Avatar for the OpenMathLib user
OpenMathLib
OpenBLAS
BlogDocsChangelog

ARM64: Enable bfloat16 kernels by default

#5396Merged
Comparing
abhishek-iitmadras:abhishekk_bfloat16
(
6356190
) with
develop
(
33b5054
)
CodSpeed Performance Gauge
0%
Untouched
62

Benchmarks

62 total
test_daxpy[100-d]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
-1%
24 µs24.2 µs
test_daxpy[100-c]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
25 µs25.1 µs
test_syrk[100-d]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
339.6 µs341.1 µs
test_dgbmv[1-100-s]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
37.4 µs37.2 µs
test_nrm2[100-dz]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
28.8 µs28.9 µs
test_dgbmv[1-1000-s]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
75.1 µs74.9 µs
test_gesdd[mn0-s]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
109.2 µs108.9 µs
test_nrm2[1000-dz]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
35.3 µs35.4 µs
test_syrk[100-c]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
472.4 µs473.6 µs
test_dot[100]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
22.3 µs22.3 µs
test_daxpy[100-s]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
23.9 µs24 µs
test_gesv[100-d]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
395.5 µs394.6 µs
test_daxpy[1000-s]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
27.4 µs27.5 µs
test_daxpy[1000-d]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
32.3 µs32.4 µs
test_dgbmv[1-100-d]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
37.9 µs38 µs
test_dgbmv[1-1000-d]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
83.7 µs83.5 µs
test_daxpy[100-z]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
25.8 µs25.7 µs
test_dot[1000]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
28.4 µs28.5 µs
test_dgbmv[1-1000-z]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
118.9 µs118.8 µs
test_gesv[100-z]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
937.9 µs938.8 µs
test_syrk[100-z]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
856.5 µs857.2 µs
test_nrm2[100-d]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
36.7 µs36.7 µs
test_gesdd[mn0-d]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
120 µs120.1 µs
test_daxpy[1000-z]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
40.5 µs40.5 µs
test_dgemv[100-s]
benchmark/pybench/benchmarks/bench_blas.py
CodSpeed Performance Gauge
0%
103.7 µs103.8 µs

Commits

Click on a commit to change the comparison range
Base
develop
33b5054
-0.07%
add neon header
0bc79da
1 year ago
by abhishek-iitmadras
+0.03%
Darwin/arm64: disable SVE/SME and fix gfortran link path
4c8dcb3
1 year ago
by abhishek-iitmadras
0%
fix gfortran link path in dynamic_arch.yml
6356190
1 year ago
by abhishek-iitmadras
© 2026 CodSpeed Technology
Home Terms Privacy Docs