Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features

[Luc, Piotr]

On Mon, 2016-10-17 at 09:55 +0200, Thomas Gleixner wrote:
> On Sun, 16 Oct 2016, Borislav Petkov wrote:
> > 
> > > 
> > > The spec can be found in Intel Software Developer Manual or in
> > > Instruction Set Extensions Programming Reference. See
> > > https://software.intel.com/sites/default/files/managed/69/78/3194
> > > 33-025.pdf.
> > 
> > > 
> > > +/* Intel-defined CPU features, CPUID level 0x00000007:0 (edx),
> > > word 18 */
> > > +#define X86_FEATURE_AVX512_4VNNIWÂÂ(18*32+2) /* AVX-512 Neural
> > > Network Instructions */
> > > +#define X86_FEATURE_AVX512_4FMAPSÂÂ(18*32+3) /* AVX-512 Multiply
> > > Accumulation Single precision */
> > 
> > This is getting ridiculous: we keep adding new leafs to
> > ->x86_capability, thus bloating cpuinfo_x86 but then it is not even
> > worth it - this patch defines only two bits.
> 
> What's worse is that the Instruction Set Extensions Programming
> Reference
> manual says:
> 
> CPUID.(EAX=07H, ECX=0):EDX[bit 02] AVX512_4FMAPS
> CPUID.(EAX=07H, ECX=0):EBX[bit 03] AVX512_4VNNIW
> 
> So AVX512_4VNNIW is in EBX not EDX. What's correct here? The manual
> or the patch?
> 
> I'm going to zap it.
> 
The manual contains the typo in Âtable 2.1 on page 2.2.
Please compare it to the detailed description of CPUID in table 4.8 on
page 2-16.
There manual groups both new bits under EDX:

EDX Â ÂBits 01 - 00: Reserved
   ÂBit 02: AVX512_4VNNIW (Vector instructions for deep learning
enhanced word variable precision.)
   ÂBit 03: AVX512_4FMAPS (Vector instructions for deep learning
floating-point single precision.)
   ÂBits 31-04: Reserved

The typo was acknowledged and is going to be fixed in next version of
the document.Â

Regards,
Piotr