Commit 3e18214b authored by Brice Videau's avatar Brice Videau
Browse files

Added missing cuda headers.

parent d9a6b13a
Pipeline #11571 failed with stage
......@@ -70,13 +70,29 @@ EXTRA_DIST += \
cudart_model.rb
CUDA_EXTRA_HDR = \
$(srcdir)/include/cuda_runtime_api.h
$(srcdir)/include/builtin_types.h \
$(srcdir)/include/driver_types.h \
$(srcdir)/include/texture_types.h \
$(srcdir)/include/surface_types.h \
$(srcdir)/include/crt/host_defines.h \
$(srcdir)/include/device_types.h \
$(srcdir)/include/cuda_device_runtime_api.h \
$(srcdir)/include/cuda_runtime_api.h \
$(srcdir)/include/vector_types.h
CUDA_HDR = \
$(srcdir)/include/cuda.h \
$(CUDA_EXTRA_HDR)
MODIFIED_CUDA_EXTRA_HDR = \
modified_include/cuda_runtime_api.h
modified_include/builtin_types.h \
modified_include/driver_types.h \
modified_include/texture_types.h \
modified_include/surface_types.h \
modified_include/crt/host_defines.h \
modified_include/device_types.h \
modified_include/cuda_device_runtime_api.h \
modified_include/cuda_runtime_api.h \
modified_include/vector_types.h
MODIFIED_CUDA_HDR = \
modified_include/cuda.h \
$(MODIFIED_CUDA_EXTRA_HDR)
......
/*
* Copyright 1993-2014 NVIDIA Corporation. All rights reserved.
*
* NOTICE TO LICENSEE:
*
* This source code and/or documentation ("Licensed Deliverables") are
* subject to NVIDIA intellectual property rights under U.S. and
* international Copyright laws.
*
* These Licensed Deliverables contained herein is PROPRIETARY and
* CONFIDENTIAL to NVIDIA and is being provided under the terms and
* conditions of a form of NVIDIA software license agreement by and
* between NVIDIA and Licensee ("License Agreement") or electronically
* accepted by Licensee. Notwithstanding any terms or conditions to
* the contrary in the License Agreement, reproduction or disclosure
* of the Licensed Deliverables to any third party without the express
* written consent of NVIDIA is prohibited.
*
* NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE
* LICENSE AGREEMENT, NVIDIA MAKES NO REPRESENTATION ABOUT THE
* SUITABILITY OF THESE LICENSED DELIVERABLES FOR ANY PURPOSE. IT IS
* PROVIDED "AS IS" WITHOUT EXPRESS OR IMPLIED WARRANTY OF ANY KIND.
* NVIDIA DISCLAIMS ALL WARRANTIES WITH REGARD TO THESE LICENSED
* DELIVERABLES, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY,
* NONINFRINGEMENT, AND FITNESS FOR A PARTICULAR PURPOSE.
* NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE
* LICENSE AGREEMENT, IN NO EVENT SHALL NVIDIA BE LIABLE FOR ANY
* SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, OR ANY
* DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,
* WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
* ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE
* OF THESE LICENSED DELIVERABLES.
*
* U.S. Government End Users. These Licensed Deliverables are a
* "commercial item" as that term is defined at 48 C.F.R. 2.101 (OCT
* 1995), consisting of "commercial computer software" and "commercial
* computer software documentation" as such terms are used in 48
* C.F.R. 12.212 (SEPT 1995) and is provided to the U.S. Government
* only as a commercial end item. Consistent with 48 C.F.R.12.212 and
* 48 C.F.R. 227.7202-1 through 227.7202-4 (JUNE 1995), all
* U.S. Government End Users acquire the Licensed Deliverables with
* only those rights set forth herein.
*
* Any use of the Licensed Deliverables in individual and commercial
* software must include, in the user documentation and internal
* comments to the code, the above Disclaimer and U.S. Government End
* Users Notice.
