Re: 2.6.9-rc1-mm1

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Thursday 26 of August 2004 16:45, Felipe Alfaro Solana wrote:
> On Thursday 26 August 2004 16:36, Rafael J. Wysocki wrote:
> > I think the problem is that relatively not so many people run -mm, and
> > even less people try to use them for a longer time.  Also, there
> > sometimes are some issues with -mm that must be sorted out first, but
> > then there's not much time left for testing the scheduler before the next
> > -mm.
>
> I think this is the main reason of existence for -mm kernels: find
> problems, sort them out and fix them.

That's the point.  You don't pay attention to the differences between 
schedulers if there are more serious problems, do you?

> I've been running -mm kernels since
> 2.5.80+ and all problems I have had were resolved in a timely manner.

I agree, but it's a different thing. :-)

> What I think is that Con's scheduler is the one that needs to get into -mm
> kernels to give it more exposure. Currently, it has a very limited
> audience.

IMHO, -mm could stick for a while with one of the alternative schedulers so 
that it gets more testing.

Regards,
RJW

-- 
For a successful technology, reality must take precedence over public 
relations, for nature cannot be fooled.
					-- Richard P. Feynman