diff mbox series

[v3,3/7] x86: KVM: Advertise AMX-FP16 CPUID to user space

Message ID 20221110015252.202566-4-jiaxi.chen@linux.intel.com (mailing list archive)
State New, archived
Headers show
Series x86: KVM: Advertise CPUID of new Intel platform | expand

Commit Message

Jiaxi Chen Nov. 10, 2022, 1:52 a.m. UTC
Latest Intel platform Granite Rapids has introduced a new instruction -
AMX-FP16, which performs dot-products of two FP16 tiles and accumulates
the results into a packed single precision tile. This instrucion needs
no additional enabling on top of the existing kernel AMX enabling.

AMX-FP16 adds FP16 capability and also allows a FP16 GPU trained model
to run faster without loss of accuracy or added SW overhead.

The bit definition:
CPUID.(EAX=7,ECX=1):EAX[bit 21]

This CPUID is exposed to user space. Besides, there is no other VMX
control for this instruction.

Signed-off-by: Jiaxi Chen <jiaxi.chen@linux.intel.com>
---
 arch/x86/kvm/cpuid.c         | 2 +-
 arch/x86/kvm/reverse_cpuid.h | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)
diff mbox series

Patch

diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index b388ef52f8c8..19ef02d5b11b 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -657,7 +657,7 @@  void kvm_set_cpu_caps(void)
 		kvm_cpu_cap_set(X86_FEATURE_SPEC_CTRL_SSBD);
 
 	kvm_cpu_cap_init_scattered(CPUID_7_1_EAX,
-		F(AVX_VNNI) | F(AVX512_BF16) | F(CMPCCXADD)
+		F(AVX_VNNI) | F(AVX512_BF16) | F(CMPCCXADD) | F(AMX_FP16)
 	);
 
 	kvm_cpu_cap_mask(CPUID_D_1_EAX,
diff --git a/arch/x86/kvm/reverse_cpuid.h b/arch/x86/kvm/reverse_cpuid.h
index 24f570ddb225..05fd43ebd226 100644
--- a/arch/x86/kvm/reverse_cpuid.h
+++ b/arch/x86/kvm/reverse_cpuid.h
@@ -29,6 +29,7 @@  enum kvm_only_cpuid_leafs {
 #define X86_FEATURE_AVX_VNNI            KVM_X86_FEATURE(CPUID_7_1_EAX, 4)
 #define X86_FEATURE_AVX512_BF16         KVM_X86_FEATURE(CPUID_7_1_EAX, 5)
 #define X86_FEATURE_CMPCCXADD           KVM_X86_FEATURE(CPUID_7_1_EAX, 7)
+#define X86_FEATURE_AMX_FP16            KVM_X86_FEATURE(CPUID_7_1_EAX, 21)
 
 struct cpuid_reg {
 	u32 function;