By: -.- (blarg.delete@this.mailinator.com), May 29, 2022 5:28 pm
Room: Moderated Discussions
Michael S (already5chosen.delete@this.yahoo.com) on May 29, 2022 1:38 pm wrote:
> And while not everything is matmul, a lot of "dense" compute-intensive
> things can be turned into likes of matmul.
> And if your workload is not "dense" then you are probably limited by bandwidth of one or another cache/memory
> level and can't take advantage of amount of FLOPs provided by good old AVX+FMA, much less so by AVX-512.
What about something like JSON parsing? Definitely not "dense" or mat-mul transformable, and not fast enough to be bandwidth limited, but gets notable speedups with AVX-512.
> And while not everything is matmul, a lot of "dense" compute-intensive
> things can be turned into likes of matmul.
> And if your workload is not "dense" then you are probably limited by bandwidth of one or another cache/memory
> level and can't take advantage of amount of FLOPs provided by good old AVX+FMA, much less so by AVX-512.
What about something like JSON parsing? Definitely not "dense" or mat-mul transformable, and not fast enough to be bandwidth limited, but gets notable speedups with AVX-512.