Julien Jerphanion 459624cec2 Add a WASM SIMD128 4x4 DTRMM microkernel ...

Generic scalar DTRMM was the remaining Level-3 gap (~15 GFLOPS vs
~23 DGEMM). Keep SIMD for double only; single-precision TRMM already
auto-vectorized and a shared S+D kernel slowed SGEMM.

Signed-off-by: Julien Jerphanion <git@jjerphan.xyz>

2026-08-20 14:57:36 +02:00

..

gemmkernel_4x4_wasm128.c

wasm: add a 4x4 SIMD128 GEMM microkernel for SGEMM/DGEMM

2026-08-19 14:17:22 +02:00

gemmkernel_wasm128.c

Add WebAssembly SIMD GEMM kernels

2026-03-17 05:51:54 +09:00

KERNEL

wasm: add a 4x4 SIMD128 GEMM microkernel for SGEMM/DGEMM

2026-08-19 14:17:22 +02:00

KERNEL.WASM128_GENERIC

Add a WASM SIMD128 4x4 DTRMM microkernel

2026-08-20 14:57:36 +02:00

trmmkernel_4x4_wasm128.c

Add a WASM SIMD128 4x4 DTRMM microkernel

2026-08-20 14:57:36 +02:00

trsm_kernel_LN_wasm128.c

wasm: gate TRSM call counters behind OPENBLAS_WASM_TRSM_PROFILE

2026-08-19 14:16:28 +02:00

trsm_kernel_LT_wasm128.c

wasm: gate TRSM call counters behind OPENBLAS_WASM_TRSM_PROFILE

2026-08-19 14:16:28 +02:00

trsm_kernel_RN_wasm128.c

wasm: gate TRSM call counters behind OPENBLAS_WASM_TRSM_PROFILE

2026-08-19 14:16:28 +02:00

trsm_kernel_RT_wasm128.c

wasm: gate TRSM call counters behind OPENBLAS_WASM_TRSM_PROFILE

2026-08-19 14:16:28 +02:00