[Mono-devel-list] SSAPRE: this time it's for real ;-)

[Paolo Molaro]

On 11/26/04 Willibald Krenn wrote:
> The problem is: If I go ahead and do something in that area and propose 
> a patch on the list, it won't be included because the JIT team has come 
> up with a completely distinct solution to the problem. :-((

I don't remember any significant contribution that was turned down
because there was already someone working on similar code:
following the mono lists, the public blogs and the cvs/svn archive
is all that's needed to follow the development of mono. The only major
feature that was developed for a few months in private was the new jit, 
but people knew about it and it's not like we were offered many working 
jit implementations in the meantime;-)

> Everything happens behind closed curtains - you are the only one who 

Some code was posted back in August, hardly behind closed curtains:
http://lists.ximian.com/archives/public/mono-devel-list/2004-August/007566.html
Recompilation with full optimizations is effective when there is a 
sensible speed difference between default and fully-optimized code:
right now the difference is usually small, so it's not a priority
to implement it (though I want to get the statistical profiler in before
mono 1.2).
If you plan to work on a feature to be included in mono svn you should
explain your proposal on the list, so we can comment if there is some
obvious misdesign or some obvious improvement that can be made.
Of course you're also free to write the code first: a working 
implementation is sure going to get people more interested than
just talk;-)

lupus

-- 
-----------------------------------------------------------------
lupus at debian.org                                     debian/rules
lupus at ximian.com                             Monkeys do it better