[Info-vax] Compile error (Pthread on VMS)?

[Stephen Hoffman]

On 2013-02-20 11:38:55 +0000, Joukj said:

> ...In the end I solved the problem by sending signals from/to the 
> threads by means of the system services SYS$ENQ and SYS$DEQ. (a little 
> shooting at mouse using a cannon), but I could use the extra 
> possibilties these routines gave me.

Yeah; but from reports from under some of the other rocks I looked 
underneath, the pthread_kill call reportedly had some reliability 
issues on at least Tru64 Unix configurations, too.

Digressions....

Given its status, the sys$sigprc call might not work all that well, 
either.  (But this whole matter is fodder for a discussion with some of 
the folks in VMS Engineering, and the OP has better access than most.  
Whether that's for suggestions, or to get the POSIX threads 
documentation updated to show some examples of compilation and linking.)

Other alternatives to POSIX threads in general?  Look at KP Threads on 
VMS, if you're writing native code.  Blocks if you have access to 
those, though the VMS C and C++ compilers don't.  Or yes, the 
ill-documented POSIX threading on VMS.

One of the folks from HP Labs has a paper 
<http://www.hpl.hp.com/techreports/2004/HPL-2004-209.html> that might 
be worth a read, if you're interested in some of the "fun" here.  But 
if you're interested in the other sort of "fun" here, when you see a 
programming problem and think ..."I know, I'll use threads."  Now have 
problems two you.  :-)

-- 
Pure Personal Opinion | HoffmanLabs LLC