Loading examples/heat-equation/heatEquationSolver_impl.h +3 −3 Original line number Diff line number Diff line Loading @@ -266,9 +266,9 @@ assemblyLinearSystem( const RealType& time, u, matrix, b ); //matrix.print( cout ); //cout << endl << b << endl; //cout << endl << u << endl; matrix.print( cout ); cout << endl << b << endl; cout << endl << u << endl; //abort(); } Loading src/core/tnlCuda.h +4 −0 Original line number Diff line number Diff line Loading @@ -94,7 +94,11 @@ class tnlCuda static __device__ Index getInterleaving( const Index index ); #endif #ifdef HAVE_CUDA static bool checkDevice( const char* file_name, int line ); #else static bool checkDevice( const char* file_name, int line ) { return false;}; #endif }; Loading src/implementation/core/CMakeLists.txt +1 −0 Original line number Diff line number Diff line Loading @@ -29,6 +29,7 @@ IF( BUILD_CUDA ) ${tnl_implementation_core_cuda_CUDA__SOURCES} ${tnl_implementation_core_vectors_CUDA__SOURCES} ${common_SOURCES} ${CURRENT_DIR}/tnlCuda.cu PARENT_SCOPE ) ENDIF() Loading src/implementation/core/tnlCuda.cpp +0 −365 Original line number Diff line number Diff line Loading @@ -41,371 +41,6 @@ int tnlCuda::getNumberOfGrids( const int blocks, return roundUpDivision( blocks, gridSize ); } bool tnlCuda::checkDevice( const char* file_name, int line ) { #ifdef HAVE_CUDA cudaError error = cudaGetLastError(); if( error == cudaSuccess ) return true; cerr << "CUDA ERROR(" << error << ") at line " << line << " in " << file_name << ":" << endl; switch( error ) { case cudaErrorMissingConfiguration: cerr << "The device function being invoked (usually via ::cudaLaunch()) was not " << endl << "previously configured via the ::cudaConfigureCall() function. " << endl; break; case cudaErrorMemoryAllocation: cerr << "The API call failed because it was unable to allocate enough memory to " << endl << "perform the requested operation. " << endl; break; case cudaErrorInitializationError: cerr << "The API call failed because the CUDA driver and runtime could not be " << endl << "initialized. " << endl; break; case cudaErrorLaunchFailure: cerr << "An exception occurred on the device while executing a kernel. Common " << endl << "causes include dereferencing an invalid device pointer and accessing " << endl << "out of bounds shared memory. The device cannot be used until " << endl << "::cudaThreadExit() is called. All existing device memory allocations " << endl << "are invalid and must be reconstructed if the program is to continue " << endl << "using CUDA. " << endl; break; case cudaErrorPriorLaunchFailure: cerr << "This indicated that a previous kernel launch failed. This was previously " << endl << "used for device emulation of kernel launches. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorLaunchTimeout: cerr << "This indicates that the device kernel took too long to execute. This can " << endl << "only occur if timeouts are enabled - see the device property " << endl << "ref ::cudaDeviceProp::kernelExecTimeoutEnabled \"kernelExecTimeoutEnabled\" " << endl << "for more information. The device cannot be used until ::cudaThreadExit() " << endl << "is called. All existing device memory allocations are invalid and must be " << endl << "reconstructed if the program is to continue using CUDA. " << endl; break; case cudaErrorLaunchOutOfResources: cerr << "This indicates that a launch did not occur because it did not have " << endl << "appropriate resources. Although this error is similar to " << endl << "::cudaErrorInvalidConfiguration, this error usually indicates that the " << endl << "user has attempted to pass too many arguments to the device kernel, or the " << endl << "kernel launch specifies too many threads for the kernel's register count. " << endl; break; case cudaErrorInvalidDeviceFunction: cerr << "The requested device function does not exist or is not compiled for the " << endl << "proper device architecture. " << endl; break; case cudaErrorInvalidConfiguration: cerr << "This indicates that a kernel launch is requesting resources that can " << endl << "never be satisfied by the current device. Requesting more shared memory " << endl << "per block than the device supports will trigger this error, as will " << endl << "requesting too many threads or blocks. See ::cudaDeviceProp for more " << endl << "device limitations. " << endl; break; case cudaErrorInvalidDevice: cerr << "This indicates that the device ordinal supplied by the user does not " << endl << "correspond to a valid CUDA device. " << endl; break; case cudaErrorInvalidValue: cerr << "This indicates that one or more of the parameters passed to the API call " << endl << "is not within an