Conversation
This commit implements a specialized SIMD optimization for `offset == 6` in `decompress_bmi2`. By using `_mm_shuffle_epi8` with precomputed cyclic masks (`OFFSET6_MASKS`), we can replicate the 6-byte repeating pattern into 16-byte vectors and process data in 48-byte chunks (LCM of 6 and 16). This avoids the slow scalar fallback loop for offsets < 8. Benchmark results (bench_decompress_offset6_micro): - Baseline: ~2.06 GiB/s - Optimized: ~10.75 GiB/s - Improvement: +423% Also added `bench_decompress_offset6_micro` to `benches/bench_main.rs` to verify and track this optimization. Co-authored-by: 404Setup <[email protected]>
|
👋 Jules, reporting for duty! I'm here to lend a hand with this pull request. When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down. I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job! For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with New to Jules? Learn more at jules.google/docs. For security, I will only act on instructions from the user who triggered this task. |
404Setup
deleted the
bolt/optimize-decompress-offset6-6778083682881966153
branch
1 participant
Implemented AVX2/SIMD optimization for LZ77 decompression with offset 6.
OFFSET6_MASKStable.else if offset == 6block insrc/decompress/x86.rs.bench_decompress_offset6_microbenchmark inbenches/bench_main.rs.PR created automatically by Jules for task 6778083682881966153 started by @404Setup