Re: [DotGNU]CVM to machine code

[Paolo Molaro]

On 07/27/03 Gopal V wrote:
> If memory serves me right, Paolo Molaro wrote:
> > I should also note that a current ilrun doesn't execute the code as mono
> > and the MS runtime do (they print some TRACE messages that ilrun does
> > not), so I guess ilrun is missing something yet (though I don't think it
> > would change the run time much).
> 
> pnetlib is missing decent switched TraceListener support :)

Ok, maybe you want to revisit the performance numbers when ilrun
executes the code as expected (like mono and the MS engine do).

> > Total jit time for pnetmark is 0.26 seconds, for a test that lasts about
> > 50 seconds: it's irrelevant.
> 
> -precompile ? :)

No, if you run with --profile, it will tell you how much time is spent
jitting.

> > Mono is about 3 times faster than the fastest ilrun build, the others
> > are 5 or more times slower.
> 
> Compare apples to oranges , for taste :)

Well, I didn't start comparing interpreters and jitters, you did:-)
They are just not in the same league.
I just provided the hard numbers that people can reproduce (though, as
other posts have shown, it's also easy to get deceiving numbers out of
benchmarks by not enabling optimizations or not using an optimized
build, or many other ways).

lupus

-- 
-----------------------------------------------------------------
address@hidden                                     debian/rules
address@hidden                             Monkeys do it better