NexMing wrote:

Hi @jeanPerier 

> While I agree that leveraging the existing memref infra makes sense, and as 
> @rscottmanley mentioned, we are actually also working on enabling converting 
> part of FIR to MLIR memref/scf, I think the more generic MLIR optimizations 
> passes are the one that are dialect independent and relies on interfaces.

I agree with your proposal. I think we are largely aligned on the overall work 
of converting FIR to MLIR memref/scf, and I also agree that establishing the 
pipeline at this stage is too premature. If I revert the handling of fir.cg, do 
you think the current PR is still worth submitting (it seems to overlap with 
your work)? I need to rely on the work of converting FIR to MLIR for further 
optimization exploration.

> I would more than welcome some help to enable the mem2reg with fir.delcare, I 
> think what is needed is a fir.declare_value that can be used to keep track of 
> the promoted values (very similar to what is done for 
> [llvm.dbg.declare](https://github.com/llvm/llvm-project/blob/6e01ea4babc8b5f41cc1541ab73af4388b1744f3/mlir/include/mlir/Dialect/LLVMIR/LLVMIntrinsicOps.td#L710))
>  so that we can later emit debug info for it in the DebugInfo pass. Would you 
> be interested to work on this?

I’d be very happy to help with this work. Do you mean enabling mem2reg at the 
FIR level and converting `fir.declare` to `fir.declare_value` during the FIR to 
MLIR lowering process?

https://github.com/llvm/llvm-project/pull/168703
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to