Randall, If you replace the PetscStrncpy(b,a,n+1); with PetscStrncpy(b,a,n);
#define FIXCHAR(a,n,b) \ {\ if (a == PETSC_NULL_CHARACTER_Fortran) { \ b = a = 0; \ } else { \ while((n > 0) && (a[n-1] == ' ')) n--; \ *ierr = PetscMalloc((n+1)*sizeof(char),&b); \ if (*ierr) return; \ *ierr = PetscStrncpy(b,a,n+1); \ if (*ierr) return; \ b[n] = 0; \ } \ } in include/petsc-private/fortranimpl.h and recompile the PETSc libraries does the problem go away? Let me know and I’ll fix the code. Thanks Barry On Aug 2, 2014, at 2:20 PM, Randall Mackie <rlmackie...@gmail.com> wrote: > The attached small program, basically a call to PetscPrintf, gives the > following valgrind errors: > > [rmackie ~/tst_petsc_problem] ./cmd_test > ==24812== Invalid read of size 1 > ==24812== at 0x4C2E500: __GI_strncpy (in > /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) > ==24812== by 0x4EF5B4D: PetscStrncpy (str.c:335) > ==24812== by 0x4F334E0: petscprintf_ (zmprintf.c:54) > ==24812== by 0x400F28: MAIN__ (test.F90:11) > ==24812== by 0x400F9A: main (test.F90:16) > ==24812== Address 0x8cef9bb is 0 bytes after a block of size 11 alloc'd > ==24812== at 0x4C2AB80: malloc (in > /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) > ==24812== by 0x6828BB4: ??? (in > /usr/lib/x86_64-linux-gnu/libgfortran.so.3.0.0) > ==24812== by 0x68FBC97: _gfortran_string_trim (in > /usr/lib/x86_64-linux-gnu/libgfortran.so.3.0.0) > ==24812== by 0x400EFD: MAIN__ (test.F90:11) > ==24812== by 0x400F9A: main (test.F90:16) > ==24812== > > > The program was compiled with PETSc V3.5.1 and --download-mpich. > > The same program compiled with V3.4.4 and --download-mpich does not give > these errors. > > It seems like an innocuous error, and I can certainly create a suppression > file for it, but I thought I'd report it anyway. > I get the same error on 2 different systems and I've tried 3 different > gfortran versions, all with the same result. > > Program, configure commands, etc, attached. > > Randy M. > > > <makefile><test.F90><valgrind_out.txt><cmd_configure><cmd_test>