martin-frbg:issue5818 - Branch - OpenMathLib/OpenBLAS - CodSpeed

Blog Docs Changelog

Fix misoptimization of C/ZGETF2 by Clang on Apple M

#5820Merged

Comparing

martin-frbg:issue5818

(

1f2ae72

) with

develop

(

804a77c

)

0%

Untouched: 62

Benchmarks

62 total

test_syrk[100-d]

benchmark/pybench/benchmarks/bench_blas.py

+1%

338.1 µs336.3 µs

test_daxpy[100-c]

benchmark/pybench/benchmarks/bench_blas.py

0%

21.6 µs21.5 µs

test_daxpy[1000-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

23.9 µs23.8 µs

test_daxpy[1000-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

28.4 µs28.3 µs

test_syrk[100-c]

benchmark/pybench/benchmarks/bench_blas.py

0%

470.4 µs469.1 µs

test_daxpy[100-z]

benchmark/pybench/benchmarks/bench_blas.py

0%

22.3 µs22.3 µs

test_daxpy[1000-z]

benchmark/pybench/benchmarks/bench_blas.py

0%

37 µs36.9 µs

test_gesdd[mn0-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

108.1 µs107.9 µs

test_gesdd[mn0-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

119.2 µs119 µs

test_daxpy[1000-c]

benchmark/pybench/benchmarks/bench_blas.py

0%

29.1 µs29.1 µs

test_nrm2[1000-dz]

benchmark/pybench/benchmarks/bench_blas.py

0%

31.9 µs31.9 µs

test_syrk[100-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

210.2 µs210.2 µs

test_dgemv[1000-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

7 ms7 ms

test_gemm[100-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

468 µs467.9 µs

test_gemm[100-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

270 µs270 µs

test_syev[50-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

1.3 ms1.3 ms

test_gesv[1000-c]

benchmark/pybench/benchmarks/bench_blas.py

0%

188.6 ms188.6 ms

test_dgemv[100-z]

benchmark/pybench/benchmarks/bench_blas.py

0%

229.1 µs229.1 µs

test_dgemv[1000-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

13.9 ms13.9 ms

test_gemm[100-z]

benchmark/pybench/benchmarks/bench_blas.py

0%

1.2 ms1.2 ms

test_dgemv[1000-z]

benchmark/pybench/benchmarks/bench_blas.py

0%

26.3 ms26.3 ms

test_syrk[100-z]

benchmark/pybench/benchmarks/bench_blas.py

0%

853.2 µs853.2 µs

test_gemm[1000-d]

benchmark/pybench/benchmarks/bench_blas.py

0%

239.4 ms239.4 ms

test_syrk[1000-c]

benchmark/pybench/benchmarks/bench_blas.py

0%

227.5 ms227.5 ms

test_syrk[1000-s]

benchmark/pybench/benchmarks/bench_blas.py

0%

65.4 ms65.4 ms

Commits

Click on a commit to change the comparison range

Base

develop

804a77c

-0.11%

Make temp1 volatile to keep Clang on AppleM from optimizing out the division guards

1f2ae72

1 day ago

by martin-frbg

© 2026 CodSpeed Technology

Home Terms Privacy Docs