Skip to content

Comment on Introducing SIMD.jsparent

Comments

You (and a lot of people) make it sound like its magic but it's not - http://blog.stuffedcow.net/2014/01/x86-memory-disambiguation...

It's not magic. But it's not what that blog post is talking about.

On some of these processors, 128 bytes of stack or so is not really "memory" (in the sense of being stored with memory), so spilling is not that bad.

That's the magic I'm talking about because it's not true; memory is memory and stack memory isn't treated specially by the processor. What it does have is a store buffer, which applies to all memory accesses and is what store forwarding uses to bypass L1.

I'm simply going to disagree with you on this one, because i can't make my evidence public :)

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.