x86/hvm: Conditionally leave CPUID Faulting active in HVM context
authorAndrew Cooper <andrew.cooper3@citrix.com>
Mon, 9 Jan 2017 13:42:02 +0000 (13:42 +0000)
committerAndrew Cooper <andrew.cooper3@citrix.com>
Tue, 24 Jan 2017 10:38:35 +0000 (10:38 +0000)
commit9bd6b01f9d466a5836633599934d6637b965133f
tree813150d51c08d925a91dd1165b6940cb59950607
parente225a1c7c06037e4f938efa43d4407e7abb088c1
x86/hvm: Conditionally leave CPUID Faulting active in HVM context

If the hardware supports faulting, and the guest has chosen to use it, leave
faulting active in HVM context.

It is more efficient to have hardware convert CPUID to a #GP fault (which we
don't intercept), than to take a VMExit and have Xen re-inject a #GP fault.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
xen/arch/x86/cpu/intel.c
xen/arch/x86/hvm/vmx/vmx.c