| Issue |
139785
|
| Summary |
[MLIR] Existing Python bindings issues
|
| Labels |
mlir
|
| Assignees |
makslevental
|
| Reporter |
makslevental
|
I plan to propose a partial rewrite/refactor of the Python bindings later this year. This github issue collects up some of the problems I've noticed/identified and some associated PRs/issues
1. Decoupled/divorced lifetimes between C++/C/Python; succinctly this revolves around `liveOperations`
- https://github.com/llvm/llvm-project/pull/93339
- https://github.com/llvm/llvm-project/issues/69730
- https://github.com/llvm/llvm-project/issues/92344
- https://github.com/llvm/llvm-project/issues/93337
- https://github.com/llvm/llvm-project/issues/63916
- half-hearted attempt fix via removal: https://github.com/llvm/llvm-project/pull/92631
- unfinished weak-ref system: https://github.com/llvm/llvm-project/pull/97340
2. The CMake build system is "mysterious and important", which is to say it's abstruse and inflexible; succinctly this affects modularity vis-a-vis piecemeal dialect registration/distribution
- https://github.com/llvm/llvm-project/issues/74245
- https://github.com/llvm/llvm-project/blob/5e83a5b4752da6631d79c446f21e5d128b5c5495/mlir/examples/standalone/python/CMakeLists.txt#L43
- https://github.com/llvm/llvm-project/pull/74252
- https://github.com/llvm/llvm-project/pull/124878
- https://github.com/llvm/llvm-project/issues/108253
![]()
_______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
