Page MenuHomePhabricator

[mlir] [VectorOps] Improve vector.constant_mask lowering

Authored by aartbik on Jun 18 2020, 10:00 PM.



Use direct vector constants for the 1-D case. This approach
scales much better than generating elaborate insertion operations
that are eventually folded into a constant. We could of course
generalize the 1-D case to higher ranks, but this simplification
already helps in scaling some microbenchmarks that would formerly
crash on the intermediate IR length.

Diff Detail

Event Timeline

aartbik created this revision.Jun 18 2020, 10:00 PM
This revision is now accepted and ready to land.Jun 19 2020, 9:49 AM

@mehdi_amini : my first use of "aart-land" to submit this :-)

This revision was automatically updated to reflect the committed changes.