Repository navigation

arm64: NEON div and max/min, Go 1.27 mnemonics for sqrt/add/sub/mul; parallelFor findings - #30

Merged

tannevaled merged 4 commits into

mainfrom

arm64/vfsqrt-mnemonic

Oct 7, 2026

Merged

arm64: NEON div and max/min, Go 1.27 mnemonics for sqrt/add/sub/mul; parallelFor findings#30
tannevaled merged 4 commits into
mainfrom
arm64/vfsqrt-mnemonic

Commits on Oct 7, 2026
- arm64: VFSQRT/VFADD/VFSUB/VFMUL mnemonics, NEON div and max/min kernels (Go 1.27)
  
  tannevaled
  and
  claude
  committed
  e12b82c
- bench: DivInto without zero divisors
  
  tannevaled
  and
  claude
  committed
  ee5e247
- docs: arm64 NEON div/max/min and mnemonics; parallelFor scheduling measurements replace the cache-affinity guess
  
  tannevaled
  and
  claude
  committed
  8beb837
- arm64: below 8 elements max/min use the scalar oracle; unrolled reducers off arm64 (dead there)
  
  tannevaled
  and
  claude
  committed
  86e384e