[petsc-dev] Seg fault after changeset aa6a8e1

[Satish Balay]

==13017== Invalid read of size 8
==13017==    at 0x4FD9477: PetscViewerASCIIAddTab (filev.c:302)
==13017==    by 0x5E3507E: KSPMonitorTrueResidualNorm (iterativ.c:258)
==13017==    by 0x5E1E8BB: KSPMonitor (itfunc.c:1595)
==13017==    by 0x5D65159: KSPGMRESCycle (gmres.c:137)
==13017==    by 0x5D66179: KSPSolve_GMRES (gmres.c:236)
==13017==    by 0x5E131EE: KSPSolve (itfunc.c:585)
==13017==    by 0x402754: main (ex2.c:198)
==13017==  Address 0x298 is not stack'd, malloc'd or (recently) free'd


git bisect says - the following commit is causing this change.

f55353a2a7c46e5e81f955495d9d64b6042b9555 is the first bad commit
commit f55353a2a7c46e5e81f955495d9d64b6042b9555
Author: Barry Smith <bsmith at mcs.anl.gov>
Date:   Thu Jul 30 20:16:03 2015 -0500

    initial locks with concurrencykit and OpenMP
    
    cleanup of many monitors that used nonthread safe constructors when no viewer passed in
    
    cleanup of examples that generated unused variable messages when compiled with --with-log=0



Satish


On Tue, 18 Aug 2015, Garth N. Wells wrote:

> I get a seg fault on master since changeset aa6a8e1 (
> https://bitbucket.org/petsc/petsc/commits/aa6a8e194ee3df4b7fc8e67a3c63e52c695f2de4?at=master)
> when I call  KSPMonitorSet(ksp, KSPMonitorTrueResidualNorm, NULL, NULL). Is
> there a problem with the changeset?
> 
> Garth
>