Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 2 Jul 2001 15:02:28 -0500
From:      Alfred Perlstein <bright@sneakerz.org>
To:        Julian Elischer <julian@elischer.org>
Cc:        "Michael C . Wu" <keichii@peorth.iteration.net>, "E.B. Dreger" <eddy+public+spam@noc.everquick.net>, smp@FreeBSD.ORG
Subject:   Re: per cpu runqueues, cpu affinity and cpu binding.
Message-ID:  <20010702150228.S84523@sneakerz.org>
In-Reply-To: <Pine.BSF.4.21.0107021420570.13213-100000@InterJet.elischer.org>; from julian@elischer.org on Mon, Jul 02, 2001 at 02:24:17PM -0700
References:  <20010702141113.Q84523@sneakerz.org> <Pine.BSF.4.21.0107021420570.13213-100000@InterJet.elischer.org>

next in thread | previous in thread | raw e-mail | index | archive | help
* Julian Elischer <julian@elischer.org> [010702 14:58] wrote:
> 
> 
> On Mon, 2 Jul 2001, Alfred Perlstein wrote:
> 
> > * Julian Elischer <julian@elischer.org> [010702 13:38] wrote:
> > > 
> > > At USENIX we decided to proceed with the KSE work.
> > 
> > yay!
> > 
> > > 
> > > As a side issue I plan on NOT ALLOWING multiple KSEs (thread carriers?)
> > > from the same thread group in the same process to be on the same
> > > processor. SO load balancing and processor affinity will not
> > > apply to the thread-carrying entities (KSEs). Of course the userland
> > > thread scheduler has the ultimate say as to which processor
> > > a thread is scheduled on.
> > 
> > Actually, this may cause some performance problems, when you have
> > a shared address space you can avoid tlb shootdowns when a process's
> > address space changes, you also share the cache, lastly there's
> > some rumor about a new CPU archetecture that runs multple threads
> > on the same CPU at the same time.  Just food for thought.
> 
> If you select to run 2 thread carriers (see other mail on nomenclature)> 
> (KSEs) then you have specifically asked for 2 processors worth of
> concurrency so we ASSUME you know what you are doing.. If you want to run 
> all the threads on a single processor to get better cache activity, then
> you should't ASK to run on 2  (or more) processors.

Agreed, however don't forget about the multiple thread execution
units that may become available, meaning that as long as you share
an address space you can run two (or more) threads in parrallel on
a single processor.  You wouldn't want to preclude us of taking
advantage of that if it becomes available.

-- 
-Alfred Perlstein [alfred@freebsd.org]
Ok, who wrote this damn function called '??'?
And why do my programs keep crashing in it?

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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010702150228.S84523>