This is the mail archive of the
glibc-bugs@sourceware.org
mailing list for the glibc project.
[Bug libc/12123] SIGBUS on strstr_sse42 due to bad alignment
- From: "viriketo at gmail dot com" <sourceware-bugzilla at sourceware dot org>
- To: glibc-bugs at sources dot redhat dot com
- Date: Thu, 28 Oct 2010 20:10:35 +0000
- Subject: [Bug libc/12123] SIGBUS on strstr_sse42 due to bad alignment
- Auto-submitted: auto-generated
- References: <bug-12123-131@http.sourceware.org/bugzilla/>
http://sourceware.org/bugzilla/show_bug.cgi?id=12123
--- Comment #20 from LluÃs Batlle <viriketo at gmail dot com> 2010-10-28 20:10:04 UTC ---
(In reply to comment #19)
> (In reply to comment #18)
> > (In reply to comment #15)
> > > strstr.os was compiled with -mpreferred-stack-boundary=2 which implies
> > > -mincoming-stack-boundary=2, so it is a compiler bug if it assumes bigger
> > > alignment.
> >
> > I see '-mincoming-stack-boundary' appears only in gcc 4.4. Is gcc 4.4 a
> > requirement for building glibc?
> >
> > (In reply to comment #17)
> > We use gcc 4.3 to build glibc in i686-linux, and we use gcc 4.5.1 to build
>
> You need gcc 4.4 or above to properly align the stack in 32bit.
Ok, thank you. I see.
I tried building glibc with gcc 4.5.1, and strstr_sse42 starts properly:
1121d0: 55 push %ebp
1121d1: 89 e5 mov %esp,%ebp
1121d3: 83 e4 f0 and $0xfffffff0,%esp
Couldn't the configure script check the gcc version? Otherwise gcc 4.3 silently
builds a broken glibc.
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.