*/
/*******************************************************************************
* *
* *
* *
*******************************************************************************/
#include "device_types.h"
#if !defined(__CUDACC_RTC__)
#define EXCLUDE_FROM_RTC
#include "driver_types.h"
#undef EXCLUDE_FROM_RTC
#endif /* !__CUDACC_RTC__ */
#include "surface_types.h"
#include "texture_types.h"
#include "vector_types.h"
/*
* Copyright 1993-2017 NVIDIA Corporation. All rights reserved.
*
* NOTICE TO LICENSEE:
*
* This source code and/or documentation ("Licensed Deliverables") are
* subject to NVIDIA intellectual property rights under U.S. and
* international Copyright laws.
*
* These Licensed Deliverables contained herein is PROPRIETARY and
* CONFIDENTIAL to NVIDIA and is being provided under the terms and
* conditions of a form of NVIDIA software license agreement by and
* between NVIDIA and Licensee ("License Agreement") or electronically
* accepted by Licensee. Notwithstanding any terms or conditions to
* the contrary in the License Agreement, reproduction or disclosure
* of the Licensed Deliverables to any third party without the express
* written consent of NVIDIA is prohibited.
*
* NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE
* LICENSE AGREEMENT, NVIDIA MAKES NO REPRESENTATION ABOUT THE
* SUITABILITY OF THESE LICENSED DELIVERABLES FOR ANY PURPOSE. IT IS
* PROVIDED "AS IS" WITHOUT EXPRESS OR IMPLIED WARRANTY OF ANY KIND.
* NVIDIA DISCLAIMS ALL WARRANTIES WITH REGARD TO THESE LICENSED
* DELIVERABLES, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY,
* NONINFRINGEMENT, AND FITNESS FOR A PARTICULAR PURPOSE.
* NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE
* LICENSE AGREEMENT, IN NO EVENT SHALL NVIDIA BE LIABLE FOR ANY
* SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, OR ANY
* DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,
* WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
* ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE
* OF THESE LICENSED DELIVERABLES.
*
* U.S. Government End Users. These Licensed Deliverables are a
* "commercial item" as that term is defined at 48 C.F.R. 2.101 (OCT
* 1995), consisting of "commercial computer software" and "commercial
* computer software documentation" as such terms are used in 48
* C.F.R. 12.212 (SEPT 1995) and is provided to the U.S. Government
* only as a commercial end item. Consistent with 48 C.F.R.12.212 and
* 48 C.F.R. 227.7202-1 through 227.7202-4 (JUNE 1995), all
* U.S. Government End Users acquire the Licensed Deliverables with
* only those rights set forth herein.
*
* Any use of the Licensed Deliverables in individual and commercial
* software must include, in the user documentation and internal
* comments to the code, the above Disclaimer and U.S. Government End
* Users Notice.
*/
#if !defined(__CUDA_INCLUDE_COMPILER_INTERNAL_HEADERS__)
#if defined(_MSC_VER)
#pragma message("crt/host_defines.h is an internal header file and must not be used directly. Please use cuda_runtime_api.h or cuda_runtime.h instead.")
#else
#warning "crt/host_defines.h is an internal header file and must not be used directly. Please use cuda_runtime_api.h or cuda_runtime.h instead."
#endif
#define __CUDA_INCLUDE_COMPILER_INTERNAL_HEADERS__
#define __UNDEF_CUDA_INCLUDE_COMPILER_INTERNAL_HEADERS_HOST_DEFINES_H__
#endif
#if !defined(__HOST_DEFINES_H__)
#define __HOST_DEFINES_H__
/* CUDA JIT mode (__CUDACC_RTC__) also uses GNU style attributes */
#if defined(__GNUC__) || (defined(__PGIC__) && defined(__linux__)) || defined(__CUDA_LIBDEVICE__) || defined(__CUDACC_RTC__)
#if defined(__CUDACC_RTC__)
#define __volatile__ volatile
#endif /* __CUDACC_RTC__ */
#define __no_return__ \
__attribute__((noreturn))
#if defined(__CUDACC__) || defined(__CUDA_ARCH__) || defined(__CUDA_LIBDEVICE__)
/* gcc allows users to define attributes with underscores,
e.g., __attribute__((__noinline__)).
