Skip to content
This repository has been archived by the owner on Nov 17, 2023. It is now read-only.

fix shared_storage free #11159

Merged
merged 6 commits into from
Jun 6, 2018
Merged

fix shared_storage free #11159

merged 6 commits into from
Jun 6, 2018

Conversation

zhreshold
Copy link
Member

Description

Fix ndarray free on cpu_shared.

Original problem:

Using gluon.data.DataLoader with num_workers > 0 if ndarray was not originally on cpu_shared will cause

IOError: [Errno 9] Bad file descriptor

because file is already closed.

Checklist

Essentials

Please feel free to remove inapplicable items for your PR.

  • The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to the relevant JIRA issue created (except PRs with tiny changes)
  • Changes are complete (i.e. I finished coding on this PR)
  • All changes have test coverage:
  • Unit tests are added for small changes to verify correctness (e.g. adding a new operator)
  • Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore)
  • Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL)
  • Code is well-documented:
  • For user-facing API changes, API doc string has been updated.
  • For new C++ functions in header files, their functionalities and arguments are documented.
  • For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable
  • Check the API doc at http://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html
  • To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change

Changes

  • Feature1, tests, (and when applicable, API doc)
  • Feature2, tests, (and when applicable, API doc)

Comments

  • If this change is a backward incompatible change, why must this change be made.
  • Interesting edge cases to note here

@@ -215,12 +215,12 @@ void CPUSharedStorageManager::FreeImpl(const Storage::Handle& handle) {
<< "Failed to unmap shared memory. munmap failed with error "
<< strerror(errno);

if (count == 0) {
Copy link
Member

@szha szha Jun 5, 2018

Choose a reason for hiding this comment

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

the closing curly brace needs to be moved out of #ifdef __linux__ too

CHECK_EQ(close(handle.shared_id), 0)
<< "Failed to close shared memory. close failed with error "
<< strerror(errno);
CHECK_EQ(close(handle.shared_id), 0)
Copy link
Contributor

Choose a reason for hiding this comment

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

if(count==0) shouldnt' be moved. Skip close if shared_id == -1 instead

Copy link
Member Author

Choose a reason for hiding this comment

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

You mean close on first FreeImpl and skip if already closed?

@zhreshold
Copy link
Member Author

@piiswrong Can you have a second look?

@piiswrong piiswrong merged commit 318c689 into apache:master Jun 6, 2018
zheng-da pushed a commit to zheng-da/incubator-mxnet that referenced this pull request Jun 28, 2018
* fix shared_storage free

* fix bracket

* make local ref

* cpplint

* fix tests

* fix tests
XinYao1994 pushed a commit to XinYao1994/incubator-mxnet that referenced this pull request Aug 29, 2018
* fix shared_storage free

* fix bracket

* make local ref

* cpplint

* fix tests

* fix tests
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants