Control.Parallel.OpenCL.Program

Types

type CLProgram = Ptr ()

data CLBuildStatus

• CL_BUILD_NONE. The build status returned if no build has been performed on the specified program object for device.
• CL_BUILD_ERROR. The build status returned if the last call to clBuildProgram on the specified program object for device generated an error.
• CL_BUILD_SUCCESS. The build status retrned if the last call to clBuildProgram on the specified program object for device was successful.
• CL_BUILD_IN_PROGRESS. The build status returned if the last call to clBuildProgram on the specified program object for device has not finished.

Constructors

CL_BUILD_NONE

CL_BUILD_ERROR

CL_BUILD_SUCCESS

CL_BUILD_IN_PROGRESS

Instances

Enum CLBuildStatus

Show CLBuildStatus

type CLKernel = Ptr ()

Program Functions

clCreateProgramWithSource :: CLContext -> String -> IO CLProgram

Creates a program object for a context, and loads the source code specified by the text strings in the strings array into the program object. The devices associated with the program object are the devices associated with context.

OpenCL allows applications to create a program object using the program source or binary and build appropriate program executables. This allows applications to determine whether they want to use the pre-built offline binary or load and compile the program source and use the executable compiled/linked online as the program executable. This can be very useful as it allows applications to load and build program executables online on its first instance for appropriate OpenCL devices in the system. These executables can now be queried and cached by the application. Future instances of the application launching will no longer need to compile and build the program executables. The cached executables can be read and loaded by the application, which can help significantly reduce the application initialization time.

An OpenCL program consists of a set of kernels that are identified as functions declared with the __kernel qualifier in the program source. OpenCL programs may also contain auxiliary functions and constant data that can be used by __kernel functions. The program executable can be generated online or offline by the OpenCL compiler for the appropriate target device(s).

clCreateProgramWithSource returns a valid non-zero program object if the program object is created successfully. Otherwise, it throws one of the following CLError exceptions:

• CL_INVALID_CONTEXT if context is not a valid context.
• CL_OUT_OF_HOST_MEMORY if there is a failure to allocate resources required by the OpenCL implementation on the host.

clCreateProgramWithBinary :: CLContext -> [CLDeviceID] -> [[Word8]] -> IO (CLProgram, [CLError])

Creates a program object for a context, and loads specified binary data into the program object.

The program binaries specified by binaries contain the bits that describe the program executable that will be run on the device(s) associated with context. The program binary can consist of either or both of device-specific executable(s), and/or implementation-specific intermediate representation (IR) which will be converted to the device-specific executable.

OpenCL allows applications to create a program object using the program source or binary and build appropriate program executables. This allows applications to determine whether they want to use the pre-built offline binary or load and compile the program source and use the executable compiled/linked online as the program executable. This can be very useful as it allows applications to load and build program executables online on its first instance for appropriate OpenCL devices in the system. These executables can now be queried and cached by the application. Future instances of the application launching will no longer need to compile and build the program executables. The cached executables can be read and loaded by the application, which can help significantly reduce the application initialization time.

Returns a valid non-zero program object and a list of CLError values whether the program binary for each device specified in device_list was loaded successfully or not. It is list of the same length the list of devices with CL_SUCCESS if binary was successfully loaded for device specified by same position; otherwise returns CL_INVALID_VALUE if length of binary is zero or CL_INVALID_BINARY if program binary is not a valid binary for the specified device.

The function can throw on of the following CLError exceptions:

• CL_INVALID_CONTEXT if context is not a valid context.
• CL_INVALID_VALUE if the device list is empty; or if lengths or binaries are empty.
• CL_INVALID_DEVICE if OpenCL devices listed in the device list are not in the list of devices associated with context.
• CL_INVALID_BINARY if an invalid program binary was encountered for any device.
• CL_OUT_OF_HOST_MEMORY if there is a failure to allocate resources required by the OpenCL implementation on the host.

clRetainProgram :: CLProgram -> IO Bool

Increments the program reference count. clRetainProgram returns True if the function is executed successfully. It returns False if program is not a valid program object.

