Re: Cancelling threads and C++ (LinuxThreads)

[Richard Henderson <rth@twiddle.net>]

On Mon, Jun 28, 1999 at 10:03:20AM -0400, george@moberg.com wrote:
> This work-around works.  However, I've corresponded with Dave Butenhof, who
> wrote "Programming with POSIX Threads".  His opinion was that this solution
> isn't portable, and that the library implementing pthreads should see to it
> that destructors for stack-based objects are called without my program having
> to do anything special.

As quality-of-implementation, that's true.  But pedanticly, the standard
(being for C) doesn't say anything about C++, and so relying on the
implementation to call destructors is also not portable.

>    Is a patch to LinuxThreads to implement the throw/catch mechanism
>    for cancelling threads appropriate?  This would include the -fexceptions
>    compile option for the entire C library.

I would not be opposed to it.  Drepper would make the final decision.

>    Would this require that a single mechanism for exceptions be used,
>    rather than the two that are available in egcs now?  (-fnew-exceptions
>    vs. -fsjlj-exceptions)

Actually, the two mechanisms are -fsjlj-exceptions and -fno-sjlj-exceptions.
-fnew-exceptions does something different, and is forward compatible with
-fno-sjlj-exceptions.

But yes, you would be limited to one of -f{no-}sjlj-exceptions.


r~