NVSHMEM
3.2.5
简介
主要特性
通信传输
NVSHMEM 的优势
GPU 发起的通信和强扩展
使用 NVSHMEM
NVSHMEM 示例程序
使用 NVSHMEM InfiniBand GPUDirect Async Transport
将 NVSHMEM 与 MPI 或 OpenSHMEM 结合使用
编译 NVSHMEM 程序
运行 NVSHMEM 程序
通信模型
数据一致性
多进程 GPU 支持
构建 NVSHMEM 应用程序/库
NVSHMEM 和 CUDA 模型
CUDA 执行模型
CUDA 中的工作提交
CUDA 抽象机
非本地操作和 CUDA 执行模型
CUDA 流和循环依赖关系
CUDA 流顺序和执行资源
CUDA 流和虚假循环依赖关系
内核内同步
使用 NVSHMEM 协同内核启动确保安全的非本地操作
隐式异步 cudaMemcpy
内存模型
对称对象的指针
操作顺序
原子性保证
NVSHMEM 和 OpenSHMEM 之间的差异
阻塞式获取操作的顺序
可见性保证
执行模型
NVSHMEM 操作的进度
调用 NVSHMEM 操作
库常量
库句柄
环境变量
标准选项
引导选项
附加选项
集合操作选项
传输选项
NVTX 选项
NVSHMEM API
API 概述
不支持的 OpenSHMEM 1.3 API
不支持通过远程网络传输的 OpenSHMEM 1.3 API
支持的 OpenSHMEM API(OpenSHMEM 1.4 和 1.5)
用于 CPU 线程的 NVSHMEM API 扩展
用于 GPU 线程的 NVSHMEM API 扩展
库设置、退出和查询
NVSHMEM_INIT
NVSHMEMX_INIT_ATTR
NVSHMEMX_HOSTLIB_INIT_ATTR
NVSHMEMX_HOSTLIB_FINALIZE
NVSHMEMX_GET_UNIQUE_ID
NVSHMEMX_SET_ATTR_UNIQUEID_ARGS
NVSHMEMX_CUMODULE_INIT
NVSHMEMX_INIT_STATUS
NVSHMEM_MY_PE
NVSHMEM_N_PES
NVSHMEM_FINALIZE
NVSHMEM_GLOBAL_EXIT
NVSHMEM_PTR
NVSHMEMX_MC_PTR
NVSHMEM_INFO_GET_VERSION
NVSHMEM_INFO_GET_NAME
NVSHMEMX_VENDOR_GET_VERSION_INFO
线程支持
NVSHMEM_INIT_THREAD
NVSHMEM_QUERY_THREAD
内核启动例程
NVSHMEMX_COLLECTIVE_LAUNCH
NVSHMEMX_COLLECTIVE_LAUNCH_QUERY_GRIDSIZE
内存管理
NVSHMEM_MALLOC、NVSHMEM_FREE、NVSHMEM_ALIGN
NVSHMEM_CALLOC
内存注册
NVSHMEMX_BUFFER_REGISTER
NVSHMEMX_BUFFER_UNREGISTER
NVSHMEMX_BUFFER_UNREGISTER_ALL
团队管理
预定义和应用程序定义的团队
团队句柄
线程安全
集体排序
团队创建
NVSHMEM_TEAM_MY_PE
NVSHMEM_TEAM_N_PES
NVSHMEM_TEAM_CONFIG_T
NVSHMEM_TEAM_GET_CONFIG
NVSHMEM_TEAM_TRANSLATE_PE
NVSHMEM_TEAM_SPLIT_STRIDED
NVSHMEM_TEAM_SPLIT_2D
NVSHMEM_TEAM_DESTROY
远程内存访问
阻塞式 RMA
NVSHMEM_PUT
NVSHMEM_P
NVSHMEM_IPUT
NVSHMEM_GET
NVSHMEM_G
NVSHMEM_IGET
非阻塞式 RMA
NVSHMEM_PUT_NBI
NVSHMEM_GET_NBI
原子内存操作
NVSHMEM_ATOMIC_FETCH
NVSHMEM_ATOMIC_SET
NVSHMEM_ATOMIC_COMPARE_SWAP
NVSHMEM_ATOMIC_SWAP
NVSHMEM_ATOMIC_FETCH_INC
NVSHMEM_ATOMIC_INC
NVSHMEM_ATOMIC_FETCH_ADD
NVSHMEM_ATOMIC_ADD
NVSHMEM_ATOMIC_FETCH_AND
NVSHMEM_ATOMIC_AND
NVSHMEM_ATOMIC_FETCH_OR
NVSHMEM_ATOMIC_OR
NVSHMEM_ATOMIC_FETCH_XOR
NVSHMEM_ATOMIC_XOR
信号操作
信号操作的原子性保证
可用的信号操作符
