* [PATCH 0/2] Enumerate and expose AVX_VNNI feature
@ 2021-01-05 0:49 Yang Zhong
2021-01-05 0:49 ` [PATCH 1/2] Enumerate AVX Vector Neural Network instructions Yang Zhong
` (2 more replies)
0 siblings, 3 replies; 10+ messages in thread
From: Yang Zhong @ 2021-01-05 0:49 UTC (permalink / raw)
To: linux-kernel, x86, kvm
Cc: tglx, mingo, bp, hpa, tony.luck, pbonzini, seanjc, vkuznets,
wanpengli, jmattson, joro, kyung.min.park, yang.zhong
A processor supports AVX_VNNI instructions if CPUID.(EAX=7,ECX=1):EAX[bit 4]
is present.
This series includes kernel and kvm patches, kernel patch define this
new cpu feature bit and kvm expose this bit to guest. When this bit is
enabled on cpu or vcpu, the cpu feature flag is shown as "avx_vnni" in
/proc/cpuinfo of host and guest.
Detailed information on the instruction and CPUID feature flag can be
found in the latest "extensions" manual [1].
Reference:
[1]. https://software.intel.com/content/www/us/en/develop/download/intel-architecture-instruction-set-extensions-programming-reference.html
Kyung Min Park (1):
Enumerate AVX Vector Neural Network instructions
Yang Zhong (1):
KVM: Expose AVX_VNNI instruction to guset
arch/x86/include/asm/cpufeatures.h | 1 +
arch/x86/kvm/cpuid.c | 2 +-
2 files changed, 2 insertions(+), 1 deletion(-)
--
2.29.2.334.gfaefdd61ec
^ permalink raw reply [flat|nested] 10+ messages in thread* [PATCH 1/2] Enumerate AVX Vector Neural Network instructions
2021-01-05 0:49 [PATCH 0/2] Enumerate and expose AVX_VNNI feature Yang Zhong
@ 2021-01-05 0:49 ` Yang Zhong
2021-01-05 11:47 ` Paolo Bonzini
2021-01-05 0:49 ` [PATCH 2/2] KVM: Expose AVX_VNNI instruction to guset Yang Zhong
2021-01-21 15:02 ` [PATCH 0/2] Enumerate and expose AVX_VNNI feature Paolo Bonzini
2 siblings, 1 reply; 10+ messages in thread
From: Yang Zhong @ 2021-01-05 0:49 UTC (permalink / raw)
To: linux-kernel, x86, kvm
Cc: tglx, mingo, bp, hpa, tony.luck, pbonzini, seanjc, vkuznets,
wanpengli, jmattson, joro, kyung.min.park, yang.zhong
From: Kyung Min Park <kyung.min.park@intel.com>
Add AVX version of the Vector Neural Network (VNNI) Instructions.
A processor supports AVX VNNI instructions if CPUID.0x07.0x1:EAX[4] is
present. The following instructions are available when this feature is
present.
1. VPDPBUS: Multiply and Add Unsigned and Signed Bytes
2. VPDPBUSDS: Multiply and Add Unsigned and Signed Bytes with Saturation
3. VPDPWSSD: Multiply and Add Signed Word Integers
4. VPDPWSSDS: Multiply and Add Signed Integers with Saturation
The only in-kernel usage of this is kvm passthrough. The CPU feature
flag is shown as "avx_vnni" in /proc/cpuinfo.
This instruction is currently documented in the latest "extensions"
manual (ISE). It will appear in the "main" manual (SDM) in the future.