Consider a non-CUDA source file (e.g. .cpp) that has the
above attribute specification, and includes this header file. In that case,
defining __noinline__ as below would cause a gcc compilation error.
Hence, only define __noinline__ when the code is being processed
by a CUDA compiler component.
*/
#define __noinline__ \
__attribute__((noinline))
#endif /* __CUDACC__ || __CUDA_ARCH__ || __CUDA_LIBDEVICE__ */
#define __forceinline__ \
__inline__ __attribute__((always_inline))
#define __align__(n) \
__attribute__((aligned(n)))
#define __thread__ \
__thread
#define __import__
#define __export__
#define __cdecl
#define __annotate__(a) \
__attribute__((a))
#define __location__(a) \
__annotate__(a)
#define CUDARTAPI
#elif defined(_MSC_VER)
#if _MSC_VER >= 1400
#define __restrict__ \
__restrict
#else /* _MSC_VER >= 1400 */
#define __restrict__
#endif /* _MSC_VER >= 1400 */
#define __inline__ \
__inline
#define __no_return__ \
__declspec(noreturn)
#define __noinline__ \
__declspec(noinline)
#define __forceinline__ \
__forceinline
#define __align__(n) \
__declspec(align(n))
#define __thread__ \
__declspec(thread)
#define __import__ \
__declspec(dllimport)
#define __export__ \
__declspec(dllexport)
#define __annotate__(a) \
__declspec(a)
#define __location__(a) \
__annotate__(__##a##__)
#define CUDARTAPI \
__stdcall
#else /* __GNUC__ || __CUDA_LIBDEVICE__ || __CUDACC_RTC__ */
#define __inline__
#if !defined(__align__)
#error --- !!! UNKNOWN COMPILER: please provide a CUDA compatible definition for '__align__' !!! ---
#endif /* !__align__ */
#if !defined(CUDARTAPI)
#error --- !!! UNKNOWN COMPILER: please provide a CUDA compatible definition for 'CUDARTAPI' !!! ---
#endif /* !CUDARTAPI */
#endif /* __GNUC__ || __CUDA_LIBDEVICE__ || __CUDACC_RTC__ */
#if (defined(__GNUC__) && (__GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 3 && !defined(__clang__)))) || \
(defined(_MSC_VER) && _MSC_VER < 1900) || \
(!defined(__GNUC__) && !defined(_MSC_VER))
#define __specialization_static \
static
#else /* (__GNUC__ && (__GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 3 && !__clang__))) ||
(_MSC_VER && _MSC_VER < 1900) ||
(!__GNUC__ && !_MSC_VER) */
#define __specialization_static
#endif /* (__GNUC__ && (__GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 3 && !__clang__))) ||
(_MSC_VER && _MSC_VER < 1900) ||
(!__GNUC__ && !_MSC_VER) */
#if !defined(__CUDACC__) && !defined(__CUDA_LIBDEVICE__)
#undef __annotate__
#define __annotate__(a)
#else /* !__CUDACC__ && !__CUDA_LIBDEVICE__ */
#define __launch_bounds__(...) \
__annotate__(launch_bounds(__VA_ARGS__))
#endif /* !__CUDACC__ && !__CUDA_LIBDEVICE__ */
#if defined(__CUDACC__) || defined(__CUDA_LIBDEVICE__) || \
defined(__GNUC__) || defined(_WIN64)
#define __builtin_align__(a) \
__align__(a)
#else /* __CUDACC__ || __CUDA_LIBDEVICE__ || __GNUC__ || _WIN64 */
#define __builtin_align__(a)
#endif /* __CUDACC__ || __CUDA_LIBDEVICE__ || __GNUC__ || _WIN64 */
#if defined(__CUDACC__) || !defined(__host__)
#define __host__ \
__location__(host)
#endif /* defined(__CUDACC__) || !defined(__host__) */
#if defined(__CUDACC__) || !defined(__device__)
#define __device__ \
__location__(device)
#endif /* defined(__CUDACC__) || !defined(__device__) */
#if defined(__CUDACC__) || !defined(__global__)
#define __global__ \
__location__(global)
#endif /* defined(__CUDACC__) || !defined(__global__) */
#if defined(__CUDACC__) || !defined(__shared__)
#define __shared__ \
__location__(shared)
#endif /* defined(__CUDACC__) || !defined(__shared__) */
