Kenneth Heafield

WebRTC the video / voice call standard? We used to have a REST server but it was janky, particularly with regard to security. https://www.zdnet.com/article/zoom-defends-use-of-local-web-server-on-macs-after-security-report/

The end goal is matrix multiply. Native matrix multiply implementations have highly-optimized implementations that are separate for x86 and ARM. If I'm honest about how I would use these instructions,...

What about just having an ARM USDOT / x86 VNNI wrapper that always does unsigned * signed? On pre-VNNI x86 it lowers to `pmaddubsw` (which gets saturated 16 bit results)...

I'm after 8-bit signed GEMM too for my project @browsermt, ultimately for quantized neural networks, which appears to be @bjacob's use case as well. As you mentioned in https://github.com/WebAssembly/simd/issues/224, GEMM...

The WebNN people are proposing to add GEMM to the browser, including 8-bit: https://webmachinelearning.github.io/webnn/#api-neuralnetworkcontext-gemm .

The operation can be emulated with SSSE3 and even SSE2 if necessary (but I think WebAssembly already assumes SSSE3). Usually, 8-bit GEMM is implemented on pre-VNNI Intel as `vpmaddubsw`, `vpmaddwd`,...

@lars-t-hansen Sorry what I meant is WebAssembly already has instructions that map to SSSE3 and later on Intel, which is the highest version required by `vpmaddubsw`, `vpmaddwd`, and `vpaddd` (perhaps...

To add some specific numbers: https://bugzilla.mozilla.org/show_bug.cgi?id=1746631 A machine translation application compiled to WebAssembly https://github.com/browsermt/bergamot-translator . Speed measured in words translated per second (wps). Heavy user of 8-bit integer matrix multiplication....

@Maratyszcza WebAssembly SIMD 128-bit. I've updated my comment.

I think we just install the generated configuration header with the rest of the headers.