Messages in this thread | | | Date | Thu, 18 Jun 2009 00:34:03 -0400 (EDT) | From | Steven Rostedt <> | Subject | Re: problem with function_graph self-test? |
| |
On Wed, 17 Jun 2009, Jake Edge wrote:
> you don't run into this under your "normal" configs?
Doing some google research, I came across this:
http://www.mail-archive.com/gcc@gcc.gnu.org/msg33101.html
and
http://gcc.gnu.org/ml/gcc-patches/2008-06/msg00025.html
and if I disable OPTIMIZE_FOR_SIZE I get this:
00000420 <timer_stats_update_stats>: 420: 55 push %ebp 421: 89 e5 mov %esp,%ebp 423: 83 e4 e0 and $0xffffffe0,%esp
Which gives me the right results. Perhaps we need to make FUNCTION_GRAPH depend on !OPTIMIZE_FOR_SIZE, or perhaps if it is only i386 then depend on !(OPTIMIZE_FOR_SIZE && X86_32) :-/
-- Steve
| |