NVIDIA / NVIDIA/cudf

[FEA] Document Compilation Procedure for Generated PTX Code in UDF Unit Tests

Open
#17,967 0 comments 0 reactions 0 assignees View on GitHub
feature request Java libcudf numba
Dominant language
C++
Stars
9.8k
Forks
1.1k
Avg merge
3d 6m
Merged PRs (30d)
278

Description

**Is your feature request related to a problem? Please describe.**
Some UDF unit tests include generated PTX code compiled either by Numba or CUDA+NVCC.
A couple of them don't have their source code, compile flags, or compilation process documented, thus, making it difficult to work with for people new to them.

Some of them are:
- https://github.com/rapidsai/cudf/blob/428dc188cab5a51c1e15fb90c93a231ad95b7be2/cpp/tests/streams/binaryop_test.cpp#L96
- https://github.com/rapidsai/cudf/blob/428dc188cab5a51c1e15fb90c93a231ad95b7be2/cpp/tests/jit/parse_ptx_function.cpp#L54
- https://github.com/rapidsai/cudf/blob/428dc188cab5a51c1e15fb90c93a231ad95b7be2/cpp/tests/transform/integration/unary_transform_test.cpp#L79
- https://github.com/rapidsai/cudf/blob/428dc188cab5a51c1e15fb90c93a231ad95b7be2/cpp/tests/streams/transform_test.cpp#L70
- https://github.com/rapidsai/cudf/blob/428dc188cab5a51c1e15fb90c93a231ad95b7be2/cpp/tests/binaryop/binop-generic-ptx-test.cpp#L55
- https://github.com/rapidsai/cudf/blob/428dc188cab5a51c1e15fb90c93a231ad95b7be2/cpp/tests/binaryop/binop-generic-ptx-test.cpp#L55
- https://github.com/rapidsai/cudf/blob/428dc188cab5a51c1e15fb90c93a231ad95b7be2/cpp/tests/rolling/grouped_rolling_test.cpp#L67
- https://github.com/rapidsai/cudf/blob/428dc188cab5a51c1e15fb90c93a231ad95b7be2/cpp/tests/rolling/rolling_test.cpp#L1244
- https://github.com/rapidsai/cudf/blob/428dc188cab5a51c1e15fb90c93a231ad95b7be2/java/src/test/java/ai/rapids/cudf/ColumnVectorTest.java#L67

**Describe the solution you'd like**
For the **full** source code, flags, and environment to be documented. Either in a separate doc and having them referenced in the source code or directly embedded as comments in the source code.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.