[Numpy-discussion] test_multiarray.TestIO.test_ascii segmentation fault with Python2.7

Bruce Southey bsouthey@gmail....
Fri Dec 18 11:15:34 CST 2009


On 12/15/2009 10:59 AM, Charles R Harris wrote:
>
>
> On Tue, Dec 15, 2009 at 9:51 AM, Pauli Virtanen <pav+sp@iki.fi 
> <mailto:pav%2Bsp@iki.fi>> wrote:
>
>     Tue, 15 Dec 2009 10:36:03 -0600, Bruce Southey wrote:
>     [clip]
>     > Program received signal SIGSEGV, Segmentation fault. setup_context
>     > (registry=<value optimized out>, module=<value optimized out>,
>     > lineno=<value optimized out>, filename=<value optimized out>,
>     > stack_level=<value optimized out>)
>     >      at Python/_warnings.c:449
>     > 449         PyFrameObject *f = PyThreadState_GET()->frame; (gdb) bt
>     > #0  setup_context (registry=<value optimized out>, module=<value
>     > optimized out>, lineno=<value optimized out>, filename=<value
>     optimized
>     > out>, stack_level=<value optimized out>)
>     >      at Python/_warnings.c:449
>     > #1  do_warn (registry=<value optimized out>, module=<value optimized
>     > out>, lineno=<value optimized out>, filename=<value optimized out>,
>     > stack_level=<value optimized out>)
>     >      at Python/_warnings.c:593
>     > #2  0x0000000000493c81 in PyErr_WarnEx (category=0x760720,
>     text=<value
>     > optimized out>, stack_level=1) at Python/_warnings.c:719 #3
>     > 0x00000000004c8e94 in PyOS_ascii_strtod (nptr=0x7ffff7f08914 "1
>     , 2 , 3
>     > , 4", endptr=0x7fffffffdb28) at Python/pystrtod.c:282 #4
>     > 0x00007ffff2954151 in NumPyOS_ascii_strtod (s=0x7ffff7f08914 "1
>     , 2 , 3
>     > , 4", endptr=0x7fffffffdb28) at
>     numpy/core/src/multiarray/numpyos.c:527
>
>     Looks like it's trying to raise a deprecation warning after
>     PyArray_FromString has released GIL. So that was the reason why it
>     caused
>     a segfault also in 3.1.
>
>     PyOS_ascii_strtod was deprecated in 2.7 and in 3.1. Apparently, we now
>     *must* do something like
>
>     #if PY_VERSION_HEX >= 0x02060000
>        return PyOS_string_to_double(s, endptr, NULL);
>     #else
>        return PyOS_ascii_strtod(s, endptr);
>     #endif
>
>     everywhere the function is used.
>
>     It also seems that this needs to be backported to Numpy 1.4.x...
>
>     (Note to self: this is also the origin of the crashes in scipy/
>     lambertw... GIL must be reacquired before raising any warnings.)
>
>
> Would it be appropriate to put macros for all these in config.h or 
> some other common spot? Having all the python version dependencies in 
> one spot might make it easier to keep current. I've been thinking of 
> moving the numpy deprecation macro for that reason.
>
> Chuck
>
There are two places that appear to use PyOS_ascii_strtod in numpy:
1) core/SConscript -  'Define the function PyOS_ascii_strod if not 
available'
There is no definition for Python 3K and I do not know the purpose of 
this function.

2)  core/src/multiarray/numpyos.c - defines NumPyOS_ascii_strtod:
Here the usage is already within an '#if defined(NPY_PY3K)'  for Python 
3K. In the numpyos.c with numpy '1.5.0.dev8019' I replaced the NPY_PY3K 
with:
#if PY_VERSION_HEX >= 0x02070000

In Python 2.6 all the tests passed. But for Python 2.7 I got the 
following errors.

======================================================================
ERROR: test_buffer_hashlib (test_regression.TestRegression)
----------------------------------------------------------------------
Traceback (most recent call last):
   File 
"/usr/local/lib/python2.7/site-packages/numpy/core/tests/test_regression.py", 
line 1255, in test_buffer_hashlib
     assert_equal(md5(x).hexdigest(), '2a1dd1e1e59d0a384c26951e316cd7e6')
TypeError: object supporting the buffer API required

======================================================================
FAIL: Check formatting of complex types.
----------------------------------------------------------------------
Traceback (most recent call last):
   File "/usr/local/lib/python2.7/site-packages/nose/case.py", line 183, 
in runTest
     self.test(*self.arg)
   File 
"/usr/local/lib/python2.7/site-packages/numpy/core/tests/test_print.py", 
line 61, in check_complex_type
     err_msg='Failed str formatting for type %s' % tp)
   File "/usr/local/lib/python2.7/site-packages/numpy/testing/utils.py", 
line 305, in assert_equal
     raise AssertionError(msg)
AssertionError:
Items are not equal: Failed str formatting for type <type 'numpy.complex64'>
  ACTUAL: '-1j'
  DESIRED: '(-0-1j)'

======================================================================
FAIL: Check formatting of complex types.
----------------------------------------------------------------------
Traceback (most recent call last):
   File "/usr/local/lib/python2.7/site-packages/nose/case.py", line 183, 
in runTest
     self.test(*self.arg)
   File 
"/usr/local/lib/python2.7/site-packages/numpy/core/tests/test_print.py", 
line 61, in check_complex_type
     err_msg='Failed str formatting for type %s' % tp)
   File "/usr/local/lib/python2.7/site-packages/numpy/testing/utils.py", 
line 305, in assert_equal
     raise AssertionError(msg)
AssertionError:
Items are not equal: Failed str formatting for type <type 
'numpy.complex128'>
  ACTUAL: '-1j'
  DESIRED: '(-0-1j)'

======================================================================
FAIL: Check formatting of complex types.
----------------------------------------------------------------------
Traceback (most recent call last):
   File "/usr/local/lib/python2.7/site-packages/nose/case.py", line 183, 
in runTest
     self.test(*self.arg)
   File 
"/usr/local/lib/python2.7/site-packages/numpy/core/tests/test_print.py", 
line 61, in check_complex_type
     err_msg='Failed str formatting for type %s' % tp)
   File "/usr/local/lib/python2.7/site-packages/numpy/testing/utils.py", 
line 305, in assert_equal
     raise AssertionError(msg)
AssertionError:
Items are not equal: Failed str formatting for type <type 
'numpy.complex256'>
  ACTUAL: '-1j'
  DESIRED: '(-0-1j)'

----------------------------------------------------------------------
Ran 2494 tests in 6.932s

FAILED (KNOWNFAIL=5, errors=1, failures=3)
<nose.result.TextTestResult run=2494 errors=1 failures=3>

These last ones may be due to:

"Another format()-related change: the default precision used for 
floating-point and complex numbers was changed from 6 decimal places to 
12, which matches the precision used by str(). (Changed by Eric Smith; 
issue 5920.)"
http://docs.python.org/dev/whatsnew/2.7.html

Bruce


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.scipy.org/pipermail/numpy-discussion/attachments/20091218/d983644b/attachment.html 


More information about the NumPy-Discussion mailing list