nakagawa-fj:gemm_2d_thread_partitioning - Branch - OpenMathLib/OpenBLAS - CodSpeed

Blog Docs Changelog

Improvement of 2D thread-partitioned GEMM for M << N case

#5276Merged

Comparing

nakagawa-fj:gemm_2d_thread_partitioning

(

2351a98

) with

develop

(

a5f701c

)

0%

Untouched: 62

Benchmarks

62 total

test_nrm2[100-d]

benchmark/pybench/benchmarks/bench_blas.py

+2%

35.3 µs34.7 µs

test_dgbmv[1-100-c]

benchmark/pybench/benchmarks/bench_blas.py

0%

39.6 µs39.4 µs

test_daxpy[100-c]

benchmark/pybench/benchmarks/bench_blas.py

0%

24.6 µs24.6 µs

test_gesdd[mn0-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

109 µs108.6 µs

test_syrk[100-c]

benchmark/pybench/benchmarks/bench_blas.py

0%

473.2 µs471.8 µs

test_daxpy[100-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

23.6 µs23.6 µs

test_gesdd[mn0-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

120.1 µs119.9 µs

test_daxpy[1000-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

31.8 µs31.7 µs

test_dgbmv[1-100-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

37.2 µs37.2 µs

test_daxpy[100-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

23.5 µs23.4 µs

test_syrk[100-z]

benchmark/pybench/benchmarks/bench_blas.py

0%

856.2 µs855.4 µs

test_dgbmv[1-100-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

36.7 µs36.6 µs

test_dgemv[100-z]

benchmark/pybench/benchmarks/bench_blas.py

0%

230.7 µs230.5 µs

test_dgemv[100-c]

benchmark/pybench/benchmarks/bench_blas.py

0%

149.5 µs149.3 µs

test_dgbmv[1-1000-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

74.4 µs74.4 µs

test_dgemv[100-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

147.8 µs147.7 µs

test_gesv[100-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

256.7 µs256.6 µs

test_gesv[100-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

395.5 µs395.4 µs

test_dgbmv[1-1000-c]

benchmark/pybench/benchmarks/bench_blas.py

0%

98.6 µs98.6 µs

test_syrk[100-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

339.1 µs339 µs

test_syev[50-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

1.3 ms1.3 ms

test_gemm[100-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

272.7 µs272.7 µs

test_dgbmv[1-100-z]

benchmark/pybench/benchmarks/bench_blas.py

0%

41.4 µs41.4 µs

test_gesv[1000-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

52.6 ms52.6 ms

test_daxpy[100-z]

benchmark/pybench/benchmarks/bench_blas.py

0%

25.3 µs25.3 µs

Commits

Click on a commit to change the comparison range

Base

develop

a5f701c

+0.04%

Update 2D thread-partitioned GEMM for M << N case.

2351a98

1 year ago

by nakagawa-fj

© 2026 CodSpeed Technology

Home Terms Privacy Docs