[Bug gdb/22594] Possible access to unintended variable in "gdb/stabsread.c" line 1364

["ssbssa at sourceware dot org" <sourceware-bugzilla@sourceware.org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=22594

Hannes Domani <ssbssa at sourceware dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ssbssa at sourceware dot org

--- Comment #2 from Hannes Domani <ssbssa at sourceware dot org> ---
(In reply to Petru-Florin Mihancea from comment #0)
> While experimenting with a CodeSonar plugin we develop, we noticed a
> potential bug in file "gdb/stabsread.c" line 1364, function define_symbol 
> 
> 1359           *typedef_sym = *sym;
> 1360           SYMBOL_ACLASS_INDEX (typedef_sym) = LOC_TYPEDEF;
> 1361           SYMBOL_VALUE (typedef_sym) = valu;
> 1362           SYMBOL_DOMAIN (typedef_sym) = VAR_DOMAIN;
> 1363           if (TYPE_NAME (SYMBOL_TYPE (sym)) == 0)
> 1364             TYPE_NAME (SYMBOL_TYPE (sym)) //HERE
> 1365               = obconcat (&objfile->objfile_obstack,
> 1366                           SYMBOL_LINKAGE_NAME (sym),
> 1367                           (char *) NULL);
> 1368           add_symbol_to_list (typedef_sym, &file_symbols);
> 
> Shouldn't typedef_sym be used instead of sym (and maybe also in the previous
> line)? A similar issue might be in line 1320.
> 
> Regards,
> Petru Florin Mihancea

Because of:
          *typedef_sym = *sym;

TYPE_NAME (SYMBOL_TYPE (sym)) and TYPE_NAME (SYMBOL_TYPE (typedef_sym)) are
equivalent, so it doesn't make a difference which one is used here.

-- 
You are receiving this mail because:
You are on the CC list for the bug.