Attachment 876412 Details for Bug 1078354 – numpy_long_double_representation.patch

[patch] numpy_long_double_representation.patch

numpy_long_double_representation.patch (text/plain), 2.86 KB, created by Michel Normand on 2014-03-19 16:21:40 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Michel Normand

Created: 2014-03-19 16:21:40 UTC

Size: 2.86 KB

patch

obsolete

>From 958d2717dafff07c6155defcf559fc03ed073e8a Mon Sep 17 00:00:00 2001
>From: Michel Normand <normand@linux.vnet.ibm.com>
>Date: Wed, 19 Mar 2014 12:10:28 -0400
>Subject: [PATCH] numpy new patch to build for ppc64le archi
>
>check_long_double_ppc64le.patch patch not yet upstream
>
>Signed-off-by: Michel Normand <normand@linux.vnet.ibm.com>
>
>diff --git a/check_long_double_ppc64le.patch b/check_long_double_ppc64le.patch
>new file mode 100644
>index 0000000..6e8704b
>--- /dev/null
>+++ b/check_long_double_ppc64le.patch
>@@ -0,0 +1,33 @@
>+From c1973d167eb08a11260227374d3c7cbc8ccf0eb9 Mon Sep 17 00:00:00 2001
>+From: Michel Normand <normand@linux.vnet.ibm.com>
>+Date: Wed, 19 Mar 2014 16:52:56 +0100
>+Subject: [PATCH] long_double_representation required changes for ppc64le
>+ archi
>+
>+Changes required in this fonction for ppc64le to avoid error msg while build:
>+===
>+Traceback (most recent call last):
>+  File "setup.py", line 192, in <module>
>+    setup_package()
>+...
>+  File "numpy/core/setup_common.py", line 305, in long_double_representation
>+    raise ValueError("Unrecognized format (%s)" % saw)
>+ValueError: Unrecognized format (['001', '043', '105', '147', '211', '253', '315', '357', '000', '000', '000', '124', '064', '157', '235', '301', '000', '000', '000', '000', '000', '000', '000', '000', '376', '334', '272', '230', '166', '124', '062', '020'])
>+===
>+
>+Signed-off-by: Michel Normand <normand@linux.vnet.ibm.com>
>+
>+diff --git a/numpy/core/setup_common.py b/numpy/core/setup_common.py
>+index 2db4b43..3e92263 100644
>+--- a/numpy/core/setup_common.py
>++++ b/numpy/core/setup_common.py
>+@@ -295,6 +295,9 @@ def long_double_representation(lines):
>+                         return 'IEEE_QUAD_LE'
>+                     elif read[8:-8] == _DOUBLE_DOUBLE_BE:
>+                         return 'DOUBLE_DOUBLE_BE'
>++                    elif read[8:16] == _IEEE_DOUBLE_LE and \
>++                         read[16:24] == _BEFORE_SEQ[:8]:
>++                            return 'IEEE_DOUBLE_LE'
>+                 elif read[:16] == _BEFORE_SEQ:
>+                     if read[16:-8] == _IEEE_DOUBLE_LE:
>+                         return 'IEEE_DOUBLE_LE'
>diff --git a/numpy.spec b/numpy.spec
>index b83870c..5cbdf81 100644
>--- a/numpy.spec
>+++ b/numpy.spec
>@@ -25,6 +25,8 @@ Source0:        http://downloads.sourceforge.net/numpy/%{name}-%{version}%{?relc
> # - 524b9eaa33ec67e34eb31a208e02bb934f778096
> # - 0bb46c1448b0d3f5453d5182a17ea7ac5854ee15
> Patch0:         numpy-insecure-mktemp-use.patch
>+# 
>+Patch1:         check_long_double_ppc64le.patch
> 
> BuildRequires:  python2-devel lapack-devel python-setuptools gcc-gfortran atlas-devel python-nose
> Requires:       python-nose
>@@ -95,6 +97,7 @@ This package includes a version of f2py that works properly with NumPy.
> rm numpy/distutils/command/__init__.py && touch numpy/distutils/command/__init__.py
> 
> %patch0 -p1
>+%patch1 -p1
> 
> # Atlas 3.10 library names
> %if 0%{?fedora} >= 21

Actions: View | Diff

Attachments on bug 1078354: 876412 | 889192