pytorch_learn

Pruthvi Madugundu fbd08fb358 Introduce TORCH_DISABLE_GPU_ASSERTS (#84190 ) - Asserts for CUDA are enabled by default - Disabled for ROCm by default by setting `TORCH_DISABLE_GPU_ASSERTS` to `ON` - Can be enabled for ROCm by setting above variable to`OFF` during build or can be forcefully enabled by setting `ROCM_FORCE_ENABLE_GPU_ASSERTS:BOOL=ON` This is follow up changes as per comment in PR #81790, comment [link](https://github.com/pytorch/pytorch/pull/81790#issuecomment-1215929021) Pull Request resolved: https://github.com/pytorch/pytorch/pull/84190 Approved by: https://github.com/jeffdaily, https://github.com/malfet	1 year ago
..
External	Update NCCL to current master and remove patch step (#85367)	1 year ago

Modules	build: support DNNL_GRAPH_CPU_RUNTIME=TBB (#87512)	1 year ago

Modules_CUDA_fix	ada lovelace (arch 8.9) support (#87436)	1 year ago

public	Introduce TORCH_DISABLE_GPU_ASSERTS (#84190)	1 year ago

Allowlist.cmake	Replace whitelist with allowlist (#42067)	3 years ago

BuildVariables.cmake	Remove `Caffe2_MAIN_LIBS` (#38408)	4 years ago

Caffe2Config.cmake.in	Revert "Use FindCUDAToolkit to find cuda dependencies (#82695)"	1 year ago

Caffe2ConfigVersion.cmake.in	Forbid trailing whitespace (#53406)	3 years ago

Codegen.cmake	[retake2][mobile] Fix lightweight dispatch OOM error by introducing selective build (#80791)	1 year ago

DebugHelper.cmake	Add debug helper function to check target property (#52093)	3 years ago

Dependencies.cmake	Introduce TORCH_DISABLE_GPU_ASSERTS (#84190)	1 year ago

FlatBuffers.cmake	[pytorch][PR] Add ability for a mobile::Module to save as flatbuffer (#70201)	2 years ago

GoogleTestPatch.cmake	Update CMake and use native CUDA language support (#62445)	2 years ago

IncludeSource.cpp.in	CMake: Include instead of copying cpu kernel files (#67656)	2 years ago

Metal.cmake	[GPU] Enable Metal on macosx (#47635)	3 years ago

MiscCheck.cmake	Disable RDYNAMIC check with MSVC (#62949)	2 years ago

ProtoBuf.cmake	Update protobuf to 3.13.1 (#62571)	2 years ago

ProtoBufPatch.cmake	Update protobuf to 3.13.1 (#62571)	2 years ago

Summary.cmake	Introduce TORCH_DISABLE_GPU_ASSERTS (#84190)	1 year ago

TorchConfig.cmake.in	reorder cpuinfo and clog deps in TorchConfig.cmake (#79551)	1 year ago

TorchConfigVersion.cmake.in	Formatting cmake (to lowercase without space for if/elseif/else/endif) (#35521)	4 years ago

VulkanCodegen.cmake	make vulkan codegen windows-compatible (#85241)	1 year ago

VulkanDependencies.cmake	[Vulkan]Fix build warnings-treated-as-error on Linux. (#52781)	3 years ago

cmake_uninstall.cmake.in	Formatting cmake (to lowercase without space for if/elseif/else/endif) (#35521)	4 years ago

iOS.cmake	Forbid trailing whitespace (#53406)	3 years ago

Pruthvi Madugundu fbd08fb358 Introduce TORCH_DISABLE_GPU_ASSERTS (#84190 )

- Asserts for CUDA are enabled by default
- Disabled for ROCm by default by setting `TORCH_DISABLE_GPU_ASSERTS` to `ON`
- Can be enabled for ROCm by setting above variable to`OFF` during build or can be forcefully enabled by setting `ROCM_FORCE_ENABLE_GPU_ASSERTS:BOOL=ON`

This is follow up changes as per comment in PR #81790, comment [link](https://github.com/pytorch/pytorch/pull/81790#issuecomment-1215929021)

Pull Request resolved: https://github.com/pytorch/pytorch/pull/84190
Approved by: https://github.com/jeffdaily, https://github.com/malfet

1 year ago

External

Update NCCL to current master and remove patch step (#85367)

1 year ago