Python threading (was: Re: global interpreter lock not working as it should)

[Armin Steinhoff]

Paul Rubin <phr-n2002b at NOSPAMnightsong.com> wrote in message news:<7xu1mbj67c.fsf at ruckus.brouhaha.com>...
> a-steinhoff at web.de (Armin Steinhoff) writes:
> > > The interpreter in Python is effectively a critical region of
> > > code. Sections of the interpreter that might block are placed
> > > outside of the region so that other threads can enter it while
> > > that thread is blocked. Similarly, every thread is forced
> > > periodically to leave the region and re-enter it in order to allow
> > > the thread scheduler to re-schedule as necessary (the piece you
> > > posted).
> > 
> > IMHO ... there is nothing like a 'thread-scheduler' in the python
> > code.  All python threads are scheduled by the OS ... that means
> > there is no code in the python program which can _force_
> > periodically a thread to leave this critical section ( e.g. in the
> > middle of the execution of the 10 byte codes).
> 
> I think you're confused about the meaning of "critical section".

No .. I believe you didn't understand what I wrote!
 
> A critical section is a region of code where thread switching is
> impossible, because the code has a lock set,

No ... a thread which owns a critical section can always be preempted
by a thread with a higher priority :)

> interrupts are masked, or something like that.

Critical sections are most used to protect shared resources.

 
> When the program leaves the critical section by releasing the lock,
> thread switches become possible.

This is a wrong theory ...

>  Just because they're POSSIBLE
> doesn't mean they necessarily HAPPEN.  It just means the OS's
> scheduler can decide to switch threads there if it wants to.  It might
> not want to.

Interesting :) IMHO, scheduler rules are event (IRQ, timers) and
priority driven ... so there is no room for 'may or may not'
 
>  While in the critical section, the OS doesn't have the
> option of switching.  That's the difference.
> 
> In Python, the interpreter leaves the critical section (releases the
> GIL) every 10 byte codes.  That doesn't mean there has to be a thread
> switch then, but just that there MIGHT be one.  Do you understand now?

Try to understand what I wrote ... then please ask again.

Cheers 

Armin