Subject: Re: the path from nathanw_sa -> newlock2
To: None <firstname.lastname@example.org>
From: Bucky Katz <email@example.com>
Date: 02/13/2007 23:38:48
Andrew Doran <firstname.lastname@example.org> writes:
> One of the major drivers behind the decision to switch threading models was
> this: to make the SA implementation work and be scaleable in an SMP setting
> would have required significant reworking of the kernel components. That
> would have meant, essentially, a rewrite of the core in kern_sa.c bringing
> its own new bugs and headaches.
I have no problem with this. I am sad that I didn't know about it six
months ago and spent time fixing M:N bugs.
Life goes on.