Faster GEMV for RVV
#5826Comparing
ChipKerchner:fasterRVVGEMV
(6a23c36
) withdevelop
(91f97c6
)0%
- Untouched
- 62
Benchmarks
62 total
test_syev[200-s]
benchmark/pybench/benchmarks/bench_blas.py+1%
49.2 ms48.5 ms
test_dot[100]
benchmark/pybench/benchmarks/bench_blas.py+1%
18.1 µs18 µs
test_gesdd[mn1-s]
benchmark/pybench/benchmarks/bench_blas.py+1%
65.4 ms65 ms
test_daxpy[100-c]
benchmark/pybench/benchmarks/bench_blas.py0%
21.7 µs21.6 µs
test_dot[1000]
benchmark/pybench/benchmarks/bench_blas.py0%
24.6 µs24.5 µs
test_gesdd[mn0-s]
benchmark/pybench/benchmarks/bench_blas.py0%
108.2 µs107.9 µs
test_gesv[100-d]
benchmark/pybench/benchmarks/bench_blas.py0%
391.6 µs390.5 µs
test_daxpy[100-s]
benchmark/pybench/benchmarks/bench_blas.py0%
20.5 µs20.5 µs
test_dgbmv[1-100-s]
benchmark/pybench/benchmarks/bench_blas.py0%
34 µs33.9 µs
test_daxpy[1000-z]
benchmark/pybench/benchmarks/bench_blas.py0%
37.1 µs37 µs
test_nrm2[100-dz]
benchmark/pybench/benchmarks/bench_blas.py0%
25.4 µs25.3 µs
test_gemm[100-z]
benchmark/pybench/benchmarks/bench_blas.py0%
1.2 ms1.2 ms
test_dgbmv[1-100-d]
benchmark/pybench/benchmarks/bench_blas.py0%
34.6 µs34.5 µs
test_gesv[100-s]
benchmark/pybench/benchmarks/bench_blas.py0%
253.9 µs253.5 µs
test_daxpy[1000-s]
benchmark/pybench/benchmarks/bench_blas.py0%
24 µs23.9 µs
test_dgbmv[1-1000-s]
benchmark/pybench/benchmarks/bench_blas.py0%
71.4 µs71.3 µs
test_daxpy[1000-d]
benchmark/pybench/benchmarks/bench_blas.py0%
28.4 µs28.4 µs
test_daxpy[1000-c]
benchmark/pybench/benchmarks/bench_blas.py0%
29.2 µs29.2 µs
test_syrk[100-s]
benchmark/pybench/benchmarks/bench_blas.py0%
210.3 µs210.2 µs
test_gemm[1000-d]
benchmark/pybench/benchmarks/bench_blas.py0%
239.4 ms239.3 ms
test_gemm[100-c]
benchmark/pybench/benchmarks/bench_blas.py0%
656.4 µs656.2 µs
test_syev[50-s]
benchmark/pybench/benchmarks/bench_blas.py0%
1.3 ms1.3 ms
test_syrk[100-z]
benchmark/pybench/benchmarks/bench_blas.py0%
853.4 µs853.3 µs
test_gemm[1000-z]
benchmark/pybench/benchmarks/bench_blas.py0%
875.6 ms875.5 ms
test_daxpy[100-d]
benchmark/pybench/benchmarks/bench_blas.py0%
20.7 µs20.7 µs
Commits
Click on a commit to change the comparison rangeBase
develop
91f97c6
0%
Unroll inner loop - 2 rows at a time. Up to 1.5X faster.
6a23c36
1 day ago
by ChipKerchner