AVX512F strlen is faster nearly 5 times than the library function, and strchr nearly 4.5 times. https://github.com/WojciechMula/toys/tree/master/avx512-string … #avx512
-
-
But it makes context switch 5x slower by requiring an extra 1.5+ kB to be saved/restored, right?
3 replies 0 retweets 1 like -
related question - how does the OS know which SIMD registers (if any) to save when preempting a thread?
1 reply 0 retweets 0 likes -
Replying to @rzidane360 @RichFelker and
It seems like it must have to save the SSE/AVX context regardless of whether you use them.
2 replies 0 retweets 0 likes -
Replying to @CopperheadOS @RichFelker and
that's what I thought. So it'll save all sse/sse2/avx/avx512 registers? That seems like a lot.
2 replies 0 retweets 0 likes
Replying to @rzidane360 @CopperheadOS and
Most fpu/vector archs have a way to trap on 1st use so procs that don't use don't need saving.
11:40 AM - 5 Oct 2016
0 replies
0 retweets
3 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.