Fwd: Bug#618480: __unified_syscall bug

Hi,

maybe someone has an idea. I don’t think a function needs a
“frame pointer” but have no idea about that cfi thing.

bye,
//mirabilos
-- 
13:47⎜<tobiasu> if i were omnipotent, i would divide by zero
                all day long ;)
(thinking about http://lobacevski.tumblr.com/post/3260866481 by waga)

---------- Forwarded message ----------
From: Paul Dugan <[email protected]>
Message-ID: <[email protected]>
Date: Tue, 15 Mar 2011 16:17:50 +0100
Resent-Date: Tue, 15 Mar 2011 15:21:01 +0000
Resent-Date: Tue, 15 Mar 2011 15:21:04 +0000
Subject: Bug#618480: __unified_syscall bug

Package: dietlibc-dev
Version: 0.31-1.2
Severity: normal


Apparently __unified_syscall is implemented without:

1. no frame pointer
2. not proper cfi directives to describe how to unwind from it.

This causes no backtrace in GDB.

See post at:
http://stackoverflow.com/questions/5290257/gdb-weird-backtrace