You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Qdyn5p, version 5.06 crashed when using the Intel Fortran compiler at version 15.. while optimizations (-O3) are turned on. The crash happens in the torsion subroutine code near the entry point and is not found for either version 12.1.4 or 14.0.2 of the compiler, or when optimizations are disabled with -O0.
If other compiler dependent crashes are found, please add them to this issue
The text was updated successfully, but these errors were encountered:
In which test does it crash?
Does it crash for compilations coming from the current master branch?
I remember that discussion we had about optimal compiler options now, perhaps we should have it again?
Perhaps looking into the Release notes might help:
Hello,
the crash occurred for a basic system provided by Martin Floor (see recent message on the mailing list) under the 15.0.0 version and the 15.0.0.90 version
It crashes only if the -O3 flag was set, no crash happened for -O0.
The crash was just after entry into the torsion function, but no further cause could be found because of the optimization.
With the new master branch version, no crash happens. This is scary because it means there is a hidden bug somewhere in the code
Qdyn5p, version 5.06 crashed when using the Intel Fortran compiler at version 15.. while optimizations (-O3) are turned on. The crash happens in the torsion subroutine code near the entry point and is not found for either version 12.1.4 or 14.0.2 of the compiler, or when optimizations are disabled with -O0.
If other compiler dependent crashes are found, please add them to this issue
The text was updated successfully, but these errors were encountered: