CVS commit: pkgsrc/math/octave

["Maya Rashish" <maya%netbsd.org@localhost>]

Module Name:    pkgsrc
Committed By:   maya
Date:           Sat Nov 19 00:47:08 UTC 2016

Modified Files:
        pkgsrc/math/octave: distinfo
        pkgsrc/math/octave/patches: patch-liboctave_operators_mx-inlines.cc

Log Message:
octave: reference llvm & octave bug reports for ambiguous float
precision patch.


To generate a diff of this commit:
cvs rdiff -u -r1.48 -r1.49 pkgsrc/math/octave/distinfo
cvs rdiff -u -r1.1 -r1.2 \
    pkgsrc/math/octave/patches/patch-liboctave_operators_mx-inlines.cc

Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.

Modified files:

Index: pkgsrc/math/octave/distinfo
diff -u pkgsrc/math/octave/distinfo:1.48 pkgsrc/math/octave/distinfo:1.49
--- pkgsrc/math/octave/distinfo:1.48    Wed Sep 28 02:54:15 2016
+++ pkgsrc/math/octave/distinfo Sat Nov 19 00:47:08 2016
@@ -1,4 +1,4 @@
-$NetBSD: distinfo,v 1.48 2016/09/28 02:54:15 maya Exp $
+$NetBSD: distinfo,v 1.49 2016/11/19 00:47:08 maya Exp $
 
 SHA1 (octave-4.0.3.tar.xz) = c798346a8271e0141d9dbe5610584dabb8311277
 RMD160 (octave-4.0.3.tar.xz) = 32bad5cdb6b832d651183b92f59b396940d6ec04
@@ -28,7 +28,7 @@ SHA1 (patch-libinterp_dldfcn_config-modu
 SHA1 (patch-libinterp_dldfcn_module.mk) = 7627fa6132036f26b77743d30dbbe99ed6442b42
 SHA1 (patch-libinterp_oct-conf.in.h) = 9a6b0e8d38bc2f9da916cb3d8145307e7a415f4f
 SHA1 (patch-liboctave_numeric_oct-fftw.h) = b152566ad3fcf6c5c1808f4428e6d51966ca1daf
-SHA1 (patch-liboctave_operators_mx-inlines.cc) = 3f01077700a3b8f51a3a171fdc11bafd784e3ea3
+SHA1 (patch-liboctave_operators_mx-inlines.cc) = 3002122937942a4a2bbcc6fff90398d082971d53
 SHA1 (patch-liboctave_system_oct-time.cc) = 7b8a33911c69861572e35320e13c6c777048bc6c
 SHA1 (patch-m4_acinclude.m4) = 194b8c3cf6f5e45a0251b229af9b3a47c12aa608
 SHA1 (patch-scripts_pkg_private_configure__make.m) = 7d4b620f889faa66c4c9c581ef8a7e7692b68c94

Index: pkgsrc/math/octave/patches/patch-liboctave_operators_mx-inlines.cc
diff -u pkgsrc/math/octave/patches/patch-liboctave_operators_mx-inlines.cc:1.1 pkgsrc/math/octave/patches/patch-liboctave_operators_mx-inlines.cc:1.2
--- pkgsrc/math/octave/patches/patch-liboctave_operators_mx-inlines.cc:1.1      Tue Feb 16 04:21:40 2016
+++ pkgsrc/math/octave/patches/patch-liboctave_operators_mx-inlines.cc  Sat Nov 19 00:47:08 2016
@@ -1,10 +1,13 @@
-$NetBSD: patch-liboctave_operators_mx-inlines.cc,v 1.1 2016/02/16 04:21:40 dbj Exp $
+$NetBSD: patch-liboctave_operators_mx-inlines.cc,v 1.2 2016/11/19 00:47:08 maya Exp $
 
 std::pow is required to promote numeric arguments to the smallest
 common float type. octave_int<> provides another test of template
 specializations for pow() as well as explicit cast operators.
 The combination of both leads to ambiguous resolutions.
 
+https://savannah.gnu.org/bugs/?43298
+https://llvm.org/bugs/show_bug.cgi?id=21083
+
 --- liboctave/operators/mx-inlines.cc.orig     2015-05-23 14:21:53.000000000 +0000
 +++ liboctave/operators/mx-inlines.cc
 @@ -307,7 +307,6 @@ inline void F (size_t n, R *r, X x, cons