acceptable range of values. " << endl; break; case cudaErrorInvalidPitchValue: cerr << "This indicates that one or more of the pitch-related parameters passed " << endl << "to the API call is not within the acceptable range for pitch. " << endl; break; case cudaErrorInvalidSymbol: cerr << "This indicates that the symbol name/identifier passed to the API call " << endl << "is not a valid name or identifier. " << endl; break; case cudaErrorMapBufferObjectFailed: cerr << "This indicates that the buffer object could not be mapped. " << endl; break; case cudaErrorUnmapBufferObjectFailed: cerr << "This indicates that the buffer object could not be unmapped. " << endl; break; case cudaErrorInvalidHostPointer: cerr << "This indicates that at least one host pointer passed to the API call is " << endl << "not a valid host pointer. " << endl; break; case cudaErrorInvalidDevicePointer: cerr << "This indicates that at least one device pointer passed to the API call is " << endl << "not a valid device pointer. " << endl; break; case cudaErrorInvalidTexture: cerr << "This indicates that the texture passed to the API call is not a valid " << endl << "texture. " << endl; break; case cudaErrorInvalidTextureBinding: cerr << "This indicates that the texture binding is not valid. This occurs if you " << endl << "call ::cudaGetTextureAlignmentOffset() with an unbound texture. " << endl; break; case cudaErrorInvalidChannelDescriptor: cerr << "This indicates that the channel descriptor passed to the API call is not " << endl << "valid. This occurs if the format is not one of the formats specified by " << endl << "::cudaChannelFormatKind, or if one of the dimensions is invalid. " << endl; break; case cudaErrorInvalidMemcpyDirection: cerr << "This indicates that the direction of the memcpy passed to the API call is " << endl << "not one of the types specified by ::cudaMemcpyKind. " << endl; break; case cudaErrorAddressOfConstant: cerr << "This indicated that the user has taken the address of a constant variable, " << endl << "which was forbidden up until the CUDA 3.1 release. " << endl << "This error return is deprecated as of CUDA 3.1. Variables in constant " << endl << "memory may now have their address taken by the runtime via " << endl << "::cudaGetSymbolAddress(). " << endl; break; case cudaErrorTextureFetchFailed: cerr << "This indicated that a texture fetch was not able to be performed. " << endl << "This was previously used for device emulation of texture operations. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorTextureNotBound: cerr << "This indicated that a texture was not bound for access. " << endl << "This was previously used for device emulation of texture operations. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorSynchronizationError: cerr << "This indicated that a synchronization operation had failed. " << endl << "This was previously used for some device emulation functions. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorInvalidFilterSetting: cerr << "This indicates that a non-float texture was being accessed with linear " << endl << "filtering. This is not supported by CUDA. " << endl; break; case cudaErrorInvalidNormSetting: cerr << "This indicates that an attempt was made to read a non-float texture as a " << endl << "normalized float. This is not supported by CUDA. " << endl; break; case cudaErrorMixedDeviceExecution: cerr << "Mixing of device and device emulation code was not allowed. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorCudartUnloading: cerr << "This indicated an issue with calling API functions during the unload " << endl << "process of the CUDA runtime in prior releases. " << endl << "This error return is deprecated as of CUDA 3.2. " << endl; break; case cudaErrorUnknown: cerr << "This indicates that an unknown internal error has occurred. " << endl; break; case cudaErrorNotYetImplemented: cerr << "This indicates that the API call is not yet implemented. Production " << endl << "releases of CUDA will never return this error. " << endl; break; case cudaErrorMemoryValueTooLarge: cerr << "This indicated that an emulated device pointer exceeded the 32-bit address " << endl << "range. