Search results

Use AVX instead of SSE for x86-64 machines

SIMD

Status: Open.

#9 In SwayamInSync/QBLAS;

· SwayamInSync opened on Dec 6, 2025

CI: Switch to macos-15 runner

Status: Open.

#7 In SwayamInSync/QBLAS;

· SwayamInSync opened on Jul 13, 2025

Perf: Inside column-major GEMV, reduction can be optimized using SIMD loads and FMA

SIMD

Status: Open.

#6 In SwayamInSync/QBLAS;

· SwayamInSync opened on Jul 13, 2025

SIMD: <code>QuadVector</code> class initialization is based on Machine rather than available SIMD instructions

SIMD

Status: Open.

#5 In SwayamInSync/QBLAS;

· SwayamInSync opened on Jul 12, 2025

SIMD: Vector memory alignment can be optimized to support higher Quads

SIMD

Status: Open.

#4 In SwayamInSync/QBLAS;

· SwayamInSync opened on Jul 12, 2025

ENH: Performance Improvement on non-ARM machines

Status: Open.

#3 In SwayamInSync/QBLAS;

· SwayamInSync opened on Jul 10, 2025