Contains definitions of APIs for HIP runtime. More...

#include <math.h>
#include <string.h>
#include <stddef.h>
#include <hip/hip_runtime_api.h>
#include <grid_launch.h>
#include <hip/hcc_detail/host_defines.h>
#include <hip/hcc_detail/math_functions.h>
#include <hip/hcc_detail/device_functions.h>

Macros
#define	NVCC_COMPAT

#define	CUDA_SUCCESS hipSuccess

#define	ADDRESS_SPACE_1 __attribute__((address_space(1)))

#define	ADDRESS_SPACE_3 __attribute__((address_space(3)))

#define	hipLaunchParm grid_launch_parm

#define	launch_bounds_impl0(requiredMaxThreadsPerBlock) __attribute__((amdgpu_flat_work_group_size(1, requiredMaxThreadsPerBlock)))

#define	launch_bounds_impl1(requiredMaxThreadsPerBlock, minBlocksPerMultiprocessor)

#define	select_impl_(_1, _2, impl_, ...) impl_

#define	__launch_bounds__(...)

#define	clock_t long long int

#define	hipThreadIdx_x (hc_get_workitem_id(0))

#define	hipThreadIdx_y (hc_get_workitem_id(1))

#define	hipThreadIdx_z (hc_get_workitem_id(2))

#define	hipBlockIdx_x (hc_get_group_id(0))

#define	hipBlockIdx_y (hc_get_group_id(1))

#define	hipBlockIdx_z (hc_get_group_id(2))

#define	hipBlockDim_x (hc_get_group_size(0))

#define	hipBlockDim_y (hc_get_group_size(1))

#define	hipBlockDim_z (hc_get_group_size(2))

#define	hipGridDim_x (hc_get_num_groups(0))

#define	hipGridDim_y (hc_get_num_groups(1))

#define	hipGridDim_z (hc_get_num_groups(2))

#define	__syncthreads() hc_barrier(CLK_LOCAL_MEM_FENCE)

#define	HIP_KERNEL_NAME(...) (__VA_ARGS__)

#define	HIP_SYMBOL(X) #X

#define	HIP_DYNAMIC_SHARED(type, var)

#define	HIP_DYNAMIC_SHARED_ATTRIBUTE

Functions
__device__ long long int	clock64 ()

__device__ clock_t	clock ()

__device__ void	abort ()

__device__ int	atomicAdd (int *address, int val)

__device__ unsigned int	atomicAdd (unsigned int *address, unsigned int val)

__device__ unsigned long long int	atomicAdd (unsigned long long int *address, unsigned long long int val)

__device__ float	atomicAdd (float *address, float val)

__device__ int	atomicSub (int *address, int val)

__device__ unsigned int	atomicSub (unsigned int *address, unsigned int val)

__device__ int	atomicExch (int *address, int val)

__device__ unsigned int	atomicExch (unsigned int *address, unsigned int val)

__device__ unsigned long long int	atomicExch (unsigned long long int *address, unsigned long long int val)

__device__ float	atomicExch (float *address, float val)

__device__ int	atomicMin (int *address, int val)

__device__ unsigned int	atomicMin (unsigned int *address, unsigned int val)

__device__ unsigned long long int	atomicMin (unsigned long long int *address, unsigned long long int val)

__device__ int	atomicMax (int *address, int val)

__device__ unsigned int	atomicMax (unsigned int *address, unsigned int val)

__device__ unsigned long long int	atomicMax (unsigned long long int *address, unsigned long long int val)

__device__ int	atomicCAS (int *address, int compare, int val)

__device__ unsigned int	atomicCAS (unsigned int *address, unsigned int compare, unsigned int val)

__device__ unsigned long long int	atomicCAS (unsigned long long int *address, unsigned long long int compare, unsigned long long int val)

__device__ int	atomicAnd (int *address, int val)

__device__ unsigned int	atomicAnd (unsigned int *address, unsigned int val)

__device__ unsigned long long int	atomicAnd (unsigned long long int *address, unsigned long long int val)

__device__ int	atomicOr (int *address, int val)

__device__ unsigned int	atomicOr (unsigned int *address, unsigned int val)

__device__ unsigned long long int	atomicOr (unsigned long long int *address, unsigned long long int val)

__device__ int	atomicXor (int *address, int val)

__device__ unsigned int	atomicXor (unsigned int *address, unsigned int val)

__device__ unsigned long long int	atomicXor (unsigned long long int *address, unsigned long long int val)

__device__ unsigned int	atomicInc (unsigned int *address, unsigned int val)

__device__ unsigned int	atomicDec (unsigned int *address, unsigned int val)

__device__ int	__all (int input)

__device__ int	__any (int input)

__device__ unsigned long long int	__ballot (int input)

__device__ int	__shfl (int input, int lane, int width)

__device__ int	__shfl_up (int input, unsigned int lane_delta, int width)

__device__ int	__shfl_down (int input, unsigned int lane_delta, int width)

__device__ int	__shfl_xor (int input, int lane_mask, int width)

__device__ float	__shfl (float input, int lane, int width)

__device__ float	__shfl_up (float input, unsigned int lane_delta, int width)

__device__ float	__shfl_down (float input, unsigned int lane_delta, int width)

__device__ float	__shfl_xor (float input, int lane_mask, int width)

__device__ unsigned	__hip_ds_bpermute (int index, unsigned src)

__device__ float	__hip_ds_bpermutef (int index, float src)

__device__ unsigned	__hip_ds_permute (int index, unsigned src)

__device__ float	__hip_ds_permutef (int index, float src)

__device__ unsigned	__hip_ds_swizzle (unsigned int src, int pattern)

__device__ float	__hip_ds_swizzlef (float src, int pattern)

__device__ int	__hip_move_dpp (int src, int dpp_ctrl, int row_mask, int bank_mask, bool bound_ctrl)

__host__ __device__ int	min (int arg1, int arg2)

__host__ __device__ int	max (int arg1, int arg2)

__device__ void *	__get_dynamicgroupbaseptr ()

	__attribute__ ((const)) __device__ void __hip_hc_threadfence() __asm("__llvm_fence_sc_dev")

__device__ void	__threadfence_system (void)
	threadfence_system makes writes to pinned system memory visible on host CPU. More...

__device__ void *	__hip_hc_memcpy (void dst, const void src, size_t size)

__device__ void *	__hip_hc_memset (void *ptr, uint8_t val, size_t size)

__device__ void *	__hip_hc_malloc (size_t)

__device__ void *	__hip_hc_free (void *ptr)

Variables
int	HIP_TRACE_API

Detailed Description

Contains definitions of APIs for HIP runtime.

Macro Definition Documentation

#define __launch_bounds__ ( ... )

Value:

select_impl_(\

__VA_ARGS__, launch_bounds_impl1, launch_bounds_impl0)(__VA_ARGS__)

#define HIP_DYNAMIC_SHARED	(	type,
		var
	)

Value:

type* var = \

(type*)__get_dynamicgroupbaseptr(); \

extern shared

#define launch_bounds_impl1	(	requiredMaxThreadsPerBlock,
		minBlocksPerMultiprocessor
	)

Value:

__attribute__((amdgpu_flat_work_group_size(1, requiredMaxThreadsPerBlock),\

amdgpu_waves_per_eu(minBlocksPerMultiprocessor)))

Macros

Functions

Variables

Detailed Description

Macro Definition Documentation