Closed ghost closed 6 years ago
@Godsoft029 Yes, it is the intended behaviour. The globalSize must be a multiple of localSize. If your localSize is 1 then globalSize must be at least 1, if globalSize is 10, then you will have 10 groups of localSize 1.
@Godsoft029 I'm going to close this ticket and label it as a question. Please feel free to re-open if you feel the ticket is still valid, and even if you keep it closed feel free to continue commenting on this ticket, thanks.
It seems like this error happens everytime Kernel.cleanUpArrays()
is called, which probably is not intended? It also happens in the CleanUpArraysDemo
class in the aparapi-examples repo.
OpenCL version: 2.1 Windows 10 x64 Happens on both AMD and AMD
Taking a look at the error codes it is clear that it doesnt want to execute 0-range-kernels. Is this intended behaviour?
Error Log:
after clEnqueueNDRangeKernelWOOOT5 , globalSize[0] = 0, localSize[0] = 1 !!!!!!! clEnqueueNDRangeKernel() failed invalid work group size Jun 29, 2018 1:44:52 PM com.aparapi.internal.kernel.KernelRunner fallBackToNextDevice WARNUNG: Device failed for NeuralNetKernel, devices={AMD<GPU>|AMD<CPU>|Java Alternative Algorithm|Java Thread Pool}: OpenCL execution seems to have failed (runKernelJNI returned -54) com.aparapi.internal.exception.AparapiException: OpenCL execution seems to have failed (runKernelJNI returned -54) at com.aparapi.internal.kernel.KernelRunner.executeOpenCL(KernelRunner.java:1231) at com.aparapi.internal.kernel.KernelRunner.executeInternalInner(KernelRunner.java:1690) at com.aparapi.internal.kernel.KernelRunner.executeInternalOuter(KernelRunner.java:1351) at com.aparapi.internal.kernel.KernelRunner.execute(KernelRunner.java:1342) at com.aparapi.Kernel.execute(Kernel.java:2857) at com.aparapi.Kernel.execute(Kernel.java:2814) at com.aparapi.Kernel.execute(Kernel.java:2789) at com.aparapi.internal.kernel.KernelRunner.cleanUpArrays(KernelRunner.java:208) at com.aparapi.Kernel.cleanUpArrays(Kernel.java:2883) at com.godsoft.fourwin.nn.NeuralNetKernel.computeBlocking(NeuralNetKernel.java:57) at com.godsoft.fourwin.nn.NeuralNetwork.computeOutput(NeuralNetwork.java:62) at com.godsoft.fourwin.MainClass$1.run(MainClass.java:67) at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) at java.util.concurrent.FutureTask.run(FutureTask.java:266) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) at java.lang.Thread.run(Thread.java:748)