KA.__synchronize, add GLOBAL_MEM_FENCE semantic

[vchuravy]

Mirror of JuliaGPU/OpenCL.jl#338

[github-actions]

Benchmark Results

	main	6377284...	main / 6377284...
saxpy/default/Float32/1024	0.0468 ± 0.029 ms	0.0446 ± 0.029 ms	1.05 ± 0.94
saxpy/default/Float32/1048576	0.459 ± 0.024 ms	0.471 ± 0.027 ms	0.974 ± 0.075
saxpy/default/Float32/16384	0.0654 ± 0.027 ms	0.0576 ± 0.026 ms	1.13 ± 0.7
saxpy/default/Float32/2048	0.0456 ± 0.027 ms	0.0451 ± 0.027 ms	1.01 ± 0.86
saxpy/default/Float32/256	0.0507 ± 0.029 ms	0.0494 ± 0.029 ms	1.03 ± 0.84
saxpy/default/Float32/262144	0.16 ± 0.025 ms	0.156 ± 0.025 ms	1.03 ± 0.23
saxpy/default/Float32/32768	0.0635 ± 0.027 ms	0.062 ± 0.026 ms	1.02 ± 0.61
saxpy/default/Float32/4096	0.0497 ± 0.024 ms	0.0482 ± 0.023 ms	1.03 ± 0.71
saxpy/default/Float32/512	0.0587 ± 0.029 ms	0.0535 ± 0.029 ms	1.1 ± 0.81
saxpy/default/Float32/64	0.0449 ± 0.029 ms	0.0442 ± 0.028 ms	1.01 ± 0.92
saxpy/default/Float32/65536	0.0814 ± 0.027 ms	0.0761 ± 0.026 ms	1.07 ± 0.5
saxpy/default/Float64/1024	0.0471 ± 0.029 ms	0.0452 ± 0.03 ms	1.04 ± 0.94
saxpy/default/Float64/1048576	0.6 ± 0.071 ms	0.567 ± 0.066 ms	1.06 ± 0.18
saxpy/default/Float64/16384	0.0713 ± 0.027 ms	0.0598 ± 0.026 ms	1.19 ± 0.68
saxpy/default/Float64/2048	0.0453 ± 0.025 ms	0.0454 ± 0.027 ms	0.999 ± 0.82
saxpy/default/Float64/256	0.0575 ± 0.029 ms	0.0534 ± 0.029 ms	1.08 ± 0.8
saxpy/default/Float64/262144	0.172 ± 0.024 ms	0.166 ± 0.027 ms	1.03 ± 0.22
saxpy/default/Float64/32768	0.0737 ± 0.026 ms	0.0671 ± 0.025 ms	1.1 ± 0.57
saxpy/default/Float64/4096	0.0501 ± 0.025 ms	0.0475 ± 0.023 ms	1.06 ± 0.73
saxpy/default/Float64/512	0.0569 ± 0.03 ms	0.0477 ± 0.029 ms	1.19 ± 0.96
saxpy/default/Float64/64	0.0438 ± 0.029 ms	0.0438 ± 0.029 ms	0.999 ± 0.94
saxpy/default/Float64/65536	0.0916 ± 0.026 ms	0.0854 ± 0.026 ms	1.07 ± 0.45
saxpy/static workgroup=(1024,)/Float32/1024	0.0514 ± 0.029 ms	0.0451 ± 0.029 ms	1.14 ± 0.97
saxpy/static workgroup=(1024,)/Float32/1048576	0.459 ± 0.023 ms	0.449 ± 0.024 ms	1.02 ± 0.074
saxpy/static workgroup=(1024,)/Float32/16384	0.0613 ± 0.026 ms	0.0541 ± 0.026 ms	1.13 ± 0.72
saxpy/static workgroup=(1024,)/Float32/2048	0.0443 ± 0.027 ms	0.0444 ± 0.027 ms	0.999 ± 0.86
saxpy/static workgroup=(1024,)/Float32/256	0.0541 ± 0.027 ms	0.0474 ± 0.026 ms	1.14 ± 0.85
saxpy/static workgroup=(1024,)/Float32/262144	0.159 ± 0.026 ms	0.155 ± 0.026 ms	1.03 ± 0.24
saxpy/static workgroup=(1024,)/Float32/32768	0.0633 ± 0.026 ms	0.0607 ± 0.026 ms	1.04 ± 0.62
saxpy/static workgroup=(1024,)/Float32/4096	0.048 ± 0.025 ms	0.0455 ± 0.024 ms	1.05 ± 0.78
saxpy/static workgroup=(1024,)/Float32/512	0.0573 ± 0.027 ms	0.0525 ± 0.028 ms	1.09 ± 0.78
saxpy/static workgroup=(1024,)/Float32/64	0.0513 ± 0.026 ms	0.0508 ± 0.025 ms	1.01 ± 0.71
saxpy/static workgroup=(1024,)/Float32/65536	0.0757 ± 0.026 ms	0.0724 ± 0.026 ms	1.05 ± 0.52
saxpy/static workgroup=(1024,)/Float64/1024	0.0449 ± 0.029 ms	0.0436 ± 0.029 ms	1.03 ± 0.95
saxpy/static workgroup=(1024,)/Float64/1048576	0.646 ± 0.11 ms	0.502 ± 0.055 ms	1.29 ± 0.26
saxpy/static workgroup=(1024,)/Float64/16384	0.0667 ± 0.026 ms	0.055 ± 0.025 ms	1.21 ± 0.73
saxpy/static workgroup=(1024,)/Float64/2048	0.0448 ± 0.027 ms	0.0448 ± 0.027 ms	1 ± 0.85
saxpy/static workgroup=(1024,)/Float64/256	0.0536 ± 0.027 ms	0.0449 ± 0.027 ms	1.2 ± 0.93
saxpy/static workgroup=(1024,)/Float64/262144	0.168 ± 0.026 ms	0.169 ± 0.028 ms	0.998 ± 0.23
saxpy/static workgroup=(1024,)/Float64/32768	0.0685 ± 0.026 ms	0.063 ± 0.025 ms	1.09 ± 0.6
saxpy/static workgroup=(1024,)/Float64/4096	0.0486 ± 0.026 ms	0.0456 ± 0.024 ms	1.06 ± 0.8
saxpy/static workgroup=(1024,)/Float64/512	0.0524 ± 0.028 ms	0.0479 ± 0.028 ms	1.1 ± 0.87
saxpy/static workgroup=(1024,)/Float64/64	0.05 ± 0.027 ms	0.0483 ± 0.027 ms	1.03 ± 0.81
saxpy/static workgroup=(1024,)/Float64/65536	0.0865 ± 0.027 ms	0.0827 ± 0.027 ms	1.05 ± 0.47
time_to_load	1.39 ± 0.019 s	1.36 ± 0.019 s	1.02 ± 0.02

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

Attention: Patch coverage is 0% with 1 line in your changes missing coverage. Please review.

Project coverage is 47.70%. Comparing base (eb8ea8e) to head (6377284).
Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
src/pocl/backend.jl	0.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #618   +/-   ##
=======================================
  Coverage   47.70%   47.70%           
=======================================
  Files          22       22           
  Lines        1723     1723           
=======================================
  Hits          822      822           
  Misses        901      901           

☔ 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.