Jeff Fifield

> This partially overlaps with #9040. Have asked @shaurjai to review if all the features in this PR are covered in #9040. It looks like the use cases from this...

How hard is it to generalize from 2d?

Seems like red-flag that the IR is left in a bad state somewhere. Do you have a failing (fixed) test you can add to the PR?

> @fifield I finally got around to creating a fairly minimal test. I can confirm without this PR, the added test still fails on current MLIR-AIE. Sorry it took so...

> > @fifield I finally got around to creating a fairly minimal test. I can confirm without this PR, the added test still fails on current MLIR-AIE. Sorry it took...

> * Clean up WriteRTP/Buffer APIs e.g., code like here: It's not clear to me that we still need the RTP ops, given that `npu.write32` and `npu.blockwrite` offer essentially the...

Without this [I get errors](https://github.com/fifield/mlir-aie/actions/runs/11243558802/job/31259719824#step:4:2024) like: ``` [139/591] Generating mm.o FAILED: aie_kernels/mm.o /workspace/build/aie_kernels/mm.o cd /workspace/build/aie_kernels && /workspace/mlir-aie/tools/chess-clang/xchesscc_wrapper AIE2 -aietools /workspace/venv -I /workspace/venv/include -c /workspace/mlir-aie/aie_kernels/mm.cc -o /workspace/build/aie_kernels/mm.o Error: Couldn't open "/workspace/venv/data/aie_ml/lib/me.prx":...

Currently failing because of this update to upload action: > Due to how Artifacts are created in this new version, it is no longer possible to upload to the same...

> A minimal example would be a chain of two compute tiles, each sending a packet back and forth, incrementing the value stored within until one of tiles detects that...

> The follow-up question would then be: how involved would it be to add functionality to either the cpp or python APIs to emit the header read/write behavior in MLIR?...