Signed-off-by: Kyung Min Park <kyung.min.park@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
---
arch/x86/include/asm/cpufeatures.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h
index f5ef2d5b9231..d10d9962bd9b 100644
--- a/arch/x86/include/asm/cpufeatures.h
+++ b/arch/x86/include/asm/cpufeatures.h
@@ -293,6 +293,7 @@
#define X86_FEATURE_PER_THREAD_MBA (11*32+ 7) /* "" Per-thread Memory Bandwidth Allocation */
/* Intel-defined CPU features, CPUID level 0x00000007:1 (EAX), word 12 */
+#define X86_FEATURE_AVX_VNNI (12*32+ 4) /* AVX VNNI instructions */
#define X86_FEATURE_AVX512_BF16 (12*32+ 5) /* AVX512 BFLOAT16 instructions */
/* AMD-defined CPU features, CPUID level 0x80000008 (EBX), word 13 */
--
2.29.2.334.gfaefdd61ec
^ permalink raw reply related [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] Enumerate AVX Vector Neural Network instructions
2021-01-05 0:49 ` [PATCH 1/2] Enumerate AVX Vector Neural Network instructions Yang Zhong
@ 2021-01-05 11:47 ` Paolo Bonzini
2021-01-05 12:14 ` Borislav Petkov
0 siblings, 1 reply; 10+ messages in thread
From: Paolo Bonzini @ 2021-01-05 11:47 UTC (permalink / raw)
To: Yang Zhong, linux-kernel, kvm, bp
Cc: tglx, mingo, hpa, tony.luck, seanjc, vkuznets, wanpengli,
jmattson, joro, kyung.min.park, x86
On 05/01/21 01:49, Yang Zhong wrote:
> From: Kyung Min Park <kyung.min.park@intel.com>
>
> Add AVX version of the Vector Neural Network (VNNI) Instructions.
>
> A processor supports AVX VNNI instructions if CPUID.0x07.0x1:EAX[4] is
> present. The following instructions are available when this feature is
> present.
> 1. VPDPBUS: Multiply and Add Unsigned and Signed Bytes
> 2. VPDPBUSDS: Multiply and Add Unsigned and Signed Bytes with Saturation
> 3. VPDPWSSD: Multiply and Add Signed Word Integers
> 4. VPDPWSSDS: Multiply and Add Signed Integers with Saturation
>
> The only in-kernel usage of this is kvm passthrough. The CPU feature
> flag is shown as "avx_vnni" in /proc/cpuinfo.
>
> This instruction is currently documented in the latest "extensions"
> manual (ISE). It will appear in the "main" manual (SDM) in the future.
>
> Signed-off-by: Kyung Min Park <kyung.min.park@intel.com>
> Signed-off-by: Yang Zhong <yang.zhong@intel.com>
> Reviewed-by: Tony Luck <tony.luck@intel.com>
> ---
> arch/x86/include/asm/cpufeatures.h | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h
> index f5ef2d5b9231..d10d9962bd9b 100644
> --- a/arch/x86/include/asm/cpufeatures.h
> +++ b/arch/x86/include/asm/cpufeatures.h
> @@ -293,6 +293,7 @@
> #define X86_FEATURE_PER_THREAD_MBA (11*32+ 7) /* "" Per-thread Memory Bandwidth Allocation */
>
> /* Intel-defined CPU features, CPUID level 0x00000007:1 (EAX), word 12 */
> +#define X86_FEATURE_AVX_VNNI (12*32+ 4) /* AVX VNNI instructions */
> #define X86_FEATURE_AVX512_BF16 (12*32+ 5) /* AVX512 BFLOAT16 instructions */
>
> /* AMD-defined CPU features, CPUID level 0x80000008 (EBX), word 13 */
>
Boris, is it possible to have a topic branch for this patch?
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] Enumerate AVX Vector Neural Network instructions
2021-01-05 11:47 ` Paolo Bonzini
@ 2021-01-05 12:14 ` Borislav Petkov
2021-01-12 2:13 ` Yang Zhong
0 siblings, 1 reply; 10+ messages in thread
From: Borislav Petkov @ 2021-01-05 12:14 UTC (permalink / raw)
To: Paolo Bonzini
Cc: Yang Zhong, linux-kernel, kvm, tglx, mingo, hpa, tony.luck,
seanjc, vkuznets, wanpengli, jmattson, joro, kyung.min.park, x86
On Tue, Jan 05, 2021 at 12:47:23PM +0100, Paolo Bonzini wrote:
> On 05/01/21 01:49, Yang Zhong wrote:
> > From: Kyung Min Park <kyung.min.park@intel.com>
> >
> > Add AVX version of the Vector Neural Network (VNNI) Instructions.
> >
> > A processor supports AVX VNNI instructions if CPUID.0x07.0x1:EAX[4] is
> > present. The following instructions are available when this feature is
> > present.
> > 1. VPDPBUS: Multiply and Add Unsigned and Signed Bytes
> > 2. VPDPBUSDS: Multiply and Add Unsigned and Signed Bytes with Saturation
> > 3. VPDPWSSD: Multiply and Add Signed Word Integers
> > 4. VPDPWSSDS: Multiply and Add Signed Integers with Saturation
> >
> > The only in-kernel usage of this is kvm passthrough. The CPU feature
> > flag is shown as "avx_vnni" in /proc/cpuinfo.
