Re: 2.4.10pre10aa1 (first spin to the vm rewrite included)

Re: 2.4.10pre10aa1 (first spin to the vm rewrite included)

[Olaf Zaplinski]

Andrea Arcangeli wrote:
> 
> Some of the the main features of 2.4.10pre10aa1 are:
> [...]

I would like to patch 2.4.10pre10 and try it, if I only could find its
.tar.bz2... where can I get it? I looked for it in
http://www.kernel.org/pub/linux/kernel/v2.4/test-kernels/ and
http://www.kernel.org/pub/linux/kernel/testing/

Olaf

Re: 2.4.10pre10aa1 (first spin to the vm rewrite included)

[Mads Martin Joergensen]

* Olaf Zaplinski <o.zaplinski@mediascape.de> [Sep 18. 2001 17:00]:
> Andrea Arcangeli wrote:
> > 
> > Some of the the main features of 2.4.10pre10aa1 are:
> > [...]
> 
> I would like to patch 2.4.10pre10 and try it, if I only could find its
> .tar.bz2... where can I get it? I looked for it in
> http://www.kernel.org/pub/linux/kernel/v2.4/test-kernels/ and
> http://www.kernel.org/pub/linux/kernel/testing/

You've to grab linux-2.4.9.tar.bz2 and apply the patch in the
linux/kernel/testing directory.

-- 
Mads Martin Joergensen, http://mmj.dk
"Why make things difficult, when it is possible to make them cryptic
 and totally illogic, with just a little bit more effort."
                                -- A. P. J.

Re: 2.4.10pre10aa1 (first spin to the vm rewrite included)

[Dipankar Sarma]

In article <20010917180914.I713@athlon.random> andrea@suse.de wrote:
> Some of the the main features of 2.4.10pre10aa1 are:

> o	rcu core included (thought it will be rewritten, I agreed with
> 	Dipankar that the per-cpu scheduling sequence number seems
> 	the best approch, such number can serve also as a statistic
> 	to usespace infact, it is very similar to Rusty's patch but it
> 	doesn't add any branch to the schedule fast path, he's rewriting
> 	the patch at the moment and I'll include it in the next release)

Sorry, in the latest patch I folded the per-cpu context switch counter into 
per-cpu counters for other quiescent states - user mode code and idle loop
in order to save on an extra compare making a single counter :-)
So, it is not much of a user space statistics anymore.

I can put that counter back, but I think eventually it may anyway
make sense to start using per-cpu statistics counters. As a matter
of fact we are working on a framework to support these. If we do
use per-cpu statistics counters later, RCU can make use of the
per-cpu context switch statistics counter.

Thanks
Dipankar
-- 
Dipankar Sarma  <dipankar@in.ibm.com> Project: http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.