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorInvalidResourceHandle: cerr << "This indicates that a resource handle passed to the API call was not " << endl << "valid. Resource handles are opaque types like ::cudaStream_t and " << endl << "::cudaEvent_t. " << endl; break; case cudaErrorNotReady: cerr << "This indicates that asynchronous operations issued previously have not " << endl << "completed yet. This result is not actually an error, but must be indicated " << endl << "differently than ::cudaSuccess (which indicates completion). Calls that " << endl << "may return this value include ::cudaEventQuery() and ::cudaStreamQuery(). " << endl; break; case cudaErrorInsufficientDriver: cerr << "This indicates that the installed NVIDIA CUDA driver is older than the " << endl << "CUDA runtime library. This is not a supported configuration. Users should " << endl << "install an updated NVIDIA display driver to allow the application to run. " << endl; break; case cudaErrorSetOnActiveProcess: cerr << "This indicates that the user has called ::cudaSetDevice(), " << endl << "::cudaSetValidDevices(), ::cudaSetDeviceFlags(), " << endl << "::cudaD3D9SetDirect3DDevice(), ::cudaD3D10SetDirect3DDevice, " << endl << "::cudaD3D11SetDirect3DDevice(), * or ::cudaVDPAUSetVDPAUDevice() after " << endl << "initializing the CUDA runtime by calling non-device management operations " << endl << "(allocating memory and launching kernels are examples of non-device " << endl << "management operations). This error can also be returned if using " << endl << "runtime/driver interoperability and there is an existing ::CUcontext " << endl << "active on the host thread. " << endl; break; case cudaErrorInvalidSurface: cerr << "This indicates that the surface passed to the API call is not a valid " << endl << "surface. " << endl; break; case cudaErrorNoDevice: cerr << "This indicates that no CUDA-capable devices were detected by the installed " << endl << "CUDA driver. " << endl; break; case cudaErrorECCUncorrectable: cerr << "This indicates that an uncorrectable ECC error was detected during " << endl << "execution. " << endl; break; case cudaErrorSharedObjectSymbolNotFound: cerr << "This indicates that a link to a shared object failed to resolve. " << endl; break; case cudaErrorSharedObjectInitFailed: cerr << "This indicates that initialization of a shared object failed. " << endl; break; case cudaErrorUnsupportedLimit: cerr << "This indicates that the ::cudaLimit passed to the API call is not " << endl << "supported by the active device. " << endl; break; case cudaErrorDuplicateVariableName: cerr << "This indicates that multiple global or constant variables (across separate " << endl << "CUDA source files in the application) share the same string name. " << endl; break; case cudaErrorDuplicateTextureName: cerr << "This indicates that multiple textures (across separate CUDA source " << endl << "files in the application) share the same string name. " << endl; break; case cudaErrorDuplicateSurfaceName: cerr << "This indicates that multiple surfaces (across separate CUDA source " << endl << "files in the application) share the same string name. " << endl; break; case cudaErrorDevicesUnavailable: cerr << "This indicates that all CUDA devices are busy or unavailable at the current " << endl << "time. Devices are often busy/unavailable due to use of " << endl << "::cudaComputeModeExclusive or ::cudaComputeModeProhibited. They can also " << endl << "be unavailable due to memory constraints on a device that already has " << endl << "active CUDA work being performed. " << endl; break; case cudaErrorInvalidKernelImage: cerr << "This indicates that the device kernel image is invalid. " << endl; break; case cudaErrorNoKernelImageForDevice: cerr << "This indicates that there is no kernel image available that is suitable " << endl << "for the device. This can occur when a user specifies code generation " << endl << "options for a particular CUDA source file that do not include the " << endl << "corresponding device configuration. " << endl; break; case cudaErrorIncompatibleDriverContext: cerr << "This indicates that the current context is not compatible with this " << endl << "version of the CUDA Runtime. This can only occur if you are using CUDA " << endl << "Runtime/Driver interoperability and have created an existing Driver " << endl << "context using an older API. Please see \ref CUDART_DRIVER " << endl << "\"Interactions with the CUDA Driver API\" for more information. " << endl; break; case cudaErrorStartupFailure: cerr << "This indicates an internal startup failure in the CUDA runtime. " << endl; break; case cudaErrorApiFailureBase: cerr << "Any unhandled CUDA driver error is added to this value and returned via " << endl << "the runtime. Production releases of CUDA should not return such errors. " << endl; break; } return false; #else /* HAVE_CUDA */ return true; #endif /* HAVE_CUDA */ } /*size_t tnlCuda::getFreeMemory() { Loading src/implementation/core/tnlCuda.cu 0 → 100644 +379 −0 File added.Preview size limit exceeded, changes collapsed. Show changes Loading
