[uClibc]execve segfaults

Erik Andersen andersen at codepoet.org
Wed Jun 26 23:33:50 UTC 2002


On Thu Jun 27, 2002 at 12:17:47AM +0200, Ronald Wahl wrote:
> On Wed, 26 Jun 2002 16:09:01 -0600, Erik Andersen wrote:
> 
> > Hmm.  Well for starters, you can rule out the shared lib loader 
> > by compiling busybox staticly (edit busybox/Makefile and set 
> > DOSTATIC = true) and then testing if you still see the problem.
> 
> The problem occurs only with dynamically linked programs. So I think it
> is _in_ the shared lib loader. Maybe I should try the loader from glibc
> to be sure. 

I have limited access to one powerpc system out on the net, 
so I'll try it out there now and see how it behaves.

    <sound of compiling in the distance>

Ok.  Things compiled up just fine.  Here are a few details about
the system:

$ uname -a
Linux voltaire 2.4.18-powerpc-smp #1 SMP Thu Mar 14 19:59:10 EST 2002 ppc unknown
$ cat /proc/cpuinfo 
[------snip------]
total bogomips  : 1985.74
machine         : PowerMac3,3
motherboard     : PowerMac3,3 MacRISC Power Macintosh
L2 cache        : 1024K unified
memory          : 320MB
pmac-generation : NewWorld

Testing out the latest and greatest busybox with the latest
and greatest uClibc, this is what I get:
 
$ ldd ./busybox
        libc.so.0 => /home/andersee/usr/powerpc-linux-uclibc/lib/libc.so.0 (0x0x30016000)
        ld-uClibc.so.0 => /home/andersee/usr/powerpc-linux-uclibc/lib/ld-uClibc.so.0 (0x0x30000000)
$ ln -s busybox date
$ ls -l ./date
lrwxrwxrwx    1 andersee Debian          7 Jun 26 19:24 ./date -> busybox*
$ ./date 
Wed Jun 26 23:24:28 UTC 2002
$ ./date -u
Wed Jun 26 23:24:33 UTC 2002
$ ./date -u -u
Wed Jun 26 23:24:35 UTC 2002
$ ./date -u -u -u
Wed Jun 26 23:24:39 UTC 2002
$ test1= ./date -u -u -u
Wed Jun 26 23:24:45 UTC 2002
$ test1= test2= ./date -u -u -u
Wed Jun 26 23:24:51 UTC 2002


So everything works just perfectly for me on this powerpc system.
Wish I could help you, but i'm just not seeing the problem,

 -Erik

--
Erik B. Andersen             http://codepoet-consulting.com/
--This message was written using 73% post-consumer electrons--



More information about the uClibc mailing list