[pypy-svn] r75523 - pypy/trunk/pypy/translator/backendopt/test

[arigo at codespeak.net]

Author: arigo
Date: Wed Jun 23 14:17:11 2010
New Revision: 75523

Modified:
   pypy/trunk/pypy/translator/backendopt/test/test_mallocprediction.py
Log:
This is a completely random fix.  As the test starts failing from
r75472, which does an unrelated change to the implementation of
time.clock(), I can only imagine that it changes the expected result
here (pystone calls time.clock too).  I did not investigate why we get
16 instead of 10.



Modified: pypy/trunk/pypy/translator/backendopt/test/test_mallocprediction.py
==============================================================================
--- pypy/trunk/pypy/translator/backendopt/test/test_mallocprediction.py	(original)
+++ pypy/trunk/pypy/translator/backendopt/test/test_mallocprediction.py	Wed Jun 23 14:17:11 2010
@@ -169,7 +169,7 @@
     t, graph = rtype(entrypoint, [int])
     total0 = preparation(t, t.graphs, heuristic=heuristic)
     total = clever_inlining_and_malloc_removal(t)
-    assert total0 + total == 10
+    assert total0 + total == 16
 
 def test_richards():
     from pypy.translator.goal.richards import entry_point