experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Allow unsigned integers in HLO_DimensionTensor and HLO_StaticShapeTensor

Browse Source 
Follow up after enabling unsigned integers in the dialect.

PiperOrigin-RevId: 309134014
Change-Id: Ie458c93ec69a5ed1be183fdd208dc27d3de61e1a
This commit is contained in:
Smit Hinsu 2020-04-29 17:47:54 -07:00 committed by TensorFlower Gardener
parent af76ffe37f
commit 3dc9712013
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
tensorflow/compiler/mlir/xla/ir/hlo_ops_base.td
View File

@ -64,14 +64,14 @@ def HLO_TensorOrTuple : AnyTypeOf<[HLO_Tensor, HLO_Tuple]>;
// Dynamic representation of a shape vector as a tensor.
def HLO_DimensionTensor : ShapedContainerType<
[Index, AnySignlessInteger],
[Index, HLO_Pred, HLO_Int],
And<[IsTensorTypePred, HasAnyRankOfPred<[1]>]>,
"a 1D tensor of dimensions">;
// In general, static shaped tensor constraints should be avoided unless
// it is for a legacy op which is only correct with static shapes.
def HLO_StaticShapeTensor : StaticShapeTensorOf<[
AnyFloat, AnySignlessInteger, HLO_Complex]>;
AnyFloat, HLO_Pred, HLO_Int, HLO_Complex]>;
//===----------------------------------------------------------------------===//
// XLA on tensors combined type definitions.