Skip to content
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.

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

Accessing component from locality where it was not created segfaults #1898

Closed
ltroska opened this issue Dec 8, 2015 · 1 comment
Closed

Comments

@ltroska
Copy link
Contributor

ltroska commented Dec 8, 2015

The following minimal example http://pastebin.com/RmNccZX6 segfaults if using >= 2 localities on v0.9.11 release.
Output when using hpxrun.py -l 2 ./a.out

num locs 3
invoked get_data on 0 getting size 4
applying from 0 on {0000000200000000, 0000000000000000}
applying from 0 on {0000000300000000, 0000000000000000}
received on 1:
Process 0 failed with an unexpected error code of -11 (expected 0)
Process 1 failed with an unexpected error code of -15 (expected 0)

running it individually I get on locality 0:

num locs 2
invoked get_data on 0 getting size 4
applying from 0 on {0000000200000000, 0000000000000000}
received on 1:
Segmentation fault

@ltroska ltroska changed the title Accessing component from locality where it was not located segfaults Accessing component from locality where it was not created located segfaults Dec 8, 2015
@ltroska ltroska changed the title Accessing component from locality where it was not created located segfaults Accessing component from locality where it was not created segfaults Dec 8, 2015
sithhell added a commit that referenced this issue Dec 8, 2015
We currently run into a situation where we would need to wait on futures before
actual serialization after the first round of awaiting has been completed. In
addition, the future itself doesn't give its held value the chance to be awaited.
@hkaiser hkaiser added this to the 0.9.12 milestone Dec 8, 2015
@hkaiser
Copy link
Member

hkaiser commented Dec 8, 2015

Initial investigation shows that this test reveals two issues:

sithhell added a commit that referenced this issue Dec 8, 2015
We currently run into a situation where we would need to wait on futures before
actual serialization after the first round of awaiting has been completed. In
addition, the future itself doesn't give its held value the chance to be awaited.
sithhell added a commit that referenced this issue Dec 9, 2015
Fixing a problem with credit splitting as revealed by #1898
@hkaiser hkaiser closed this as completed Dec 10, 2015
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants