[ET Trac] #2842: Silo captures incorrect LIBS values on second CST run

[Roland Haas]

#2842: Silo captures incorrect LIBS values on second CST run

 Reporter: Roland Haas
   Status: new
Milestone: 
  Version: 
     Type: bug
 Priority: major
Component: EinsteinToolkit thorn

In git hash [49ef0de](https://github.com/EinsteinToolkit/ExternalLibraries-Silo/commits/49ef0de20fc9c4f5b6690bdbbdd8afc0af746194) "Silo: capture option list LIBS and LIBDIRS" of [ExternalLibraries-Silo](https://github.com/EinsteinToolkit/ExternalLibraries-Silo) code was added to capture LIBS for systems where extra libraries are required to make simple code compile. This introduced a different sort of bug.

It only captures the correct `LIBS` once when CST for PETSc runs in a "clean" configuration. Afterwards `LIBS` is set to the full set of ExternalLibraries in the makefile before CST runs and thus it captures all libraries.

A correct fix would seem to require changes to the flesh [720ef037](https://bitbucket.org/cactuscode/cactus/commits/720ef0372df4089a51e777fd8ad7593c9ceb906c) "Cactus: keep track of ExternalLibs separately from Libs" of [cactus](https://bitbucket.org/cactuscode/cactus)

Related to #2821 which has the same issue with LIBS.

--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2842/silo-captures-incorrect-libs-values-on
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.einsteintoolkit.org/pipermail/trac/attachments/20250104/fe84440d/attachment.htm>