Add Stream for fetch op handle#16600
Merged
chengduoZH merged 2 commits intoPaddlePaddle:developfrom Apr 2, 2019
Merged
Conversation
chengduoZH
force-pushed
the
add_delay_ops_for_threaded_executor
branch
from
7e0a41c to
f2bed8f
Compare
test=develop
chengduoZH
force-pushed
the
add_delay_ops_for_threaded_executor
branch
from
f2bed8f to
90b3e94
Compare
chengduoZH
force-pushed
the
add_delay_ops_for_threaded_executor
branch
from
747322e to
1804b19
Compare
chengduoZH
commented
Apr 2, 2019
| #ifdef PADDLE_WITH_CUDA | ||
| TensorCopySync(t, cpu, &tensors_[i]); | ||
| TensorCopy(t, cpu, *dev_ctxes_.at(t.place()), &tensors_[i]); | ||
| dev_ctxes_.at(t.place())->Wait(); |
Contributor
Author
There was a problem hiding this comment.
Don't use default stream here, it may make program slow. https://devblogs.nvidia.com/gpu-pro-tip-cuda-7-streams-simplify-concurrency/
Yancey0623
reviewed
Apr 2, 2019
| void ReduceSSAGraphBuilder::InsertPostprocessOps(ir::Graph *result) const { | ||
| if (UseGPU()) { | ||
| if (strategy_.fuse_broadcast_op_) { | ||
| if (strategy_.fuse_broadcast_ops_) { |
Contributor
There was a problem hiding this comment.
Maybe we need to let users know when to turn this flag on? If it always works better with Reduce strategy, please keep it on. In additional, we can update the best practices of GPU distributed training.
Contributor
Author
There was a problem hiding this comment.
Maybe we need to let users know when to turn this flag on?
I have added the doc for fuse_broadcast_ops in pybind.cc.
In additional, we can update the best practices of GPU distributed training.
I quite agree with you that we should have a best practice doc.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.