Re: GIL in alternative implementations

Path	csiph.com!x330-a1.tempe.blueboxinc.net!usenet.pasdenom.info!news.albasani.net!news2.arglkargh.de!news.musoftware.de!wum.musoftware.de!fu-berlin.de!uni-berlin.de!not-for-mail
From	Marc Christiansen <usenet@solar-empire.de>
Newsgroups	comp.lang.python
Subject	Re: GIL in alternative implementations
Date	Sat, 28 May 2011 11:55:34 +0200
Lines	62
Sender	<tolot@jupiter.solar-empire.de>
Message-ID	<mdt6b8-css.ln1@pluto.solar-empire.de> (permalink)
References	<4DE015EA.4040600@gmail.com> <mailman.2185.1306548406.9059.python-list@python.org>
X-Trace	news.uni-berlin.de UeL651HE/WkCuBhvtMuT/QalOCmZ6a1JJkVI46YLpFf90G15s=
X-Orig-Path	not-for-mail
User-Agent	tin/1.9.6-20101126 ("Burnside") (UNIX) (Linux/2.6.37-gentoo-r4 (x86_64))
Xref	x330-a1.tempe.blueboxinc.net comp.lang.python:6461

Show key headers only | View raw

Daniel Kluev <dan.kluev@gmail.com> wrote:
> test.py:
> 
> from threading import Thread
> class X(object):
>    pass
> obj = X()
> obj.x = 0
> 
> def f(*args):
>   for i in range(10000):
>       obj.x += 1
> 
> threads = []
> for i in range(100):
>    t = Thread(target=f)
>    threads.append(t)
>    t.start()
> 
> for t in threads:
>    while t.isAlive():
>        t.join(1)
> 
> print(obj.x)
> 
>> python test.py
> 1000000
>> pypy test.py
> 1000000
>> jython-2.5 test.py
> 19217
>> ipy test.py
> 59040
> 
> Not that this thing is reasonable to do in real code, but cpython and
> other implementations with GIL at least give you some safety margin.
> 
Sure? ;)
 > for p in python2.4 python2.5 python2.6 python2.7 python3.1 python3.2; do echo $p; $p /tmp/test.py; $p /tmp/test.py; done
 python2.4
 525369
 736202
 python2.5
 449496
 551023
 python2.6
 903405
 937335
 python2.7
 885834
 910144
 python3.1
 866557
 766842
 python3.2
 1000000
 1000000

So even CPython (at least < 3.2) isn't safe. And I wouldn't rely on 3.2
not to break.

Marc

Thread

Re: GIL in alternative implementations Daniel Kluev <dan.kluev@gmail.com> - 2011-05-28 13:06 +1100
  Re: GIL in alternative implementations Marc Christiansen <usenet@solar-empire.de> - 2011-05-28 11:55 +0200
    Re: GIL in alternative implementations Wolfgang Rohdewald <wolfgang@rohdewald.de> - 2011-05-28 13:16 +0200
  Re: GIL in alternative implementations John Nagle <nagle@animats.com> - 2011-05-28 09:39 -0700
    Re: GIL in alternative implementations Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2011-05-28 17:05 +0000
      Re: GIL in alternative implementations "Gabriel Genellina" <gagsl-py2@yahoo.com.ar> - 2011-06-07 01:03 -0300
        Re: GIL in alternative implementations Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2011-06-07 07:09 +0000
        Re: GIL in alternative implementations Jean-Paul Calderone <calderone.jeanpaul@gmail.com> - 2011-06-07 05:07 -0700

csiph-web