Evgeny Mankov
@@ -33,34 +33,34 @@ THE SOFTWARE.
|
||||
#include <hip_common.h>
|
||||
|
||||
typedef struct {
|
||||
// 32-bit Atomics:
|
||||
unsigned hasGlobalInt32Atomics : 1; ///< 32-bit integer atomics for global memory
|
||||
unsigned hasGlobalFloatAtomicExch : 1; ///< 32-bit float atomic exch for global memory
|
||||
unsigned hasSharedInt32Atomics : 1; ///< 32-bit integer atomics for shared memory
|
||||
unsigned hasSharedFloatAtomicExch : 1; ///< 32-bit float atomic exch for shared memory
|
||||
unsigned hasFloatAtomicAdd : 1; ///< 32-bit float atomic add in global and shared memory
|
||||
// 32-bit Atomics
|
||||
unsigned hasGlobalInt32Atomics : 1; ///< 32-bit integer atomics for global memory.
|
||||
unsigned hasGlobalFloatAtomicExch : 1; ///< 32-bit float atomic exch for global memory.
|
||||
unsigned hasSharedInt32Atomics : 1; ///< 32-bit integer atomics for shared memory.
|
||||
unsigned hasSharedFloatAtomicExch : 1; ///< 32-bit float atomic exch for shared memory.
|
||||
unsigned hasFloatAtomicAdd : 1; ///< 32-bit float atomic add in global and shared memory.
|
||||
|
||||
// 64-bit Atomics:
|
||||
unsigned hasGlobalInt64Atomics : 1; ///< 64-bit integer atomics for global memory
|
||||
unsigned hasSharedInt64Atomics : 1; ///< 64-bit integer atomics for shared memory
|
||||
// 64-bit Atomics
|
||||
unsigned hasGlobalInt64Atomics : 1; ///< 64-bit integer atomics for global memory.
|
||||
unsigned hasSharedInt64Atomics : 1; ///< 64-bit integer atomics for shared memory.
|
||||
|
||||
// Doubles
|
||||
unsigned hasDoubles : 1; ///< double-precision floating point.
|
||||
unsigned hasDoubles : 1; ///< Double-precision floating point.
|
||||
|
||||
// Warp cross-lane operations:
|
||||
unsigned hasWarpVote : 1; ///< warp vote instructions (__any, __all)
|
||||
unsigned hasWarpBallot : 1; ///< warp ballot instructions (__ballot)
|
||||
unsigned hasWarpShuffle : 1; ///< warp shuffle operations. (__shfl_*)
|
||||
unsigned hasFunnelShift : 1; ///< funnel two words into one, with shift&mask caps
|
||||
// Warp cross-lane operations
|
||||
unsigned hasWarpVote : 1; ///< Warp vote instructions (__any, __all).
|
||||
unsigned hasWarpBallot : 1; ///< Warp ballot instructions (__ballot).
|
||||
unsigned hasWarpShuffle : 1; ///< Warp shuffle operations. (__shfl_*).
|
||||
unsigned hasFunnelShift : 1; ///< Funnel two words into one with shift&mask caps.
|
||||
|
||||
// Sync
|
||||
unsigned hasThreadFenceSystem : 1; ///< __threadfence_system
|
||||
unsigned hasSyncThreadsExt : 1; ///< __syncthreads_count, syncthreads_and, syncthreads_or
|
||||
unsigned hasThreadFenceSystem : 1; ///< __threadfence_system.
|
||||
unsigned hasSyncThreadsExt : 1; ///< __syncthreads_count, syncthreads_and, syncthreads_or.
|
||||
|
||||
// Misc
|
||||
unsigned hasSurfaceFuncs : 1; ///< Surface functions
|
||||
unsigned has3dGrid : 1; ///< Grid and group dims are 3D (rather than 2D)
|
||||
unsigned hasDynamicParallelism : 1; ///< Dynamic parallelism
|
||||
unsigned hasSurfaceFuncs : 1; ///< Surface functions.
|
||||
unsigned has3dGrid : 1; ///< Grid and group dims are 3D (rather than 2D).
|
||||
unsigned hasDynamicParallelism : 1; ///< Dynamic parallelism.
|
||||
} hipDeviceArch_t;
|
||||
|
||||
|
||||
@@ -72,28 +72,25 @@ typedef struct {
|
||||
*
|
||||
*/
|
||||
typedef struct hipDeviceProp_t {
|
||||
char name[256]; ///< Device name
|
||||
size_t totalGlobalMem; ///< Size of global memory region (in bytes)
|
||||
size_t sharedMemPerBlock; ///< Size of shared memory region (in bytes)
|
||||
int regsPerBlock ; ///< registers per block
|
||||
int warpSize ; ///< warp size
|
||||
int maxThreadsPerBlock; ///< max work items per work group or workgroup max size
|
||||
int maxThreadsDim[3]; ///< max number of threads in each dimension (XYZ) of a block
|
||||
int maxGridSize[3]; ///< max grid dimensions (XYZ)
|
||||
int clockRate ; ///< max clock frequency of the multiProcessors, in khz.
