pytorch/caffe2/core
Emilio Castillo f9ec86a6c6 External stream (#59527)
Summary:
Previous is https://github.com/pytorch/pytorch/issues/57781

We add now two CUDA bindings to avoid using ctypes to fix a windows issue.
However, we use ctypes to allocate the stream and create its pointer
(we can do this with a 0-dim tensor too if it feels better).

CC. ezyang rgommers ngimel mruberry

Pull Request resolved: https://github.com/pytorch/pytorch/pull/59527

Reviewed By: albanD

Differential Revision: D29053062

Pulled By: ezyang

fbshipit-source-id: 661e7e58de98b1bdb7a0871808cd41d91fe8f13f
2021-06-14 13:46:11 -07:00
..
hip
nomnigraph Disallow versionless Python shebangs (#58275) 2021-05-14 08:26:02 -07:00
__init__.py
allocator.cc
allocator.h
blob.h
blob_gpu_test.cc
blob_serialization.cc [Caffe2] [Easy] Fix comment on caffe2_serialize_using_bytes_as_holder to reflect correct types 2021-05-20 18:03:34 -07:00
blob_serialization.h [caffe2] add a SerializationOptions field for the save operator (#53402) 2021-03-11 13:02:58 -08:00
blob_serialization_gpu.cc
blob_serializer_base.h [caffe2] add a SerializationOptions field for the save operator (#53402) 2021-03-11 13:02:58 -08:00
blob_stats.cc
blob_stats.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
blob_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
CMakeLists.txt
common.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
common.h [caffe2] Avoid attempt to use undefined preprocessor directive 2021-06-05 09:22:52 -07:00
common_cudnn.cc
common_cudnn.h
common_gpu.cc
common_gpu.h Introducing TORCH_CUDA_CPP_API and TORCH_CUDA_CU_API to the code (#50627) 2021-01-21 19:09:11 -08:00
common_omp.h
common_test.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
context.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
context.h External stream (#59527) 2021-06-14 13:46:11 -07:00
context_base.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
context_base.h External stream (#59527) 2021-06-14 13:46:11 -07:00
context_gpu.cu use device methods (#52899) 2021-03-02 20:14:23 -08:00
context_gpu.h Don't switch streams when running Caffe2 ops from c10. (#55121) 2021-04-08 13:21:11 -07:00
context_gpu_test.cc
context_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
cudnn_wrappers.h
db.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
db.h [caffe2] allow passing options to the DB in Save operations (#55935) 2021-04-15 14:45:47 -07:00
distributions_stubs.h
event.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
event.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
event_cpu.h
event_gpu.cc
event_gpu_test.cc
event_test.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
export_c10_op_to_caffe2.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
export_c10_op_to_caffe2.h Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
export_caffe2_op_to_c10.h Don't switch streams when running Caffe2 ops from c10. (#55121) 2021-04-08 13:21:11 -07:00
flags.h
graph.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
graph.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
graph_test.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
init.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
init.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
init_denormals.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
init_intrinsics_check.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
init_omp.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
init_test.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
int8_serialization.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
logging.h
macros.h
macros.h.in Move USE_NUMPY to more appropriate targets (#51143) 2021-01-27 15:44:12 -08:00
memonger.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
memonger.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
module.cc
module.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
module_test.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
net.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
net.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
net_async_base.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
net_async_base.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
net_async_scheduling.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
net_async_scheduling.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
net_async_task.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
net_async_task.h
net_async_task_future.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
net_async_task_future.h
net_async_task_graph.cc
net_async_task_graph.h
net_async_tracing.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
net_async_tracing.h Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
net_async_tracing_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
net_dag_utils.cc
net_dag_utils.h
net_dag_utils_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
net_gpu_test.cc
net_parallel.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
net_parallel.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
net_simple.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
net_simple.h [Caffe2] Implement BlackBoxPredictor::BenchmarkIndividualOps (#52903) 2021-02-27 19:49:22 -08:00
net_simple_refcount.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
net_simple_refcount.h
net_simple_refcount_test.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
net_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
numa.cc
numa.h
observer.h
observer_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
operator.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
operator.h Don't switch streams when running Caffe2 ops from c10. (#55121) 2021-04-08 13:21:11 -07:00
operator_gpu_test.cc
operator_gradient.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
operator_schema.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
operator_schema.h [codemod] Enforce proper use of emplacy functions 2021-05-04 20:58:18 -07:00
operator_schema_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
operator_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
parallel_net_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
plan_executor.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
plan_executor.h
plan_executor_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
prof_dag_counters.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
prof_dag_counters.h
qtensor.cc
qtensor.h Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
qtensor_serialization.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
qtensor_serialization.h
scope_guard.h
serialization_test.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
static_tracepoint.h
static_tracepoint_elfx86.h
stats.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
stats.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
stats_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
storage.h
tensor.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
tensor.h [caffe2] Don't call TensorImpl::size() in dim32() (#53852) 2021-03-12 16:22:29 -08:00
tensor_impl.h
tensor_int8.cc
tensor_int8.h
test_utils.cc [StaticRuntime] Replace StorageImpl with TensorImpl in MemoryPlanner (#56447) 2021-04-20 23:04:01 -07:00
test_utils.h Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
timer.h
timer_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
transform.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
transform.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
transform_test.cc [clang-tidy] Exclude cppcoreguidelines-avoid-magic-numbers (#57841) 2021-05-07 20:02:33 -07:00
types.cc
types.h Renaming CAFFE2_API to TORCH_API (#49496) 2020-12-18 10:54:50 -08:00
workspace.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00
workspace.h [caffe2] Avoid some double (and triple) lookups in workspace (#53319) 2021-03-04 22:57:02 -08:00
workspace_test.cc Make PyTorch code-base clang-tidy compliant (#56892) 2021-04-28 14:10:25 -07:00