RFC: Killing some warnings

[Andrew Cagney]

> Personally, I'd prefer to see these changed to avoid the casts to (void *)
> and (int).  See the implementation in lin-thread.c (and also elsewhere)
> for an example.

That is correct - you can't robustly cast (int) <-> (void*) on some
architectures.  Plenty of examples of how to handle this cleanly.  See
sol-thread.c:restore_inferior_thread() for instance.

> (If we ever get around to merging my pid/tid/lwp megapatch, these are
> cleaned up everywhere.  I.e, there is one implementation that all files
> which need to do this sort of thing use.)

Yes.  One way or the other I intend making it happen in the next few
months :-)

	Andrew