Re: SCO: "thread creation is about a thousand times faster than

From: yodaiken@fsmlabs.com
Date: Mon Aug 28 2000 - 08:32:16 EST


On Sun, Aug 27, 2000 at 11:26:53PM -0700, Marty Fouts wrote:
>
> So when a 'process' has tons'o'threads from the point-of-view of a
> programmer, even though it is really tons'o'clone()s from the point of view
> of an OS implementation, what *does* 'send the signal to the process' mean?

In my humble opinion, it's correct here to take advantage of the POSIX
distinction between thread identifier and pid. That is, a signal sent to
the pid of the root process is a signal sent to the threaded process, but
kill(pidx,n) where pidx is the processid of a thread should be
equivalent to pthread_kill(pidx,n). That is, if you use the pid of a
thread, you are specifying the thread, not the thread-group (process).

The introduction of the variable resources forks was key to coming up
with any solution.

-- 
---------------------------------------------------------
Victor Yodaiken 
Finite State Machine Labs: The RTLinux Company.
 www.fsmlabs.com  www.rtlinux.com

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Thu Aug 31 2000 - 21:00:21 EST