Skip to content

Commit

Permalink
Fix unsigned integers for collapse/expand
Browse files Browse the repository at this point in the history
The `tensor.expand` and `tensor.collapse` operations need to have
their signedness cleaned up.
  • Loading branch information
rsuderman committed Oct 26, 2023
1 parent 41c2b6d commit 5f7030d
Showing 1 changed file with 2 additions and 0 deletions.
Original file line number Diff line number Diff line change
Expand Up @@ -529,6 +529,8 @@ struct ConvertStableHloToIreeInputDialects final
ml_program::GlobalLoadConstOp::getOperationName(),
ml_program::GlobalStoreOp::getOperationName(),
tensor::FromElementsOp::getOperationName(),
tensor::CollapseShapeOp::getOperationName(),
tensor::ExpandShapeOp::getOperationName(),
arith::IndexCastUIOp::getOperationName()}) {
patterns.add<GenericTypeConvert>(opName, *typeConverter, context);
}
Expand Down

0 comments on commit 5f7030d

Please sign in to comment.