I didn’t see this when I was doing my last round of testing. Maybe there’s some incompatibility with gnuplot-5, since that’s the only plotting-related item that changed significantly since I last updated Octave. I haven’t seen any runtime effect, though.
> On Feb 2, 2015, at 8:24 AM, Jack Howarth <howarth.at.f...@gmail.com> wrote: > > Alexander, > I am seeing the same failure on 10.10 with the stock octave364 > packaging... > > scripts/plot/allchild.m ................................panic: > Segmentation fault: 11 -- stopping myself... > make[1]: *** [check] Segmentation fault: 11 > make: *** [check] Error 2 > ### execution of /tmp/fink.bq1M1 failed, exit code 2 > ### execution of /tmp/fink.xMqKn failed, exit code 2 > > Jack > > On Mon, Feb 2, 2015 at 10:38 AM, Jack Howarth <howarth.at.f...@gmail.com> > wrote: >> Alexander, >> Oddly, I am seeing the following failure in 10.7 with this patch… >> >> scripts/plot/__gnuplot_drawnow__.m ..................... PASS 1/1 >> scripts/plot/__plt_get_axis_arg__.m .................... PASS 1/1 >> panic: Segmentation fault: 11 -- stopping myself... >> Makefile:2127: recipe for target 'check' failed >> make[1]: *** [check] Segmentation fault: 11 >> make[1]: Leaving directory >> '/sw/src/fink.build/octave-3.6.4-14/octave-3.6.4/test' >> Makefile:2549: recipe for target 'check' failed >> >> I'll jump back to 10.10 to verify that the current package passes >> 'fink -m' there. >> Jack >> >> On Mon, Feb 2, 2015 at 10:02 AM, Jack Howarth <howarth.at.f...@gmail.com> >> wrote: >>> Alexander, >>> Why not just restore the existing X11_INCFLAGS to functionality with… >>> >>> --- octave-3.6.4.info.orig 2015-02-01 16:11:35.000000000 -0500 >>> +++ octave-3.6.4.info 2015-02-02 09:58:28.000000000 -0500 >>> @@ -2,7 +2,7 @@ >>> Package: octave%type_pkg[-blas]%type_pkg[-x11] >>> Type: -blas (. -atlas -ref), oct (3.6.4), gcc (4.9), -x11 (boolean), >>> lapack (3.5.0) >>> Version: 3.6.4 >>> -Revision: 13 >>> +Revision: 14 >>> Maintainer: Alexander Hansen <alexkhan...@users.sourceforge.net> >>> >>> BuildDepends: << >>> @@ -181,6 +181,10 @@ >>> >>> # Patch configure not to link like Puma on Yosemite >>> perl -pi -e 's/(10\.\[012\])\*/\1\,.\*/' configure >>> + >>> + # Fix X11 include >>> + perl -pi -e >>> 's,X11_INCFLAGS=\"\$x_includes\",X11_INCFLAGS=\"-I\$x_includes\",' >>> configure >>> + perl -pi -e 's,\$XTRA_CXXFLAGS \$FT2_CFLAGS,\$XTRA_CXXFLAGS >>> \$FT2_CFLAGS \$X11_INCFLAGS,' configure >>> << >>> >>> GCC: 4.0 >>> @@ -229,7 +233,8 @@ >>> >>> osversion=`uname -r | cut -d. -f1` >>> export CFLAGS='-O3 -MD' >>> - export CXXFLAGS="-O3 -MD -I/usr/X11/include" >>> + #export CXXFLAGS="-O3 -MD -I/usr/X11/include" >>> + export CXXFLAGS="-O3 -MD" >>> if [ $osversion -ge 13 ] ; then >>> export CXXFLAGS="$CXXFLAGS -std=c++0x" >>> fi >>> >>> This produces the desired… >>> >>> /bin/sh ../libtool --tag=CXX --mode=compile oct-cxx -DHAVE_CONFIG_H >>> -I. -I.. -I../libgnu -I../libgnu -I../libcruft/misc >>> -I/sw/include -I/sw/include -O3 -MD -DHAVE_CONFIG_H -I/sw/include >>> -I/sw/include/freetype2 -I/usr/X11/include -I/sw/include -Wall -W >>> -Wshadow -Wold-style-cast -Wformat -Wpointer-arith -Wwrite-strings >>> -Wcast-align -Wcast-qual -O3 -MD -D_THREAD_SAFE -pthread -O3 -MD -MT >>> liboctave_la-dir-ops.lo -MD -MP -MF .deps/liboctave_la-dir-ops.Tpo -c >>> -o liboctave_la-dir-ops.lo `test -f 'dir-ops.cc' || echo >>> './'`dir-ops.cc >>> mv -f .deps/liboctave_la-CollocWt.Tpo .deps/liboctave_la-CollocWt.Plo >>> >>> Jack >>> >>> On Sun, Feb 1, 2015 at 11:37 PM, Alexander Hansen >>> <alexanderk.han...@gmail.com> wrote: >>>> >>>>> On Feb 1, 2015, at 2:29 PM, Jack Howarth <howarth.at.f...@gmail.com> >>>>> wrote: >>>>> >>>>> Alexander, >>>>> It looks like the octave packages in fink may have to return to >>>>> using flag-sort. The build of octave-3.6.4 on 10.7 is broken with the >>>>> failure… >>>>> >>>>> /usr/X11/include/ft2build.h:56:10: fatal error: >>>>> 'freetype/config/ftheader.h' file not found >>>>> >>>>> This is because the CONFIGURE_CXXFLAG picks up -I/usr/X11/include and >>>>> are placed before the XTRA_CXXFLAG containing the >>>>> -I/sw/include/freetype2 when assigning AM_CXXFLAGS throughout the >>>>> Makefile.in's in octave. >>>>> Jack >>>> >>>> >>>> It picks it up because I EXPLICITLY put that in in the CompileScript >>>> because without it the 10.7 build failed because it couldn’t find >>>> X11/Xlib.h. >>>> >>>> I don’t reproduce this on 10.10. I’m not inclined to revert the packaging >>>> on other platforms for this corner case of the next OS slated for >>>> end-of-life. >>>> -- >>>> Alexander Hansen, Ph.D. >>>> Fink User Liaison >>>> -- Alexander Hansen, Ph.D. Fink User Liaison ------------------------------------------------------------------------------ Dive into the World of Parallel Programming. The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net/ _______________________________________________ Fink-devel mailing list Fink-devel@lists.sourceforge.net List archive: http://news.gmane.org/gmane.os.apple.fink.devel Subscription management: https://lists.sourceforge.net/lists/listinfo/fink-devel
