[Bug lto/47274] [4.6 regression] ICE in lto_varpool_replace_node, at lto-symtab.c:306
jakub at gcc dot gnu.org
gcc-bugzilla@gcc.gnu.org
Thu Feb 3 10:59:00 GMT 2011
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47274
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution| |FIXED
--- Comment #36 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-02-03 10:58:56 UTC ---
I guess the remaining two places could use sizeof, but the question is whether
we can portably use uint32_t in lto-streamer-out.c (lto-plugin uses it, but
nothing in gcc/*). So perhaps it could use unsigned int instead of uint32_t
for the slot no (or int).
And lto-plugin
memcpy (&entry->size, p, sizeof (uint64_t));
p += 8;
memcpy (&aux->slot, p, sizeof (uint32_t));
p += 4;
could increment by the sizeof too. Anyway, I'm not aware of any sizeof (int)
!= 4 host gcc would support, so I consider this bug fixed.
More information about the Gcc-bugs
mailing list