[Stackless] Stackless python 2.6.6

Anselm Kruis a.kruis at science-computing.de
Tue Sep 28 10:47:19 CEST 2010

Hi Richard,

I just did a 'make test' and 'make teststackless' and got almost identical 
results with and without my patch applied.
Therefore I guess, that the patch isn't completely wrong. I'm testing on an old
SUSE LINUX 10.1 (X86-64) dual cpu system. gcc is 3.4.6.

In make teststackless one test fails with and without my patch:

FAIL: Test that the last runnable tasklet cannot be blocked on a channel.
Traceback (most recent call last):
   File "./test_channel.py", line 95, in testMainTaskletBlockingWithoutASender
     self.failUnless(stackless.getruncount() == 1, "Leakage from other tests, 
with tasklets still in the scheduler.")
AssertionError: Leakage from other tests, with tasklets still in the scheduler.

And in 'make test' a few tests are disabled and the following failures occur 
without my patch:

2 tests failed:
     test_sys test_urllib2_localnet

With my patch applied I get:

3 tests failed:
     test_lib2to3 test_sys test_urllib2_localnet

test_lib2to3 is caused by a missing file (Lib/lib2to3/fixes/fix_exitfunc.py) in 
my patch. A new version of the patch is available under


Of the make test failures only the test_sys failure seems to be stackless related.

test test_sys failed -- Traceback (most recent call last):
line 713, in test_objecttypes
     check(newstyleclass, s)
line 474, in check_sizeof
     self.assertEqual(result, size, msg)
AssertionError: wrong size for <type 'type'>: got 992, expected 988

As far as I looked into the changes between 265 and 266, I didn't see any real 
structural differences. Only massive white space changes.


Am 28.09.2010 01:45, schrieb Richard Tew:
> On Tue, Sep 28, 2010 at 12:04 AM, Anselm Kruis
> <a.kruis at science-computing.de>  wrote:
>> I'm currently working on a project, that saves and transfers the state of
>> executing programs using by pickling tasklets. Because we started a few
>> month ago, we currently using stackless python 2.6.5. Now that 2.6.6 is out,
>> I thought why not to try to build a stackless 2.6.6. It wasn't to difficult,
>> besides a lot of white space changes.
> Hi Anselm,
> We had a patch prepared when 2.6.6 was released, but the problem was
> that the Stackless unit tests or the Python standard library unit
> tests did not pass.  It seemed that there were structural changes in
> 2.6.6 which prevented them from doing so.  Did you run both these unit
> test suites?
> Cheers,
> Richard.
> _______________________________________________
> Stackless mailing list
> Stackless at stackless.com
> http://www.stackless.com/mailman/listinfo/stackless

  Dipl. Phys. Anselm Kruis                       science + computing ag
  Senior Solution Architect                      Ingolstädter Str. 22
  email A.Kruis at science-computing.de             80807 München, Germany
  phone +49 89 356386 874  fax 737               www.science-computing.de

Vorstand/Board of Management:
Dr. Bernd Finkbeiner, Dr. Roland Niemeier, 
Dr. Arno Steitz, Dr. Ingrid Zech
Vorsitzender des Aufsichtsrats/
Chairman of the Supervisory Board:
Michel Lepert
Sitz/Registered Office: Tuebingen
Registergericht/Registration Court: Stuttgart
Registernummer/Commercial Register No.: HRB 382196 

More information about the Stackless mailing list