From: matz@... (Yukihiro Matsumoto)
Date: 2003-12-24T05:17:38+09:00
Subject: Re: --enable-pthread still segfaults...

Hi,

In message "--enable-pthread still segfaults..."
    on 03/12/24, Nathaniel Talbott <nathaniel@talbott.ws> writes:

|I've finally been able to test my application under load using the 
|latest Ruby (1.8.1p4) with --enable-pthread (for those just tuning in, 
|see [ruby-core:1932] and the following thread). Thankfully, it doesn't 
|hang any longer! Unfortunately, it does still segfault. I haven't yet 
|had a chance to reduce the error, but with the nearing release of Ruby 
|1.8.1, I wanted to let you know that there is a segfault as quickly as 
|possible. Currently all I have is this backtrace:
|
|   #0  0x4064a0ca in ?? ()
|   #1  0x400e81e1 in siglongjmp () from /lib/libc.so.6
|   #2  0x400e814c in siglongjmp () from /lib/libc.so.6
|   #3  0x40022426 in longjmp () from /lib/libpthread.so.0

With --enable-pthread, longjmp() should not be called.  Show us your
config.h.  I'd like to know if USE_CONTEXT is defined (thus
rb_jump_context() is used) in eval.c.

							matz.