#if defined(__CUDACC__) || !defined(__constant__)
#define __constant__ \
__location__(constant)
#endif /* defined(__CUDACC__) || !defined(__constant__) */
#if defined(__CUDACC__) || !defined(__managed__)
#define __managed__ \
__location__(managed)
#endif /* defined(__CUDACC__) || !defined(__managed__) */
#if !defined(__CUDACC__)
#define __device_builtin__
#define __device_builtin_texture_type__
#define __device_builtin_surface_type__
#define __cudart_builtin__
#else /* defined(__CUDACC__) */
#define __device_builtin__ \
__location__(device_builtin)
#define __device_builtin_texture_type__ \
__location__(device_builtin_texture_type)
#define __device_builtin_surface_type__ \
__location__(device_builtin_surface_type)
#define __cudart_builtin__ \
__location__(cudart_builtin)
#endif /* !defined(__CUDACC__) */
#endif /* !__HOST_DEFINES_H__ */
#if defined(__UNDEF_CUDA_INCLUDE_COMPILER_INTERNAL_HEADERS_HOST_DEFINES_H__)
#undef __CUDA_INCLUDE_COMPILER_INTERNAL_HEADERS__
#undef __UNDEF_CUDA_INCLUDE_COMPILER_INTERNAL_HEADERS_HOST_DEFINES_H__
#endif
/*
* Copyright 1993-2018 NVIDIA Corporation. All rights reserved.
*
* NOTICE TO LICENSEE:
*
* This source code and/or documentation ("Licensed Deliverables") are
* subject to NVIDIA intellectual property rights under U.S. and
* international Copyright laws.
*
* These Licensed Deliverables contained herein is PROPRIETARY and
* CONFIDENTIAL to NVIDIA and is being provided under the terms and
* conditions of a form of NVIDIA software license agreement by and
* between NVIDIA and Licensee ("License Agreement") or electronically
* accepted by Licensee. Notwithstanding any terms or conditions to
* the contrary in the License Agreement, reproduction or disclosure
* of the Licensed Deliverables to any third party without the express
* written consent of NVIDIA is prohibited.
*
* NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE
* LICENSE AGREEMENT, NVIDIA MAKES NO REPRESENTATION ABOUT THE
* SUITABILITY OF THESE LICENSED DELIVERABLES FOR ANY PURPOSE. IT IS
* PROVIDED "AS IS" WITHOUT EXPRESS OR IMPLIED WARRANTY OF ANY KIND.
* NVIDIA DISCLAIMS ALL WARRANTIES WITH REGARD TO THESE LICENSED
* DELIVERABLES, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY,
* NONINFRINGEMENT, AND FITNESS FOR A PARTICULAR PURPOSE.
* NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE
* LICENSE AGREEMENT, IN NO EVENT SHALL NVIDIA BE LIABLE FOR ANY
* SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES, OR ANY
* DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,
* WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
* ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE
* OF THESE LICENSED DELIVERABLES.
*
* U.S. Government End Users. These Licensed Deliverables are a
* "commercial item" as that term is defined at 48 C.F.R. 2.101 (OCT
* 1995), consisting of "commercial computer software" and "commercial
* computer software documentation" as such terms are used in 48
* C.F.R. 12.212 (SEPT 1995) and is provided to the U.S. Government
* only as a commercial end item. Consistent with 48 C.F.R.12.212 and
* 48 C.F.R. 227.7202-1 through 227.7202-4 (JUNE 1995), all
* U.S. Government End Users acquire the Licensed Deliverables with
* only those rights set forth herein.
*
* Any use of the Licensed Deliverables in individual and commercial
* software must include, in the user documentation and internal
* comments to the code, the above Disclaimer and U.S. Government End
* Users Notice.
