issue with pthreads on Fedora 9

Hi,

I wonder if anybody on the list has faced an issue like mine (searching 
the list archive did not help):

Running a relatively complex application on Fedora 9 (the app is a 
LispWorks5.0.2 console image with our code delivered with 
:multiprocessing t, and it indeed uses threads), I get
"ibgcc_s.so.1 must be installed for pthread_cancel to work" on stderr, 
and it exits -11.

The same binary file works fine on Fedora 8. (I'll raise it to support 
once I get a minimal standalone app for debugging, unless there is a 
simple way to fix the issue).

Thanks,
Mike

RE: issue with pthreads on Fedora 9

Hi Michael,

Google is your friend -- seems to be a problem of Linux with particular versions of libraries on AMD64?

http://serversupportforum.de/forum/linux/11415-libstdc-so-5-a.html

David

____________________________________________________________
FREE 3D EARTH SCREENSAVER - Watch the Earth right on your desktop!
Check it out at http://www.inbox.com/earth

Re: issue with pthreads on Fedora 9

david tolpin wrote:
> Google is your friend -- seems to be a problem of Linux with
> particular versions of libraries on AMD64?

Thanks, David,
unfortunately, it's not the case: neither AMD, nor 64bit. There are 
better guesses about a bug in gcc3.3, but ...

Regards,
Mike

Re: issue with pthreads on Fedora 9

Michael Sulyaev wrote:
> ... once I get a minimal standalone app for debugging  ...

Ok, here it is:

$ cat bug.lisp
(defun start ()
   (print "Hello LW!"))

(progn
   (compile 'start)
   (deliver 'start "bug" 0 :multiprocessing t)
   (quit))

$ lw502 -init bug.lisp
....
Total Size 24363K, Allocated 21216K, Free 2799K
Shaking stage : Saving image
Delivery successful - bug

# Copying the file to a Fedora 9 machine, and running
$./bug
[msulyaev@vfc9 ~]$ ./bug

"Hello LW!"libgcc_s.so.1 must be installed for pthread_cancel to work

Error: Signal 6 [code 0] at 110416 {inside foreign code}
eax        0 ; ebx     34B3 ; ecx     34B5 ; edx        6
esp  172DBF8 ; ebp  172DC04 ; esi FFFFFFFF ; edi   CDCFF4
   1 (abort) Quit process.

Type :b for backtrace, :c <option number> to proceed,  or :? for other 
options

CL-USER 1 : 1 > :b
Call to (SUBFUNCTION MP::PROCESS-SG-FUNCTION MP::INITIALIZE-PROCESS-STACK)

CL-USER 2 : 1 > (quit)
Segmentation fault


Help is much appreciated!

Regards,
Mike