Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 12 Jul 2002 16:37:25 -0700
From:      Alfred Perlstein <bright@mu.org>
To:        Matthew Dillon <dillon@apollo.backplane.com>
Cc:        Alan Cox <alc@cs.rice.edu>, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG
Subject:   Re: cvs commit: src/sys/kern kern_synch.c src/sys/sys proc.h src/sys/vm vm_zeroidle.c src/sys/i386/i386 genassym.c pmap.c
Message-ID:  <20020712233725.GQ97638@elvis.mu.org>
In-Reply-To: <200207122221.g6CMLZwi063791@apollo.backplane.com>
References:  <200207122017.g6CKH7fx096339@freefall.freebsd.org> <20020712205516.GD5265@cs.rice.edu> <200207122221.g6CMLZwi063791@apollo.backplane.com>

next in thread | previous in thread | raw e-mail | index | archive | help
* Matthew Dillon <dillon@apollo.backplane.com> [020712 15:48] wrote:
> 
> :I think we should try to take this a step further.  Specifically, we
> :should recognize when processors are running in user mode in order
> :to avoid IPIs on kernel pmap changes.  On reentry to the kernel,
> :the processor would then do whatever is necessary to resync.  If
> :too many kernel pmap changes accumulate, then you force a single
> :IPI that performs the changes in bulk.
> :
> :As a start, it would be interesting to know how many IPIs could be
> :avoided by such a scheme.  
> :
> :Alan
> 
>     We need to run statistics to see where the IPIs are coming from.
>     A deferral mechanism is only going to be more efficient if its
>     simple and it might be better to instead just knock down the 
>     subsystems responsible for the majority of IPIs within the 
>     subsystem itself.

I'm quite sure Alan's method is described in Vahalia's(sp?) book as
an optimization the SVR4.2 (the SVR with mp) took that helped quite
a bit.

-- 
-Alfred Perlstein [alfred@freebsd.org]
'Instead of asking why a piece of software is using "1970s technology,"
 start asking why software is ignoring 30 years of accumulated wisdom.'
Tax deductible donations for FreeBSD: http://www.freebsdfoundation.org/

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020712233725.GQ97638>