On Tue, Jul 30, 2013 at 6:29 PM, Jonathan S. Shapiro <[email protected]>wrote:

> No no. The dynload gets a relocation at dynload time. The only way you can
> get an offset mismatch is if you have a version mismatch across the
> subsystems where some object structure changed in a significant way. That's
> a case where the dynload needs to fail anyway.
>

How are interfaces vtable's reached from instance pointers? Or are you not
admitting interfaces?

If two classes A and B implement interfaces X and Y, a static offset
computation could give us vtables for A=0, B=0, X=1, Y=1. However, if you
dynload a module with a class that implements X and Y, you have to redo all
offsets. Unless you are doing constant dynamic resolution of interfaces,
which sounds slow.
_______________________________________________
bitc-dev mailing list
[email protected]
http://www.coyotos.org/mailman/listinfo/bitc-dev

Reply via email to