zzy34407230
/
mindspore2022
Not watched
1
0
Fork 0
Code
Releases 22 Wiki evaluate Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain HPC
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
25174 Commits
18 Branches
584 MB
125 Download
 
 
 
 
 
 
Tree: c2c48a7ca2
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'c2c48a7ca2'
${ noResults }
mindspore2022/mindspore/ccsrc/runtime/device/gpu
i-robot 9852cced86 !17839 Fix device memory can not release in PyNative mode 4 years ago
..
distribution GPU supports p2p nccl interfaces 5 years ago
mpi code warning clean 5 years ago
blocking_queue.cc fix codecheck and pclint 4 years ago
blocking_queue.h add input data type check for ps cache mode 5 years ago
cuda_common.h add more dtypes support for gatherdgrad and other bugfix 5 years ago
cuda_driver.cc fix an allreduce bug with two streams sync problem 4 years ago
cuda_driver.h fix an allreduce bug with two streams sync problem 4 years ago
cuda_env_checker.cc search nvcc in entire PATH 5 years ago
cuda_env_checker.h search nvcc in entire PATH 5 years ago
gpu_bucket.cc unify runtime for PyNative distributed mode 4 years ago
gpu_bucket.h unify runtime for PyNative distributed mode 4 years ago
gpu_buffer_mgr.cc add error log when set device id failed 5 years ago
gpu_buffer_mgr.h add push opt logic 5 years ago
gpu_common.h Unify runtime support profiling 4 years ago
gpu_device_address.cc fix repeated release device resource of actor runtime 4 years ago
gpu_device_address.h add host_format for device_info of MetaTensor. 4 years ago
gpu_device_manager.cc fix repeated release device resource of actor runtime 4 years ago
gpu_device_manager.h fix an allreduce bug with two streams sync problem 4 years ago
gpu_event.cc PyNative AllReduce Bucket 5 years ago
gpu_event.h PyNative AllReduce Bucket 5 years ago
gpu_kernel_build.cc move the akg kernel build timer into AkgKernelBuilder::AkgKernelParallelBuild, so that it can time the Ascend kernel builder 5 years ago
gpu_kernel_build.h add hardware abstract layer 5 years ago
gpu_kernel_runtime.cc fix core dump when destroy device context in PyNative mode 4 years ago
gpu_kernel_runtime.h fix profiler pclint&codex 4 years ago
gpu_launch_kernel.cc add hardware abstract layer 5 years ago
gpu_launch_kernel.h add op atomic clean to clear input addr in launch allreduce 5 years ago
gpu_launch_mul.cc add op_mul fusion based on allreduce fusion in pynative mode 5 years ago
gpu_launch_mul.h add op atomic clean to clear input addr in launch allreduce 5 years ago
gpu_memory_allocator.cc fix codecheck and pclint 4 years ago
gpu_memory_allocator.h Unified code style 5 years ago
gpu_memory_copy_manager.cc fix an allreduce bug with two streams sync problem 4 years ago
gpu_memory_copy_manager.h fix an allreduce bug with two streams sync problem 4 years ago
gpu_memory_manager.cc add the continue memory alloc of communication kernel for actor runtime 5 years ago
gpu_memory_manager.h profiler memory 5 years ago
gpu_stream_assign.cc actor runtimie supports allreduce multi-stream 4 years ago
gpu_stream_assign.h actor runtimie supports allreduce multi-stream 4 years ago
kernel_info_setter.cc fix gpu reduce precision 4 years ago
kernel_info_setter.h 1.Optimize bias add grad kernel 4 years ago
queue_common.h add trace for gpu error/excpt log 5 years ago
readme.md mindspore path adjust 5 years ago
trt_loader.cc trt operator 4 years ago
trt_loader.h gpu inference mixed precision 5 years ago

readme.md

gpu

Delete task

Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.

cancel
confirm

阿对对队

C++ Python Text C Unity3D Asset other