use POCL with fixes

[vchuravy]

Following JuliaGPU/OpenCL.jl#298 (comment)

[github-actions]

Benchmark Results

	main	d630770...	main/d630770724df2c...
saxpy/default/Float16/1024	0.0603 ± 0.026 ms	0.0513 ± 0.027 ms	1.17
saxpy/default/Float16/1048576	0.895 ± 0.022 ms	0.891 ± 0.025 ms	1.01
saxpy/default/Float16/16384	0.0692 ± 0.027 ms	0.06 ± 0.025 ms	1.15
saxpy/default/Float16/2048	0.0512 ± 0.023 ms	0.0461 ± 0.019 ms	1.11
saxpy/default/Float16/256	0.0639 ± 0.027 ms	0.0542 ± 0.029 ms	1.18
saxpy/default/Float16/262144	0.272 ± 0.024 ms	0.269 ± 0.025 ms	1.01
saxpy/default/Float16/32768	0.0834 ± 0.027 ms	0.0735 ± 0.025 ms	1.14
saxpy/default/Float16/4096	0.0642 ± 0.023 ms	0.0523 ± 0.024 ms	1.23
saxpy/default/Float16/512	0.0611 ± 0.027 ms	0.0467 ± 0.028 ms	1.31
saxpy/default/Float16/64	0.062 ± 0.028 ms	0.044 ± 0.029 ms	1.41
saxpy/default/Float16/65536	0.112 ± 0.027 ms	0.103 ± 0.026 ms	1.09
saxpy/default/Float32/1024	0.0596 ± 0.027 ms	0.0474 ± 0.029 ms	1.26
saxpy/default/Float32/1048576	0.481 ± 0.035 ms	0.47 ± 0.052 ms	1.02
saxpy/default/Float32/16384	0.0587 ± 0.026 ms	0.0543 ± 0.024 ms	1.08
saxpy/default/Float32/2048	0.0489 ± 0.023 ms	0.0448 ± 0.02 ms	1.09
saxpy/default/Float32/256	0.0603 ± 0.028 ms	0.0451 ± 0.029 ms	1.34
saxpy/default/Float32/262144	0.164 ± 0.034 ms	0.15 ± 0.033 ms	1.1
saxpy/default/Float32/32768	0.0629 ± 0.026 ms	0.0605 ± 0.024 ms	1.04
saxpy/default/Float32/4096	0.0614 ± 0.026 ms	0.0498 ± 0.025 ms	1.23
saxpy/default/Float32/512	0.0618 ± 0.028 ms	0.0439 ± 0.029 ms	1.41
saxpy/default/Float32/64	0.0584 ± 0.028 ms	0.0427 ± 0.029 ms	1.37
saxpy/default/Float32/65536	0.0764 ± 0.028 ms	0.0739 ± 0.027 ms	1.03
saxpy/default/Float64/1024	0.057 ± 0.027 ms	0.0461 ± 0.028 ms	1.24
saxpy/default/Float64/1048576	0.516 ± 0.041 ms	0.508 ± 0.041 ms	1.02
saxpy/default/Float64/16384	0.0589 ± 0.026 ms	0.0548 ± 0.025 ms	1.07
saxpy/default/Float64/2048	0.0481 ± 0.023 ms	0.0445 ± 0.02 ms	1.08
saxpy/default/Float64/256	0.0612 ± 0.028 ms	0.0453 ± 0.029 ms	1.35
saxpy/default/Float64/262144	0.176 ± 0.028 ms	0.17 ± 0.029 ms	1.03
saxpy/default/Float64/32768	0.0661 ± 0.026 ms	0.0624 ± 0.025 ms	1.06
saxpy/default/Float64/4096	0.0612 ± 0.026 ms	0.0498 ± 0.025 ms	1.23
saxpy/default/Float64/512	0.0606 ± 0.028 ms	0.0434 ± 0.029 ms	1.39
saxpy/default/Float64/64	0.0585 ± 0.028 ms	0.0424 ± 0.029 ms	1.38
saxpy/default/Float64/65536	0.0888 ± 0.027 ms	0.0828 ± 0.026 ms	1.07
saxpy/static workgroup=(1024,)/Float16/1024	0.0582 ± 0.026 ms	0.0457 ± 0.027 ms	1.27
saxpy/static workgroup=(1024,)/Float16/1048576	0.902 ± 0.028 ms	0.894 ± 0.027 ms	1.01
saxpy/static workgroup=(1024,)/Float16/16384	0.0654 ± 0.026 ms	0.0571 ± 0.024 ms	1.15
saxpy/static workgroup=(1024,)/Float16/2048	0.0541 ± 0.024 ms	0.0471 ± 0.022 ms	1.15
saxpy/static workgroup=(1024,)/Float16/256	0.0622 ± 0.026 ms	0.0432 ± 0.027 ms	1.44
saxpy/static workgroup=(1024,)/Float16/262144	0.273 ± 0.026 ms	0.266 ± 0.027 ms	1.03