clReleaseProgram :: CLProgram -> IO Bool

Decrements the program reference count. The program object is deleted after all kernel objects associated with program have been deleted and the program reference count becomes zero. clReleseProgram returns True if the function is executed successfully. It returns False if program is not a valid program object.

clUnloadCompiler :: IO ()

Allows the implementation to release the resources allocated by the OpenCL compiler. This is a hint from the application and does not guarantee that the compiler will not be used in the future or that the compiler will actually be unloaded by the implementation. Calls to clBuildProgram after clUnloadCompiler will reload the compiler, if necessary, to build the appropriate program executable.

clBuildProgram :: CLProgram -> [CLDeviceID] -> String -> IO ()

Builds (compiles and links) a program executable from the program source or binary. OpenCL allows program executables to be built using the source or the binary. The build options are categorized as pre-processor options, options for math intrinsics, options that control optimization and miscellaneous options. This specification defines a standard set of options that must be supported by an OpenCL compiler when building program executables online or offline. These may be extended by a set of vendor- or platform-specific options.

• Preprocessor Options

These options control the OpenCL preprocessor which is run on each program source before actual compilation. -D options are processed in the order they are given in the options argument to clBuildProgram.

-D name

Predefine name as a macro, with definition 1.

-D name=definition

The contents of definition are tokenized and processed as if they appeared during translation phase three in a `#define' directive. In particular, the definition will be truncated by embedded newline characters.

-I dir

Add the directory dir to the list of directories to be searched for header files.

• Math Intrinsics Options

These options control compiler behavior regarding floating-point arithmetic. These options trade off between speed and correctness.

-cl-single-precision-constant

Treat double precision floating-point constant as single precision constant.

-cl-denorms-are-zero

This option controls how single precision and double precision denormalized numbers are handled. If specified as a build option, the single precision denormalized numbers may be flushed to zero and if the optional extension for double precision is supported, double precision denormalized numbers may also be flushed to zero. This is intended to be a performance hint and the OpenCL compiler can choose not to flush denorms to zero if the device supports single precision (or double precision) denormalized numbers.

This option is ignored for single precision numbers if the device does not support single precision denormalized numbers i.e. CL_FP_DENORM bit is not set in clGetDeviceSingleFPConfig.

This option is ignored for double precision numbers if the device does not support double precision or if it does support double precison but CL_FP_DENORM bit is not set in clGetDeviceDoubleFPConfig.

This flag only applies for scalar and vector single precision floating-point variables and computations on these floating-point variables inside a program. It does not apply to reading from or writing to image objects.

• Optimization Options

These options control various sorts of optimizations. Turning on optimization flags makes the compiler attempt to improve the performance and/or code size at the expense of compilation time and possibly the ability to debug the program.

-cl-opt-disable

This option disables all optimizations. The default is optimizations are enabled.

-cl-strict-aliasing

This option allows the compiler to assume the strictest aliasing rules.

The following options control compiler behavior regarding floating-point arithmetic. These options trade off between performance and correctness and must be specifically enabled. These options are not turned on by default since it can result in incorrect output for programs which depend on an exact implementation of IEEE 754 rules/specifications for math functions.

-cl-mad-enable

Allow a * b + c to be replaced by a mad. The mad computes a * b + c with reduced accuracy. For example, some OpenCL devices implement mad as truncate the result of a * b before adding it to c.

-cl-no-signed-zeros

Allow optimizations for floating-point arithmetic that ignore the signedness of zero. IEEE 754 arithmetic specifies the behavior of distinct +0.0 and -0.0 values, which then prohibits simplification of expressions such as x+0.0 or 0.0*x (even with -clfinite-math only). This option implies that the sign of a zero result isn't significant.

-cl-unsafe-math-optimizations

Allow optimizations for floating-point arithmetic that (a) assume that arguments and results are valid, (b) may violate IEEE 754 standard and (c) may violate the OpenCL numerical compliance requirements as defined in section 7.4 for single-precision floating-point, section 9.3.9 for double-precision floating-point, and edge case behavior in section 7.5. This option includes the -cl-no-signed-zeros and -cl-mad-enable options.

-cl-finite-math-only

Allow optimizations for floating-point arithmetic that assume that arguments and results are not NaNs or . This option may violate the OpenCL numerical compliance requirements defined in in section 7.4 for single-precision floating-point, section 9.3.9 for double-precision floating-point, and edge case behavior in section 7.5.

