[Python-checkins] cpython: Last attempt to get compiler recursion crasher to fail reliably across (original) (raw)

nick.coghlan python-checkins at python.org
Sat Mar 19 06:23:25 CET 2011


http://hg.python.org/cpython/rev/9c6dcb5d8f01 changeset: 68673:9c6dcb5d8f01 parent: 68670:cfcff2683c71 user: Nick Coghlan <ncoghlan at gmail.com> date: Sat Mar 19 15:22:59 2011 +1000 summary: Last attempt to get compiler recursion crasher to fail reliably across platforms before giving up and skipping it as unreliably platform dependent

files: Lib/test/crashers/compiler_recursion.py

diff --git a/Lib/test/crashers/compiler_recursion.py b/Lib/test/crashers/compiler_recursion.py --- a/Lib/test/crashers/compiler_recursion.py +++ b/Lib/test/crashers/compiler_recursion.py @@ -9,5 +9,5 @@

e.g. '1*'*10**5+'1' will die in compiler_visit_expr

The exact limit to destroy the stack will vary by platform

-# but 1M should do the trick most places -compile('()'*10**6, '?', 'exec') +# but 10M should do the trick even with huge stack allocations +compile('()'*10**7, '?', 'exec')

-- Repository URL: http://hg.python.org/cpython



More information about the Python-checkins mailing list