examples/heat-equation/heatEquationSolver_impl.h +3 −3 Original line number Diff line number Diff line Loading @@ -266,9 +266,9 @@ assemblyLinearSystem( const RealType& time, u, matrix, b ); //matrix.print( cout ); //cout << endl << b << endl; //cout << endl << u << endl; matrix.print( cout ); cout << endl << b << endl; cout << endl << u << endl; //abort(); } Loading
src/core/tnlCuda.h +4 −0 Original line number Diff line number Diff line Loading @@ -94,7 +94,11 @@ class tnlCuda static __device__ Index getInterleaving( const Index index ); #endif #ifdef HAVE_CUDA static bool checkDevice( const char* file_name, int line ); #else static bool checkDevice( const char* file_name, int line ) { return false;}; #endif }; Loading
src/implementation/core/CMakeLists.txt +1 −0 Original line number Diff line number Diff line Loading @@ -29,6 +29,7 @@ IF( BUILD_CUDA ) ${tnl_implementation_core_cuda_CUDA__SOURCES} ${tnl_implementation_core_vectors_CUDA__SOURCES} ${common_SOURCES} ${CURRENT_DIR}/tnlCuda.cu PARENT_SCOPE ) ENDIF() Loading
src/implementation/core/tnlCuda.cpp +0 −365 Original line number Diff line number Diff line Loading @@ -41,371 +41,6 @@ int tnlCuda::getNumberOfGrids( const int blocks, return roundUpDivision( blocks, gridSize ); } bool tnlCuda::checkDevice( const char* file_name, int line ) { #ifdef HAVE_CUDA cudaError error = cudaGetLastError(); if( error == cudaSuccess ) return true; cerr << "CUDA ERROR(" << error << ") at line " << line << " in " << file_name << ":" << endl; switch( error ) { case cudaErrorMissingConfiguration: cerr << "The device function being invoked (usually via ::cudaLaunch()) was not " << endl << "previously configured via the ::cudaConfigureCall() function. " << endl; break; case cudaErrorMemoryAllocation: cerr << "The API call failed because it was unable to allocate enough memory to " << endl << "perform the requested operation. " << endl; break; case cudaErrorInitializationError: cerr << "The API call failed because the CUDA driver and runtime could not be " << endl << "initialized. " << endl; break; case cudaErrorLaunchFailure: cerr << "An exception occurred on the device while executing a kernel. Common " << endl << "causes include dereferencing an invalid device pointer and accessing " << endl << "out of bounds shared memory. The device cannot be used until " << endl << "::cudaThreadExit() is called. All existing device memory allocations " << endl << "are invalid and must be reconstructed if the program is to continue " << endl << "using CUDA. " << endl; break; case cudaErrorPriorLaunchFailure: cerr << "This indicated that a previous kernel launch failed. This was previously " << endl << "used for device emulation of kernel launches. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorLaunchTimeout: cerr << "This indicates that the device kernel took too long to execute. This can " << endl << "only occur if timeouts are enabled - see the device property " << endl << "ref ::cudaDeviceProp::kernelExecTimeoutEnabled \"kernelExecTimeoutEnabled\" " << endl << "for more information. The device cannot be used until ::cudaThreadExit() " << endl << "is called. All existing device memory allocations are invalid and must be " << endl << "reconstructed if the program is to continue using CUDA. " << endl; break; case cudaErrorLaunchOutOfResources: cerr << "This indicates that a launch did not occur because it did not have " << endl << "appropriate resources. Although this error is similar to " << endl << "::cudaErrorInvalidConfiguration, this error usually indicates that the " << endl << "user has attempted to pass too many arguments to the device kernel, or the " << endl << "kernel launch specifies too many threads for the kernel's register count. " << endl; break; case cudaErrorInvalidDeviceFunction: cerr << "The requested device function does not exist or is not compiled for the " << endl << "proper device architecture. " << endl; break; case cudaErrorInvalidConfiguration: cerr << "This indicates that a kernel launch is requesting resources that can " << endl << "never be satisfied by the current device. Requesting more shared memory " << endl << "per block than the device supports will trigger this error, as will " << endl << "requesting too many threads or blocks. See ::cudaDeviceProp for more " << endl << "device limitations. " << endl; break; case cudaErrorInvalidDevice: cerr << "This indicates that the device ordinal supplied by the user does not " << endl << "correspond to a valid CUDA device. " << endl; break; case cudaErrorInvalidValue: cerr << "This indicates that one or more of the parameters passed to the API call " << endl << "is not within an acceptable range of values. " << endl; break; case