[ANN]Draft rough plan for 3.6!
marcus at ira.uka.de
Tue Apr 15 08:50:21 UTC 2003
On Tue, Apr 15, 2003 at 09:36:35AM +0200, Daniel Vainsencher wrote:
> It seems we can get both 1 & 2 without almost any new work. Because they
> clean stuff up and simplify life on the image side, I think they're
> worthy even without giving us speed. Heck, I'd take a 5% overall hit
> without blinking, if it gets me rid of the Magical Compiled Methods.
> Andreas seems to have suggested it is "slow" (not clear whether he meant
> significantly slower than now, or just not faster). Does anyone have
> benchmarks of the ClosureCompiler work with patched VMs?
Yes, I just did "[(1 to: 1000000) do: [:i | i]] timeToRun" on a self-build
patched Vm, both closurecompiled and not:
With BlockClosures: 1440
No BlockClosures: 948
So it's 50% slower. It would be interesting how this translates to
real world stuff. I will do some macroBenchmarks for a completly
recompiled image later.
Marcus Denker marcus at ira.uka.de -- Squeak! http://squeak.de
More information about the Squeak-dev