------- Comment #6 from tglx at linutronix dot de 2009-11-20 00:52 -------
I changed the summary to match the real problem.
Further info:
While testing various kernel configs we found out that the problem
comes and goes. Finally I started to compare the gcc command line
options and after some fiddling it turned out that the following
minimal deltas change the code generator behaviour:
Bad: -march=pentium-mmx -Wa,-mtune=generic32
Good: -march=i686 -mtune=generic -Wa,-mtune=generic32
Good: -march=pentium-mmx -mtune-generic -Wa,-mtune=generic32
The good ones produce:
650: 55 push %ebp
651: 89 e5 mov %esp,%ebp
653: 83 e4 f0 and $0xfffffff0,%esp
The bad one:
000005f0 <timer_stats_update_stats>:
5f0: 57 push %edi
5f1: 8d 7c 24 08 lea 0x8(%esp),%edi
5f5: 83 e4 f0 and $0xfffffff0,%esp
5f8: ff 77 fc pushl -0x4(%edi)
5fb: 55 push %ebp
5fc: 89 e5 mov %esp,%ebp
It's worse code for no reason and breaks the kernel assumption of ebp + 4
pointing to the real return address on the stack.
--
tglx at linutronix dot de changed:
What |Removed |Added
----------------------------------------------------------------------------
Summary|16 byte stack alignment on |stack alignment happens
|random Linux kernel |_before_ mcount "push %ebp
|functions |..." depending on -mtune
| |flags
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42109
