On Mon, Apr 14, 2008 at 6:04 PM, Rodrigo Kumpera <[EMAIL PROTECTED]> wrote: > Other thing, how does sharing and inlining play together? I guess inlined > generic code > uses rgctx a bit less as it has precise type information at hand. If so, > wouldn't > more aggressively inlining generic code reduce the need for rgctx?
Yes, but the difference is negligible. In FSharp the number of RGCTXs that need to be creates goes down to 647 (from 651). On the other hand, more RGCTXs (most of which are of course never needed) are allocated when code is inlined (i.e. more classes/vtables which require RGCTXs are initialized). I'm not quite sure why that is, though. It wouldn't be an issue when we do RGCTX allocation lazily anyway. I should also mention that we currently don't allow inlining in shared code, for the simple reason that it's not implemented yet. Mark _______________________________________________ Mono-devel-list mailing list [email protected] http://lists.ximian.com/mailman/listinfo/mono-devel-list