NVSHMEM_PUT_SIGNAL
NVSHMEM_PUT_SIGNAL_NBI
NVSHMEM_SIGNAL_FETCH
NVSHMEMX_SIGNAL
NVSHMEMX_SIGNAL_OP
集体通信
基于团队的集合操作
隐式团队集合操作
从基于团队的集合操作返回的错误代码
NVSHMEM_BARRIER_ALL
NVSHMEM_BARRIER
NVSHMEM_SYNC
NVSHMEM_SYNC_ALL
NVSHMEM_ALLTOALL
NVSHMEM_BROADCAST
NVSHMEM_FCOLLECT
NVSHMEM_REDUCTIONS
AND
OR
XOR
MAX
MIN
SUM
PROD
点对点同步
NVSHMEM_WAIT_UNTIL
NVSHMEM_WAIT_UNTIL_ALL
NVSHMEM_WAIT_UNTIL_ANY
NVSHMEM_WAIT_UNTIL_SOME
NVSHMEM_WAIT_UNTIL_ALL_VECTOR
NVSHMEM_WAIT_UNTIL_ANY_VECTOR
NVSHMEM_WAIT_UNTIL_SOME_VECTOR
NVSHMEM_TEST
NVSHMEM_TEST_ALL
NVSHMEM_TEST_ANY
NVSHMEM_TEST_SOME
NVSHMEM_TEST_ALL_VECTOR
NVSHMEM_TEST_ANY_VECTOR
NVSHMEM_TEST_SOME_VECTOR
NVSHMEM_SIGNAL_WAIT_UNTIL
内存排序
NVSHMEM_FENCE
NVSHMEM_QUIET
示例
基于属性的初始化示例
集体启动示例
流上示例
线程组示例
块放置示例
环形广播示例
环形 Allreduce 示例
故障排除和常见问题解答
一般常见问题解答
先决条件常见问题解答
运行 NVSHMEM 程序常见问题解答
与 MPI 互操作性常见问题解答
与 OpenSHMEM 互操作性常见问题解答
GPU-GPU 互连常见问题解答
NVSHMEM API 使用常见问题解答
调试常见问题解答
其他常见问题解答
NVSHMEM SLA
NVIDIA 软件开发工具包许可协议
1. 许可。
2. 限制。
3. 所有权。
4. 无担保。
5. 责任限制。
6. 终止。
7. 通用。
NVIDIA 软件开发工具包软件许可协议的 NVSHMEM 补充条款
致谢
通知
商标
版权
NVSHMEM
文档
»
索引
索引
N
N
nvshmem_align (C 函数)
nvshmem_alltoallmem (C 函数)
,
[1]
nvshmem_barrier (C 函数)
,
[1]
nvshmem_barrier_all (C 函数)
,
[1]
NVSHMEM_BARRIER_DISSEM_KVAL (C 变量)
NVSHMEM_BARRIER_TG_DISSEM_KVAL (C 变量)
NVSHMEM_BCAST_ALGO (C 变量)
NVSHMEM_BOOTSTRAP (C 变量)
NVSHMEM_BOOTSTRAP_MPI_PLUGIN (C 变量)
NVSHMEM_BOOTSTRAP_PLUGIN (C 变量)
NVSHMEM_BOOTSTRAP_PMI (C 变量)
NVSHMEM_BOOTSTRAP_PMI2_PLUGIN (C 变量)
NVSHMEM_BOOTSTRAP_PMI_PLUGIN (C 变量)
NVSHMEM_BOOTSTRAP_PMIX_PLUGIN (C 变量)
NVSHMEM_BOOTSTRAP_SHMEM_PLUGIN (C 变量)
NVSHMEM_BOOTSTRAP_UID_PLUGIN (C 变量)
NVSHMEM_BOOTSTRAP_UID_SESSION_ID (C 变量)
NVSHMEM_BOOTSTRAP_UID_SOCK_FAMILY (C 变量)
NVSHMEM_BOOTSTRAP_UID_SOCK_IFNAME (C 变量)
nvshmem_broadcastmem (C 函数)
,
[1]
nvshmem_calloc (C 函数)
NVSHMEM_CMP_EQ (C 变量)
NVSHMEM_CMP_GE (C 变量)
NVSHMEM_CMP_GT (C 变量)
NVSHMEM_CMP_LE (C 变量)
NVSHMEM_CMP_LT (C 变量)
NVSHMEM_CMP_NE (C 变量)
NVSHMEM_CUDA_LIMIT_STACK_SIZE (C 变量)
NVSHMEM_CUMEM_GRANULARITY (C 变量)
NVSHMEM_DEBUG (C 变量)
NVSHMEM_DEBUG_FILE (C 变量)
NVSHMEM_DISABLE_CUDA_VMM (C 变量)
