Skip to content

Add lock to record_event. #11346

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jun 11, 2018
Merged

Add lock to record_event. #11346

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions paddle/fluid/platform/device_context.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@ limitations under the License. */
#pragma once

#include <memory>
#include <mutex> // NOLINT
#include <string>
#include <unordered_map>
#include <vector>
Expand Down Expand Up @@ -100,6 +101,7 @@ class CUDADeviceContext : public DeviceContext {

template <typename Callback>
void RecordEvent(cudaEvent_t ev, Callback callback) {
std::lock_guard<std::mutex> guard(mtx_);
callback();
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

How about replacing std::mutex with std::recursive_mutex? Because, if the callback() contains another RecordEvent() calling, it will be hang again.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This situation should not happen.

PADDLE_ENFORCE(cudaEventRecord(ev, stream_));
}
Expand All @@ -116,6 +118,8 @@ class CUDADeviceContext : public DeviceContext {
int compute_capability;
int multi_process;
int max_threads_per_mp;

std::mutex mtx_;
};

template <>
Expand Down