This fixes the build on FreeBSD 10.x and 11-CURRENT on i386.
While here, completely remove the OPENMP option. The implementation was
incomplete anyway because disabling it would not turn off OpenMP in
darktable if it was built with GCC.
Reviewed by: kwm
Approved by: kwm
Differential Revision: https://reviews.freebsd.org/D4130
Use Clang 3.7 from Ports to enable OpenMP. This brings a major
performance boost for people using FreeBSD 10.x and -CURRENT.
`patch-src_CMakeLists.txt` is a new patch to disable `-ffast-math` when
Clang is used. Without this, isnan() always returns false, even if
`-fno-finite-math-only` is set. According to `clang --help`, this flag
provides no optimization anyway so we don't loose anything. darktable
2.0, which is almost ready, needs this patch too otherwise it crashes
during startup because of the isnan() misbehavior. The patch is not
committed upstream yet because I would prefer to understand what's
wrong.
Reviewed by: kwm
Approved by: kwm
Differential Revision: https://reviews.freebsd.org/D3922
Replace gtk20 dependancy with gdk-pixbuf2 and pango, and drop the
gtk-engines2 run dependacy.
Update ports to take the dependacy changes into account.
PR: 202378
Exp-run by: antoine@
instructions) use. Particularly, saying "SSE extensions" is superfluous,
as SSE abbreviation itself means "Streaming SIMD Extensions". Correct
way to refer to particular (CPU-agnostic) technology is "SIMD instruction
set", which is not limited to MMX, SSE family, AVX, NEON, etc.
Since "instruction set" looks a bit too formal in the option description,
use more casual and shorter word "instructions".
The patch "patch-tools_generate__prefs.xsl" was committed upstream and
is no longer needed.
Differential Revision: https://reviews.freebsd.org/D2385
Reviewed by: kwm
Approved by: kwm
The patch was committed upstream. The commit message has a detailed
explanation:
9d77a28e54
Differential Revision: https://reviews.freebsd.org/D2308
Reviewed by: kwm
Approved by: kwm
The release notes of this new stable branch are available on the
darktable website:
http://www.darktable.org/2014/12/released-darktable-1-6/
The following changes was made to the options:
o LIBSECRET is a new option, turned off by default. When selected,
darktable can use libsecret as a password storage backend. This
can be useful for GNOME 3 users and it may be turned on by
default in the future as the dependency is light.
o SLIDESHOW is turned on by default: the features received
enhancements in 1.6.0.
o SQUISH is turned on by default. This library is used to compress
thumbnail and it's recommanded by the darktable developers.
PR: 195820
Differential Revision: https://reviews.freebsd.org/D1311
Submitted by: martin@sugioarto.com (earlier version)
Reviewed by: antoine
Approved by: antoine
the 32 ports that still use it. Bump PORTREVISION on their dependent
ports except the ones that depend on these:
audio/libogg
audio/libvorbis
devel/pcre
ftp/curl
graphics/jpeg
graphics/libart_lgpl
graphics/tiff
textproc/expat2
textproc/libxslt
In these cases the same trick as in the recent gettext update is used.
The ports install a symlink with the old library version. When enough
of their dependent ports have had regular updates the remaining ones can
get a PORTREVISION bump and the links can be removed.
Also remove the devel/pcre dependency from USE_GNOME=glib20. It causes
over 2200 packages to depend on devel/pcre while less than 200 actually
link with it. The glib20 package still depends on devel/pcre so this
should not make a difference for ports with USE_GNOME=glib20. Also,
libdata/pkgconfig/glib-2.0.pc lists pcre as a private library so
USE_GNOME=glib20 should not propagate it.
PR: 195724
Exp-run by: antoine
Approved by: portmgr (antoine)
GCC 4.2 in FreeBSD 8.X/9.X base is now too old to compile OpenEXR, so
GCC-based systems will upgrade to the default ports compiler (GCC 4.7
currently.)
Add two patches to OpenEXR to permit building it in a live system with
the older OpenEXR version installed. Bug report filed to upstream Github
at https://github.com/openexr/openexr/issues/130
Couple OpenEXR more tightly to ilmbase and require its exact .so
version.
Add UPDATING note, and bump PORTREVISION of all dependent ports.
