Date: Mon, 27 Mar 2000 13:54:15 -0800 From: Alfred Perlstein <bright@wintelcom.net> To: Matthew Dillon <dillon@apollo.backplane.com> Cc: Daniel Eischen <eischen@vigrid.com>, Nate Williams <nate@yogotech.com>, nms@otdel-1.org, freebsd-current@FreeBSD.ORG Subject: Re: Is there spinlocks/semaphores available for drivers? Message-ID: <20000327135414.M21029@fw.wintelcom.net> In-Reply-To: <200003272008.MAA42799@apollo.backplane.com>; from dillon@apollo.backplane.com on Mon, Mar 27, 2000 at 12:08:33PM -0800 References: <Pine.SUN.3.91.1000327142313.16202A@pcnet1.pcnet.com> <200003272008.MAA42799@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
* Matthew Dillon <dillon@apollo.backplane.com> [000327 12:36] wrote:
>
> :
> :There's a paper that describes how Solaris transitioned from spl()s
> :to mutexes. ISTR they created one mutex for each splxxx. I'll have
> :to find this and re-read it.
> :
> :--
> :Dan Eischen
>
> I think we're using a slightly different mechanism... our spl*()'s
> are actually interrupt bit masks. That is, any single spl*() call
> may mask several interrupt sources. Turning a mask with five or six
> bits set in it into a set of mutexes is a very expensive proposition.
I think you're thinking this:
/-----int 1
spl -<---> int 2
\-----int 3
spl messing with several mutexes, instead consider:
int 1 >---\
int 2 >---->-- splmutex
int 3 >---/
Where a single mutex is shared by several interrupts.
There's also this to consider:
proc0: splaaa(); splbbb();
proc1: splbbb(); splaaa();
deadlock. Which needs to be worked out somehow.
--
-Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org]
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000327135414.M21029>