devanbenz commented on code in PR #10136: URL: https://github.com/apache/arrow-rs/pull/10136#discussion_r4087856384
########## arrow-buffer/src/util/bit_util.rs: ########## @@ -19,6 +19,45 @@ use crate::bit_chunk_iterator::BitChunks; +/// Parallel bit extract: for each set bit in `mask`, extract the +/// corresponding bit from `value` and pack them contiguously into the low +/// bits of the return value. +/// +/// Equivalent to the x86 BMI2 `PEXT` instruction. When compiled with the +/// `bmi2` target feature enabled (for example `-C target-cpu=x86-64-v3`) +/// this lowers to the hardware `pext` instruction; otherwise it falls back +/// to a portable scalar loop. +// +// Replace with `value.compress(mask)` when `uint_gather_scatter_bits` is +// stabilised: <https://github.com/rust-lang/rust/issues/149069> +#[inline] +pub fn compress(value: u64, mask: u64) -> u64 { + #[cfg(all(target_arch = "x86_64", target_feature = "bmi2"))] + { + // SAFETY: the `bmi2` target feature is statically enabled for this + // build, so the `pext` instruction is guaranteed to be available. + unsafe { std::arch::x86_64::_pext_u64(value, mask) } + } + + #[cfg(not(all(target_arch = "x86_64", target_feature = "bmi2")))] + { + let mut mask = mask; + let mut result = 0_u64; + let mut dest_bit = 1_u64; + while mask != 0 { Review Comment: I actually attempted to get claude to generate this code and benchmark during an earlier iteration (the exact section from Hackers Delight -- which, I will be keeping a PDF copy on my machine for future references. Super cool book!). Unfortunately on my machine it caused a regression for a few of the kernels compared to this existing code :thinking: Could be interesting to give it another shot. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