saxpy/static workgroup=(1024,)/Float16/32768	0.0794 ± 0.026 ms	0.0707 ± 0.023 ms	1.12
saxpy/static workgroup=(1024,)/Float16/4096	0.0577 ± 0.026 ms	0.0475 ± 0.026 ms	1.21
saxpy/static workgroup=(1024,)/Float16/512	0.0615 ± 0.026 ms	0.0437 ± 0.027 ms	1.41
saxpy/static workgroup=(1024,)/Float16/64	0.0605 ± 0.027 ms	0.0416 ± 0.028 ms	1.45
saxpy/static workgroup=(1024,)/Float16/65536	0.111 ± 0.026 ms	0.0991 ± 0.025 ms	1.12
saxpy/static workgroup=(1024,)/Float32/1024	0.058 ± 0.027 ms	0.0443 ± 0.028 ms	1.31
saxpy/static workgroup=(1024,)/Float32/1048576	0.467 ± 0.031 ms	0.456 ± 0.051 ms	1.02
saxpy/static workgroup=(1024,)/Float32/16384	0.0539 ± 0.025 ms	0.0516 ± 0.024 ms	1.04
saxpy/static workgroup=(1024,)/Float32/2048	0.0463 ± 0.023 ms	0.0437 ± 0.021 ms	1.06
saxpy/static workgroup=(1024,)/Float32/256	0.0592 ± 0.026 ms	0.0442 ± 0.027 ms	1.34
saxpy/static workgroup=(1024,)/Float32/262144	0.161 ± 0.035 ms	0.145 ± 0.033 ms	1.11
saxpy/static workgroup=(1024,)/Float32/32768	0.0583 ± 0.025 ms	0.0567 ± 0.024 ms	1.03
saxpy/static workgroup=(1024,)/Float32/4096	0.0562 ± 0.026 ms	0.0457 ± 0.026 ms	1.23
saxpy/static workgroup=(1024,)/Float32/512	0.0598 ± 0.026 ms	0.0434 ± 0.026 ms	1.38
saxpy/static workgroup=(1024,)/Float32/64	0.0573 ± 0.026 ms	0.0436 ± 0.027 ms	1.31
saxpy/static workgroup=(1024,)/Float32/65536	0.0717 ± 0.028 ms	0.0702 ± 0.027 ms	1.02
saxpy/static workgroup=(1024,)/Float64/1024	0.0591 ± 0.027 ms	0.0431 ± 0.029 ms	1.37
saxpy/static workgroup=(1024,)/Float64/1048576	0.506 ± 0.038 ms	0.508 ± 0.039 ms	0.996
saxpy/static workgroup=(1024,)/Float64/16384	0.0562 ± 0.026 ms	0.0528 ± 0.024 ms	1.06
saxpy/static workgroup=(1024,)/Float64/2048	0.047 ± 0.023 ms	0.0437 ± 0.022 ms	1.08
saxpy/static workgroup=(1024,)/Float64/256	0.0614 ± 0.026 ms	0.0419 ± 0.027 ms	1.47
saxpy/static workgroup=(1024,)/Float64/262144	0.172 ± 0.029 ms	0.166 ± 0.031 ms	1.03
saxpy/static workgroup=(1024,)/Float64/32768	0.0641 ± 0.025 ms	0.061 ± 0.024 ms	1.05
saxpy/static workgroup=(1024,)/Float64/4096	0.054 ± 0.026 ms	0.0459 ± 0.025 ms	1.18
saxpy/static workgroup=(1024,)/Float64/512	0.0606 ± 0.026 ms	0.0422 ± 0.027 ms	1.43
saxpy/static workgroup=(1024,)/Float64/64	0.056 ± 0.026 ms	0.0415 ± 0.026 ms	1.35
saxpy/static workgroup=(1024,)/Float64/65536	0.0855 ± 0.027 ms	0.0787 ± 0.026 ms	1.09
time_to_load	1.33 ± 0.0064 s	1.34 ± 0.0064 s	0.994

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).

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 47.52%. Comparing base (31937c5) to head (d630770).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #589   +/-   ##
=======================================
  Coverage   47.52%   47.52%           
=======================================
  Files          22       22           
  Lines        1715     1715           
=======================================
  Hits          815      815           
  Misses        900      900           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.