add launch bounds, enable more tests (#18909)
authorJ M Dieterich <dieterich@ogolem.org>
Fri, 5 Apr 2019 17:11:43 +0000 (10:11 -0700)
committerFacebook Github Bot <facebook-github-bot@users.noreply.github.com>
Fri, 5 Apr 2019 17:17:15 +0000 (10:17 -0700)
commite45e3634d6dff9384d899a6a3623572342af6032
tree7bb309a0f366e23a2c13d959f6f61ecf7825c83e
parent1d263ed92a11941bbb856114a68d2dbfa7a95e3f
add launch bounds, enable more tests (#18909)

Summary:
Add launch bounds annotations for ROCm arising from maxThreadsPerBlock and apply threads use.

Enable tests that now work.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/18909

Differential Revision: D14801490

Pulled By: ezyang

fbshipit-source-id: b81c97fc783a2627bc7e31b32036a364cfe40cc7
21 files changed:
aten/src/ATen/native/cuda/BatchLinearAlgebra.cu
aten/src/ATen/native/cuda/SortingCommon.cuh
aten/src/ATen/native/cuda/SummaryOps.cu
aten/src/ATen/native/cuda/TensorFactories.cu
aten/src/ATen/native/cuda/TensorTransformations.cu
aten/src/ATen/native/sparse/cuda/SparseCUDAApplyUtils.cuh
aten/src/ATen/native/sparse/cuda/SparseCUDATensorMath.cu
aten/src/THC/THCApply.cuh
aten/src/THC/THCTensorMath.cuh
aten/src/THC/THCTensorRandom.cuh
aten/src/THC/THCTensorScatterGather.cu
aten/src/THCUNN/SpatialUpSamplingBilinear.cu
aten/src/THCUNN/SpatialUpSamplingNearest.cu
aten/src/THCUNN/TemporalUpSamplingLinear.cu
aten/src/THCUNN/TemporalUpSamplingNearest.cu
aten/src/THCUNN/VolumetricUpSamplingNearest.cu
test/test_autograd.py
test/test_cuda.py
test/test_nn.py
test/test_sparse.py
test/test_torch.py