You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hama.apache.org by "Thomas Jungblut (JIRA)" <ji...@apache.org> on 2011/06/17 17:58:47 UTC
[jira] [Created] (HAMA-404) LocalBSPRunner returns wrong superstep
numbers
LocalBSPRunner returns wrong superstep numbers
----------------------------------------------
Key: HAMA-404
URL: https://issues.apache.org/jira/browse/HAMA-404
Project: Hama
Issue Type: Bug
Affects Versions: 0.2.0
Reporter: Thomas Jungblut
Assignee: Thomas Jungblut
Priority: Critical
Fix For: 0.3.0
Superstep is a threadlocal variable in the LocalGroom, so each thread is resposible for itself to increment the superstep after calling sync.
In the old logic it was a non-local variable which where just the first thread should increment it.
Remove the "only first thread updates" logic and let each thread increment the superstep.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HAMA-404) LocalBSPRunner returns wrong superstep
numbers
Posted by "Thomas Jungblut (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HAMA-404?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Thomas Jungblut updated HAMA-404:
---------------------------------
Attachment: HAMA-404_v00.patch
This fixes it.
It is build on windows, so I hope that hudson won't cry again.
> LocalBSPRunner returns wrong superstep numbers
> ----------------------------------------------
>
> Key: HAMA-404
> URL: https://issues.apache.org/jira/browse/HAMA-404
> Project: Hama
> Issue Type: Bug
> Affects Versions: 0.2.0
> Reporter: Thomas Jungblut
> Assignee: Thomas Jungblut
> Priority: Critical
> Fix For: 0.3.0
>
> Attachments: HAMA-404_v00.patch
>
>
> Superstep is a threadlocal variable in the LocalGroom, so each thread is resposible for itself to increment the superstep after calling sync.
> In the old logic it was a non-local variable which where just the first thread should increment it.
> Remove the "only first thread updates" logic and let each thread increment the superstep.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HAMA-404) LocalBSPRunner returns wrong superstep
numbers
Posted by "Thomas Jungblut (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HAMA-404?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Thomas Jungblut updated HAMA-404:
---------------------------------
Status: Patch Available (was: Open)
> LocalBSPRunner returns wrong superstep numbers
> ----------------------------------------------
>
> Key: HAMA-404
> URL: https://issues.apache.org/jira/browse/HAMA-404
> Project: Hama
> Issue Type: Bug
> Affects Versions: 0.2.0
> Reporter: Thomas Jungblut
> Assignee: Thomas Jungblut
> Priority: Critical
> Fix For: 0.3.0
>
> Attachments: HAMA-404_v00.patch
>
>
> Superstep is a threadlocal variable in the LocalGroom, so each thread is resposible for itself to increment the superstep after calling sync.
> In the old logic it was a non-local variable which where just the first thread should increment it.
> Remove the "only first thread updates" logic and let each thread increment the superstep.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HAMA-404) LocalBSPRunner returns wrong
superstep numbers
Posted by "Edward J. Yoon (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HAMA-404?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13051489#comment-13051489 ]
Edward J. Yoon commented on HAMA-404:
-------------------------------------
I just committed this, Thanks Thomas!
> LocalBSPRunner returns wrong superstep numbers
> ----------------------------------------------
>
> Key: HAMA-404
> URL: https://issues.apache.org/jira/browse/HAMA-404
> Project: Hama
> Issue Type: Bug
> Affects Versions: 0.2.0
> Reporter: Thomas Jungblut
> Assignee: Thomas Jungblut
> Priority: Critical
> Fix For: 0.3.0
>
> Attachments: HAMA-404_v00.patch
>
>
> Superstep is a threadlocal variable in the LocalGroom, so each thread is resposible for itself to increment the superstep after calling sync.
> In the old logic it was a non-local variable which where just the first thread should increment it.
> Remove the "only first thread updates" logic and let each thread increment the superstep.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HAMA-404) LocalBSPRunner returns wrong superstep
numbers
Posted by "Edward J. Yoon (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HAMA-404?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Edward J. Yoon updated HAMA-404:
--------------------------------
Resolution: Fixed
Status: Resolved (was: Patch Available)
> LocalBSPRunner returns wrong superstep numbers
> ----------------------------------------------
>
> Key: HAMA-404
> URL: https://issues.apache.org/jira/browse/HAMA-404
> Project: Hama
> Issue Type: Bug
> Affects Versions: 0.2.0
> Reporter: Thomas Jungblut
> Assignee: Thomas Jungblut
> Priority: Critical
> Fix For: 0.3.0
>
> Attachments: HAMA-404_v00.patch
>
>
> Superstep is a threadlocal variable in the LocalGroom, so each thread is resposible for itself to increment the superstep after calling sync.
> In the old logic it was a non-local variable which where just the first thread should increment it.
> Remove the "only first thread updates" logic and let each thread increment the superstep.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira