NVIDIA/nvbench
1
0
Fork 0
mirror of https://github.com/NVIDIA/nvbench.git synced 2026-04-20 06:48:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

cuda.nvbench -> cuda.bench

Browse Source 
Per PR review suggestion:
   - `cuda.parallel`    - device-wide algorithms/Thrust
   - `cuda.cooperative` - Cooperative algorithsm/CUB
   - `cuda.bench`       - Benchmarking/NVBench
This commit is contained in:
Oleksandr Pavlyk
2025-08-04 13:42:43 -05:00
parent c2a2acc9b6
commit b5e4b4ba31
19 changed files with 136 additions and 140 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
python/examples/cupy_extract.py
View File

@@ -16,18 +16,18 @@
import sys
import cuda.nvbench as nvbench
import cuda.bench as bench
import cupy as cp
def as_cp_ExternalStream(
cs: nvbench.CudaStream, dev_id: int | None = -1
cs: bench.CudaStream, dev_id: int | None = -1
) -> cp.cuda.ExternalStream:
h = cs.addressof()
return cp.cuda.ExternalStream(h, dev_id)
def cupy_extract_by_mask(state: nvbench.State):
def cupy_extract_by_mask(state: bench.State):
n_cols = state.get_int64("numCols")
n_rows = state.get_int64("numRows")
@@ -48,7 +48,7 @@ def cupy_extract_by_mask(state: nvbench.State):
mask = cp.ones((n_cols, n_rows), dtype=bool_dt)
_ = X[mask]
def launcher(launch: nvbench.Launch):
def launcher(launch: bench.Launch):
with as_cp_ExternalStream(launch.get_stream(), dev_id):
_ = X[mask]
@@ -56,8 +56,8 @@ def cupy_extract_by_mask(state: nvbench.State):
if __name__ == "__main__":
b = nvbench.register(cupy_extract_by_mask)
b = bench.register(cupy_extract_by_mask)
b.add_int64_axis("numCols", [1024, 2048, 4096, 2 * 4096])
b.add_int64_axis("numRows", [1024, 2048, 4096, 2 * 4096])
nvbench.run_all_benchmarks(sys.argv)
bench.run_all_benchmarks(sys.argv)