More KernelInterface interface tests

[christiangnrd]

No description provided.

[github-actions]

Benchmark Results

Show table

	main	73fdef9...	main / 73fdef9...
saxpy/default/Float32/1024	0.0689 ± 0.03 ms	0.05 ± 0.031 ms	1.38 ± 1.1
saxpy/default/Float32/1048576	0.485 ± 0.02 ms	0.481 ± 0.023 ms	1.01 ± 0.065
saxpy/default/Float32/16384	0.0608 ± 0.031 ms	0.0579 ± 0.027 ms	1.05 ± 0.72
saxpy/default/Float32/2048	0.0688 ± 0.028 ms	0.0561 ± 0.03 ms	1.23 ± 0.83
saxpy/default/Float32/256	0.0491 ± 0.031 ms	0.0443 ± 0.025 ms	1.11 ± 0.93
saxpy/default/Float32/262144	0.172 ± 0.029 ms	0.164 ± 0.028 ms	1.05 ± 0.25
saxpy/default/Float32/32768	0.0666 ± 0.029 ms	0.0629 ± 0.023 ms	1.06 ± 0.6
saxpy/default/Float32/4096	0.0674 ± 0.026 ms	0.058 ± 0.026 ms	1.16 ± 0.69
saxpy/default/Float32/512	0.0653 ± 0.031 ms	0.0444 ± 0.028 ms	1.47 ± 1.2
saxpy/default/Float32/64	0.0463 ± 0.029 ms	0.043 ± 0.023 ms	1.08 ± 0.89
saxpy/default/Float32/65536	0.0816 ± 0.03 ms	0.0783 ± 0.028 ms	1.04 ± 0.53
saxpy/default/Float64/1024	0.0672 ± 0.03 ms	0.0499 ± 0.032 ms	1.35 ± 1
saxpy/default/Float64/1048576	0.551 ± 0.11 ms	0.511 ± 0.067 ms	1.08 ± 0.25
saxpy/default/Float64/16384	0.0574 ± 0.026 ms	0.0546 ± 0.021 ms	1.05 ± 0.63
saxpy/default/Float64/2048	0.0542 ± 0.03 ms	0.0531 ± 0.032 ms	1.02 ± 0.84
saxpy/default/Float64/256	0.0486 ± 0.031 ms	0.0422 ± 0.023 ms	1.15 ± 0.96
saxpy/default/Float64/262144	0.189 ± 0.032 ms	0.183 ± 0.031 ms	1.04 ± 0.25
saxpy/default/Float64/32768	0.0685 ± 0.027 ms	0.0655 ± 0.023 ms	1.05 ± 0.56
saxpy/default/Float64/4096	0.0691 ± 0.025 ms	0.0551 ± 0.026 ms	1.25 ± 0.74
saxpy/default/Float64/512	0.0558 ± 0.031 ms	0.0429 ± 0.025 ms	1.3 ± 1
saxpy/default/Float64/64	0.0462 ± 0.029 ms	0.0421 ± 0.022 ms	1.1 ± 0.9
saxpy/default/Float64/65536	0.0873 ± 0.028 ms	0.0843 ± 0.027 ms	1.04 ± 0.47
saxpy/static workgroup=(1024,)/Float32/1024	0.0662 ± 0.03 ms	0.0478 ± 0.031 ms	1.38 ± 1.1
saxpy/static workgroup=(1024,)/Float32/1048576	0.419 ± 0.025 ms	0.414 ± 0.031 ms	1.01 ± 0.098
saxpy/static workgroup=(1024,)/Float32/16384	0.0572 ± 0.03 ms	0.0542 ± 0.026 ms	1.06 ± 0.75
saxpy/static workgroup=(1024,)/Float32/2048	0.0658 ± 0.028 ms	0.0535 ± 0.03 ms	1.23 ± 0.87
saxpy/static workgroup=(1024,)/Float32/256	0.0485 ± 0.029 ms	0.0438 ± 0.024 ms	1.11 ± 0.9
saxpy/static workgroup=(1024,)/Float32/262144	0.15 ± 0.029 ms	0.139 ± 0.028 ms	1.08 ± 0.3
saxpy/static workgroup=(1024,)/Float32/32768	0.0608 ± 0.026 ms	0.0583 ± 0.02 ms	1.04 ± 0.57
saxpy/static workgroup=(1024,)/Float32/4096	0.0658 ± 0.026 ms	0.0524 ± 0.027 ms	1.26 ± 0.8
saxpy/static workgroup=(1024,)/Float32/512	0.0612 ± 0.029 ms	0.045 ± 0.03 ms	1.36 ± 1.1
saxpy/static workgroup=(1024,)/Float32/64	0.0468 ± 0.027 ms	0.0428 ± 0.016 ms	1.09 ± 0.76
saxpy/static workgroup=(1024,)/Float32/65536	0.074 ± 0.029 ms	0.0712 ± 0.026 ms	1.04 ± 0.56
saxpy/static workgroup=(1024,)/Float64/1024	0.0648 ± 0.03 ms	0.0464 ± 0.031 ms	1.4 ± 1.1
saxpy/static workgroup=(1024,)/Float64/1048576	0.499 ± 0.094 ms	0.482 ± 0.084 ms	1.04 ± 0.27
saxpy/static workgroup=(1024,)/Float64/16384	0.0546 ± 0.024 ms	0.0524 ± 0.02 ms	1.04 ± 0.62
saxpy/static workgroup=(1024,)/Float64/2048	0.0531 ± 0.029 ms	0.0527 ± 0.031 ms	1.01 ± 0.81
saxpy/static workgroup=(1024,)/Float64/256	0.0484 ± 0.029 ms	0.0433 ± 0.025 ms	1.12 ± 0.92
saxpy/static workgroup=(1024,)/Float64/262144	0.182 ± 0.031 ms	0.177 ± 0.028 ms	1.03 ± 0.24
saxpy/static workgroup=(1024,)/Float64/32768	0.0659 ± 0.025 ms	0.0636 ± 0.023 ms	1.04 ± 0.54
saxpy/static workgroup=(1024,)/Float64/4096	0.064 ± 0.025 ms	0.0512 ± 0.026 ms	1.25 ± 0.79
saxpy/static workgroup=(1024,)/Float64/512	0.0601 ± 0.029 ms	0.0446 ± 0.029 ms	1.35 ± 1.1
saxpy/static workgroup=(1024,)/Float64/64	0.0469 ± 0.029 ms	0.0428 ± 0.024 ms	1.1 ± 0.91
saxpy/static workgroup=(1024,)/Float64/65536	0.084 ± 0.026 ms	0.0809 ± 0.023 ms	1.04 ± 0.44
time_to_load	0.886 ± 0.0093 s	0.887 ± 0.0047 s	0.999 ± 0.012

Benchmark Plots

A plot of the benchmark results have been uploaded as an artifact to the workflow run for this PR.
Go to "Actions"->"Benchmark a pull request"->[the most recent run]->"Artifacts" (at the bottom).

[vchuravy]

Should we define this alias in the package?

[christiangnrd]

Maybe? However I was starting to think that we should maybe have the get_X_size and get_X_id intrinsics take an optional type so people can have their kernels use 4-byte indexingwhen they know they won't need more and it might be annoying to do after we've defined this