NVSHMEM_DISABLE_GDRCOPY (C 变量)
NVSHMEM_DISABLE_IB_NATIVE_ATOMICS (C 变量)
NVSHMEM_DISABLE_LOCAL_ONLY_PROXY (C 变量)
NVSHMEM_DISABLE_NCCL (C 变量)
NVSHMEM_DISABLE_NVLS (C 变量)
NVSHMEM_DISABLE_P2P (C 变量)
NVSHMEM_ENABLE_NIC_PE_MAPPING (C 变量)
NVSHMEM_FCOLLECT_LL_THRESHOLD (C 变量)
nvshmem_fcollectmem (C 函数)
,
[1]
nvshmem_fence (C 函数)
,
[1]
nvshmem_finalize (C 函数)
nvshmem_free (C 函数)
nvshmem_getmem (C 函数)
,
[1]
nvshmem_getmem_nbi (C 函数)
,
[1]
nvshmem_getSIZE (C 函数)
,
[1]
nvshmem_getSIZE_nbi (C 函数)
,
[1]
nvshmem_global_exit (C 函数)
,
[1]
NVSHMEM_HCA_LIST (C 变量)
NVSHMEM_HCA_PE_MAPPING (C 变量)
NVSHMEM_IB_ADDR_FAMILY (C 变量)
NVSHMEM_IB_ADDR_RANGE (C 变量)
NVSHMEM_IB_DISABLE_DMABUF (C 变量)
NVSHMEM_IB_ENABLE_IBGDA (C 变量)
NVSHMEM_IB_GID_INDEX (C 变量)
NVSHMEM_IB_ROCE_VERSION_NUM (C 变量)
NVSHMEM_IB_SL (C 变量)
NVSHMEM_IB_TRAFFIC_CLASS (C 变量)
NVSHMEM_IBGDA_DCI_MAP_BY (C 变量)
NVSHMEM_IBGDA_ENABLE_MULTI_PORT (C 变量)
NVSHMEM_IBGDA_FORCE_NIC_BUF_MEMTYPE (C 变量)
NVSHMEM_IBGDA_NIC_HANDLER (C 变量)
NVSHMEM_IBGDA_NUM_DCI (C 变量)
NVSHMEM_IBGDA_NUM_DCT (C 变量)
NVSHMEM_IBGDA_NUM_FETCH_SLOTS_PER_DCI (C 变量)
NVSHMEM_IBGDA_NUM_FETCH_SLOTS_PER_RC (C 变量)
NVSHMEM_IBGDA_NUM_RC_PER_PE (C 变量)
NVSHMEM_IBGDA_NUM_REQUESTS_IN_BATCH (C 变量)
NVSHMEM_IBGDA_NUM_SHARED_DCI (C 变量)
NVSHMEM_IBGDA_RC_MAP_BY (C 变量)
nvshmem_igetSIZE (C 函数)
,
[1]
NVSHMEM_INFO (C 变量)
nvshmem_info_get_name (C 函数)
,
[1]
nvshmem_info_get_version (C 函数)
,
[1]
nvshmem_init (C 函数)
nvshmem_init_thread (C 函数)
nvshmem_iputSIZE (C 函数)
,
[1]
NVSHMEM_LIBFABRIC_PROVIDER (C 变量)
NVSHMEM_MAJOR_VERSION (C 变量)
nvshmem_malloc (C 函数)
NVSHMEM_MAX_MEMORY_PER_GPU (C 变量)
NVSHMEM_MAX_NAME_LEN (C 变量)
NVSHMEM_MAX_TEAMS (C 变量)
NVSHMEM_MINOR_VERSION (C 变量)
nvshmem_my_pe (C 函数)
,
[1]
nvshmem_n_pes (C 函数)
,
[1]
NVSHMEM_NVTX (C 变量)
NVSHMEM_PROXY_REQUEST_BATCH_MAX (C 变量)
nvshmem_ptr (C 函数)
,
[1]
nvshmem_putmem (C 函数)
,
[1]
nvshmem_putmem_nbi (C 函数)
,
[1]
nvshmem_putmem_signal (C 函数)
,
[1]
nvshmem_putmem_signal_nbi (C 函数)
,
[1]
nvshmem_putSIZE (C 函数)
,
[1]
nvshmem_putSIZE_nbi (C 函数)
,
[1]
nvshmem_putSIZE_signal (C 函数)
