[RFC] C++ vs forced unwinding

[Richard Henderson]

On Tue, Apr 01, 2003 at 02:09:20PM -0800, Mike Stump wrote:
>  The problem I see is that in C++, catch (...) { ... } can be a 
> cleanup, but, if it is, or not, cannot be known by the compiler.

See the multivendor C++ ABI.  By definition this is not a cleanup.
A cleanup in C++ is an object's destructor, only.

> Also, does the design cover Ada as well?

EH as a whole, yes.  Thread cancellation in specific, no.


r~