[Stackless] Stackless status

Just van Rossum just at letterror.com
Sun Oct 8 18:17:33 CEST 2000


At 4:53 PM +0300 08-10-2000, Christian Tismer wrote:
>What memory-leak problem are you referring to? Stackless
>Python itself has no leak, when just used to replace
>the Standard distribution. It is said that the continuation
>module shows a leak under special circumstances with
>exception handlers, but I couldn't reproduce them yet.
>
>If anybody can produce such a leak under Win32, please
>let me know and send me the code.

Dunno about Win32, but the following script leaks on my machine (most of
the time, but not always!):


import continuation

class LeakMonitor:
	instances = 0
	def __init__(self):
		LeakMonitor.instances = LeakMonitor.instances + 1
	def __del__(self):
		LeakMonitor.instances = LeakMonitor.instances - 1


def leaker():
	continuation.return_current()
	lm = LeakMonitor()
	for i in range(10000):
		try:
			1/0
		except:
			pass
	continuation.uthread_reset()
	print "done!"


co = leaker()

# minimal uthread scheduler
while co:
	co = continuation.timeslice(co)

print "Number of instances left (should be zero):", LeakMonitor.instances


(The stackless version I have is somewhat newer than 1.1)

Just


_______________________________________________
Stackless mailing list
Stackless at starship.python.net
http://starship.python.net/mailman/listinfo/stackless



More information about the Stackless mailing list