From: KOSAKI Motohiro <kosaki.motohiro@...>
Date: 2012-02-03T08:39:58+09:00
Subject: [ruby-core:42335] Re: [RFC/PATCH] reinit signal_thread_list_lock atfork

2012/2/2 KOSAKI Motohiro <kosaki.motohiro@gmail.com>:
> 2012/2/2 Eric Wong <normalperson@yhbt.net>:
>> KOSAKI Motohiro <kosaki.motohiro@gmail.com> wrote:
>>> > signal_thread_list_lock may be held in the timer thread during
>>> > fork(),
>>>
>>> I don't think it's possible. �see,
>>
>> OK, then is there any chance this could be th->interrupt_lock can be
>> held by the main thread?
>
> It's possible. OK, we might need to reinitialize it.

Err. no. fork caller must be not taken interrupt_lock and other
thread's interrupt_lock was destroyed with thread itself. So, current
code is correct.