> >
> > This instruction is currently documented in the latest "extensions"
> > manual (ISE). It will appear in the "main" manual (SDM) in the future.
> >
> > Signed-off-by: Kyung Min Park <kyung.min.park@intel.com>
> > Signed-off-by: Yang Zhong <yang.zhong@intel.com>
> > Reviewed-by: Tony Luck <tony.luck@intel.com>
> > ---
> > arch/x86/include/asm/cpufeatures.h | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h
> > index f5ef2d5b9231..d10d9962bd9b 100644
> > --- a/arch/x86/include/asm/cpufeatures.h
> > +++ b/arch/x86/include/asm/cpufeatures.h
> > @@ -293,6 +293,7 @@
> > #define X86_FEATURE_PER_THREAD_MBA (11*32+ 7) /* "" Per-thread Memory Bandwidth Allocation */
> > /* Intel-defined CPU features, CPUID level 0x00000007:1 (EAX), word 12 */
> > +#define X86_FEATURE_AVX_VNNI (12*32+ 4) /* AVX VNNI instructions */
> > #define X86_FEATURE_AVX512_BF16 (12*32+ 5) /* AVX512 BFLOAT16 instructions */
> > /* AMD-defined CPU features, CPUID level 0x80000008 (EBX), word 13 */
> >
>
> Boris, is it possible to have a topic branch for this patch?
Just take it through your tree pls.
Acked-by: Borislav Petkov <bp@suse.de>
Thx.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH 1/2] Enumerate AVX Vector Neural Network instructions
2021-01-05 12:14 ` Borislav Petkov
@ 2021-01-12 2:13 ` Yang Zhong
2021-01-13 12:54 ` Paolo Bonzini
0 siblings, 1 reply; 10+ messages in thread
From: Yang Zhong @ 2021-01-12 2:13 UTC (permalink / raw)
To: Paolo Bonzini
Cc: Borislav Petkov, linux-kernel, kvm, tglx, mingo, hpa, tony.luck,
seanjc, vkuznets, wanpengli, jmattson, joro, kyung.min.park, x86,
yang.zhong
On Tue, Jan 05, 2021 at 01:14:56PM +0100, Borislav Petkov wrote:
> On Tue, Jan 05, 2021 at 12:47:23PM +0100, Paolo Bonzini wrote:
> > On 05/01/21 01:49, Yang Zhong wrote:
> > > From: Kyung Min Park <kyung.min.park@intel.com>
> > >
> > > Add AVX version of the Vector Neural Network (VNNI) Instructions.
> > >
> > > A processor supports AVX VNNI instructions if CPUID.0x07.0x1:EAX[4] is
> > > present. The following instructions are available when this feature is
> > > present.
> > > 1. VPDPBUS: Multiply and Add Unsigned and Signed Bytes
> > > 2. VPDPBUSDS: Multiply and Add Unsigned and Signed Bytes with Saturation
> > > 3. VPDPWSSD: Multiply and Add Signed Word Integers
> > > 4. VPDPWSSDS: Multiply and Add Signed Integers with Saturation
> > >
> > > The only in-kernel usage of this is kvm passthrough. The CPU feature
> > > flag is shown as "avx_vnni" in /proc/cpuinfo.
> > >
> > > This instruction is currently documented in the latest "extensions"
> > > manual (ISE). It will appear in the "main" manual (SDM) in the future.
> > >
> > > Signed-off-by: Kyung Min Park <kyung.min.park@intel.com>
> > > Signed-off-by: Yang Zhong <yang.zhong@intel.com>
> > > Reviewed-by: Tony Luck <tony.luck@intel.com>
> > > ---
> > > arch/x86/include/asm/cpufeatures.h | 1 +
> > > 1 file changed, 1 insertion(+)
> > >
> > > diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h
> > > index f5ef2d5b9231..d10d9962bd9b 100644
> > > --- a/arch/x86/include/asm/cpufeatures.h
> > > +++ b/arch/x86/include/asm/cpufeatures.h
> > > @@ -293,6 +293,7 @@
> > > #define X86_FEATURE_PER_THREAD_MBA (11*32+ 7) /* "" Per-thread Memory Bandwidth Allocation */
> > > /* Intel-defined CPU features, CPUID level 0x00000007:1 (EAX), word 12 */
> > > +#define X86_FEATURE_AVX_VNNI (12*32+ 4) /* AVX VNNI instructions */
> > > #define X86_FEATURE_AVX512_BF16 (12*32+ 5) /* AVX512 BFLOAT16 instructions */
> > > /* AMD-defined CPU features, CPUID level 0x80000008 (EBX), word 13 */
> > >
> >
> > Boris, is it possible to have a topic branch for this patch?