*/
#if !defined(__CUDA_DEVICE_RUNTIME_API_H__)
#define __CUDA_DEVICE_RUNTIME_API_H__
/*******************************************************************************
* *
* *
* *
*******************************************************************************/
#if !defined(__CUDACC_RTC__)
#if !defined(__CUDACC_RDC__) && !defined(__CUDACC_EWP__) && defined(__CUDA_ARCH__) && (__CUDA_ARCH__ >= 350) && !defined(__CUDADEVRT_INTERNAL__)
#if defined(__cplusplus)
extern "C" {
#endif
struct cudaFuncAttributes;
#if defined(_WIN32)
#define __NV_WEAK__ __declspec(nv_weak)
#else
#define __NV_WEAK__ __attribute__((nv_weak))
#endif
__device__ __NV_WEAK__ cudaError_t CUDARTAPI cudaMalloc(void **p, size_t s)
{
return cudaErrorUnknown;
}
__device__ __NV_WEAK__ cudaError_t CUDARTAPI cudaFuncGetAttributes(struct cudaFuncAttributes *p, const void *c)
{
return cudaErrorUnknown;
}
__device__ __NV_WEAK__ cudaError_t CUDARTAPI cudaDeviceGetAttribute(int *value, enum cudaDeviceAttr attr, int device)
{
return cudaErrorUnknown;
}
__device__ __NV_WEAK__ cudaError_t CUDARTAPI cudaGetDevice(int *device)
{
return cudaErrorUnknown;
}
__device__ __NV_WEAK__ cudaError_t CUDARTAPI cudaOccupancyMaxActiveBlocksPerMultiprocessor(int *numBlocks, const void *func, int blockSize, size_t dynamicSmemSize)
{
return cudaErrorUnknown;
}
__device__ __NV_WEAK__ cudaError_t CUDARTAPI cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(int *numBlocks, const void *func, int blockSize, size_t dynamicSmemSize, unsigned int flags)
{
return cudaErrorUnknown;
}
#undef __NV_WEAK__
#if defined(__cplusplus)
}
#endif
#endif /* !defined(__CUDACC_RDC__) && !defined(__CUDACC_EWP__) && defined(__CUDA_ARCH__) && (__CUDA_ARCH__ >= 350) && !defined(__CUDADEVRT_INTERNAL__) */
#endif /* !defined(__CUDACC_RTC__) */
#if defined(__cplusplus) && defined(__CUDACC__) /* Visible to nvcc front-end only */
#if !defined(__CUDA_ARCH__) || (__CUDA_ARCH__ >= 350) // Visible to SM>=3.5 and "__host__ __device__" only
#include "driver_types.h"
#include "crt/host_defines.h"
extern "C"
{
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaDeviceGetAttribute(int *value, enum cudaDeviceAttr attr, int device);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaDeviceGetLimit(size_t *pValue, enum cudaLimit limit);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaDeviceGetCacheConfig(enum cudaFuncCache *pCacheConfig);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaDeviceGetSharedMemConfig(enum cudaSharedMemConfig *pConfig);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaDeviceSynchronize(void);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaGetLastError(void);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaPeekAtLastError(void);
extern __device__ __cudart_builtin__ const char* CUDARTAPI cudaGetErrorString(cudaError_t error);
extern __device__ __cudart_builtin__ const char* CUDARTAPI cudaGetErrorName(cudaError_t error);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaGetDeviceCount(int *count);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaGetDevice(int *device);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaStreamCreateWithFlags(cudaStream_t *pStream, unsigned int flags);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaStreamDestroy(cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaStreamWaitEvent(cudaStream_t stream, cudaEvent_t event, unsigned int flags);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaStreamWaitEvent_ptsz(cudaStream_t stream, cudaEvent_t event, unsigned int flags);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaEventCreateWithFlags(cudaEvent_t *event, unsigned int flags);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaEventRecord(cudaEvent_t event, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaEventRecord_ptsz(cudaEvent_t event, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaEventDestroy(cudaEvent_t event);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaFuncGetAttributes(struct cudaFuncAttributes *attr, const void *func);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaFree(void *devPtr);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMalloc(void **devPtr, size_t size);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemcpyAsync(void *dst, const void *src, size_t count, enum cudaMemcpyKind kind, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemcpyAsync_ptsz(void *dst, const void *src, size_t count, enum cudaMemcpyKind kind, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemcpy2DAsync(void *dst, size_t dpitch, const void *src, size_t spitch, size_t width, size_t height, enum cudaMemcpyKind kind, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemcpy2DAsync_ptsz(void *dst, size_t dpitch, const void *src, size_t spitch, size_t width, size_t height, enum cudaMemcpyKind kind, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemcpy3DAsync(const struct cudaMemcpy3DParms *p, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemcpy3DAsync_ptsz(const struct cudaMemcpy3DParms *p, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemsetAsync(void *devPtr, int value, size_t count, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemsetAsync_ptsz(void *devPtr, int value, size_t count, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemset2DAsync(void *devPtr, size_t pitch, int value, size_t width, size_t height, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemset2DAsync_ptsz(void *devPtr, size_t pitch, int value, size_t width, size_t height, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemset3DAsync(struct cudaPitchedPtr pitchedDevPtr, int value, struct cudaExtent extent, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaMemset3DAsync_ptsz(struct cudaPitchedPtr pitchedDevPtr, int value, struct cudaExtent extent, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaRuntimeGetVersion(int *runtimeVersion);
/**
* \ingroup CUDART_EXECUTION
* \brief Obtains a parameter buffer
*
* Obtains a parameter buffer which can be filled with parameters for a kernel launch.
* Parameters passed to ::cudaLaunchDevice must be allocated via this function.
*
* This is a low level API and can only be accessed from Parallel Thread Execution (PTX).
* CUDA user code should use <<< >>> to launch kernels.
*
* \param alignment - Specifies alignment requirement of the parameter buffer
* \param size - Specifies size requirement in bytes
*
* \return
* Returns pointer to the allocated parameterBuffer
* \notefnerr
*
* \sa cudaLaunchDevice
*/
extern __device__ __cudart_builtin__ void * CUDARTAPI cudaGetParameterBuffer(size_t alignment, size_t size);
/**
* \ingroup CUDART_EXECUTION
* \brief Launches a specified kernel
*
* Launches a specified kernel with the specified parameter buffer. A parameter buffer can be obtained
* by calling ::cudaGetParameterBuffer().
*
* This is a low level API and can only be accessed from Parallel Thread Execution (PTX).
* CUDA user code should use <<< >>> to launch the kernels.
*
* \param func - Pointer to the kernel to be launched
* \param parameterBuffer - Holds the parameters to the launched kernel. parameterBuffer can be NULL. (Optional)
* \param gridDimension - Specifies grid dimensions
* \param blockDimension - Specifies block dimensions
* \param sharedMemSize - Specifies size of shared memory
* \param stream - Specifies the stream to be used
*
* \return
* ::cudaSuccess, ::cudaErrorInvalidDevice, ::cudaErrorLaunchMaxDepthExceeded, ::cudaErrorInvalidConfiguration,
* ::cudaErrorStartupFailure, ::cudaErrorLaunchPendingCountExceeded, ::cudaErrorLaunchOutOfResources
* \notefnerr
* \n Please refer to Execution Configuration and Parameter Buffer Layout from the CUDA Programming
* Guide for the detailed descriptions of launch configuration and parameter layout respectively.
*
* \sa cudaGetParameterBuffer
*/
extern __device__ __cudart_builtin__ void * CUDARTAPI cudaGetParameterBufferV2(void *func, dim3 gridDimension, dim3 blockDimension, unsigned int sharedMemSize);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaLaunchDevice_ptsz(void *func, void *parameterBuffer, dim3 gridDimension, dim3 blockDimension, unsigned int sharedMemSize, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaLaunchDeviceV2_ptsz(void *parameterBuffer, cudaStream_t stream);
#if defined(CUDA_API_PER_THREAD_DEFAULT_STREAM) && defined(__CUDA_ARCH__)
// When compiling for the device and per thread default stream is enabled, add
// a static inline redirect to the per thread stream entry points.