-cl-fast-relaxed-math

Sets the optimization options -cl-finite-math-only and -cl-unsafe-math-optimizations. This allows optimizations for floating-point arithmetic that may violate the IEEE 754 standard and the OpenCL numerical compliance requirements defined in the specification in section 7.4 for single-precision floating-point, section 9.3.9 for double-precision floating-point, and edge case behavior in section 7.5. This option causes the preprocessor macro __FAST_RELAXED_MATH__ to be defined in the OpenCL program.

• Options to Request or Suppress Warnings

Warnings are diagnostic messages that report constructions which are not inherently erroneous but which are risky or suggest there may have been an error. The following languageindependent options do not enable specific warnings but control the kinds of diagnostics produced by the OpenCL compiler.

-w

Inhibit all warning messages.

-Werror

Make all warnings into errors.

clBuildProgram can throw the following CLError exceptions when fails:

• CL_INVALID_PROGRAM if program is not a valid program object.
• CL_INVALID_DEVICE if OpenCL devices listed in device_list are not in the list of devices associated with program.
• CL_INVALID_BINARY if program is created with clCreateWithProgramWithBinary and devices listed in device_list do not have a valid program binary loaded.
• CL_INVALID_BUILD_OPTIONS if the build options specified by options are invalid.
• CL_INVALID_OPERATION if the build of a program executable for any of the devices listed in device_list by a previous call to clBuildProgram for program has not completed.
• CL_COMPILER_NOT_AVAILABLE if program is created with clCreateProgramWithSource and a compiler is not available i.e. clGetDeviceCompilerAvailable is set to False.
• CL_BUILD_PROGRAM_FAILURE if there is a failure to build the program executable. This error will be returned if clBuildProgram does not return until the build has completed.
• CL_INVALID_OPERATION if there are kernel objects attached to program.
• CL_OUT_OF_HOST_MEMORY if there is a failure to allocate resources required by the OpenCL implementation on the host.

clGetProgramReferenceCount :: CLProgram -> IO CLuint

Return the program reference count. The reference count returned should be considered immediately stale. It is unsuitable for general use in applications. This feature is provided for identifying memory leaks.

This function execute OpenCL clGetProgramInfo with CL_PROGRAM_REFERENCE_COUNT.

clGetProgramContext :: CLProgram -> IO CLContext

Return the context specified when the program object is created.

This function execute OpenCL clGetProgramInfo with CL_PROGRAM_CONTEXT.

clGetProgramNumDevices :: CLProgram -> IO CLuint

Return the number of devices associated with program.

This function execute OpenCL clGetProgramInfo with CL_PROGRAM_NUM_DEVICES.

clGetProgramDevices :: CLProgram -> IO [CLDeviceID]

Return the list of devices associated with the program object. This can be the devices associated with context on which the program object has been created or can be a subset of devices that are specified when a progam object is created using clCreateProgramWithBinary.

This function execute OpenCL clGetProgramInfo with CL_PROGRAM_DEVICES.

clGetProgramSource :: CLProgram -> IO String

Return the program source code specified by clCreateProgramWithSource. The source string returned is a concatenation of all source strings specified to clCreateProgramWithSource with a null terminator. The concatenation strips any nulls in the original source strings. The actual number of characters that represents the program source code including the null terminator is returned in param_value_size_ret.

This function execute OpenCL clGetProgramInfo with CL_PROGRAM_SOURCE.

clGetProgramBinarySizes :: CLProgram -> IO [CSize]

Returns an array that contains the size in bytes of the program binary for each device associated with program. The size of the array is the number of devices associated with program. If a binary is not available for a device(s), a size of zero is returned.

This function execute OpenCL clGetProgramInfo with CL_PROGRAM_BINARY_SIZES.

clGetProgramBinaries :: CLProgram -> IO [[Word8]]

Return the program binaries for all devices associated with program. For each device in program, the binary returned can be the binary specified for the device when program is created with clCreateProgramWithBinary or it can be the executable binary generated by clBuildProgram. If program is created with clCreateProgramWithSource, the binary returned is the binary generated by clBuildProgram. The bits returned can be an implementation-specific intermediate representation (a.k.a. IR) or device specific executable bits or both. The decision on which information is returned in the binary is up to the OpenCL implementation.

