simdjson disables 256-bit LASX on Loongson 3A6000: 128-bit LSX is 6.7% faster
lemire · x · 2026-11-10
- Daniel Lemire reports that on a Loongson 3A6000 (LA664 core), simdjson's LASX (256-bit SIMD) kernel is slower than the LSX (128-bit) kernel, reproduced with both GCC 15 and LLVM.
- Benchmarks (2.5 GHz, clang 21, DOM parse, best of 3): LASX is on geomean 6.7% slower overall and 13.9% slower in stage 1; forcing LSX with GCC 15 is 3–18% faster per file.
- simdjson has therefore disabled LASX runtime dispatching by default — the LSX kernel is used even when the CPU supports LASX, unless built with -DSIMDJSONDISABLELASXRUNTIMEDISPATCH=0 or -mlasx.
- The team is asking whether this holds on other LoongArch processors and welcomes a faster LASX kernel.
More from Infra
- Chrome 153 ships native tab recorder: 120fps 4K AV1 capture, no ffmpeg needed — Vjeux · 2026-11-10
- AWS EKS now offers per-Pod hardware isolation with Nitro Sandbox — tekbog · 2026-11-10
- AI capex needs to grow 5-10x before oversupply, argues Rob LeClerc — robleclerc · 2026-11-10
- Vera launches gyms, benchmarks and agents for cloud infrastructure operations — tianyin_xu · 2026-11-10
- NVIDIA paper: masking ~10 'intrusion heads' recovers 57% of distractor accuracy loss — rohanpaul_ai · 2026-11-10
- Memory pricing itself out of demand: PC BOM share hits 35-38%, shipments plunge — tengyanAI · 2026-11-10