This is an automated email from the ASF dual-hosted git repository.
apeforest pushed a change to branch benchmark
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 0cbee04 Fix the index_t with int comparisoon
add 51c2065 Pointwise fusion for GPU (#15167)
No new revisions were added by this update.
Summary of changes:
docs/static_site/src/pages/api/faq/env_var.md | 25 +-
src/common/exec_utils.cc | 79 ++
src/common/exec_utils.h | 19 +
src/executor/exec_pass.h | 42 ++
src/executor/graph_executor.cc | 48 +-
src/executor/infer_graph_attr_pass.cc | 287 ++++---
src/executor/pointwise_fusion_pass.cc | 308 ++++++++
src/executor/simple_partition_pass.h | 445 +++++++++++
src/imperative/cached_op.cc | 292 +++++---
src/imperative/cached_op.h | 2 -
src/imperative/imperative.cc | 4 +-
src/operator/fusion/fused_op-inl.h | 999 +++++++++++++++++++++++++
src/operator/fusion/fused_op.cc | 307 ++++++++
src/operator/fusion/fused_op.cu | 746 ++++++++++++++++++
src/operator/fusion/fused_op.h | 203 +++++
src/operator/mshadow_op.h | 2 +-
src/operator/tensor/elemwise_unary_op_basic.cc | 2 +-
src/storage/pooled_storage_manager.h | 4 +-
tests/python/gpu/test_fusion.py | 223 ++++++
tests/python/unittest/test_gluon.py | 41 +
20 files changed, 3862 insertions(+), 216 deletions(-)
create mode 100644 src/common/exec_utils.cc
create mode 100644 src/executor/pointwise_fusion_pass.cc
create mode 100644 src/executor/simple_partition_pass.h
create mode 100644 src/operator/fusion/fused_op-inl.h
create mode 100644 src/operator/fusion/fused_op.cc
create mode 100644 src/operator/fusion/fused_op.cu
create mode 100644 src/operator/fusion/fused_op.h
create mode 100644 tests/python/gpu/test_fusion.py