To find out which device the program binary in the array refers to, use the clGetProgramDevices query to get the list of devices. There is a one-to-one correspondence between the array of data returned by clGetProgramBinaries and array of devices returned by clGetProgramDevices.

This function execute OpenCL clGetProgramInfo with CL_PROGRAM_BINARIES.

clGetProgramBuildStatus :: CLProgram -> CLDeviceID -> IO CLBuildStatus

Returns the build status of program for a specific device as given by device.

This function execute OpenCL clGetProgramBuildInfo with CL_PROGRAM_BUILD_STATUS.

clGetProgramBuildOptions :: CLProgram -> CLDeviceID -> IO String

Return the build options specified by the options argument in clBuildProgram for device. If build status of program for device is CL_BUILD_NONE, an empty string is returned.

This function execute OpenCL clGetProgramBuildInfo with CL_PROGRAM_BUILD_OPTIONS.

clGetProgramBuildLog :: CLProgram -> CLDeviceID -> IO String

Return the build log when clBuildProgram was called for device. If build status of program for device is CL_BUILD_NONE, an empty string is returned.

This function execute OpenCL clGetProgramBuildInfo with CL_PROGRAM_BUILD_LOG.

Kernel Functions

clCreateKernel :: CLProgram -> String -> IO CLKernel

Creates a kernal object. A kernel is a function declared in a program. A kernel is identified by the __kernel qualifier applied to any function in a program. A kernel object encapsulates the specific __kernel function declared in a program and the argument values to be used when executing this __kernel function.

clCreateKernel returns a valid non-zero kernel object if the kernel object is created successfully. Otherwise, it throws one of the following CLError exceptions:

• CL_INVALID_PROGRAM if program is not a valid program object.
• CL_INVALID_PROGRAM_EXECUTABLE if there is no successfully built executable for program.
• CL_INVALID_KERNEL_NAME if kernel_name is not found in program.
• CL_INVALID_KERNEL_DEFINITION if the function definition for __kernel function given by kernel_name such as the number of arguments, the argument types are not the same for all devices for which the program executable has been built.
• CL_OUT_OF_HOST_MEMORY if there is a failure to allocate resources required by the OpenCL implementation on the host.

clCreateKernelsInProgram :: CLProgram -> IO [CLKernel]

Creates kernel objects for all kernel functions in a program object. Kernel objects are not created for any __kernel functions in program that do not have the same function definition across all devices for which a program executable has been successfully built.

Kernel objects can only be created once you have a program object with a valid program source or binary loaded into the program object and the program executable has been successfully built for one or more devices associated with program. No changes to the program executable are allowed while there are kernel objects associated with a program object. This means that calls to clBuildProgram return CL_INVALID_OPERATION if there are kernel objects attached to a program object. The OpenCL context associated with program will be the context associated with kernel. The list of devices associated with program are the devices associated with kernel. Devices associated with a program object for which a valid program executable has been built can be used to execute kernels declared in the program object.

clCreateKernelsInProgram will return the kernel objects if the kernel objects were successfully allocated, throws CL_INVALID_PROGRAM if program is not a valid program object, throws CL_INVALID_PROGRAM_EXECUTABLE if there is no successfully built executable for any device in program and throws CL_OUT_OF_HOST_MEMORY if there is a failure to allocate resources required by the OpenCL implementation on the host.

clRetainKernel :: CLKernel -> IO Bool

Increments the program program reference count. clRetainKernel returns True if the function is executed successfully. clCreateKernel or clCreateKernelsInProgram do an implicit retain.

clReleaseKernel :: CLKernel -> IO Bool

Decrements the kernel reference count. The kernel object is deleted once the number of instances that are retained to kernel become zero and the kernel object is no longer needed by any enqueued commands that use kernel. clReleaseKernel returns True if the function is executed successfully.

clSetKernelArg :: Integral a => CLKernel -> CLuint -> a -> Ptr b -> IO ()

Used to set the argument value for a specific argument of a kernel.

