See output of regrtest below. It was run on an Intel Mac OS X 10.4

A similar failure is observed on the py3k branch.

./python.exe Lib/test/regrtest.py -v test_math

test_math
testAcos (test.test_math.MathTests) ... ok
testAcosh (test.test_math.MathTests) ... ok
testAsin (test.test_math.MathTests) ... ok
testAsinh (test.test_math.MathTests) ... ok
testAtan (test.test_math.MathTests) ... ok
testAtan2 (test.test_math.MathTests) ... ok
testAtanh (test.test_math.MathTests) ... ok
testCeil (test.test_math.MathTests) ... ok
testConstants (test.test_math.MathTests) ... ok
testCopysign (test.test_math.MathTests) ... ok
testCos (test.test_math.MathTests) ... ok
testCosh (test.test_math.MathTests) ... ok
testDegrees (test.test_math.MathTests) ... ok
testExp (test.test_math.MathTests) ... ok
testFabs (test.test_math.MathTests) ... ok
testFactorial (test.test_math.MathTests) ... ok
testFloor (test.test_math.MathTests) ... ok
testFmod (test.test_math.MathTests) ... ok
testFrexp (test.test_math.MathTests) ... ok
testFsum (test.test_math.MathTests) ... ok
testHypot (test.test_math.MathTests) ... ok
testIsinf (test.test_math.MathTests) ... ok
testIsnan (test.test_math.MathTests) ... ok
testLdexp (test.test_math.MathTests) ... ok
testLog (test.test_math.MathTests) ... ok
testLog10 (test.test_math.MathTests) ... ok
testLog1p (test.test_math.MathTests) ... ok
testModf (test.test_math.MathTests) ... ok
testPow (test.test_math.MathTests) ... ok
testRadians (test.test_math.MathTests) ... ok
testSin (test.test_math.MathTests) ... ok
testSinh (test.test_math.MathTests) ... ok
testSqrt (test.test_math.MathTests) ... ok
testTan (test.test_math.MathTests) ... ok
testTanh (test.test_math.MathTests) ... ok
test_exceptions (test.test_math.MathTests) ... ok
test_mtestfile (test.test_math.MathTests) ... FAIL
test_testfile (test.test_math.MathTests) ... ok
test_trunc (test.test_math.MathTests) ... ok
Doctest: ieee754.txt ... ok

======================================================================
FAIL: test_mtestfile (test.test_math.MathTests)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/Users/salman/svn/python/trunk/Lib/test/test_math.py", line
1016, in test_mtestfile
    '\n  '.join(failures))
AssertionError: Failures in test_mtestfile:
  expm10302:expm1(709.7827128932839): expected 1.797693134682424e+308,
got 'OverflowError'

----------------------------------------------------------------------
Ran 40 tests in 2.273s

FAILED (failures=1)
test test_math failed -- Traceback (most recent call last):
  File "/Users/salman/svn/python/trunk/Lib/test/test_math.py", line
1016, in test_mtestfile
    '\n  '.join(failures))
AssertionError: Failures in test_mtestfile:
  expm10302:expm1(709.7827128932839): expected 1.797693134682424e+308,
got 'OverflowError'

1 test failed:
    test_math

Salman Haq also reported (on IRC) that the configure output shows:

checking for expm1... yes

so this appears to be a problem with the platform's implementation of 
expm1, rather than with Python's expm1 code (which is only used when the 
platform version isn't available).

This seems to be specific to OS X/Intel:  I'm not seeing any failure on my 
ancient iBook G4 (OS X 10.4.11).

Salman Haq, could you find out roughly where expm1 starts overflowing on 
your machine?  E.g., do all of the following overflow for you?

>>> from math import expm1
>>> expm1(709.78271289338397)
1.7976931348622732e+308
>>> expm1(709.782712893)
1.797693134172102e+308
>>> expm1(709.7827)
1.7976699566638014e+308
>>> expm1(709.7)
1.6549840276802644e+308

I'm just trying to determine whether the 10.4 expm1 is so broken that 
Python should work around it, or whether it's only slightly broken.  In 
the latter case we can just live with it, and either weaken the tests 
slightly or explicitly skip that particularly test on OS X 10.4.

Only the argument with one decimal place precision does NOT overflow.

Python 2.7a1+ (trunk:76872, Dec 21 2009, 09:54:29) 
[GCC 4.0.1 (Apple Computer, Inc. build 5370)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> from math import expm1
>>> expm1(709.78271289338397)     
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
OverflowError: math range error
>>> expm1(709.782712893)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
OverflowError: math range error
>>> expm1(709.7827)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
OverflowError: math range error
>>> expm1(709.782)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
OverflowError: math range error
>>> expm1(709.78)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
OverflowError: math range error
>>> expm1(709.7)
1.6549840276802644e+308

Thanks for that.  I'd call that pretty broken then:  the result of 
expm1(709.78) is effectively wrong by 24.4 trillion ulps (perhaps that 
should be 24.4 teraulps?).  I guess someone just coded a hard cutoff value 
somewhere between 709.7 and 709.78.

But since anyone getting this close to overflow is asking for trouble 
anyway, perhaps the easiest solution is just to change the test to test a 
value not quite as close to the overflow boundary.

Here's a patch (against trunk).

Fixed in r77116 (trunk), r77117 (py3k).