Re: toolchain/59549: gdb is not ctype(3) safe

["Taylor R Campbell via gnats" <gnats-admin%NetBSD.org@localhost>]

The following reply was made to PR toolchain/59549; it has been noted by GNATS.

From: Taylor R Campbell <riastradh%NetBSD.org@localhost>
To: Thomas Klausner <wiz%NetBSD.org@localhost>
Cc: gnats-bugs%NetBSD.org@localhost, netbsd-bugs%NetBSD.org@localhost
Subject: Re: toolchain/59549: gdb is not ctype(3) safe
Date: Sat, 26 Jul 2025 12:50:03 +0000

 > #7  0x00000000006d15f9 in symbol_find_demangled_name () at /usr/src/exter=
 nal/gpl3/gdb/lib/libgdb/../../dist/gdb/symtab.c:961
 
    948    if (gsymbol->language () !=3D language_unknown)
    949      {
    950        const struct language_defn *lang =3D language_def (gsymbol->l=
 anguage ());
    951 =20
    952        lang->sniff_from_mangled_name (mangled, &demangled);
    953        return demangled;
    954      }
    955 =20
    956    for (i =3D language_unknown; i < nr_languages; ++i)
    957      {
    958        enum language l =3D (enum language) i;
    959        const struct language_defn *lang =3D language_def (l);
    960 =20
 =3D> 961        if (lang->sniff_from_mangled_name (mangled, &demangled))
 
 If gdb doesn't know the language _of this symbol_ for demangling, it
 will try all the languages' demangling methods, whether or not there's
 any specific reason for Ada to be involved in this.
 
 We should just patch gdb locally and file an upstream bug, basic case
 of clear undefined behaviour that is triggered in real-world use.
 Christos's patch looks fine to me (but I did not check whether there
 are any missing cases).
 
 I suggest filing a separate PR for the wip/Guitar issue.