Bug 299 - segmentation fault in libpthread.so

Bug#:

299 alias:

Hardware:

Reporter:

Andrija Pantovic <andrija.pantovic@ubs.com>

Host:

Target:

Build:

Product:

Add CC:

Component:

Version:

CC:

Remove selected CCs

Status:

RESOLVED

Priority:

Resolution:

INVALID

Severity:

Assigned To:

Richard Henderson <rth@gcc.gnu.org>

Target Milestone:

Flags:			Requestee:
	backport		()
	examined		()
	testsuite		()

Summary:

Keywords:

Attachment	Description	Type	Created	Actions
Create a New Attachment (proposed patch, testcase, etc.)			View All

Bug 299 depends on:			Show dependency tree Show dependency graph
Bug 299 blocks:			Show dependency tree Show dependency graph

Additional Comments:

Leave as RESOLVED INVALID
Reopen bug
Mark bug as VERIFIED

View Bug Activity | Format For Printing

Description:

Last confirmed: 0000-00-00 00:00

Opened: 2004-08-05 05:52

quite few applications are seghfaulting in libpthread 
mozilla and java in the very same function (mozilla version 1.4-1.7 
backtrace): 
Program received signal SIGSEGV, Segmentation fault. 
[Switching to Thread 2199039104016 (LWP 31886)] 
0x00000200001d7880 in __pthread_init_static_tls () 
   from /lib/tls/libpthread.so.0 
 
java backtrace: 
Program received signal SIGSEGV, Segmentation fault. 
[Switching to Thread 2199023375600 (LWP 29760)] 
0x0000020000049880 in __pthread_init_static_tls () 
   from /lib/tls/libpthread.so.0

------- Additional Comment #1 From Ulrich Drepper 2004-08-05 06:06 -------

What does "alpha-linux" mean?  Alpha processors?

------- Additional Comment #2 From Andrija Pantovic 2004-08-05 06:22 -------

Yes. 
OS: Linux 
Distribution: Fedora Core 2 
Platform: Alpha

------- Additional Comment #3 From Ulrich Drepper 2004-08-05 07:39 -------

The best you can hope for is that Richard shows interest.  Richard, feel free to
suspend the bug if you don't want to look at it.

------- Additional Comment #4 From Richard Henderson 2004-08-05 08:10 -------

Glibc from cvs generally works for me on my debian-hybrid systems.  There
are some thread cancellation backtrace problems still, but that's not the
problem you're reporting here.

You'll have to follow the rules for bug submission and give a standalone
test case that shows the problem.  And against glibc cvs, not something
modified for Fedora.

------- Additional Comment #5 From Andrija Pantovic 2004-08-06 05:34 -------

What CVS snapshot are you using on your Debian ? 
What patches, configuration options, what headers and what compiler did you 
use to build it ?

------- Additional Comment #6 From Richard Henderson 2004-08-06 05:53 -------

What's on my system is irrelevant to this PR.

------- Additional Comment #7 From Andrija Pantovic 2004-08-17 14:11 -------

Will report bug (actualy few of them) properly - against current CVS snapshot 
with proper gdb output ...

------- Additional Comment #8 From Nikolay Zhuravlev 2006-06-15 10:42 -------

It was not proven that the bug exists in glibc