[Bug lto/49277] [4.7 Regression] 456.hmmer in SPEC CPU 2006 failed to build

["rguenther at suse dot de" <gcc-bugzilla at gcc dot gnu dot org>]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49277

--- Comment #5 from rguenther at suse dot de <rguenther at suse dot de> 2011-06-07 09:00:06 UTC ---
On Mon, 6 Jun 2011, hjl.tools at gmail dot com wrote:

> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49277
> 
> H.J. Lu <hjl.tools at gmail dot com> changed:
> 
>            What    |Removed                     |Added
> ----------------------------------------------------------------------------
>              Status|WAITING                     |NEW
> 
> --- Comment #4 from H.J. Lu <hjl.tools at gmail dot com> 2011-06-06 22:15:34 UTC ---
> It depends on the glibc header files.  This testcase:
> 
> http://www.kernel.org/pub/linux/devel/gcc/lto/pr49277.tar.bz2

Can you bisect the set of object files (adding -r -nostdlib to the link)
and provide preprocessed source for the minimal set of object files that
trigger at least one such error?

The object files are not compatible with the three versions of lto1
I have around.

It looks like a bug in some optimization pass (at least I didn't see
hmmer mixing -f[un]signed-char which would make my patch suspicious).

Thanks.