Re: [PATCH v1 8/8] kvm:cpuid Report CET SHSTK and IBT support in CPUID.(EAX=0x7,ECX=0).

[Sean Christopherson]

On Wed, Dec 26, 2018 at 04:15:32PM +0800, Yang Weijiang wrote:
> Guest OS can query CET SHSTK and IBT support by CPUID.(EAX=0x7,ECX=0),
> in return, ECX[bit 7] corresponds to SHSTK feature, and EDX[bit 20]
> corresponds to IBT feature.
> 
> Signed-off-by: Zhang Yi Z <yi.z.zhang@xxxxxxxxxxxxxxx>
> Signed-off-by: Yang Weijiang <weijiang.yang@xxxxxxxxx>
> ---
>  arch/x86/kvm/cpuid.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
> index 55c282f71f93..d7b60b3c4326 100644
> --- a/arch/x86/kvm/cpuid.c
> +++ b/arch/x86/kvm/cpuid.c
> @@ -67,6 +67,8 @@ u64 kvm_supported_xcr0(void)
>  
>  #define F(x) bit(X86_FEATURE_##x)
>  
> +#define CET_IBT_BIT bit(20)

Properly define X86_FEATURE_CET_IBT if it's not being defined by some
other series, even if KVM is the only immediate user of the bit.  Linux
reserves all of word 18 for CPUID_7_0_EDX, i.e. defining the bit won't
bloat some random scattered word.

> +
>  /* For scattered features from cpufeatures.h; we currently expose none */
>  #define KF(x) bit(KVM_CPUID_BIT_##x)
>  
> @@ -407,12 +409,12 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
>  		F(AVX512VBMI) | F(LA57) | F(PKU) | 0 /*OSPKE*/ |
>  		F(AVX512_VPOPCNTDQ) | F(UMIP) | F(AVX512_VBMI2) | F(GFNI) |
>  		F(VAES) | F(VPCLMULQDQ) | F(AVX512_VNNI) | F(AVX512_BITALG) |
> -		F(CLDEMOTE);
> +		F(CLDEMOTE) | F(SHSTK);
>  
>  	/* cpuid 7.0.edx*/
>  	const u32 kvm_cpuid_7_0_edx_x86_features =
>  		F(AVX512_4VNNIW) | F(AVX512_4FMAPS) | F(SPEC_CTRL) |
> -		F(SPEC_CTRL_SSBD) | F(ARCH_CAPABILITIES);
> +		F(SPEC_CTRL_SSBD) | F(ARCH_CAPABILITIES) | CET_IBT_BIT;
>  
>  	/* all calls to cpuid_count() should be made on the same cpu */
>  	get_cpu();
> -- 
> 2.17.1
>