On 04/03/06 Willibald Krenn wrote: > So this kind of functionality is still on your whishlist? As I was working > on exactly that kind of stuff, response/interest seemed *very* little.
We have always been interested in this feature. We'll have it eventually, but the implementations details are important: what is fine for a research/thesis project may not be right for a production VM. In particular I think the mono implementation must have a very low overhead (especially in terms of memory and system resources used): the best effects will be achieved if the optimization is turned on by default. > However, the initial experiments I conducted tend to support recompilation > at least for SciMark: From 178,29 +/- 0,12 [normal mode] to 204,00 +/- 1,57 > with recompilation enabled. (204,86 +/- 0,18 was maximum when always > applying all optimizations upfront.) Yep, thanks to your work we know it's worthwhile to implement it. lupus -- ----------------------------------------------------------------- [EMAIL PROTECTED] debian/rules [EMAIL PROTECTED] Monkeys do it better _______________________________________________ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
