[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
status of building octave on cygwin
From: |
Marco Atzeri |
Subject: |
status of building octave on cygwin |
Date: |
Fri, 12 Feb 2010 20:56:23 +0000 (GMT) |
Hi,
just in case someone else want to build the development
source on cygwin, or I forget again something ;-)
I thought to write a small status report:
the changeset f0ac2fa91733 (fix incorrect strncmpi change)
build fine on cygwin and pass almost all the tests.
The failed tests are on src/data.cc and are due to a
missing functionality on cygwin handling NaN on complex.
Currently I am setting with:
configure --enable-shared \
--enable-float-truncate \
CC=gcc-4 F77=gfortran-4 CXX=g++-4 CPP=cpp-4 \
lt_cv_deplibs_check_method=pass_all \
LDFLAGS=-no-undefined
"--enable-float-truncate" is needed for
http://thread.gmane.org/gmane.comp.gnu.octave.bugs/12361/focus=12404
without it one of the quadgk test will fail as
a=a could result false for truncation problem with
complex numbers.
"lt_cv_deplibs_check_method=pass_all" is needed to bypass
libtool wrong detection of system capabilities and
to allow shared libs building.
The segfault problem I reported on
https://www-old.cae.wisc.edu/pipermail/octave-maintainers/2009-December/014376.html
was due to a dtor issue on cygwin.
On cygwin snapshot 2010-02-04 the problem is solved.
(snapshot from 2010-02-05 to 2010-02-10 are broken and
not usable)
The current cygwin package fltk-1.1.8r5648-1 is not
usable for octave as it is a win32 instead of X11 version.
It builds but doesn't work.
Building with a X11 version like cygports 1.1.9-10,
is fine and the fltk seems functional.
I have not made extended test, but I have a
fltk-1.1.10 version if someone is interested.
The fltk cygwin package should be updated to a
X11 fltk-1.1.10, by the package maintainer, in
the near future.
Regards
Marco
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- status of building octave on cygwin,
Marco Atzeri <=