|
||||
|
||||
size_t totalConstMem; ///< Size of shared memory region (in bytes)
|
||||
int major ; ///< Major compute capability. On HCC, this is an approximation and features may differ from CUDA CC. See the arch feature flags for portable ways to query feature caps.
|
||||
int minor; ///< Minor compute capability. On HCC, this is an approximation and features may differ from CUDA CC. See the arch feature flags for portable ways to query feature caps.
|
||||
int multiProcessorCount; ///< number of multi-processors (compute units)
|
||||
int l2CacheSize; ///< L2 cache size
|
||||
int maxThreadsPerMultiProcessor; ///< Maximum resident threads per multi-processor
|
||||
int computeMode; ///< Compute mode
|
||||
|
||||
int clockInstructionRate ; ///< Frequency in khz of the timer used by the device-side "clock*" instructions. New for HIP.
|
||||
|
||||
hipDeviceArch_t arch; ///< Architectural feature flags. New for HIP.
|
||||
int concurrentKernels; ///< Device can possibly execute multiple kernels concurrently
|
||||
char name[256]; ///< Device name.
|
||||
size_t totalGlobalMem; ///< Size of global memory region (in bytes).
|
||||
size_t sharedMemPerBlock; ///< Size of shared memory region (in bytes).
|
||||
int regsPerBlock; ///< Registers per block.
|
||||
int warpSize; ///< Warp size.
|
||||
int maxThreadsPerBlock; ///< Max work items per work group or workgroup max size.
|
||||
int maxThreadsDim[3]; ///< Max number of threads in each dimension (XYZ) of a block.
|
||||
int maxGridSize[3]; ///< Max grid dimensions (XYZ).
|
||||
int clockRate; ///< Max clock frequency of the multiProcessors, in khz.
|
||||
size_t totalConstMem; ///< Size of shared memory region (in bytes).
|
||||
int major; ///< Major compute capability. On HCC, this is an approximation and features may differ from CUDA CC. See the arch feature flags for portable ways to query feature caps.
|
||||
int minor; ///< Minor compute capability. On HCC, this is an approximation and features may differ from CUDA CC. See the arch feature flags for portable ways to query feature caps.
|
||||
int multiProcessorCount; ///< Number of multi-processors (compute units).
|
||||
int l2CacheSize; ///< L2 cache size.
|
||||
int maxThreadsPerMultiProcessor; ///< Maximum resident threads per multi-processor.
|
||||
int computeMode; ///< Compute mode.
|
||||
int clockInstructionRate; ///< Frequency in khz of the timer used by the device-side "clock*" instructions. New for HIP.
|
||||
hipDeviceArch_t arch; ///< Architectural feature flags. New for HIP.
|
||||
int concurrentKernels; ///< Device can possibly execute multiple kernels concurrently.
|
||||
} hipDeviceProp_t;
|
||||
|
||||
|
||||
@@ -111,19 +108,18 @@ typedef struct hipDeviceProp_t {
|
||||
* @ingroup Enumerations
|
||||
*/
|
||||
typedef enum hipError_t {
|
||||
hipSuccess = 0 ///< Successful completion.
|
||||
,hipErrorMemoryAllocation ///< Memory allocation error.
|
||||
,hipErrorMemoryFree ///< Memory free error.
|
||||
,hipErrorUnknownSymbol ///< Unknown symbol
|
||||
,hipErrorOutOfResources ///< Out of resources error
|
||||
,hipErrorInvalidValue ///< One or more of the parameters passed to the API call is NULL or not in an acceptable range.
|
||||
,hipErrorInvalidResourceHandle ///< Resource handle (hipEvent_t or hipStream_t) invalid.
|
||||
,hipErrorInvalidDevice ///< DeviceID must be in range 0...#compute-devices.
|
||||
,hipErrorNoDevice ///< Call to hipGetDeviceCount returned 0 devices
|
||||
,hipErrorNotReady ///< indicates that asynchronous operations enqueued earlier are not ready. This is not actually an error, but is used to distinguish from hipSuccess (which indicates completion). APIs that return this error include hipEventQuery and hipStreamQuery.
|
||||
|
||||
,hipErrorUnknown ///< Unknown error
|
||||
,hipErrorTbd ///< Marker that more error codes are needed.
|
||||
hipSuccess = 0 ///< Successful completion.
|
||||
,hipErrorMemoryAllocation ///< Memory allocation error.
|
||||
,hipErrorMemoryFree ///< Memory free error.
|
||||
,hipErrorUnknownSymbol ///< Unknown symbol.
|
||||
,hipErrorOutOfResources ///< Out of resources error.
|
||||
,hipErrorInvalidValue ///< One or more of the parameters passed to the API call is NULL or not in an acceptable range.
|
||||
,hipErrorInvalidResourceHandle ///< Resource handle (hipEvent_t or hipStream_t) invalid.
|
||||
,hipErrorInvalidDevice ///< DeviceID must be in range 0...#compute-devices.
