[Mlir-commits] [mlir] [mlir][gpu] Deprecate gpu::Serialization* passes. (PR #65857)

[Fabian Mora]

fabianmcg wrote:

> Let me know if you need help here. It seems like removing the 'convert-gpu-to-nvvm' addind `TargetAttr` would be enough. We are already using "TargetAttr" in one test.
> 
> FWIW, I'm about to land for 4 more tests. I will make sure to use the new workflow.

@grypp  Yes I need help please, both tests there still refer to `gpu-to-cubin` and I don't have access to a H100. To dump the ptx see this test: [dump-ptx.mlir](https://github.com/llvm/llvm-project/blob/main/mlir/test/Integration/Dialect/SparseTensor/GPU/CUDA/dump-ptx.mlir) 

> Can you document this in https://mlir.llvm.org/deprecation/ as well?
Yes I'll change the message there from migration to deprecation. I won't push the patch today, I'll first do a PSA on discourse on Monday and then wait for the migration of the `sm_90` tests.


https://github.com/llvm/llvm-project/pull/65857