[Mono-list] Performance comparisons

[Paolo Molaro]

On 01/16/03 Gopal V wrote:
> >                 blt.s      ?L2066
> 
> The CSCC compiler makes use of a few assembler optimisations like this 
> jump squashing ... and therefore generates the "small" jumps instead of
> full jumps which gives very significant reductions in file size ....

Very significant reduction is about 3% in corlib, for example, so
not very significant (though, eventually, it'll get implemented in mono
as well).

> Comparing interpreters of the projects might have shocking results ...
> So, I will not comment on that ... Rhys's papers on the Pnet engine shall
> cover the portability vs speed issues ... 

A portable interpreter is nice, but it will always be slow and
retro-fitting partial native code generation in it may not be the better
choice for maintainance. The x86 specific code in pnet is 3300+ lines
and in the new mono jitter it's 3700 lines, so at the end of the day,
it's better to have 500 lines of code more to maintain and be from 2 to
10+ times faster in the general case:-)

lupus

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