Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1501780
| From | Ingo Molnar <mingo@kernel.org> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features |
| Date | 2016-10-17 10:50 +0200 |
| Message-ID | <stbMm-19o-13@gated-at.bofh.it> (permalink) |
| References | <srw8x-7Wn-3@gated-at.bofh.it> <ssRND-59Z-23@gated-at.bofh.it> <ssUs9-6Vd-1@gated-at.bofh.it> <staZY-A3-3@gated-at.bofh.it> <stbt1-12r-45@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
* Luc, Piotr <Piotr.Luc@intel.com> wrote: > 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_4VNNIW (18*32+2) /* AVX-512 Neural > > > > Network Instructions */ > > > > +#define X86_FEATURE_AVX512_4FMAPS (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 table 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 Bits 01 - 00: Reserved > Bit 02: AVX512_4VNNIW (Vector instructions for deep learning > enhanced word variable precision.) > Bit 03: AVX512_4FMAPS (Vector instructions for deep learning > floating-point single precision.) > Bits 31-04: Reserved > > The typo was acknowledged and is going to be fixed in next version of > the document. All of this should be pointed out in the changelog. I've zapped the commit for the time being - let's iterate this once more, ok? Thanks, Ingo
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features tip-bot for Piotr Luc <tipbot@zytor.com> - 2016-10-16 13:30 +0200
Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features Borislav Petkov <bp@alien8.de> - 2016-10-16 16:20 +0200
Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features hpa@zytor.com - 2016-10-16 18:10 +0200
Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features Borislav Petkov <bp@alien8.de> - 2016-10-16 18:40 +0200
Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features hpa@zytor.com - 2016-10-16 20:50 +0200
Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features Borislav Petkov <bp@alien8.de> - 2016-10-17 00:40 +0200
Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features "Luc, Piotr" <Piotr.Luc@intel.com> - 2016-10-17 12:00 +0200
Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features Thomas Gleixner <tglx@linutronix.de> - 2016-10-17 10:00 +0200
Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features "Luc, Piotr" <Piotr.Luc@intel.com> - 2016-10-17 10:30 +0200
Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features Ingo Molnar <mingo@kernel.org> - 2016-10-17 10:50 +0200
Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features "Luc, Piotr" <Piotr.Luc@intel.com> - 2016-10-17 12:00 +0200
Re: [tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features Thomas Gleixner <tglx@linutronix.de> - 2016-10-17 12:00 +0200
[PATCH v2] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features. Piotr Luc <piotr.luc@intel.com> - 2016-10-17 17:10 +0200
Re: [v2] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features. "Luc, Piotr" <Piotr.Luc@intel.com> - 2016-10-18 15:10 +0200
Re: [v2] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features. Ingo Molnar <mingo@kernel.org> - 2016-10-18 15:10 +0200
[PATCH v3] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features. Piotr Luc <piotr.luc@intel.com> - 2016-10-18 17:10 +0200
[tip:x86/urgent] x86/cpufeature: Add AVX512_4VNNIW and AVX512_4FMAPS features tip-bot for Piotr Luc <tipbot@zytor.com> - 2016-10-19 17:50 +0200
csiph-web