Subject: Re: SA, libpthread and page faults
To: Stephan Uphoff <ups@stups.com>
From: Noriyuki Soda <soda@sra.co.jp>
List: tech-kern
Date: 08/18/2003 16:27:16
>>>>> On Thu, 14 Aug 2003 11:38:38 -0400, Stephan Uphoff <ups@stups.com> said:
>> how about a user-kernel shared variable used to indicate if
>> any upcall handler is executing or not?
>> (iirc, freebsd kse does a similar thing)
> This would work beautifully with the current implementation
> since it only supports a single virtual processor.
> Multiple virtual processors would cause trouble since user
> level threads have no way of knowing the virtual processor they
> are running on.
You are talking about SMP case here?
(Sorry, if I'm confusing what "virtual processors" mean here)
According to the original SA paper by Anderson, the userlevel thread
scheduler always knows precise information about how many and what
lwps are assigned to the process. So, in theory, isn't it possible
that the userlevel thread scheduler know the fact?
(I'm not reading current NetBSD implemetation closely yet, sorry.)
--
soda