Re: [Qemu-devel] [PATCH 2/2] Expose tsc deadline timer cpuid to guest

[Eduardo Habkost]

Jan/Avi: ping?

I would like to get this ABI detail clarified so it can be implemented
the right way on Qemu and KVM.

My proposal is to simply add tsc-deadline to the data returned by
GET_SUPPORTED_CPUID, making KVM_CAP_TSC_DEADLINE_TIMER unnecessary.


On Fri, Mar 23, 2012 at 02:17:52PM +0000, Liu, Jinsong wrote:
> Eduardo Habkost wrote:
> > On Fri, Mar 23, 2012 at 03:49:27AM +0000, Liu, Jinsong wrote:
> >> Eduardo Habkost wrote:
> >>> [1] From Documentation/virtual/kvm/api.txt:
> >>> 
> >>> "KVM_GET_SUPPORTED_CPUID
> >>> [...]
> >>> This ioctl returns x86 cpuid features which are supported by both
> >>> the hardware and kvm.  Userspace can use the information returned
> >>> by this ioctl to construct cpuid information (for KVM_SET_CPUID2)
> >>> that is consistent with hardware, kernel, and userspace
> >>>                                   capabilities, and with
> >>> ^^^^^^^^^^^^^^^^^^^^^^^^^^ 
> >>> user requirements (for example, the user may wish to constrain cpuid
> >>> to emulate older hardware, or for feature consistency across a
> >>> cluster)."
> >> 
> >> The fixbug patch is implemented by Jan and Avi, I reply per my
> >> understanding. 
> > 
> > No problem. I hope Jan or Avi can clarify this.
> > 
> >> 
> >> I think for tsc deadline timer feature, KVM_CAP_TSC_DEADLINE_TIMER is
> >> slightly better than KVM_GET_SUPPORTED_CPUID. If use
> >> KVM_GET_SUPPORTED_CPUID, it means tsc deadline features bind to host
> >> cpuid, while it fact it could be pure software emulated by kvm
> >> (though currently it implemented as bound to hareware). For the sake
> >> of 
> >> extension, it choose KVM_CAP_TSC_DEADLINE_TIMER.
> > 
> > There's no requirement for GET_SUPPORTED_CPUID to be a subset of the
> > host CPU features. If KVM can completely emulate the feature by
> > software, then it can return the feature on GET_SUPPORTED_CPUID even
> > if the host CPU doesn't have the feature. That's the case for x2apic,
> > for example (see commit 0d1de2d901f4ba0972a3886496a44fb1d3300dbd).
> 
> 
> Jan/Avi,
> 
> Could you elaborate more your thought? 
> 
> Thanks,
> Jinsong

-- 
Eduardo