This makes it difficult to embed ruby in component framework =
applications (e.g. OpenOffice), since you can't be sure which thread =
instantiates your component.

Does this also give problems if ruby is instantiated in the main thread, =
but then called later on (rb_load_file()) from another thread?
Any way to improve this behaviour?

Bye, Oli=20

-----Original Message-----
From: Yukihiro Matsumoto [mailto:matz / ruby-lang.org]=20
Sent: Tuesday, March 04, 2008 2:17 AM
To: ruby-core / ruby-lang.org
Subject: Re: embedding Ruby 1.9.0 inside pthread

Hi,

In message "Re: embedding Ruby 1.9.0 inside pthread"
    on Tue, 4 Mar 2008 09:30:34 +0900, Nobuyoshi Nakada =
<nobu / ruby-lang.org> writes:

|At Tue, 4 Mar 2008 07:50:00 +0900,
|Suraj Kurapati wrote in [ruby-core:15756]:
|> I'm having trouble embedding Ruby inside a pthread because calling
|> ruby_sysinit() is segfaulting.  I created a small example to=20
|> reproduce the problem and illustrate my humble goal.
|
|You can't call ruby_init() in child threads.

Because GC needs to know system stack address, that is only taken from =
main thread.

							matz.