,
[1]
nvshmem_putSIZE_signal_nbi (C 函数)
,
[1]
nvshmem_query_thread (C 函数)
nvshmem_quiet (C 函数)
,
[1]
NVSHMEM_REDMAXLOC_ALGO (C 变量)
NVSHMEM_REDUCE_SCRATCH_SIZE (C 变量)
NVSHMEM_REMOTE_TRANSPORT (C 变量)
NVSHMEM_SIGNAL_ADD (C 变量)
nvshmem_signal_fetch (C 函数)
NVSHMEM_SIGNAL_SET (C 变量)
nvshmem_signal_wait_until (C 函数)
NVSHMEM_SYMMETRIC_SIZE (C 变量)
nvshmem_sync (C 函数)
,
[1]
nvshmem_sync_all (C 函数)
,
[1]
nvshmem_team_destroy (C 函数)
nvshmem_team_get_config (C 函数)
NVSHMEM_TEAM_INVALID (C 变量)
nvshmem_team_my_pe (C 函数)
,
[1]
nvshmem_team_n_pes (C 函数)
,
[1]
NVSHMEM_TEAM_SHARED (C 变量)
nvshmem_team_split_2d (C 函数)
nvshmem_team_split_strided (C 函数)
nvshmem_team_sync (C 函数)
,
[1]
nvshmem_team_translate_pe (C 函数)
NVSHMEM_TEAM_WORLD (C 变量)
NVSHMEM_THREAD_FUNNELED (C 变量)
NVSHMEM_THREAD_MULTIPLE (C 变量)
NVSHMEM_THREAD_SERIALIZED (C 变量)
NVSHMEM_THREAD_SINGLE (C 变量)
nvshmem_TYPENAME_alltoall (C 函数)
,
[1]
nvshmem_TYPENAME_and_reduce (C 函数)
,
[1]
nvshmem_TYPENAME_atomic_add (C 函数)
nvshmem_TYPENAME_atomic_and (C 函数)
nvshmem_TYPENAME_atomic_compare_swap (C 函数)
nvshmem_TYPENAME_atomic_fetch (C 函数)
nvshmem_TYPENAME_atomic_fetch_add (C 函数)
nvshmem_TYPENAME_atomic_fetch_and (C 函数)
nvshmem_TYPENAME_atomic_fetch_inc (C 函数)
nvshmem_TYPENAME_atomic_fetch_or (C 函数)
nvshmem_TYPENAME_atomic_fetch_xor (C 函数)
nvshmem_TYPENAME_atomic_inc (C 函数)
nvshmem_TYPENAME_atomic_or (C 函数)
nvshmem_TYPENAME_atomic_set (C 函数)
nvshmem_TYPENAME_atomic_swap (C 函数)
nvshmem_TYPENAME_atomic_xor (C 函数)
nvshmem_TYPENAME_broadcast (C 函数)
,
[1]
nvshmem_TYPENAME_fcollect (C 函数)
,
[1]
nvshmem_TYPENAME_g (C 函数)
,
[1]
nvshmem_TYPENAME_get (C 函数)
,
[1]
nvshmem_TYPENAME_get_nbi (C 函数)
,
[1]
nvshmem_TYPENAME_iget (C 函数)
,
[1]
nvshmem_TYPENAME_iput (C 函数)
,
[1]
nvshmem_TYPENAME_max_reduce (C 函数)
,
[1]
nvshmem_TYPENAME_min_reduce (C 函数)
,
[1]
nvshmem_TYPENAME_or_reduce (C 函数)
,
[1]
nvshmem_TYPENAME_p (C 函数)
,
[1]
nvshmem_TYPENAME_prod_reduce (C 函数)
,
[1]
nvshmem_TYPENAME_put (C 函数)
,
[1]
nvshmem_TYPENAME_put_nbi (C 函数)
,
[1]
nvshmem_TYPENAME_put_signal (C 函数)
,
[1]
