Skip to content

Incorrect finfo from PPC longdouble (Trac #2077) #2669

Closed

Description

@numpy-gitbot

Original ticket http://projects.scipy.org/numpy/ticket/2077 on 2012-03-10 by @matthew-brett, assigned to unknown.

In [1]: import numpy as np

In [2]: print np.finfo(np.longdouble)
/Users/mb312/dev_trees/numpy/numpy/core/machar.py:125: RuntimeWarning: overflow encountered in add
  a = a + a
/Users/mb312/dev_trees/numpy/numpy/core/machar.py:127: RuntimeWarning: invalid value encountered in subtract
  temp1 = temp - a
/Users/mb312/dev_trees/numpy/numpy/core/machar.py:136: RuntimeWarning: invalid value encountered in subtract
  itemp = int_conv(temp-a)
/Users/mb312/dev_trees/numpy/numpy/core/machar.py:160: RuntimeWarning: overflow encountered in add
  a = a + a
/Users/mb312/dev_trees/numpy/numpy/core/machar.py:162: RuntimeWarning: invalid value encountered in subtract
  temp1 = temp - a
/Users/mb312/dev_trees/numpy/numpy/core/machar.py:169: RuntimeWarning: invalid value encountered in subtract
  if any(temp-a != zero):
Machine parameters for float128
---------------------------------------------------------------------
precision= 75   resolution= 1e-75
machep=    -4   eps=        1.3817869701e-76
negep =    -4   epsneg=     1.3817869701e-76
minexp=    -1   tiny=       -1.08420217274e-19
maxexp=     1   max=        -9.22337203471e+18
nexp  =     1   min=        -max
---------------------------------------------------------------------


In [3]: print np.finfo(np.longdouble).nmant
1

In [4]: np.__version__
Out[4]: '1.7.0.dev-aae5b0a'

I think the correct answers for nexp, nmant are 11, 106 respectively; see:
https://developer.apple.com/library/mac/#documentation/Darwin/Reference/Manpages/man3/float.3.html

(np-devel)[mb312@jerry ~]$ uname -a
Darwin jerry.bic.berkeley.edu 8.11.0 Darwin Kernel Version 8.11.0: Wed Oct 10 18:26:00 PDT 2007; root:xnu-792.24.17~1/RELEASE_PPC Power Macintosh powerpc
(np-devel)[mb312@jerry ~]$ gcc -v
Using built-in specs.
Target: powerpc-apple-darwin8
Configured with: /var/tmp/gcc/gcc-5370~2/src/configure --disable-checking -enable-werror --prefix=/usr --mandir=/share/man --enable-languages=c,objc,c++,obj-c++ --program-transform-name=/^[cg][^.-]*$/s/$/-4.0/ --with-gxx-include-dir=/include/c++/4.0.0 --with-slibdir=/usr/lib --build=powerpc-apple-darwin8 --host=powerpc-apple-darwin8 --target=powerpc-apple-darwin8
Thread model: posix
gcc version 4.0.1 (Apple Computer, Inc. build 5370)

Activity

  1. numpy-gitbot commented on Oct 19, 2012

    @numpy-gitbot
    Author

    @mwiebe wrote on 2012-03-12

    I seem to recall that the finfo code is using an algorithm that automatically detects these parameters by manipulating values of the given dtype. I guess either there's something wrong with the algorithm, or the long double on that platform is extremely non-conformant.

  2. numpy-gitbot commented on Oct 19, 2012

    @numpy-gitbot
    Author

    @charris wrote on 2012-03-12

    The algorithm (which is kind of a mess) assumes an exponent/mantissa format, while the PPC long double is implemented using two doubles. I'm not even sure if eps really makes sense for it. I'd almost be inclined to hard code these values except that they may depend on compiler flags.

  3. charris commented on Feb 21, 2014

    @charris
    Member

    #3125 has some useful information.

  4. matthew-brett commented on Feb 25, 2014

    @matthew-brett
    Contributor

    Just started running tests on buildbot - failing on these double precision tests:

    http://nipy.bic.berkeley.edu/builders/numpy-py2.6-osx-10.5-ppc/builds/0/steps/shell_3/logs/stdio

    I wonder if something like a test for common precisions and backing off to hard coding would work. I'll have a think.

  5. charris commented on Feb 25, 2014

    @charris
    Member

    I'd like that, actually. The current code for determining float precision is a mess. Basically there are two types we need to deal with: IEEE and IBM, hardcoding would be pretty easy to maintain, IMHO, work better, and be much easier to comprehend. There should be a check for IBM long doubles, like for the various IEEE types, and after that a way to expose that info. Or we could just combine endianess with the known specs for the IBM type.

  6. charris commented on Feb 3, 2017

    @charris
    Member

    Should be closed by #8504.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions