httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Stein <gst...@lyra.org>
Subject Re: 2.0 performance
Date Thu, 23 Mar 2000 22:08:49 GMT
Ditto on what Dean said.

Typecasts are usually a symptom of a different problem. The other problem
should be fixed, thus removing the need for a typecast.

Per Mike's later note, the patches are based on the NPL. In that case, I'm
definitely -1. The MPL 1.1 is not bad, but has issues; the NPL has even
more :-(

I would love to hear how the MPM achieves its performance. Even if you
can't contribute source code, it would be great to contribute the ideas
(which are non-copyrightable; I'm assuming there isn't a patent in
there... :-)

Cheers,
-g

On Thu, 23 Mar 2000, Dean Gaudet wrote:
> On Thu, 23 Mar 2000, Mike Abbott wrote:
> 
> > 1.  Would you accept similar patches, with the typecasts, for 2.0 now?
> 
> -1.  i'd rather see API changes than the introduction of typecasts -- and
> 2.0 is the place to do that.  but it should be soon, before we get to the
> point of beta and freezing (dunno when that'd be).
> 
> > 2.  Would you accept similar patches, without the typecasts, for 2.0 now?
> 
> +1
> 
> > 3.  Would you accept a new, extremely high performance MPM (and maybe
> >     IOL) based on source code controlled by the Mozilla license?
> 
> as long as it's the version of the license that doesn't have patent
> encumbrance we could consider it.  but it'd be much easier if it were a
> bsd-style license.

-- 
Greg Stein, http://www.lyra.org/


Mime
View raw message