A kernel object does not update the reference count for objects such as memory, sampler objects specified as argument values by clSetKernelArg, Users may not rely on a kernel object to retain objects specified as argument values to the kernel.

Implementations shall not allow CLKernel objects to hold reference counts to CLKernel arguments, because no mechanism is provided for the user to tell the kernel to release that ownership right. If the kernel holds ownership rights on kernel args, that would make it impossible for the user to tell with certainty when he may safely release user allocated resources associated with OpenCL objects such as the CLMem backing store used with CL_MEM_USE_HOST_PTR.

clSetKernelArg throws one of the following CLError exceptions when fails:

• CL_INVALID_KERNEL if kernel is not a valid kernel object.
• CL_INVALID_ARG_INDEX if arg_index is not a valid argument index.
• CL_INVALID_ARG_VALUE if arg_value specified is NULL for an argument that is not declared with the __local qualifier or vice-versa.
• CL_INVALID_MEM_OBJECT for an argument declared to be a memory object when the specified arg_value is not a valid memory object.
• CL_INVALID_SAMPLER for an argument declared to be of type sampler_t when the specified arg_value is not a valid sampler object.
• CL_INVALID_ARG_SIZE if arg_size does not match the size of the data type for an argument that is not a memory object or if the argument is a memory object and arg_size != sizeof(cl_mem) or if arg_size is zero and the argument is declared with the __local qualifier or if the argument is a sampler and arg_size != sizeof(cl_sampler).

clSetKernelArgSto :: Storable a => CLKernel -> CLuint -> a -> IO ()

Wrap function of clSetKernelArg with Storable data.

clGetKernelFunctionName :: CLKernel -> IO String

Return the kernel function name.

This function execute OpenCL clGetKernelInfo with CL_KERNEL_FUNCTION_NAME.

clGetKernelNumArgs :: CLKernel -> IO CLuint

Return the number of arguments to kernel.

This function execute OpenCL clGetKernelInfo with CL_KERNEL_NUM_ARGS.

clGetKernelReferenceCount :: CLKernel -> IO CLuint

Return the kernel reference count. The reference count returned should be considered immediately stale. It is unsuitable for general use in applications. This feature is provided for identifying memory leaks.

This function execute OpenCL clGetKernelInfo with CL_KERNEL_REFERENCE_COUNT.

clGetKernelContext :: CLKernel -> IO CLContext

Return the context associated with kernel.

This function execute OpenCL clGetKernelInfo with CL_KERNEL_CONTEXT.

clGetKernelProgram :: CLKernel -> IO CLProgram

Return the program object associated with kernel.

This function execute OpenCL clGetKernelInfo with CL_KERNEL_PROGRAM.

clGetKernelWorkGroupSize :: CLKernel -> CLDeviceID -> IO CSize

This provides a mechanism for the application to query the work-group size that can be used to execute a kernel on a specific device given by device. The OpenCL implementation uses the resource requirements of the kernel (register usage etc.) to determine what this work-group size should be.

This function execute OpenCL clGetKernelWorkGroupInfo with CL_KERNEL_WORK_GROUP_SIZE.

clGetKernelCompileWorkGroupSize :: CLKernel -> CLDeviceID -> IO [CSize]

Returns the work-group size specified by the __attribute__((reqd_work_gr oup_size(X, Y, Z))) qualifier. See Function Qualifiers. If the work-group size is not specified using the above attribute qualifier (0, 0, 0) is returned.

This function execute OpenCL clGetKernelWorkGroupInfo with CL_KERNEL_COMPILE_WORK_GROUP_SIZE.

clGetKernelLocalMemSize :: CLKernel -> CLDeviceID -> IO CLulong

Returns the amount of local memory in bytes being used by a kernel. This includes local memory that may be needed by an implementation to execute the kernel, variables declared inside the kernel with the __local address qualifier and local memory to be allocated for arguments to the kernel declared as pointers with the __local address qualifier and whose size is specified with clSetKernelArg.

If the local memory size, for any pointer argument to the kernel declared with the __local address qualifier, is not specified, its size is assumed to be 0.

This function execute OpenCL clGetKernelWorkGroupInfo with CL_KERNEL_LOCAL_MEM_SIZE.