This is an archive of the discontinued LLVM Phabricator instance.

[mlir][sparse] support for very narrow index and pointer types
ClosedPublic

Authored by aartbik on Apr 1 2021, 4:48 PM.

Details

Summary

Rationale:
Small indices and values, when allowed by the required range of the
input tensors, can reduce the memory footprint of sparse tensors
even more. Note, however, that we must be careful zero extending
the values (since sparse tensors never use negatives for indexing),
but LLVM treats the index type as signed in most memory operations
(like the scatter and gather). This CL dots all the i's in this regard.

Diff Detail

Event Timeline

aartbik created this revision.Apr 1 2021, 4:48 PM
aartbik requested review of this revision.Apr 1 2021, 4:48 PM
bixia accepted this revision.Apr 1 2021, 5:46 PM
This revision is now accepted and ready to land.Apr 1 2021, 5:46 PM