I merged a few PRs to SIMD optimize Wasm WASI libc, but it all got stalled in st... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		ncruces 4 months ago \| parent \| context \| favorite \| on: Why we need SIMD I merged a few PRs to SIMD optimize Wasm WASI libc, but it all got stalled in str(c)spn (which is slightly more sophisticated than the rest). There wasn't much appetite for any of it on Emscripten. https://github.com/WebAssembly/wasi-libc/pulls?q=is%3Apr+opt...

jasonthorsness 4 months ago [–]

subscribed to the str(c)spn thread for the eventual explanation of why the non-simd version seemed to give the wrong answer

ncruces 4 months ago | [–]

Very likely the issue is with the test infrastructure and not with the code under test.

My own fuzzing doesn't report any inconsistencies. But fuzzing is always necessarily incomplete.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact