Page MenuHomePhabricator

[mlir][linalg][bufferize] Support tensor.from_elements
ClosedPublic

Authored by springerm on Jan 20 2022, 5:00 AM.

Details

Summary

This is mostly a copy of the existing tensor.from_elements bufferization. Once TensorInterfaceImpl.cpp is moved to the tensor dialect, the existing rewrite pattern can be deleted.

Depends On D117770

Diff Detail

Event Timeline

springerm created this revision.Jan 20 2022, 5:00 AM
springerm requested review of this revision.Jan 20 2022, 5:00 AM
nicolasvasilache requested changes to this revision.Jan 24 2022, 1:52 AM

Same discussion here as for https://reviews.llvm.org/D117770.
This jumps abstraction gaps too quickly and makes further things like vectorization unnecessary complex.
I would also reuse existing code with big TODOs and disclaimers and move to a memref.from_elements + separate lowerings.

This revision now requires changes to proceed.Jan 24 2022, 1:52 AM
This revision is now accepted and ready to land.Jan 25 2022, 12:54 AM
This revision was landed with ongoing or failed builds.Jan 25 2022, 5:21 AM
This revision was automatically updated to reflect the committed changes.