[Info-vax] pthread_kill() replacement for VMS?

[Arne Vajhøj]

On 7/28/2017 12:14 PM, John Reagan wrote:
> On Friday, July 28, 2017 at 10:58:53 AM UTC-4, Stephen Hoffman
> wrote:
>> On 2017-07-28 00:18:07 +0000, David Froble said:
>>> Well, I guess that would depend on the multi-tasking being
>>> performed, and such.   Sure, an OS provides for multiple
>>> processes.  However, a single process may need to control and
>>> coordinate multiple "things".
>> 
>> Distributed job schedulers have made that effort easier, and with
>> less bespoke code necessary.   DLM and termination mailboxes and
>> TCP and the queue manager are all good tools and can be useful
>> here, but adapting those pieces and frameworks to scheduling is a
>> fair amount of bespoke code.  But processes and process scheduling
>> and distributed schedulers are different from threading, and
>> threading is something that really doesn't exist in BASIC short of
>> KP Threads or such.  I've yet to encounter threaded BASIC code.
>> Lots of AST-based code and multiple-process code written in BASIC
>> certainly, but not threading.
> 
> The BASIC RTL is not thread-safe.  There is something in the BASIC
> release notes about it.  I don't think the COBOL RTL is thread safe
> either.  I don't about Fortran.  I know the Pascal RTL IS thread safe
> since I did that work and I know a few Pascal customers that use
> threading with Pascal.

I have a theory that languages with a strong tradition for
recursive calls are easier to make thread safe.

Arne