Tobias Schlüter - Re: [PATCH] PR fortran/24005 (original) (raw)

This is the mail archive of the gcc-patches@gcc.gnu.orgmailing list for the GCC project.

• From: Tobias Schlüter
• To: Steve Kargl
• Cc: fortran at gcc dot gnu dot org, gcc-patches at gcc dot gnu dot org
• Date: Thu, 22 Sep 2005 22:36:59 +0200
• Subject: Re: [PATCH] PR fortran/24005
• References: <20050922194925.GA59701@troutmask.apl.washington.edu>

Steve Kargl wrote:

The problem, again discussed in the PR, is that gfortran was passing a NULL pointer to strcmp.

OK for mainline?

This is not yet ok ...

--- 926,933 ---- if (p->sym == q->sym) continue; /* Duplicates OK here */

! if (p->sym->name && q->sym->name && p->sym->module && q->sym->module ! && strcmp (p->sym->name, q->sym->name) == 0 && strcmp (p->sym->module, q->sym->module) == 0) continue;

p->sym->name and q->sym->name should never be NULL. Furthermore, the module name needn't be set, as interfaces can reference symbols in the current scope, IOW the correct check would be:

if (strcmp (p->sym->name, q->sym->name) == 0 && ((!p->sym->module && !q->sym->module) || (p->sym->module && q->sym->module && strcmp (p->sym->module, q->sym->module) == 0)))

NB I remember that we stumbled across a very similar bug when I introduced the string table changes, which it turned out I had missed, because glibc's strcmp does "the right thing" with NULL arguments unlike FreeBSD's. Alternatively, (if I understand C well enough) we could simply compare the string pointers (they're allocated by gfc_get_string) like so:

if (p->sym->name == q->sym->name && p->sym->module == q->sym->module)

• Tobi

• Follow-Ups:
  • Re: [PATCH] PR fortran/24005
    * From: Steve Kargl
  • Re: [PATCH] PR fortran/24005
    * From: Paul Thomas
• References:
  • [PATCH] PR fortran/24005
    * From: Steve Kargl