Skip to content

[CustomOp] Fix ext_tensor.cast failed bug #34884

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
Aug 18, 2021
Merged

[CustomOp] Fix ext_tensor.cast failed bug #34884

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
979f55c
fix ext_tensor.cast failed bug
chenwhql Aug 13, 2021
e7ad759
remove useless deps
chenwhql Aug 13, 2021
3adafc6
fix windows cmake failed
chenwhql Aug 13, 2021
08b9c7c
try to fix windows make failed
chenwhql Aug 16, 2021
4438c8f
fix make error on windwos
chenwhql Aug 16, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions paddle/fluid/extension/src/ext_tensor.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
ext_tensor.cc
14 changes: 11 additions & 3 deletions paddle/fluid/framework/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,7 @@ function(windows_symbolic TARGET)
add_custom_command(OUTPUT ${final_path}/.${src}.cu
COMMAND ${CMAKE_COMMAND} -E copy_if_different "${final_path}/${src}.cc" "${final_path}/.${src}.cu"
COMMENT "create hidden file of ${src}.cu")
add_custom_target(${TARGET} ALL DEPENDS .${src}.cu)
add_custom_target(${TARGET} ALL DEPENDS ${final_path}/.${src}.cu)
Copy link
Contributor

@jim19930609 jim19930609 Aug 17, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LoL, so we have a file named "ext_tensor.cu" for Linux Symlink, and a hidden ".ext_tensor.cu"(actually .ext_tensor.cu.cu) for Windows Symlink. What if there were a third/fourth... platform?

Copy link
Contributor

@zhwesky2010 zhwesky2010 Aug 17, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

src is ext_tensor here. If system is based on Unix, these is no problem generally.

Copy link
Contributor Author

@chenwhql chenwhql Aug 17, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

we should remove these symlink cases later, include data_type_transform, tensor_utils, etc.

endforeach()
endfunction()

Expand Down Expand Up @@ -413,8 +413,16 @@ include_directories(${PADDLE_SOURCE_DIR}/paddle/fluid/platform)
include_directories(${PADDLE_SOURCE_DIR}/paddle/fluid/extension/include)
include_directories(${PADDLE_SOURCE_DIR}/paddle/utils)

if(WITH_ROCM)
hip_library(custom_tensor SRCS ../extension/src/ext_tensor.cc DEPS lod_tensor memory enforce)
if (WITH_GPU)
if (WIN32)
windows_symbolic(ext_tensor_cu SRCS ext_tensor.cu PATH ../extension/src)
nv_library(custom_tensor SRCS ../extension/src/.ext_tensor.cu DEPS lod_tensor memory enforce)
add_dependencies(custom_tensor ext_tensor_cu)
else()
nv_library(custom_tensor SRCS ../extension/src/ext_tensor.cu DEPS lod_tensor memory enforce)
endif(WIN32)
elseif (WITH_ROCM)
hip_library(custom_tensor SRCS ../extension/src/ext_tensor.cu DEPS lod_tensor memory enforce)
else()
cc_library(custom_tensor SRCS ../extension/src/ext_tensor.cc DEPS lod_tensor memory enforce)
endif()
Expand Down
7 changes: 7 additions & 0 deletions paddle/fluid/framework/custom_tensor_test.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -144,6 +144,13 @@ void TestCast(paddle::DataType data_type) {
t1.template mutable_data<T>();
auto t2 = t1.cast(data_type);
CHECK(t2.type() == data_type);
#ifdef PADDLE_WITH_CUDA
auto tg1 = paddle::Tensor(paddle::PlaceType::kGPU);
tg1.reshape(tensor_shape);
tg1.template mutable_data<T>();
auto tg2 = tg1.cast(data_type);
CHECK(tg2.type() == data_type);
#endif
}

void GroupTestCopy() {
Expand Down