cudaErrorInvalidPitchValue: cerr << "This indicates that one or more of the pitch-related parameters passed " << endl << "to the API call is not within the acceptable range for pitch. " << endl; break; case cudaErrorInvalidSymbol: cerr << "This indicates that the symbol name/identifier passed to the API call " << endl << "is not a valid name or identifier. " << endl; break; case cudaErrorMapBufferObjectFailed: cerr << "This indicates that the buffer object could not be mapped. " << endl; break; case cudaErrorUnmapBufferObjectFailed: cerr << "This indicates that the buffer object could not be unmapped. " << endl; break; case cudaErrorInvalidHostPointer: cerr << "This indicates that at least one host pointer passed to the API call is " << endl << "not a valid host pointer. " << endl; break; case cudaErrorInvalidDevicePointer: cerr << "This indicates that at least one device pointer passed to the API call is " << endl << "not a valid device pointer. " << endl; break; case cudaErrorInvalidTexture: cerr << "This indicates that the texture passed to the API call is not a valid " << endl << "texture. " << endl; break; case cudaErrorInvalidTextureBinding: cerr << "This indicates that the texture binding is not valid. This occurs if you " << endl << "call ::cudaGetTextureAlignmentOffset() with an unbound texture. " << endl; break; case cudaErrorInvalidChannelDescriptor: cerr << "This indicates that the channel descriptor passed to the API call is not " << endl << "valid. This occurs if the format is not one of the formats specified by " << endl << "::cudaChannelFormatKind, or if one of the dimensions is invalid. " << endl; break; case cudaErrorInvalidMemcpyDirection: cerr << "This indicates that the direction of the memcpy passed to the API call is " << endl << "not one of the types specified by ::cudaMemcpyKind. " << endl; break; case cudaErrorAddressOfConstant: cerr << "This indicated that the user has taken the address of a constant variable, " << endl << "which was forbidden up until the CUDA 3.1 release. " << endl << "This error return is deprecated as of CUDA 3.1. Variables in constant " << endl << "memory may now have their address taken by the runtime via " << endl << "::cudaGetSymbolAddress(). " << endl; break; case cudaErrorTextureFetchFailed: cerr << "This indicated that a texture fetch was not able to be performed. " << endl << "This was previously used for device emulation of texture operations. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorTextureNotBound: cerr << "This indicated that a texture was not bound for access. " << endl << "This was previously used for device emulation of texture operations. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorSynchronizationError: cerr << "This indicated that a synchronization operation had failed. " << endl << "This was previously used for some device emulation functions. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorInvalidFilterSetting: cerr << "This indicates that a non-float texture was being accessed with linear " << endl << "filtering. This is not supported by CUDA. " << endl; break; case cudaErrorInvalidNormSetting: cerr << "This indicates that an attempt was made to read a non-float texture as a " << endl << "normalized float. This is not supported by CUDA. " << endl; break; case cudaErrorMixedDeviceExecution: cerr << "Mixing of device and device emulation code was not allowed. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorCudartUnloading: cerr << "This indicated an issue with calling API functions during the unload " << endl << "process of the CUDA runtime in prior releases. " << endl << "This error return is deprecated as of CUDA 3.2. " << endl; break; case cudaErrorUnknown: cerr << "This indicates that an unknown internal error has occurred. " << endl; break; case cudaErrorNotYetImplemented: cerr << "This indicates that the API call is not yet implemented. Production " << endl << "releases of CUDA will never return this error. " << endl; break; case cudaErrorMemoryValueTooLarge: cerr << "This indicated that an emulated device pointer exceeded the 32-bit address " << endl << "range. " << endl << "This error return is deprecated as of CUDA 3.1. Device emulation mode was " << endl << "removed with the CUDA 3.1 release. " << endl; break; case cudaErrorInvalidResourceHandle: cerr << "This indicates that a resource handle passed to the API call was not " << endl << "valid. Resource handles are opaque types like ::cudaStream_t and " << endl << "::cudaEvent_t. " << endl; break; case cudaErrorNotReady: cerr << "This indicates that asynchronous operations issued previously have not " << endl << "completed yet. This result is not