|
||||
,hipErrorNoDevice ///< Call to hipGetDeviceCount returned 0 devices
|
||||
,hipErrorNotReady ///< Indicates that asynchronous operations enqueued earlier are not ready. This is not actually an error, but is used to distinguish from hipSuccess (which indicates completion). APIs that return this error include hipEventQuery and hipStreamQuery.
|
||||
,hipErrorUnknown ///< Unknown error.
|
||||
,hipErrorTbd ///< Marker that more error codes are needed.
|
||||
} hipError_t;
|
||||
|
||||
/*
|
||||
@@ -132,24 +128,24 @@ typedef enum hipError_t {
|
||||
* @ingroup Enumerations
|
||||
*/
|
||||
typedef enum hipDeviceAttribute_t {
|
||||
hipDeviceAttributeMaxThreadsPerBlock, ///< Maximum number of threads per block.
|
||||
hipDeviceAttributeMaxBlockDimX, ///< Maximum x-dimension of a block.
|
||||
hipDeviceAttributeMaxBlockDimY, ///< Maximum y-dimension of a block.
|
||||
hipDeviceAttributeMaxBlockDimZ, ///< Maximum z-dimension of a block.
|
||||
hipDeviceAttributeMaxGridDimX, ///< Maximum x-dimension of a grid.
|
||||
hipDeviceAttributeMaxGridDimY, ///< Maximum y-dimension of a grid.
|
||||
hipDeviceAttributeMaxGridDimZ, ///< Maximum z-dimension of a grid.
|
||||
hipDeviceAttributeMaxSharedMemoryPerBlock, ///< Maximum shared memory available per block in bytes.
|
||||
hipDeviceAttributeTotalConstantMemory, ///< Constant memory size in bytes.
|
||||
hipDeviceAttributeWarpSize, ///< Warp size in threads.
|
||||
hipDeviceAttributeMaxRegistersPerBlock, ///< Maximum number of 32-bit registers available to a thread block. This number is shared by all thread blocks simultaneously resident on a multiprocessor.
|
||||
hipDeviceAttributeClockRate, ///< Peak clock frequency in kilohertz.
|
||||
hipDeviceAttributeMultiprocessorCount, ///< Number of multiprocessors on the device.
|
||||
hipDeviceAttributeComputeMode, ///< Compute mode that device is currently in.
|
||||
hipDeviceAttributeL2CacheSize, ///< Size of L2 cache in bytes. 0 if the device doesn't have L2 cache.
|
||||
hipDeviceAttributeMaxThreadsPerMultiProcessor, ///< Maximum resident threads per multiprocessor.
|
||||
hipDeviceAttributeComputeCapabilityMajor, ///< Major compute capability version number.
|
||||
hipDeviceAttributeComputeCapabilityMinor, ///< Minor compute capability version number.
|
||||
hipDeviceAttributeMaxThreadsPerBlock, ///< Maximum number of threads per block.
|
||||
hipDeviceAttributeMaxBlockDimX, ///< Maximum x-dimension of a block.
|
||||
hipDeviceAttributeMaxBlockDimY, ///< Maximum y-dimension of a block.
|
||||
hipDeviceAttributeMaxBlockDimZ, ///< Maximum z-dimension of a block.
|
||||
hipDeviceAttributeMaxGridDimX, ///< Maximum x-dimension of a grid.
|
||||
hipDeviceAttributeMaxGridDimY, ///< Maximum y-dimension of a grid.
|
||||
hipDeviceAttributeMaxGridDimZ, ///< Maximum z-dimension of a grid.
|
||||
hipDeviceAttributeMaxSharedMemoryPerBlock, ///< Maximum shared memory available per block in bytes.
|
||||
hipDeviceAttributeTotalConstantMemory, ///< Constant memory size in bytes.
|
||||
hipDeviceAttributeWarpSize, ///< Warp size in threads.
|
||||
hipDeviceAttributeMaxRegistersPerBlock, ///< Maximum number of 32-bit registers available to a thread block. This number is shared by all thread blocks simultaneously resident on a multiprocessor.
|
||||
hipDeviceAttributeClockRate, ///< Peak clock frequency in kilohertz.
|
||||
hipDeviceAttributeMultiprocessorCount, ///< Number of multiprocessors on the device.
|
||||
hipDeviceAttributeComputeMode, ///< Compute mode that device is currently in.
|
||||
hipDeviceAttributeL2CacheSize, ///< Size of L2 cache in bytes. 0 if the device doesn't have L2 cache.
|
||||
hipDeviceAttributeMaxThreadsPerMultiProcessor, ///< Maximum resident threads per multiprocessor.
|
||||
hipDeviceAttributeComputeCapabilityMajor, ///< Major compute capability version number.
|
||||
hipDeviceAttributeComputeCapabilityMinor, ///< Minor compute capability version number.
|
||||
} hipDeviceAttribute_t;
|
||||
|
||||
/**
|
||||
|
||||
مرجع در شماره جدید
Block a user