From af2872f6225476566bcbbd523a74dcaba29e159e Mon Sep 17 00:00:00 2001 From: "Chang S. Bae" Date: Fri, 25 Nov 2022 20:58:41 +0800 Subject: x86: KVM: Advertise AMX-FP16 CPUID to user space 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. 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] AMX-FP16 is on an expected-dense CPUID leaf and some other bits on this leaf have kernel usages. Given that, define this feature bit like X86_FEATURE_ in kernel. Considering AMX-FP16 itself has no truly kernel usages and /proc/cpuinfo has too much unreadable flags, hide this one in /proc/cpuinfo. Advertise AMX-FP16 to KVM userspace. This is safe because there are no new VMX controls or additional host enabling required for guests to use this feature. Signed-off-by: Chang S. Bae Signed-off-by: Jiaxi Chen Acked-by: Borislav Petkov Message-Id: <20221125125845.1182922-5-jiaxi.chen@linux.intel.com> Signed-off-by: Paolo Bonzini --- arch/x86/include/asm/cpufeatures.h | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86/include/asm/cpufeatures.h') diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 5cdd57133d90..20059dc33d24 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -309,6 +309,7 @@ #define X86_FEATURE_AVX_VNNI (12*32+ 4) /* AVX VNNI instructions */ #define X86_FEATURE_AVX512_BF16 (12*32+ 5) /* AVX512 BFLOAT16 instructions */ #define X86_FEATURE_CMPCCXADD (12*32+ 7) /* "" CMPccXADD instructions */ +#define X86_FEATURE_AMX_FP16 (12*32+21) /* "" AMX fp16 Support */ /* AMD-defined CPU features, CPUID level 0x80000008 (EBX), word 13 */ #define X86_FEATURE_CLZERO (13*32+ 0) /* CLZERO instruction */ -- cgit v1.2.3