Fix threading issue in MPAS-O GM routine #376
Merged
Conversation
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
vanroekel
approved these changes
Sep 26, 2019
Contributor
vanroekel
left a comment
vanroekel
left a comment
There was a problem hiding this comment.
approved by visual inspection
Contributor
Author
|
Just a quick note -- I had intel inspector working and did some checking on memory as well. The tools points at invalid memory access in the gm code as well, particularly this loop: I believe this is because some of the arrays are dimensioned nEdges and the loop goes over nEdges+1. Also the "private(k)" is missing in the code -- I'll commit it to this PR. |
Contributor
|
@amametjanov or @philipwjones could you please look at the openMP pragmas and confirm that they are done correctly? |
Contributor
|
@mark-petersen OpenMP pragmas look ok to me - I know @amametjanov has been removing the private clauses since there is a default private clause on the top-level parallel region, so they're redundant but do no harm. |
Contributor
|
Yes, this looks good. I think the last two commits, which make loop extents match array sizes, fix the real culprits. Formatting is so much better too! |
mark-petersen
force-pushed
the
jonbob/ocn/fix-threading-issue
branch
from
70baf5d to
e5075b9
Compare
Contributor
|
I moved the commits over to ocean/develop, otherwise we will bring extra things over with the merge. |
mark-petersen
added a commit
that referenced
this pull request
Oct 5, 2019
This PR fixes a threading issue that was introduced in [E3SM PR#3057] (E3SM-Project/E3SM#3057) that brought in the initial implementation of 3D varying GM bolus. After that commit, some E3SM threading tests were observed to fail with different results. Tested with: * SMS_P12x2.T62_oQU240.CMPASO-NYF.compy_intel - BFB results 10 times * SMS_P12x2.ne4_oQU240.A_WCYCL1850.compy_intel.allactive-mach_mods - BFB * results 10 times This last test is one that was failing on sandiatoss3.
mark-petersen
added a commit
to mark-petersen/MPAS-Model
that referenced
this pull request
Mar 24, 2020
…_update_maint1.2 Fix threading issue in MPAS-O GM routine MPAS-Dev#376 This PR fixes a threading issue that was introduced in E3SM PR #3057 that brought in the initial implementation of 3D varying GM bolus. After that commit, some E3SM threading tests were observed to fail with different results. Tested with: SMS_P12x2.T62_oQU240.CMPASO-NYF.compy_intel - BFB results 10 times SMS_P12x2.ne4_oQU240.A_WCYCL1850.compy_intel.allactive-mach_mods - BFB results 10 times This last test is one that was failing on sandiatoss3.
mark-petersen
added a commit
that referenced
this pull request
Mar 24, 2020
Update maint1.2: 3D varying GM This update adds the following PRs from ocean/develop: - Adds options for 3d varying GM bolus and 2d varying phase speed #288 - Add GM bolus eddy stats #339 - Fix threading issue in MPAS-O GM routine #376 - compute landIceMask using geometric_features #447 - bug fixes for the nonlocal source term in KPP #305
jonbob
added a commit
to E3SM-Project/E3SM
that referenced
this pull request
Mar 27, 2020
#3513) Update maint1.2: add 3D varying GM This update brings in a new mpas-source submodule with the following PRs from MPAS ocean/develop: * Adds options for 3d varying GM bolus and 2d varying phase speed (MPAS-Dev/MPAS-Model#288) * Add GM bolus eddy stats (MPAS-Dev/MPAS-Model#339) * Fix threading issue in MPAS-O GM routine (MPAS-Dev/MPAS-Model#376) * Compute landIceMask using geometric_features (MPAS-Dev/MPAS-Model#447) * Bug fixes for the nonlocal source term in KPP (MPAS-Dev/MPAS-Model#305) and from E3SM master: * Change all ocean output files to single precision (E3SM #3360) [non-BFB]
ashwathsv
pushed a commit
to ashwathsv/MPAS-Model
that referenced
this pull request
Jul 21, 2020
…elop This PR fixes a threading issue that was introduced in [E3SM PR#3057] (E3SM-Project/E3SM#3057) that brought in the initial implementation of 3D varying GM bolus. After that commit, some E3SM threading tests were observed to fail with different results. Tested with: * SMS_P12x2.T62_oQU240.CMPASO-NYF.compy_intel - BFB results 10 times * SMS_P12x2.ne4_oQU240.A_WCYCL1850.compy_intel.allactive-mach_mods - BFB * results 10 times This last test is one that was failing on sandiatoss3.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 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.
This PR fixes a threading issue that was introduced in E3SM PR #3057 that brought in the initial implementation of 3D varying GM bolus. After that commit, some E3SM threading tests were observed to fail with different results.
Tested with:
This last test is one that was failing on sandiatoss3.