Bug report: could not run CMM step-2 prestress parallel.

[saitoasukakawaii]

Description

I am attempting to run a CMM simulation (Step 1: pure flow, Step 2: prestress, Step 3: CMM).

Issue: Step 2 (prestress) fails to run in parallel and is restricted to single-core execution.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
ERROR occurred, see below for more explanation
ERROR: Shells with linear triangles should be run sequentially
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Observation: Older versions supported parallel runs. Specifically, in commit 222762a, I found that the code block in READMSH.f was commented out in the working parallel version.

!        Checks for shell elements
         DO iM=1, nMsh
            IF (msh(iM)%lShl) THEN
               IF (msh(iM)%eType.NE.eType_NRB .AND.
     2             msh(iM)%eType.NE.eType_TRI3) THEN
                  err = "Shell elements can be either triangles "//
     2               "or C1-NURBS"
               END IF
               IF (msh(iM)%eType .EQ. eType_NRB) THEN
                  DO i=1, nsd-1
                     IF (msh(iM)%bs(i)%p .LE. 1) err =
     2                  "NURBS for shell elements should be p > 1"
                  END DO
               END IF
c               IF (msh(iM)%eType .EQ. eType_TRI) THEN
c                  IF (.NOT.cm%seq()) err = "Triangular shell elements"//
c     2               " should be run sequentially"
c               END IF
            END IF
         END DO

Reproduction

run step 2 prestress parallel:
mpirun -np 8 svFSI svFSI.inp

Expected behavior

run step 2 prestress parallel.

Additional context

No response

Code of Conduct

• I agree to follow this project's Code of Conduct and Contributing Guidelines

[ktbolt]

@saitoasukakawaii The code

+               IF (msh(iM)%eType .EQ. eType_TRI3) THEN
+                  IF (.NOT.cm%seq()) err = "Shells with linear "//
+     2               "triangles should be run sequentially"
+               END IF

was uncommented in
commit ac5ee22
Mon Jan 31 18:14:46 2022

There was no comment in the commit why this was done so I can only assume that there was a problem with the parallel execution using shells.