Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Linus Torvalds on AVX512

phoronix.com

Post reply on HN
Print view View original

121–124 of 124 posts

Re: Linus Torvalds on AVX512

#121

Post by TinkersW »

pjmlp wrote: post #100
ethelward wrote: post #93

Earlier quoted context omitted.

Because the academic ML community does not care about shipping product to end users not equipped in nVidia.

Except SYSCL also works on AMD and Intel, and also has a CUDA backend, but apparently you missed that part. In what concerns commercial uses of CUDA, Hollywood doesn't seem to have any problem with it, nor the car manufacturers with Jetson.

SYSCL might be an option, but it doesn't seem to have much in the way of adoption(Forum is dead)which is concerning.

It does look like Intel is supporting at least, so maybe in the future it will be a good option.

Re: Linus Torvalds on AVX512

#122

Post by blaser-waffle »

michaelt wrote: post #43
abainbridge wrote: post #12

What are the forces in chip design that are at play here? Over the last 10-15 years, fabs have continued to fit more and more logic gates per unit area, but haven't reduced the power consumption per gate as much. As a result, if you fill your modern chip with compute gates, you cannot use them all at once because the chip will melt. Or at least you can't have them all running at max clock rates. One solution is to in…

> Over the last 10-15 years, fabs have continued to fit more and more logic gates per unit area, but haven't reduced the power consumption per gate as much. Kids these days get 8 cores for a 100W TDP. When I was a boy, 100W got you a single core. And you didn't get dynamic frequency scaling, so it'd be putting out that heat all the time . (We also had to walk to school barefoot in the snow, uphill both ways)

> so it'd be putting out that heat all the time

What's the problem? My old school pentiums kept my dorm room nice and toasty. Could keep my window cracked in the winter for fresh air while gentoo compiled...

Re: Linus Torvalds on AVX512

#123

Post by gnufx »

nix23 wrote: post #115
gnufx wrote: post #110

Earlier quoted context omitted.

SVE isn't in a co-processor, I guess is the point. There's a lot more to Fugaku than SVE (whether or not you think that's a version of avx512), though. No DDR is suggestive.

if you need SVE you INTEGRATE it OR use a co-Processor Thanks for reading.

I read, but putting it in a co-processor makes little sense if you care about memory bandwidth and latency -- which you should to keep the vector unit fed. Fujitsu have carefully designed this stuff with such considerations in mind as far as I can tell.

Re: Linus Torvalds on AVX512

#124

Post by CamperBob2 »

panpanna wrote: post #113
TinkersW wrote: post #105

Earlier quoted context omitted.

He likely has ~zero hands on experience with SIMD or AVX512 as those aren't used in his precious kernel.

First of all, there is indeed SIMD code in the kernel. Check out the very beginning of your boot message. More importantly, the kernel needs to support context switch between userspace applications that use SIMD registers. So it touches a bunch of critical data structures and event handlers.

So he knows how to push and pop the registers. Okay...
Post reply on HN