This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug debug/53363] g++.dg/debug/dwarf2/thunk1.C FAILs
- From: "ro at CeBiTec dot Uni-Bielefeld.DE" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Wed, 16 May 2012 10:51:33 +0000
- Subject: [Bug debug/53363] g++.dg/debug/dwarf2/thunk1.C FAILs
- Auto-submitted: auto-generated
- References: <bug-53363-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53363
--- Comment #2 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot Uni-Bielefeld.DE> 2012-05-16 10:51:33 UTC ---
> --- Comment #1 from Jason Merrill <jason at gcc dot gnu.org> 2012-05-15 19:09:37 UTC ---
> Yes, the test should only run in 32-bit mode.
>
> For me, on i686-unknown-linux-gnu the test passes. What are the uses of LFB3
> on Solaris for?
The failure on linux is with 32-bit binaries in my case, but only for
the -m64 compilation, which shouldn't be run in the first place.
I've recompiled the file on Solaris/x86 with -dA. Those are the
instances of LFB3:
.LFB3:
.long .LFB3 / FDE initial location
.long .LFE3-.LFB3 / FDE address range
.long .LFB3@GOTOFF / FDE initial location
.long .LFE3-.LFB3 / FDE address range
I'm also attaching the complete file.
On Darwin/x86 (32-bit i386-apple-darwin11.3.0 compiler), LFB3 doesn't
occur in thunk1.s at all.
Rainer