Refine codegen #1424
Merged
Refine codegen #1424
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
guangyey
force-pushed
the
guangyey/refine
branch
5 times, most recently
from
803902e to
805d96f
Compare
|
@EikanWang This PR only intends to refine the codegen and remove redundant code. I will separate a dedicated PR to install header files when this PR lands. |
guangyey
force-pushed
the
guangyey/refine
branch
from
bb7ce08 to
53dbfaf
Compare
EikanWang
reviewed
Mar 5, 2025
cmake/Codegen.cmake
Outdated
Comment on lines
18
to
22
| if(WIN32) | ||
| set(FILE_DISPLAY_CMD type) | ||
| # replace forward slash with back slash for compatibility with 'type' command on Windows | ||
| string(REPLACE "/" "\\" RegisterXPU_PATH_BACKSLASH "${RegisterXPU_PATH}") | ||
| string(REPLACE "/" "\\" XPUFallback_PATH_BACKSLASH "${XPUFallback_PATH}") | ||
| set(REGISTER_FALLBACK_CMD ${FILE_DISPLAY_CMD} ${XPUFallback_PATH_BACKSLASH} ">>" ${RegisterXPU_PATH_BACKSLASH}) | ||
| string(REPLACE "/" "\\" RegisterXPU_GENERATED_BACKSLASH "${RegisterXPU_GENERATED}") | ||
| string(REPLACE "/" "\\" XPUFallback_TEMPLATE_BACKSLASH "${XPUFallback_TEMPLATE}") | ||
| set(REGISTER_FALLBACK_CMD ${FILE_DISPLAY_CMD} ${XPUFallback_TEMPLATE_BACKSLASH} ">>" ${RegisterXPU_GENERATED_BACKSLASH}) | ||
| else() | ||
| set(FILE_DISPLAY_CMD cat) | ||
| set(REGISTER_FALLBACK_CMD ${FILE_DISPLAY_CMD} ${XPUFallback_PATH} ">>" ${RegisterXPU_PATH}) | ||
| set(REGISTER_FALLBACK_CMD ${FILE_DISPLAY_CMD} ${XPUFallback_TEMPLATE} ">>" ${RegisterXPU_GENERATED}) | ||
| endif() |
There was a problem hiding this comment.
if(WIN32) set(FILE_DISPLAY_CMD type) else() set(FILE_DISPLAY_CMD cat) endif()
file(TO_NATIVE_PATH "${RegisterXPU_GENERATED}" RegisterXPU_GENERATED_FILE)
file(TO_NATIVE_PATH "${XPUFallback_TEMPLATE_BACKSLASH}" XPUFallback_TEMPLATE_FILE)
set(REGISTER_FALLBACK_CMD ${FILE_DISPLAY_CMD} ${XPUFallback_TEMPLATE} ">>" ${RegisterXPU_GENERATED})
EikanWang
approved these changes
Mar 5, 2025
|
By the way, @guangyey , actually, we do not want to maintain the python scripts to refine the generated files in torch-xpu-ops. It is better to refine the torch-gen in the stock pytorch. |
guangyey
force-pushed
the
guangyey/refine
branch
2 times, most recently
from
88c24a6 to
833e1f5
Compare
|
The failures are irrelevant. test_transformers_xpu.py::TestTransformersXPU::test_multiheadattention_fastpath_attn_mask_attn_mask_dim_2_key_padding_mask_dim_2_bool_xpu
test_transformers_xpu.py::TestTransformersXPU::test_multiheadattention_fastpath_attn_mask_attn_mask_dim_3_key_padding_mask_dim_2_bool_xpu
test_transformers_xpu.py::TestTransformersXPU::test_transformerencoder_fastpath_use_torchscript_False_enable_nested_tensor_False_use_autocast_False_d_model_12_xpu
test_transformers_xpu.py::TestTransformersXPU::test_transformerencoder_fastpath_use_torchscript_False_enable_nested_tensor_False_use_autocast_True_d_model_12_xpu
test_transformers_xpu.py::TestTransformersXPU::test_transformerencoder_fastpath_use_torchscript_False_enable_nested_tensor_True_use_autocast_False_d_model_12_xpu
test_transformers_xpu.py::TestTransformersXPU::test_transformerencoder_fastpath_use_torchscript_False_enable_nested_tensor_True_use_autocast_True_d_model_12_xpu
test_linalg_xpu.py::TestLinalgXPU::test_gemm_bias_offline_tunableop_xpu_bfloat16
test_meta_xpu.py::TestMetaXPU::test_dispatch_meta_outplace_nn_functional_scaled_dot_product_attention_xpu_bfloat16
test_meta_xpu.py::TestMetaXPU::test_dispatch_meta_outplace_nn_functional_scaled_dot_product_attention_xpu_float16
test_meta_xpu.py::TestMetaXPU::test_dispatch_meta_outplace_nn_functional_scaled_dot_product_attention_xpu_float32
test_meta_xpu.py::TestMetaXPU::test_dispatch_symbolic_meta_outplace_all_strides_nn_functional_scaled_dot_product_attention_xpu_float32
test_meta_xpu.py::TestMetaXPU::test_dispatch_symbolic_meta_outplace_nn_functional_scaled_dot_product_attention_xpu_bfloat16
test_meta_xpu.py::TestMetaXPU::test_dispatch_symbolic_meta_outplace_nn_functional_scaled_dot_product_attention_xpu_float16
test_meta_xpu.py::TestMetaXPU::test_dispatch_symbolic_meta_outplace_nn_functional_scaled_dot_product_attention_xpu_float32 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Motivation
Following the comments here, this PR intend to refine code related to codegen and remove redundant code.