>
> Just take it through your tree pls.
>
> Acked-by: Borislav Petkov <bp@suse.de>
>
Paolo, Boris has acked this kernel patch, and if i need send new patchset to add this
acked-by info ? or kvm tree will directly pull this patchset? thanks.
Yang
> Thx.
>
> --
> Regards/Gruss,
> Boris.
>
> https://people.kernel.org/tglx/notes-about-netiquette
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] Enumerate AVX Vector Neural Network instructions
2021-01-12 2:13 ` Yang Zhong
@ 2021-01-13 12:54 ` Paolo Bonzini
0 siblings, 0 replies; 10+ messages in thread
From: Paolo Bonzini @ 2021-01-13 12:54 UTC (permalink / raw)
To: Yang Zhong
Cc: Borislav Petkov, linux-kernel, kvm, tglx, mingo, hpa, tony.luck,
seanjc, vkuznets, wanpengli, jmattson, joro, kyung.min.park, x86
>>> Boris, is it possible to have a topic branch for this patch?
>>
>> Just take it through your tree pls.
>>
>> Acked-by: Borislav Petkov <bp@suse.de>
>>
>
> Paolo, Boris has acked this kernel patch, and if i need send new patchset to add this
> acked-by info ? or kvm tree will directly pull this patchset? thanks.
I'll take care of it shortly.
Paolo
^ permalink raw reply [flat|nested] 10+ messages in thread
* [PATCH 2/2] KVM: Expose AVX_VNNI instruction to guset
2021-01-05 0:49 [PATCH 0/2] Enumerate and expose AVX_VNNI feature Yang Zhong
2021-01-05 0:49 ` [PATCH 1/2] Enumerate AVX Vector Neural Network instructions Yang Zhong
@ 2021-01-05 0:49 ` Yang Zhong
2021-01-21 15:02 ` [PATCH 0/2] Enumerate and expose AVX_VNNI feature Paolo Bonzini
2 siblings, 0 replies; 10+ messages in thread
From: Yang Zhong @ 2021-01-05 0:49 UTC (permalink / raw)
To: linux-kernel, x86, kvm
Cc: tglx, mingo, bp, hpa, tony.luck, pbonzini, seanjc, vkuznets,
wanpengli, jmattson, joro, kyung.min.park, yang.zhong
Expose AVX (VEX-encoded) versions of the Vector Neural Network
Instructions to guest.
The bit definition:
CPUID.(EAX=7,ECX=1):EAX[bit 4] AVX_VNNI
The following instructions are available when this feature is
present in the guest.
1. VPDPBUS: Multiply and Add Unsigned and Signed Bytes
2. VPDPBUSDS: Multiply and Add Unsigned and Signed Bytes with Saturation
3. VPDPWSSD: Multiply and Add Signed Word Integers
4. VPDPWSSDS: Multiply and Add Signed Integers with Saturation
This instruction is currently documented in the latest "extensions"
manual (ISE). It will appear in the "main" manual (SDM) in the future.
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
---
arch/x86/kvm/cpuid.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index 83637a2ff605..4229b67f0a8d 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -433,7 +433,7 @@ void kvm_set_cpu_caps(void)
kvm_cpu_cap_set(X86_FEATURE_SPEC_CTRL_SSBD);
kvm_cpu_cap_mask(CPUID_7_1_EAX,
- F(AVX512_BF16)
+ F(AVX_VNNI) | F(AVX512_BF16)
);
kvm_cpu_cap_mask(CPUID_D_1_EAX,
--
2.29.2.334.gfaefdd61ec
^ permalink raw reply related [flat|nested] 10+ messages in thread
* Re: [PATCH 0/2] Enumerate and expose AVX_VNNI feature
2021-01-05 0:49 [PATCH 0/2] Enumerate and expose AVX_VNNI feature Yang Zhong
2021-01-05 0:49 ` [PATCH 1/2] Enumerate AVX Vector Neural Network instructions Yang Zhong
2021-01-05 0:49 ` [PATCH 2/2] KVM: Expose AVX_VNNI instruction to guset Yang Zhong
@ 2021-01-21 15:02 ` Paolo Bonzini
2021-01-22 6:43 ` Yang Zhong
2 siblings, 1 reply; 10+ messages in thread
From: Paolo Bonzini @ 2021-01-21 15:02 UTC (permalink / raw)
To: Yang Zhong, linux-kernel, x86, kvm
Cc: tglx, mingo, bp, hpa, tony.luck, seanjc, vkuznets, wanpengli,
jmattson, joro, kyung.min.park
On 05/01/21 01:49, Yang Zhong wrote:
> A processor supports AVX_VNNI instructions if CPUID.(EAX=7,ECX=1):EAX[bit 4]
> is present.