nvshmem_TYPENAME_put_signal_nbi (C 函数)
,
[1]
nvshmem_TYPENAME_sum_reduce (C 函数)
,
[1]
nvshmem_TYPENAME_test (C 函数)
nvshmem_TYPENAME_test_all (C 函数)
nvshmem_TYPENAME_test_all_vector (C 函数)
nvshmem_TYPENAME_test_any (C 函数)
nvshmem_TYPENAME_test_any_vector (C 函数)
nvshmem_TYPENAME_test_some (C 函数)
nvshmem_TYPENAME_test_some_vector (C 函数)
nvshmem_TYPENAME_wait (C 函数)
nvshmem_TYPENAME_wait_until (C 函数)
nvshmem_TYPENAME_wait_until_all (C 函数)
nvshmem_TYPENAME_wait_until_all_vector (C 函数)
nvshmem_TYPENAME_wait_until_any (C 函数)
nvshmem_TYPENAME_wait_until_any_vector (C 函数)
nvshmem_TYPENAME_wait_until_some (C 函数)
nvshmem_TYPENAME_wait_until_some_vector (C 函数)
nvshmem_TYPENAME_xor_reduce (C 函数)
,
[1]
NVSHMEM_VENDOR_MAJOR_VERSION (C 变量)
NVSHMEM_VENDOR_MINOR_VERSION (C 变量)
NVSHMEM_VENDOR_PATCH_VERSION (C 变量)
NVSHMEM_VENDOR_STRING (C 变量)
NVSHMEM_VENDOR_VERSION (C 变量)
NVSHMEM_VERSION (C 变量)
nvshmemx_alltoallmem_block (C 函数)
nvshmemx_alltoallmem_on_stream (C 函数)
nvshmemx_alltoallmem_warp (C 函数)
nvshmemx_barrier_all_block (C 函数)
nvshmemx_barrier_all_on_stream (C 函数)
nvshmemx_barrier_all_warp (C 函数)
nvshmemx_barrier_block (C 函数)
nvshmemx_barrier_on_stream (C 函数)
nvshmemx_barrier_warp (C 函数)
nvshmemx_broadcastmem_block (C 函数)
nvshmemx_broadcastmem_on_stream (C 函数)
nvshmemx_broadcastmem_warp (C 函数)
nvshmemx_buffer_register (C 函数)
nvshmemx_buffer_unregister (C 函数)
nvshmemx_buffer_unregister_all (C 函数)
nvshmemx_collective_launch (C 函数)
nvshmemx_collective_launch_query_gridsize (C 函数)
nvshmemx_cumodule_init (C 函数)
nvshmemx_fcollectmem_block (C 函数)
nvshmemx_fcollectmem_on_stream (C 函数)
nvshmemx_fcollectmem_warp (C 函数)
nvshmemx_get_uniqueid (C 函数)
nvshmemx_getmem_block (C 函数)
nvshmemx_getmem_nbi_block (C 函数)
nvshmemx_getmem_nbi_on_stream (C 函数)
nvshmemx_getmem_nbi_warp (C 函数)
nvshmemx_getmem_on_stream (C 函数)
nvshmemx_getmem_warp (C 函数)
nvshmemx_getSIZE_block (C 函数)
nvshmemx_getSIZE_nbi_block (C 函数)
nvshmemx_getSIZE_nbi_on_stream (C 函数)
nvshmemx_getSIZE_nbi_warp (C 函数)
nvshmemx_getSIZE_on_stream (C 函数)
nvshmemx_getSIZE_warp (C 函数)
