OpenCL

Type Members

1. final case class CommandQueue[Owner <: Singleton with OpenCL](handle: Long) extends AnyVal with MonadicCloseable[UnitContinuation] with Product with Serializable

   

2. trait CommandQueuePool extends OpenCL

   

3. final case class DeviceBuffer[Owner <: OpenCL with Singleton, Element](handle: Long) extends MonadicCloseable[UnitContinuation] with Product with Serializable

   

   A cl_mem whose CL_MEM_TYPE is buffer CL_MEM_OBJECT_BUFFER.

   A cl_mem whose CL_MEM_TYPE is buffer CL_MEM_OBJECT_BUFFER.

   handle

   The underlying cl_mem.

   Note

   comment out extends AnyVal in case of https://github.com/scala/bug/issues/10647

4. final case class DeviceId[Owner <: Singleton with OpenCL](handle: Long) extends AnyVal with Product with Serializable

   

5. trait DontReleaseEventTooEarly extends OpenCL

   

   A plug-in that retains every Event created by clEnqueueReadBuffer and waiting at least one second before releasing it.

   A plug-in that retains every Event created by clEnqueueReadBuffer and waiting at least one second before releasing it.

   Note

   This is a workaround for https://github.com/ThoughtWorksInc/Compute.scala/issues/51

6. final case class Event[Owner <: Singleton with OpenCL](handle: Long) extends AnyVal with MonadicCloseable[UnitContinuation] with Product with Serializable

   

7. trait GlobalExecutionContext extends AnyRef

   

8. trait HandleEventInExecutionContextForIntelAndAMDPlatform extends OpenCL

   
   Note

   HandleEventInExecutionContextForIntelAndAMDPlatform should be unnecessary because only OpenCL calls to create contexts or command-queues, or blocking OpenCL operations are undefined behavior, according to https://www.khronos.org/registry/OpenCL/sdk/1.2/docs/man/xhtml/clSetEventCallback.html and we don't use those forbidden functions. Our usage should be fine according to the OpenCL specification. However, AMD SDK always crashes for any reentry calls (e.g. https://travis-ci.org/Atry/DeepLearning.scala/jobs/318466522), no matter if they are blocking or not. There is also similar bug in Intel's OpenCL implementation As a workaround, always enable this HandleEventInExecutionContextForIntelAndAMDPlatform for Intel's and AMD's OpenCL implementation.

9. final case class Kernel[Owner <: OpenCL with Singleton](handle: Long) extends AnyVal with MonadicCloseable[UnitContinuation] with Product with Serializable

   

10. trait LogContextNotification extends OpenCL

    

11. final case class PlatformId[Owner <: Singleton with OpenCL](handle: Long) extends AnyVal with Product with Serializable

    

12. trait SingleThreadExecutionContext extends AnyRef

    

13. trait SuppressWarnings extends OpenCL

    

    A plug-in of Tensors to suppress warnings during compiling a OpenCL kernel for non-AMD platforms.

14. trait SynchronizedCreatingKernel extends OpenCL

    

    Make the calls to createKernels and createKernel synchronized.

    Make the calls to createKernels and createKernel synchronized.

    Note

    If you are using Intel OpenCL SDK, you will need this plug-in as a workaround

    See also

    Bug report: clCreateKernelsInProgram is not thread-safe

15. trait UnsafeMathOptimizations extends OpenCL

    

16. trait UseAllCpuDevices extends UseAllDevicesByType

    

17. trait UseAllDevices extends UseAllDevicesByType

    

18. trait UseAllDevicesByType extends OpenCL

    

19. trait UseAllGpuDevices extends UseAllDevicesByType

    

20. trait UseFirstDevice extends OpenCL