>
> This series includes kernel and kvm patches, kernel patch define this
> new cpu feature bit and kvm expose this bit to guest. When this bit is
> enabled on cpu or vcpu, the cpu feature flag is shown as "avx_vnni" in
> /proc/cpuinfo of host and guest.
>
> Detailed information on the instruction and CPUID feature flag can be
> found in the latest "extensions" manual [1].
>
> Reference:
> [1]. https://software.intel.com/content/www/us/en/develop/download/intel-architecture-instruction-set-extensions-programming-reference.html
>
>
> Kyung Min Park (1):
> Enumerate AVX Vector Neural Network instructions
>
> Yang Zhong (1):
> KVM: Expose AVX_VNNI instruction to guset
>
> arch/x86/include/asm/cpufeatures.h | 1 +
> arch/x86/kvm/cpuid.c | 2 +-
> 2 files changed, 2 insertions(+), 1 deletion(-)
>
Queued, thanks.
Paolo
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 0/2] Enumerate and expose AVX_VNNI feature
2021-01-21 15:02 ` [PATCH 0/2] Enumerate and expose AVX_VNNI feature Paolo Bonzini
@ 2021-01-22 6:43 ` Yang Zhong
0 siblings, 0 replies; 10+ messages in thread
From: Yang Zhong @ 2021-01-22 6:43 UTC (permalink / raw)
To: Paolo Bonzini
Cc: linux-kernel, x86, kvm, tglx, mingo, bp, hpa, tony.luck, seanjc,
vkuznets, wanpengli, jmattson, joro, kyung.min.park, yang.zhong
On Thu, Jan 21, 2021 at 04:02:17PM +0100, Paolo Bonzini wrote:
> On 05/01/21 01:49, Yang Zhong wrote:
> >A processor supports AVX_VNNI instructions if CPUID.(EAX=7,ECX=1):EAX[bit 4]
> >is present.
> >
> >This series includes kernel and kvm patches, kernel patch define this
> >new cpu feature bit and kvm expose this bit to guest. When this bit is
> >enabled on cpu or vcpu, the cpu feature flag is shown as "avx_vnni" in
> >/proc/cpuinfo of host and guest.
> >
> >Detailed information on the instruction and CPUID feature flag can be
> >found in the latest "extensions" manual [1].
> >
> >Reference:
> >[1]. https://software.intel.com/content/www/us/en/develop/download/intel-architecture-instruction-set-extensions-programming-reference.html
> >
> >
> >Kyung Min Park (1):
> > Enumerate AVX Vector Neural Network instructions
> >
> >Yang Zhong (1):
> > KVM: Expose AVX_VNNI instruction to guset
> >
> > arch/x86/include/asm/cpufeatures.h | 1 +
> > arch/x86/kvm/cpuid.c | 2 +-
> > 2 files changed, 2 insertions(+), 1 deletion(-)
> >
>
> Queued, thanks.
>
> Paolo
Paolo, thanks, i will send the related Qemu patch soon.
Yang
^ permalink raw reply [flat|nested] 10+ messages in thread
* [PATCH 0/2] Enumerate and expose AVX_VNNI feature
@ 2020-12-16 2:01 Yang Zhong
2020-12-16 2:01 ` [PATCH 1/2] Enumerate AVX Vector Neural Network instructions Yang Zhong
0 siblings, 1 reply; 10+ messages in thread
From: Yang Zhong @ 2020-12-16 2:01 UTC (permalink / raw)
To: linux-kernel, x86, kvm
Cc: tglx, mingo, bp, hpa, tony.luck, pbonzini, seanjc, vkuznets,
wanpengli, jmattson, joro, kyung.min.park, yang.zhong
A processor supports AVX_VNNI instructions if CPUID.(EAX=7,ECX=1):EAX[bit 4]
is present.
