This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Stack alignment


Hi Steve,

I'm assuming that you've still not received an answer about your stack 
alignment problem.  Here's another piece of information.  I recently
installed slackware-7.0 on a laptop (it comes with glibc-2.1.2) and I tried 
the stack alignment test you posted.  It worked--whether I use slackware's 
default compiler (egcs-1.1.2) or the latest gcc snapshot.  However when I 
simply take the executable over to my old machine, I get the stack alignment 
problem.  So.

1.  Either it's success is a fluke on the new machine (it's an i686 vs an i586).

2.  The bug really doesn't have anything to do with the current snapshot (i.e.,
	it's fixed) or with the older egcs-1.1.2 release.  If this is the
	case, then the culprit is glibc or one of the header .o files.  On
	my i586, these were compiled (by me) using gcc-2.95.2, while I presume
	Slackware used egcs-1.1.2.  One other hint in this direction is a 	
	problem I've been having with using the Slackware libc to build C++
	programs using gcc-2.95.2 or the snapshots.  I get an immediate SIGSEGV
	(in libc) that goes away if I use egcs to compile.
	
	
David



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]