actually an error, but must be indicated " << endl << "differently than ::cudaSuccess (which indicates completion). Calls that " << endl << "may return this value include ::cudaEventQuery() and ::cudaStreamQuery(). " << endl; break; case cudaErrorInsufficientDriver: cerr << "This indicates that the installed NVIDIA CUDA driver is older than the " << endl << "CUDA runtime library. This is not a supported configuration. Users should " << endl << "install an updated NVIDIA display driver to allow the application to run. " << endl; break; case cudaErrorSetOnActiveProcess: cerr << "This indicates that the user has called ::cudaSetDevice(), " << endl << "::cudaSetValidDevices(), ::cudaSetDeviceFlags(), " << endl << "::cudaD3D9SetDirect3DDevice(), ::cudaD3D10SetDirect3DDevice, " << endl << "::cudaD3D11SetDirect3DDevice(), * or ::cudaVDPAUSetVDPAUDevice() after " << endl << "initializing the CUDA runtime by calling non-device management operations " << endl << "(allocating memory and launching kernels are examples of non-device " << endl << "management operations). This error can also be returned if using " << endl << "runtime/driver interoperability and there is an existing ::CUcontext " << endl << "active on the host thread. " << endl; break; case cudaErrorInvalidSurface: cerr << "This indicates that the surface passed to the API call is not a valid " << endl << "surface. " << endl; break; case cudaErrorNoDevice: cerr << "This indicates that no CUDA-capable devices were detected by the installed " << endl << "CUDA driver. " << endl; break; case cudaErrorECCUncorrectable: cerr << "This indicates that an uncorrectable ECC error was detected during " << endl << "execution. " << endl; break; case cudaErrorSharedObjectSymbolNotFound: cerr << "This indicates that a link to a shared object failed to resolve. " << endl; break; case cudaErrorSharedObjectInitFailed: cerr << "This indicates that initialization of a shared object failed. " << endl; break; case cudaErrorUnsupportedLimit: cerr << "This indicates that the ::cudaLimit passed to the API call is not " << endl << "supported by the active device. " << endl; break; case cudaErrorDuplicateVariableName: cerr << "This indicates that multiple global or constant variables (across separate " << endl << "CUDA source files in the application) share the same string name. " << endl; break; case cudaErrorDuplicateTextureName: cerr << "This indicates that multiple textures (across separate CUDA source " << endl << "files in the application) share the same string name. " << endl; break; case cudaErrorDuplicateSurfaceName: cerr << "This indicates that multiple surfaces (across separate CUDA source " << endl << "files in the application) share the same string name. " << endl; break; case cudaErrorDevicesUnavailable: cerr << "This indicates that all CUDA devices are busy or unavailable at the current " << endl << "time. Devices are often busy/unavailable due to use of " << endl << "::cudaComputeModeExclusive or ::cudaComputeModeProhibited. They can also " << endl << "be unavailable due to memory constraints on a device that already has " << endl << "active CUDA work being performed. " << endl; break; case cudaErrorInvalidKernelImage: cerr << "This indicates that the device kernel image is invalid. " << endl; break; case cudaErrorNoKernelImageForDevice: cerr << "This indicates that there is no kernel image available that is suitable " << endl << "for the device. This can occur when a user specifies code generation " << endl << "options for a particular CUDA source file that do not include the " << endl << "corresponding device configuration. " << endl; break; case cudaErrorIncompatibleDriverContext: cerr << "This indicates that the current context is not compatible with this " << endl << "version of the CUDA Runtime. This can only occur if you are using CUDA " << endl << "Runtime/Driver interoperability and have created an existing Driver " << endl << "context using an older API. Please see \ref CUDART_DRIVER " << endl << "\"Interactions with the CUDA Driver API\" for more information. " << endl; break; case cudaErrorStartupFailure: cerr << "This indicates an internal startup failure in the CUDA runtime. " << endl; break; case cudaErrorApiFailureBase: cerr << "Any unhandled CUDA driver error is added to this value and returned via " << endl << "the runtime. Production releases of CUDA should not return such errors. " << endl; break; } return false; #else /* HAVE_CUDA */ return true; #endif /* HAVE_CUDA */ } /*size_t tnlCuda::getFreeMemory() { Loading
src/implementation/core/tnlCuda.cu 0 → 100644 +379 −0 File added.Preview size limit exceeded, changes collapsed. Show changes
