[fftw/f16] libquadmath exists only on x86/x86_64 and ia64
Dan Horák
sharkcz at fedoraproject.org
Tue Oct 11 08:04:12 UTC 2011
commit 98500e043d6bb836f4cf272430bf3b85d9156f17
Author: Dan Horák <dan at danny.cz>
Date: Tue Oct 11 10:03:26 2011 +0200
libquadmath exists only on x86/x86_64 and ia64
fftw.spec | 11 ++++++++---
1 files changed, 8 insertions(+), 3 deletions(-)
---
diff --git a/fftw.spec b/fftw.spec
index 2cddb20..e41be07 100644
--- a/fftw.spec
+++ b/fftw.spec
@@ -1,6 +1,6 @@
Name: fftw
Version: 3.3
-Release: 3%{?dist}
+Release: 4%{?dist}
Summary: A Fast Fourier Transform library
Group: System Environment/Libraries
License: GPLv2+
@@ -17,11 +17,13 @@ BuildRequires: gcc-gfortran
%global openmp 1
%endif
+%global quad 0
# Quad precision support only available with gcc >= 4.6
%if 0%{?fedora} >= 15
+# and only on these arches
+%ifarch %{ix86} x86_64 ia64
%global quad 1
-%else
-%global quad 0
+%endif
%endif
# For check phase
@@ -252,6 +254,9 @@ fi
%{_libdir}/libfftw3*.a
%changelog
+* Tue Oct 11 2011 Dan Horák <dan[at]danny.cz> - 3.3-4
+- libquadmath exists only on x86/x86_64 and ia64
+
* Mon Oct 10 2011 Rex Dieter <rdieter at fedoraproject.org> 3.3-3
- -devel: Provides: fftw3-devel (#744758)
- -static: Provides: fftw3-static
More information about the scm-commits
mailing list