This series includes kernel and kvm patches, kernel patch define this
new cpu feature bit and kvm expose this bit to guest. When this bit is
enabled on cpu or vcpu, the cpu feature flag is shown as "avx_vnni" in
/proc/cpuinfo of host and guest.
Detailed information on the instruction and CPUID feature flag can be
found in the latest "extensions" manual [1].
Reference:
[1]. https://software.intel.com/content/www/us/en/develop/download/intel-architecture-instruction-set-extensions-programming-reference.html
Kyung Min Park (1):
Enumerate AVX Vector Neural Network instructions
Yang Zhong (1):
KVM: Expose AVX_VNNI instruction to guset
arch/x86/include/asm/cpufeatures.h | 1 +
arch/x86/kvm/cpuid.c | 2 +-
2 files changed, 2 insertions(+), 1 deletion(-)
--
2.29.2.334.gfaefdd61ec
^ permalink raw reply [flat|nested] 10+ messages in thread
* [PATCH 1/2] Enumerate AVX Vector Neural Network instructions
2020-12-16 2:01 Yang Zhong
@ 2020-12-16 2:01 ` Yang Zhong
0 siblings, 0 replies; 10+ messages in thread
From: Yang Zhong @ 2020-12-16 2:01 UTC (permalink / raw)
To: linux-kernel, x86, kvm
Cc: tglx, mingo, bp, hpa, tony.luck, pbonzini, seanjc, vkuznets,
wanpengli, jmattson, joro, kyung.min.park, yang.zhong
From: Kyung Min Park <kyung.min.park@intel.com>
Add AVX version of the Vector Neural Network (VNNI) Instructions.
A processor supports AVX VNNI instructions if CPUID.0x07.0x1:EAX[4] is
present. The following instructions are available when this feature is
present.
1. VPDPBUS: Multiply and Add Unsigned and Signed Bytes
2. VPDPBUSDS: Multiply and Add Unsigned and Signed Bytes with Saturation
3. VPDPWSSD: Multiply and Add Signed Word Integers
4. VPDPWSSDS: Multiply and Add Signed Integers with Saturation
The only in-kernel usage of this is kvm passthrough. The CPU feature
flag is shown as "avx_vnni" in /proc/cpuinfo.
This instruction is currently documented in the latest "extensions"
manual (ISE). It will appear in the "main" manual (SDM) in the future.
Signed-off-by: Kyung Min Park <kyung.min.park@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
---
arch/x86/include/asm/cpufeatures.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h
index f5ef2d5b9231..d10d9962bd9b 100644
--- a/arch/x86/include/asm/cpufeatures.h
+++ b/arch/x86/include/asm/cpufeatures.h
@@ -293,6 +293,7 @@
#define X86_FEATURE_PER_THREAD_MBA (11*32+ 7) /* "" Per-thread Memory Bandwidth Allocation */
/* Intel-defined CPU features, CPUID level 0x00000007:1 (EAX), word 12 */
+#define X86_FEATURE_AVX_VNNI (12*32+ 4) /* AVX VNNI instructions */
#define X86_FEATURE_AVX512_BF16 (12*32+ 5) /* AVX512 BFLOAT16 instructions */
/* AMD-defined CPU features, CPUID level 0x80000008 (EBX), word 13 */
--
2.29.2.334.gfaefdd61ec
^ permalink raw reply related [flat|nested] 10+ messages in thread
end of thread, other threads:[~2021-01-22 6:55 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-01-05 0:49 [PATCH 0/2] Enumerate and expose AVX_VNNI feature Yang Zhong
2021-01-05 0:49 ` [PATCH 1/2] Enumerate AVX Vector Neural Network instructions Yang Zhong
2021-01-05 11:47 ` Paolo Bonzini
2021-01-05 12:14 ` Borislav Petkov
2021-01-12 2:13 ` Yang Zhong
2021-01-13 12:54 ` Paolo Bonzini
2021-01-05 0:49 ` [PATCH 2/2] KVM: Expose AVX_VNNI instruction to guset Yang Zhong
2021-01-21 15:02 ` [PATCH 0/2] Enumerate and expose AVX_VNNI feature Paolo Bonzini
2021-01-22 6:43 ` Yang Zhong
-- strict thread matches above, loose matches on Subject: below --
2020-12-16 2:01 Yang Zhong
2020-12-16 2:01 ` [PATCH 1/2] Enumerate AVX Vector Neural Network instructions Yang Zhong
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).