will provide it. Do you see any alternative? We need AMD CPU models that can run out of the box using today's kernels. As no existing VMs running Opteron_G* on AMD CPUs have rdtscp, I believe it makes sense to just define Opteron_G* without rdtscp. When we add SVM rdtscp support to KVM, we can add new "Opteron_G[2-5]-rdtscp" CPU models. -- Eduardo ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Qemu-devel] -cpu check, rdtscp and AMD KVM 2015-11-12 16:27 ` Eduardo Habkost @ 2015-11-12 16:35 ` Paolo Bonzini 2015-11-12 16:48 ` Eduardo Habkost 0 siblings, 1 reply; 6+ messages in thread From: Paolo Bonzini @ 2015-11-12 16:35 UTC (permalink / raw) To: Eduardo Habkost; +Cc: qemu-devel On 12/11/2015 17:27, Eduardo Habkost wrote: >> > > To simply remove rdtscp from all Opteron_G* models? > > > > Not sure this is the right thing to do... Real hardware has it, and > > going forward KVM will provide it. > > Do you see any alternative? Live with the warning, and document it in the release notes. > We need AMD CPU models that can run > out of the box using today's kernels. As no existing VMs running > Opteron_G* on AMD CPUs have rdtscp, I believe it makes sense to > just define Opteron_G* without rdtscp. > > When we add SVM rdtscp support to KVM, we can add new > "Opteron_G[2-5]-rdtscp" CPU models. Makes sense too. However, I'm a bit afraid of the interaction with libvirt. Right now, libvirt has rdtscp in the description. If we remove it from libvirt, libvirt will start adding +rdtscp to the QEMU CPU command line option, so our change will be moot. And if we do not remove it from libvirt, libvirt will not be able to start a VM with rdtscp on a fixed kernel. Is this correct? Paolo ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [Qemu-devel] -cpu check, rdtscp and AMD KVM 2015-11-12 16:35 ` Paolo Bonzini @ 2015-11-12 16:48 ` Eduardo Habkost 0 siblings, 0 replies; 6+ messages in thread From: Eduardo Habkost @ 2015-11-12 16:48 UTC (permalink / raw) To: Paolo Bonzini; +Cc: libvir-list, qemu-devel (CCing libvir-list) On Thu, Nov 12, 2015 at 05:35:59PM +0100, Paolo Bonzini wrote: > On 12/11/2015 17:27, Eduardo Habkost wrote: > >> > > To simply remove rdtscp from all Opteron_G* models? > > > > > > Not sure this is the right thing to do... Real hardware has it, and > > > going forward KVM will provide it. > > > > Do you see any alternative? > > Live with the warning, and document it in the release notes. That's an option too. But it will cause breakage when upgrading the host kernel, and force users to live with a broken setup because we don't provide CPU models that work. In addition to providing CPU models that work, I believe it's better to make QEMU CPU models match what is already happening in practice with the existing VMs. > > > We need AMD CPU models that can run > > out of the box using today's kernels. As no existing VMs running > > Opteron_G* on AMD CPUs have rdtscp, I believe it makes sense to > > just define Opteron_G* without rdtscp. > > > > When we add SVM rdtscp support to KVM, we can add new > > "Opteron_G[2-5]-rdtscp" CPU models. > > Makes sense too. > > However, I'm a bit afraid of the interaction with libvirt. Right now, > libvirt has rdtscp in the description. If we remove it from libvirt, > libvirt will start adding +rdtscp to the QEMU CPU command line option, > so our change will be moot. And if we do not remove it from libvirt, > libvirt will not be able to start a VM with rdtscp on a fixed kernel. The former is not true (the change won't be moot). The latter seems true. AFAIK, the presence of rdtscp in cpu_map.xml only does two things: * Makes libvirt never use "-cpu Opteron_G2,+rdtscp" even if the user explicitly asked for rdtscp (which is a libvirt bug that should be fixed); * Makes libvirt refuse to run Opteron_G2 in hosts that lack rdtscp (which is also a libvirt bug, because it's checking host CPUID directly instead of GET_SUPPORTED_CPUID; libvirt must let QEMU check if the feature is really available). -- Eduardo ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2015-11-12 16:48 UTC | newest] Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2015-11-12 10:53 [Qemu-devel] -cpu check, rdtscp and AMD KVM Paolo Bonzini 2015-11-12 15:25 ` Eduardo Habkost 2015-11-12 15:58 ` Paolo Bonzini 2015-11-12 16:27 ` Eduardo Habkost 2015-11-12 16:35 ` Paolo Bonzini 2015-11-12 16:48 ` Eduardo Habkost
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.