Dougall

@fay59@tech.lgbt Congratulations on FEAT_CFLT! (optional from Armv9.7-A)

Marcin Żukowski’s PivCo-Huffman paper inspired me to revisit an Oodle-like 12-stream Huffman decoder for Apple CPUs that I experimented with in 2022. The same old kernel manages a pretty flat 5GB/s to 5.1GB/s on Apple M4, which did beat PivCo-Huffman, on one or two files, at the time of the paper. But with @rygorous@mastodon.gamedev.place's and Marcin's improvements, the updated PivCo-Huffman code reliably wins. And it doesn't need 31 GPRs. Very exciting work. https://gist.github.com/dougallj/a14d3f72b57ee58a81d487b43ff2a05b

@rygorous FWIW, your writing is hugely inspirational to me. I've learned some of my favourite ideas from your posts, like the throughput optimised bit-reading technique, and the texture swizzling bit-trick. And often your blog has let me understand things that I'd tried and failed to understand before, like FFT implementations, Huffman table initialisation, SIMD transposes, and triangle rasterisation. (Thank you!) This sucks, but many people still care about understanding.

What instructions would you like to see added to x86 or ARM? Personally, I want an ADD3 or an add-and-accumulate SIMD operation. A lot of algorithms are dominated by additions, and getting two adds with one instruction could help a lot. The closest instruction is SABA (signed-absolute-difference-and-accumulate), so I'm assuming this would be practical. I also have AVX-512 embedded rounding (and exception) control envy, but idk if ARM can find the bits for that with fixed-size instructions.