nvshmemx_hostlib_finalize (C 函数)
nvshmemx_hostlib_init_attr (C 函数)
nvshmemx_igetSIZE_block (C 函数)
nvshmemx_igetSIZE_on_stream (C 函数)
nvshmemx_igetSIZE_warp (C 函数)
nvshmemx_init_attr (C 函数)
nvshmemx_init_status (C 函数)
nvshmemx_iputSIZE_block (C 函数)
nvshmemx_iputSIZE_on_stream (C 函数)
nvshmemx_iputSIZE_warp (C 函数)
nvshmemx_mc_ptr (C 函数)
,
[1]
nvshmemx_putmem_block (C 函数)
nvshmemx_putmem_nbi_block (C 函数)
nvshmemx_putmem_nbi_on_stream (C 函数)
nvshmemx_putmem_nbi_warp (C 函数)
nvshmemx_putmem_on_stream (C 函数)
nvshmemx_putmem_signal_block (C 函数)
nvshmemx_putmem_signal_nbi_block (C 函数)
nvshmemx_putmem_signal_nbi_on_stream (C 函数)
nvshmemx_putmem_signal_nbi_warp (C 函数)
nvshmemx_putmem_signal_on_stream (C 函数)
nvshmemx_putmem_signal_warp (C 函数)
nvshmemx_putmem_warp (C 函数)
nvshmemx_putSIZE_block (C 函数)
nvshmemx_putSIZE_nbi_block (C 函数)
nvshmemx_putSIZE_nbi_on_stream (C 函数)
nvshmemx_putSIZE_nbi_warp (C 函数)
nvshmemx_putSIZE_on_stream (C 函数)
nvshmemx_putSIZE_signal_block (C 函数)
nvshmemx_putSIZE_signal_nbi_block (C 函数)
nvshmemx_putSIZE_signal_nbi_on_stream (C 函数)
nvshmemx_putSIZE_signal_nbi_warp (C 函数)
nvshmemx_putSIZE_signal_on_stream (C 函数)
nvshmemx_putSIZE_signal_warp (C 函数)
nvshmemx_putSIZE_warp (C 函数)
nvshmemx_quiet_on_stream (C 函数)
nvshmemx_set_attr_uniqueid_args (C 函数)
nvshmemx_signal_op (C 函数)
nvshmemx_signal_wait_until_on_stream (C 函数)
nvshmemx_sync_all_block (C 函数)
nvshmemx_sync_all_on_stream (C 函数)
nvshmemx_sync_all_warp (C 函数)
nvshmemx_sync_block (C 函数)
nvshmemx_sync_on_stream (C 函数)
nvshmemx_sync_warp (C 函数)
NVSHMEMX_TEAM_NODE (C 变量)
nvshmemx_team_sync_block (C 函数)
nvshmemx_team_sync_on_stream (C 函数)
nvshmemx_team_sync_warp (C 函数)
nvshmemx_TYPENAME_alltoall_block (C 函数)
nvshmemx_TYPENAME_alltoall_on_stream (C 函数)
nvshmemx_TYPENAME_alltoall_warp (C 函数)
nvshmemx_TYPENAME_and_reduce_block (C 函数)
nvshmemx_TYPENAME_and_reduce_on_stream (C 函数)
nvshmemx_TYPENAME_and_reduce_warp (C 函数)
nvshmemx_TYPENAME_broadcast_block (C 函数)
nvshmemx_TYPENAME_broadcast_on_stream (C 函数)
nvshmemx_TYPENAME_broadcast_warp (C 函数)
