Tim Peters added the comment:

OK, it sounds to me like you do not have a reproducible test case, of any kind. 
 It that's true, this bug report isn't going anywhere :-(

Python isn't a memory-testing program, so it would be insanely inefficient for 
it to (for example) read up every byte it writes, just to make sure the memory 
isn't bad.  If you want to alter the source of _your_ Python to do that, fine!  
It won't go into the official distribution, but it might help you.

For the same reason, core Python won't go into the business of sorting out 
various kinds of memory addresses.  User-level code (including Python) works 
with the virtual (logical) addresses the operating system gives it.  For a hint 
of some of the difficulties in going deeper than that, here's a note from Linus 
Torvalds:

    http://www.tldp.org/LDP/khg/HyperNews/get/devices/addrxlate.html

About "If you would have the time to stitch down some test for me to execute 
with the garbage collector calls, it would be probably the best."  I'm not sure 
what you're suggesting there.  That we write a Python program that tries to 
trigger errors in your BIOS?  LOL - not likely ;-)

If you want to force a lot of allocations, deallocations, and gc runs, 
something like this will do it:

"""
def stress():
    import gc

    class C(object):
        def __init__(self):
            self.loop = self

    N = 500000
    d = dict((C(), C()) for i in xrange(N))
    j = 0
    while 1:
        for i in xrange(N // 2):
            d.popitem()
        for i in xrange(N // 2):
            d[C()] = C()
        j += 1
        if j % 10 == 0:
            print j
            gc.set_debug(gc.DEBUG_STATS)
            gc.collect()
            gc.set_debug(0)

stress()
"""

Setting N larger will make it consume more RAM.  The output is essentially 
meaningless, produced just so you know it's still running.

----------

_______________________________________
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue18843>
_______________________________________
_______________________________________________
Python-bugs-list mailing list
Unsubscribe: 
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

Reply via email to