http://qs1969.pair.com?node_id=285423

Foggy Bottoms has asked for the wisdom of the Perl Monks concerning the following question:

I previously wrote about a similar subject here and yet I still have a problem. I read the documentation given with threads, but surely I lack knowledge of some sort...

Could someone please tell methe difference between a thread and a process in Win32 and how that corresponds to forks in Linux ?

Secondly when I create a thread in Perl that corresponds to a never-ending sub (ie it's got a while (1) loop)m how can I stop it nevertheless ?

Thanks for your insight.

Replies are listed 'Best First'.
Re: returning from a thread ?
by liz (Monsignor) on Aug 21, 2003 at 11:33 UTC
    ...the difference between a thread and a process...

    If you're using Perl threads, you shouldn't have to worry about whether it is actually implemented using processes or real threads™. That's why they're "Perl threads".

    Having said that, to my knowledge, Linux is the only system where you can actually think of threads as processes, as having seperate program id's (pid's, $$). This allows modules as Thread::Signal to function in that environment. Win32 only knows about real threads and mimics fork() using threads. Other *nixes are somewhere inbetween. Some can fork() and have real threads (Solaris seems to fall in that group). But e.g. on Mac OS X (and presumably other BSD's), threads look as seperate processes, but can not be signalled.

    ...never-ending sub (ie it's got a while (1) loop)m how can I stop it nevertheless?

    Use a shared variable in the while. So: 

    while (1) {

    [download]
    becomes: 
    while ($sharedvar) {

    [download]
    and reset the shared variable in another thread when you want to have the thread in question stop.

    In that respect, you might also want to have a look at Thread::Queue::Monitored and/or Thread::Conveyor::Monitored.

    Hope this helps.

    Liz

    Update:
    I forgot one other way to do this. This will only work if you are running under Linux for now. Use Thread::Signal to activate a handler that will do a Thread::Exit. No nice cleanup that way, but definitely effective ;-)

[reply]
[d/l]
[select]
      I didn't manage to use the shared variable "thinggie" - as I don't have the shared package and since I didn't find it on the ActiveState repository...
      Instead I used a file in which I'm writing a byte (0 for inactive 1 for active)( I know you're thinking it's way too slow but it works fine anyhow...)

      However when the new thread starts up, it blocks the main program and I can't do anything else.
      I know someone already posted a node about this but I can't seem to find it anymore (bad luck :~/ )
      If you've any idea, let me know, and thanks again for your help...
[reply]
        Strange. Which version of Perl are you using on which platform? The main program should not block!

        Liz

[reply]
        Not slow, just ugly.. Very, very ugly..
[reply]
Re: returning from a thread ?
by EdwardG (Vicar) on Aug 21, 2003 at 14:30 UTC

    To answer the Win32 part of your first question:

    An application consists of one or more processes. A process, in the simplest terms, is an executing program. One or more threads run in the context of the process. A thread is the basic unit to which the operating system allocates processor time. A thread can execute any part of the process code, including parts currently being executed by another thread.

    Source: MSDN
[reply]

Back to Seekers of Perl Wisdom