Skip to content

Commit

Permalink
[XLA:CPU/GPU] Fix a problem in the random number generator.
Browse files Browse the repository at this point in the history
Add primitive type S64 to GetNumberOfElementsPerPhiloxRngSample so that the
random number generator can generate random numbers for signed 64-bit integer.

PiperOrigin-RevId: 207621810
  • Loading branch information
bixia1 authored and tensorflower-gardener committed Aug 6, 2018
1 parent 2e7ae34 commit 7cc2dae
Showing 1 changed file with 1 addition and 0 deletions.
1 change: 1 addition & 0 deletions tensorflow/compiler/xla/service/elemental_ir_emitter.cc
Original file line number Diff line number Diff line change
Expand Up @@ -1302,6 +1302,7 @@ int32 GetNumberOfElementsPerPhiloxRngSample(PrimitiveType elem_prim_ty) {
case F16:
return 4;
case U64:
case S64:
case F64:
return 2;
default:
Expand Down

0 comments on commit 7cc2dae

Please sign in to comment.