nvshmemx_TYPENAME_fcollect_block (C 函数)
nvshmemx_TYPENAME_fcollect_on_stream (C 函数)
nvshmemx_TYPENAME_fcollect_warp (C 函数)
nvshmemx_TYPENAME_get_block (C 函数)
nvshmemx_TYPENAME_get_nbi_block (C 函数)
nvshmemx_TYPENAME_get_nbi_on_stream (C 函数)
nvshmemx_TYPENAME_get_nbi_warp (C 函数)
nvshmemx_TYPENAME_get_on_stream (C 函数)
nvshmemx_TYPENAME_get_warp (C 函数)
nvshmemx_TYPENAME_iget_block (C 函数)
nvshmemx_TYPENAME_iget_on_stream (C 函数)
nvshmemx_TYPENAME_iget_warp (C 函数)
nvshmemx_TYPENAME_iput_block (C 函数)
nvshmemx_TYPENAME_iput_on_stream (C 函数)
nvshmemx_TYPENAME_iput_warp (C 函数)
nvshmemx_TYPENAME_max_reduce_block (C 函数)
nvshmemx_TYPENAME_max_reduce_on_stream (C 函数)
nvshmemx_TYPENAME_max_reduce_warp (C 函数)
nvshmemx_TYPENAME_min_reduce_block (C 函数)
nvshmemx_TYPENAME_min_reduce_on_stream (C 函数)
nvshmemx_TYPENAME_min_reduce_warp (C 函数)
nvshmemx_TYPENAME_or_reduce_block (C 函数)
nvshmemx_TYPENAME_or_reduce_on_stream (C 函数)
nvshmemx_TYPENAME_or_reduce_warp (C 函数)
nvshmemx_TYPENAME_prod_reduce_block (C 函数)
nvshmemx_TYPENAME_prod_reduce_on_stream (C 函数)
nvshmemx_TYPENAME_prod_reduce_warp (C 函数)
nvshmemx_TYPENAME_put_block (C 函数)
nvshmemx_TYPENAME_put_nbi_block (C 函数)
nvshmemx_TYPENAME_put_nbi_on_stream (C 函数)
nvshmemx_TYPENAME_put_nbi_warp (C 函数)
nvshmemx_TYPENAME_put_on_stream (C 函数)
nvshmemx_TYPENAME_put_signal_block (C 函数)
nvshmemx_TYPENAME_put_signal_nbi_block (C 函数)
nvshmemx_TYPENAME_put_signal_nbi_on_stream (C 函数)
nvshmemx_TYPENAME_put_signal_nbi_warp (C 函数)
nvshmemx_TYPENAME_put_signal_on_stream (C 函数)
nvshmemx_TYPENAME_put_signal_warp (C 函数)
nvshmemx_TYPENAME_put_warp (C 函数)
nvshmemx_TYPENAME_signal (C 函数)
nvshmemx_TYPENAME_sum_reduce_block (C 函数)
nvshmemx_TYPENAME_sum_reduce_on_stream (C 函数)
nvshmemx_TYPENAME_sum_reduce_warp (C 函数)
nvshmemx_TYPENAME_wait_on_stream (C 函数)
nvshmemx_TYPENAME_wait_until_on_stream (C 函数)
nvshmemx_TYPENAME_xor_reduce_block (C 函数)
nvshmemx_TYPENAME_xor_reduce_on_stream (C 函数)
nvshmemx_TYPENAME_xor_reduce_warp (C 函数)
nvshmemx_vendor_get_version_info (C 函数)
,
[1]