FAIL: //tensorflow/contrib/tensor_forest:scatter_add_ndim_op_test

[lukeiwanski]

System Info

  Name:						 Hawaii
  Vendor:					 Advanced Micro Devices, Inc.
  Device OpenCL C version:			 OpenCL C 2.0 
  Driver version:				 1912.5 (VM)
  Profile:					 FULL_PROFILE
  Version:					 OpenCL 2.0 AMD-APP (1912.5)
  Extensions:					 cl_khr_fp64 cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_khr_gl_depth_images cl_ext_atomic_counters_32 cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_image2d_from_buffer cl_khr_spir cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes

ComputeCpp 0.2.0

To reproduce

bazel test --config=sycl --local_test_jobs=4 -k --test_lang_filters=cc,py --action_env=LD_PRELOAD=/usr/lib/libOpenCL.so.1 --test_timeout 300,750,1200,3600 //tensorflow/contrib/tensor_forest:scatter_add_ndim_op_test

Error

ERROR: test1dim (__main__.ScatterAddNdimTest)
...
ERROR: test3dim (__main__.ScatterAddNdimTest)
...
ERROR: testIncompleteIndices (__main__.ScatterAddNdimTest)
...
ERROR: testNoUpdates (__main__.ScatterAddNdimTest)
...
FAIL: testBadInput (__main__.ScatterAddNdimTest)
...

AssertionError: Exception of type <class 'tensorflow.python.framework.errors_impl.InvalidArgumentError'>: AttrValue must not have reference type value of float_ref
         for attr 'tensor_type'
        ; NodeDef: Variable/_5 = _Recv[_start_time=0, client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/device:SYCL:0", send_device_incarnation=1, tensor_name="edge_3_Variable", tensor_type=DT_FLOAT_REF, _device="/job:localhost/replica:0/task:0/cpu:0"](^ScatterAddNdim/indices, ^ScatterAddNdim/deltas); Op<name=_Recv; signature= -> tensor:tensor_type; attr=tensor_type:type; attr=tensor_name:string; attr=send_device:string; attr=send_device_incarnation:int; attr=recv_device:string; attr=client_terminated:bool,default=false; is_stateful=true>
         [[Node: Variable/_5 = _Recv[_start_time=0, client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/device:SYCL:0", send_device_incarnation=1, tensor_name="edge_3_Variable", tensor_type=DT_FLOAT_REF, _device="/job:localhost/replica:0/task:0/cpu:0"](^ScatterAddNdim/indices, ^ScatterAddNdim/deltas)]]

[jwlawson]

The ScatterAddNdim op is an inplace version of ScatterNdAdd from Tensorflow core.

There is no GPU version of the ScatterAddNdim kernel, and I'm not too sure what is being run on the SYCL device in this test.