#4680 NORM Future : Sugar apps' pygtk main loop polls 10 times a second, always

[Zarro Boogs per Child]

#4680: Sugar apps' pygtk main loop polls 10 times a second, always
---------------------+------------------------------------------------------
  Reporter:  gnu     |       Owner:  marco             
      Type:  defect  |      Status:  new               
  Priority:  normal  |   Milestone:  Future Release    
 Component:  sugar   |     Version:  Build 623         
Resolution:          |    Keywords:  power, performance
  Verified:  0       |  
---------------------+------------------------------------------------------

Comment(by bemasc):

 The upstream thread is broken apart in the archives, starting at:
 http://mail.python.org/pipermail/python-dev/2006-September/068569.html
 http://mail.python.org/pipermail/python-dev/2006-September/068585.html
 http://mail.python.org/pipermail/python-dev/2006-September/068610.html
 http://mail.python.org/pipermail/python-dev/2006-September/068646.html
 http://mail.python.org/pipermail/python-dev/2006-September/068714.html

 There is a python bug at: http://bugs.python.org/issue1564547

 Having read through it, the bottom line is that Python upstream may make
 the changes that would allow PyGTK to remove this polling loop, but not
 before 2.6.  However, the polling loop serves only to catch UNIX signals,
 which are not normally used by PyGTK programs.  Except for catching these
 sorts of signals, it seems that multithreaded PyGTK programs will work
 perfectly without this polling loop.

 I am not at all knowledgeable about this topic, but it seems to me that we
 could probably patch PyGTK to remove this loop, or increase its timeout to
 60 seconds, and nobody would notice.  The only change would be that Ctrl-C
 and 'kill -SIGTERM' might not work to kill PyGTK programs. Once we have
 suspend-on-cpuidle, it would be a valuable power savings.

-- 
Ticket URL: <http://dev.laptop.org/ticket/4680#comment:8>
One Laptop Per Child <http://dev.laptop.org>
OLPC bug tracking system