static __inline__ __device__ __cudart_builtin__ cudaError_t CUDARTAPI
cudaLaunchDevice(void *func, void *parameterBuffer, dim3 gridDimension, dim3 blockDimension, unsigned int sharedMemSize, cudaStream_t stream)
{
return cudaLaunchDevice_ptsz(func, parameterBuffer, gridDimension, blockDimension, sharedMemSize, stream);
}
static __inline__ __device__ __cudart_builtin__ cudaError_t CUDARTAPI
cudaLaunchDeviceV2(void *parameterBuffer, cudaStream_t stream)
{
return cudaLaunchDeviceV2_ptsz(parameterBuffer, stream);
}
#else
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaLaunchDevice(void *func, void *parameterBuffer, dim3 gridDimension, dim3 blockDimension, unsigned int sharedMemSize, cudaStream_t stream);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaLaunchDeviceV2(void *parameterBuffer, cudaStream_t stream);
#endif
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaOccupancyMaxActiveBlocksPerMultiprocessor(int *numBlocks, const void *func, int blockSize, size_t dynamicSmemSize);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(int *numBlocks, const void *func, int blockSize, size_t dynamicSmemSize, unsigned int flags);
extern __device__ __cudart_builtin__ unsigned long long CUDARTAPI cudaCGGetIntrinsicHandle(enum cudaCGScope scope);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaCGSynchronize(unsigned long long handle, unsigned int flags);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaCGSynchronizeGrid(unsigned long long handle, unsigned int flags);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaCGGetSize(unsigned int *numThreads, unsigned int *numGrids, unsigned long long handle);
extern __device__ __cudart_builtin__ cudaError_t CUDARTAPI cudaCGGetRank(unsigned int *threadRank, unsigned int *gridRank, unsigned long long handle);
}
template <typename T> static __inline__ __device__ __cudart_builtin__ cudaError_t cudaMalloc(T **devPtr, size_t size);
template <typename T> static __inline__ __device__ __cudart_builtin__ cudaError_t cudaFuncGetAttributes(struct cudaFuncAttributes *attr, T *entry);
template <typename T> static __inline__ __device__ __cudart_builtin__ cudaError_t cudaOccupancyMaxActiveBlocksPerMultiprocessor(int *numBlocks, T func, int blockSize, size_t dynamicSmemSize);
template <typename T> static __inline__ __device__ __cudart_builtin__ cudaError_t cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(int *numBlocks, T func, int blockSize, size_t dynamicSmemSize, unsigned int flags);
#endif // !defined(__CUDA_ARCH__) || (__CUDA_ARCH__ >= 350)
#endif /* defined(__cplusplus) && defined(__CUDACC__) */
#endif /* !__CUDA_DEVICE_RUNTIME_API_H__ */
/*
* Copyright 1993-2018 NVIDIA Corporation. All rights reserved.
*
* NOTICE TO LICENSEE:
*
* This source code and/or documentation ("Licensed Deliverables") are
* subject to NVIDIA intellectual property rights under U.S. and
* international Copyright laws.
*
* These Licensed Deliverables contained herein is PROPRIETARY and
* CONFIDENTIAL to NVIDIA and is being provided under the terms and
* conditions of a form of NVIDIA software license agreement by and
* between NVIDIA and Licensee ("License Agreement") or electronically
* accepted by Licensee. Notwithstanding any terms or conditions to
* the contrary in the License Agreement, reproduction or disclosure
* of the Licensed Deliverables to any third party without the express
* written consent of NVIDIA is prohibited.
*
* NOTWITHSTANDING ANY TERMS OR CONDITIONS TO THE CONTRARY IN THE
* LICENSE AGREEMENT, NVIDIA MAKES NO REPRESENTATION ABOUT THE
* SUITABILITY OF THESE LICENSED DELIVERABLES FOR ANY PURPOSE. IT IS
* PROVIDED "AS IS" WITHOUT EXPRESS OR IMPLIED WARRANTY OF ANY KIND.
* NVIDIA DISCLAIMS ALL WARRANTIES WITH REGARD TO THESE LICENSED