Comparing
ChipKerchner:addRVVVectorizedPacking
(64401b4
) withdevelop
(5e43ba9
)0%
- Untouched
- 62
Benchmarks
62 total
test_dot[1000]
benchmark/pybench/benchmarks/bench_blas.py-1%
28.2 µs28.4 µs
test_daxpy[100-s]
benchmark/pybench/benchmarks/bench_blas.py-1%
23.8 µs23.9 µs
test_dgbmv[1-100-d]
benchmark/pybench/benchmarks/bench_blas.py0%
37.7 µs37.9 µs
test_dgbmv[1-100-c]
benchmark/pybench/benchmarks/bench_blas.py0%
40.1 µs40.2 µs
test_dgbmv[1-100-z]
benchmark/pybench/benchmarks/bench_blas.py0%
41.9 µs42.1 µs
test_dot[100]
benchmark/pybench/benchmarks/bench_blas.py0%
22.2 µs22.3 µs
test_nrm2[100-dz]
benchmark/pybench/benchmarks/bench_blas.py0%
28.8 µs28.7 µs
test_daxpy[1000-c]
benchmark/pybench/benchmarks/bench_blas.py0%
32.4 µs32.5 µs
test_dgbmv[1-100-s]
benchmark/pybench/benchmarks/bench_blas.py0%
37 µs37.1 µs
test_dgbmv[1-1000-s]
benchmark/pybench/benchmarks/bench_blas.py0%
74.8 µs75 µs
test_gesdd[mn0-s]
benchmark/pybench/benchmarks/bench_blas.py0%
109 µs109.2 µs
test_gesv[100-d]
benchmark/pybench/benchmarks/bench_blas.py0%
395.5 µs394.7 µs
test_dgbmv[1-1000-d]
benchmark/pybench/benchmarks/bench_blas.py0%
83.4 µs83.5 µs
test_dgbmv[1-1000-z]
benchmark/pybench/benchmarks/bench_blas.py0%
118.8 µs119 µs
test_gesdd[mn0-d]
benchmark/pybench/benchmarks/bench_blas.py0%
119.7 µs119.9 µs
test_gesv[100-z]
benchmark/pybench/benchmarks/bench_blas.py0%
938.4 µs937.1 µs
test_daxpy[100-z]
benchmark/pybench/benchmarks/bench_blas.py0%
25.6 µs25.7 µs
test_dgbmv[1-1000-c]
benchmark/pybench/benchmarks/bench_blas.py0%
99.4 µs99.5 µs
test_nrm2[100-d]
benchmark/pybench/benchmarks/bench_blas.py0%
36.7 µs36.6 µs
test_daxpy[1000-d]
benchmark/pybench/benchmarks/bench_blas.py0%
32.2 µs32.2 µs
test_gesv[100-c]
benchmark/pybench/benchmarks/bench_blas.py0%
695.3 µs695.9 µs
test_syrk[100-s]
benchmark/pybench/benchmarks/bench_blas.py0%
213.2 µs213.3 µs
test_syrk[100-d]
benchmark/pybench/benchmarks/bench_blas.py0%
339.4 µs339.6 µs
test_syev[50-d]
benchmark/pybench/benchmarks/bench_blas.py0%
1.4 ms1.4 ms
test_syrk[100-c]
benchmark/pybench/benchmarks/bench_blas.py0%
472.2 µs472.4 µs
Commits
Click on a commit to change the comparison rangeBase
develop
5e43ba9
+0.04%
Add and use vectorized packing to ZVL128B and ZVL256B. Up to 3x+ faster than generic scalar functions.
c00afc8
11 months ago
by ChipKerchner-0.11%
Disable vectorized packing for DGEMM - since it is slower than scalar.
64401b4
11 months ago
by ChipKerchner