[RFC] C++ vs forced unwinding

[Richard Henderson]

On Thu, Apr 10, 2003 at 06:11:12PM -0700, Mark Mitchell wrote:
> Does POSIX say anything about thread cleanup functions that don't
> return?  (That's essentially what will happen if the "catch (...)"
> clause decides not to rethrow.)

Unfortunately, POSIX only defines C bindings, so this isn't
covered in any standard whatsoever.

What we have is some book somewhere in which someone opined
that thread cancellation should interoperate with EH, which
generated a bunch of glibc bug reports that it didn't work.
Since we're in the middle of replacing libpthread, it seemed
a good time to implement this reasonable behaviour.

Of course, that means we have to decide the exact semantics
ourselves, and, even more unfortunately, we don't have much
experience with what works best in practice.


r~