diff options
| author |  A. Unique TensorFlower <gardener@tensorflow.org> | 2017-12-21 14:50:09 -0800 |
|---|---|---|
| committer | TensorFlower Gardener <gardener@tensorflow.org> | 2017-12-21 14:54:21 -0800 |
| commit | e21952f86328d16d1a8a4ac5e100917a251d093f (patch) | |
| tree | 1d9274230d915810cd78563f9952ee7ecf006408 /tensorflow/contrib/reduce_slice_ops | |
| parent | 7989a3aa026f93c1f45e1baf4ac37ad2a79156f9 (diff) | |
Automated g4 rollback of changelist 179782067
PiperOrigin-RevId: 179861781
Diffstat (limited to 'tensorflow/contrib/reduce_slice_ops')
| -rw-r--r-- | tensorflow/contrib/reduce_slice_ops/kernels/reduce_slice_ops_gpu.cu.cc | 11 |
1 files changed, 5 insertions, 6 deletions
diff --git a/tensorflow/contrib/reduce_slice_ops/kernels/reduce_slice_ops_gpu.cu.cc b/tensorflow/contrib/reduce_slice_ops/kernels/reduce_slice_ops_gpu.cu.cc index 501cddb8c8..8e6870fadd 100644 --- a/tensorflow/contrib/reduce_slice_ops/kernels/reduce_slice_ops_gpu.cu.cc +++ b/tensorflow/contrib/reduce_slice_ops/kernels/reduce_slice_ops_gpu.cu.cc @@ -34,9 +34,9 @@ namespace functor { __global__ void ReduceSliceDeviceKernel##reduceop( \ Cuda3DLaunchConfig config, Index indices_width, Index bound, \ const T begin, const Index *indices, const T *input, T *out) { \ - CUDA_AXIS_KERNEL_LOOP(x, config.virtual_thread_count.x, X) { \ - CUDA_AXIS_KERNEL_LOOP(y, config.virtual_thread_count.y, Y) { \ - CUDA_AXIS_KERNEL_LOOP(z, config.virtual_thread_count.z, Z) { \ + CUDA_AXIS_KERNEL_LOOP(x, config.virtual_thread_count, x) { \ + CUDA_AXIS_KERNEL_LOOP(y, config.virtual_thread_count, y) { \ + CUDA_AXIS_KERNEL_LOOP(z, config.virtual_thread_count, z) { \ Index outidx = x * config.virtual_thread_count.y * \ config.virtual_thread_count.z + \ y * config.virtual_thread_count.z + z; \ @@ -68,9 +68,8 @@ namespace functor { if (sizex * sizey * sizez == 0) { \ return; \ } \ - Cuda3DLaunchConfig config = GetCuda3DLaunchConfig( \ - sizex, sizey, sizez, d, ReduceSliceDeviceKernel##reduceop<T, Index>, \ - 0, 0); \ + Cuda3DLaunchConfig config = GetCuda3DLaunchConfig(sizex, sizey, sizez, d,\ + ReduceSliceDeviceKernel##reduceop<T, Index>, 0, 0); \ \ ReduceSliceDeviceKernel##reduceop<T, Index> \ <<<config.block_count, config.thread_per_block, 0, d.stream()>>>( \ |