SMP Documentation

Arjun Suresh Raji arjun1296 at gmail.com
Tue Mar 31 07:46:34 PDT 2015


Thanks for your comments and inputs. I am reading the FreeBSD docs. and
planning to acquire a freebsd kernal design book. And I need a dev system
to be setup before I jump in start experimenting. As of know I am going
through the changes being made.
Happened to go through

http://bsd.slashdot.org/story/03/10/03/0158217/irc-forum-with-matthew-dillon-of-dragonfly-bsd
Wil be following up in the efnet network as the freenode one is quiet
inactive.

--
Arjun S R
http://about.me/arjunsr

On Wed, Mar 18, 2015 at 5:06 PM, Atte Peltomaki <atte.peltomaki at iki.fi>
wrote:

> On Mon, Jan 19, 2015 at 04:35:29PM +0530, Arjun Suresh Raji wrote:
> > Hi ,
> > I have just started using DragonFly BSD.
> > I am just curious if I could find some documentation on how Dfly differs
> > with FreeBSD on SMP initiatives. When going through some documents I read
> > words like big kernel lock and fine-grained locks. Apart from the buzz
> > words I was not able get any documentation on those.
> >
> > And how different kernels behave with different sets of workload?
> >
> > Can anyone point me to some documents on this.
> > Kindly help me.
>
> As Matt said, DragonFly BSD SMP model isn't comprehensively documented,
> but if you aren't familiar with the subject, I recommend checking out
> following books and texts:
>
> - UNIX Systems for Modern Architectures: Symmetric Multiprocessing and
>   Caching for Kernel Programmers. This book explains the fundamental
>   problems in SMP and explains different solutions.
>
> - Solaris Internals (2nd. edition). Explains the Solaris lightweight
>   kernel threading model, which shares some similarities with dfly.
>
> - FreeBSD SMPng Design Document:
>   https://www.freebsd.org/doc/en_US.ISO8859-1/books/arch-handbook/smp.html
>
> - The Design and Implementation of the FreeBSD Operating System. Also
>   covers FreeBSD SMP implementation and fine-grained locking.
>
> Note that the subject of SMP is very complex, it's not possible to
> accurately evaluate an implementation's performance without actually
> testing it.
>
> --
> Atte Peltomäki
>      atte.peltomaki at iki.fi <> http://kameli.org
> "Your effort to remain what you are is what limits you"
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.dragonflybsd.org/pipermail/kernel/attachments/20150331/25efa65c/attachment-0005.html>


More information about the Kernel mailing list