Some parallel adaptive longtests fail after hybrid-assemble was merged into trunk.
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Fluidity |
Fix Released
|
High
|
Lawrence Mitchell |
Bug Description
After the hybrid-assemble branch was merged into the trunk in May (at r3984) three of my parallel adaptive (multiphase flow) longtests started to silently fail (unfortunately there are currently no checks for solver convergence, and the tests still show as "passing" on buildbot even after the Velocity solve fails after just a few timesteps):
mphase_
mphase_
mphase_
I do not have OpenMP support enabled, and as far as I know the trunk does not configure with OpenMP either. These tests all run fine using the previous trunk revision (r3983), but each one fails after just a few timesteps with the hybrid-assemble contributions in r3984. The test I am currently looking at (mphase_
After changing the following line in assemble/
assemble_
to:
assemble_
the mphase_
Related branches
- Stephan Kramer: Approve
-
Diff: 76 lines (+25/-9)2 files modifiedassemble/Momentum_DG.F90 (+5/-1)
femtools/Parallel_fields.F90 (+20/-8)
Changed in fluidity: | |
importance: | Undecided → High |
assignee: | nobody → Lawrence Mitchell (wence) |
status: | New → In Progress |
Changed in fluidity: | |
status: | In Progress → Fix Committed |
Changed in fluidity: | |
status: | Fix Committed → Fix Released |
On 10/08/12 16:15, Christian Jacobs wrote:
> Public bug reported:
...
> After changing the following line in assemble/ Momentum_ DG.F90: or.element_ neighbour_ owned(U, ele) or.element_ owned(U, ele)
> assemble_element = .not.dg.
> to:
> assemble_element = .not.dg.
I was convinced at the time that element_ neighbour_ owned(U, ele) would
return .true. if ele was owned. i.e. it's a superset of
element_owned(U, ele). However, it's possible that for some corner
cases this may not be true.
Does it work if you change the test to:
.not.dg .or. (element_owned(U, ele) .or. element_ neighbour_ owned(U, ele))
?
Lawrence