[Numpy-discussion] RE: Python 2.2 seriously crippled for numerical computation?

[Paul Rubin]

huaiyu at gauss.almadan.ibm.com (Huaiyu Zhu) writes:
> 2. Is there a configure option to guarantee -lieee?
> 
>    Changing pyconfig.h.in to #define HAVE_LIBIEEE 1 does not make -lieee
>    appear in the Makefile, even though it get into pyconfig.h.

The configure script should probably be changed to test for the presence
of libieee and use it if available.

> 4. Is there an easier way to debug such problems?
> 
>    Maybe a C test program should be included in the configure script or in
>    the Tools directory so that people can easily report various conditions?
>    Such environmental testing is essential if Python is defining its own
>    numerical behavior indenpendent of IEEE or C. (but see below)

Does anybody care about running Python on non-ieee platforms these
days?  If not, it should try to use ieee arithmetic all the time.  The
exception might be some embedded platforms with software floating
point, but for those systems it should be acceptable to leave the
overflow/underflow behavior up to the floating point emulation system
if there even is one.

> 5. How is 1e200**2 handled?
> 
>    Since both 1e-200**2 and 1e200**2 produce the same errno all the time,
>    but Python still raises OverflowError for 1e200**2 when linked with
>    -lieee, there must be a separate mechanism at work.  What is that and how
>    can I override it?  I know this is by design, but I think the design is
>    dumb (to put it politely). 

What do you want it to do?