[XLA:GPU] Don't crash if a GTE feeds into a bitcast.
authorJustin Lebar <jlebar@google.com>
Wed, 21 Mar 2018 17:47:18 +0000 (10:47 -0700)
committerTensorFlower Gardener <gardener@tensorflow.org>
Wed, 21 Mar 2018 17:49:59 +0000 (10:49 -0700)
commitd7cb36a6876e02540c13f31f468a84f54c8591d4
tree48030b1545e283856cf2c6ca73bc213a31107f55
parente50fb3f561f1bfcd0a5fb457c69d50da64c789f8
[XLA:GPU] Don't crash if a GTE feeds into a bitcast.

GTE and bitcast are sort of "implicitly fused", so we have to handle
them in this way.

PiperOrigin-RevId: 189931422
tensorflow/compiler/xla/service/gpu/ir_emitter_unnested.cc
tensorflow/compiler/xla/tests/BUILD
tensorflow/compiler/xla/tests/hlo_test_base.cc
tensorflow/compiler/xla/tests/hlo_test_base.h
tensorflow/compiler/xla/tests/tuple_test.cc