Proto-STAGE hugin-devel, and mark it IGNORE because hugin is newer.
Approved by: portmgr (implicit for bumping PORTREVISION on unstaged ports)
Use variables prefixed by option names. This removes the need to use .if
blocks.
Use OPTIONS_SUB=yes and %%OPTION%% markers in pkg-plist. This fixes
pkg-plist for at least the LUA option [1].
Remove the post-patch target. It's not needed anymore, as the changes
were committed upstream.
Replace the usage of bsd.port.pre.mk/bsd.port.post.mk by
bsd.port.options.mk/bsd.port.mk.
Fix the build when WITH_DEBUG=1 is set. darktable's CMakeLists.txt
throws a fatal error if CMAKE_BUILD_TYPE is set to Debug and suggests to
use RelWithDebInfo: we now follow this advice.
Sort USES and options-prefixed variables alphabetically.
PR: 190516 [1]
Phabricator: D178
Submitted by: Matthieu Volat <mazhe@alkumuna.eu> (previous patch to fix [1])
Reviewed by: bapt@
Approved by: bapt@
The following changes were made to options:
o Remove GCC option (see below)
o Remove KWALLET option (unused)
o Rename FACEBOOK option to FB_PICASA, because both export features
depend on json-glib.
o Add COLORD option (color management)
o Add GRAPHMAGICK option (support GraphicsMagick's image formats)
o Add LUA option (embed Lua to add scripting)
o Add OPENEXR option (support HDR image format)
o Add SQUISH option (use libsquish to compress thumbnails)
o Add WEBP option (support WebP image format)
The patch enables more options by default, because I think it's more
sensible for a photographer using binary packages:
o COLORD
o FB_PICASA
o FLICKR
o GEO
o GPHOTO (already on)
o LUA
o NLS
o OPENEXR
o OPENJPEG
o RAWSPEED (already on)
o WEBP
The GCC option was removed because the usage of GCC along with libc++
in FreeBSD 10 and -CURRENT is unstable (this isn't a general truth,
just valid in the case of darktable). In FreeBSD up-to 9, the usage of
lang/gcc is mandatory anyway, because GCC 4.2.1 support is best effort
only (upstream).
Ninja is now used to build darktable, instead of make. This is not mandatory
of course, but brings a 33% build time improvement (at least for me).
pkg-plist is generated with "make makeplist" this time. That's why the
diff seems large. In fact most of the content remains the same, just with
a different order.
All patches were committed upstream and are removed, except one which
comes from upstream but isn't part of release 1.4.2.
The port is now "stagified".
Finally, I take maintainership of it.
PR: ports/186979
Phabricator: D107
Reviewed by: danfe@ (previous version), antoine@
Approved by: antoine@
GCC 4.6.4 to GCC 4.7.3. This entails updating the lang/gcc port as
well as changing the default in Mk/bsd.default-versions.mk.
Part II, Bump PORTREVISIONs.
PR: 182136
Supported by: Christoph Moench-Tegeder <cmt@burggraben.net> (fixing many ports)
Tested by: bdrewery (two -exp runs)
Remove patch to add -I/usr/local/include in freetype-config --cflags. If
ports need extra headers they should look for them, and not get them via
a side-effect.
Freetype had a header resuffle in 2.5.1, patch ports to use the new header
style.
Thanks go to bdrewery for the two exp-runs and rakuco for helping me with
some troublesome cmake ports.
PR: ports/184587
- Update to 2.0.1
- Change master sites to SAVANNAH
- Change maintainer email to @FreeBSD.org
- Remove conflict with non existent Port
- USES pathfix pkgconfig
- Add executable
- Add DOCS Option
- Support STAGEDIR and add OPTIONS_SUB
- Use pathfix instead of simple patches
- Adjust patches
- Change WWW
graphics/OpenEXR
- Update to 2.0.1
- Change master sites to SAVANNAH
- Change maintainer email to @FreeBSD.org
- Use the new format for LIB_DEPENDS
- USES gmake pathfix pkgconfig
- Add DOCS and EXAMPLES Options
- Support STAGEDIR and add OPTIONS_SUB
- Change REINPLACE_CMD
- Add extra patch for EXAMPLES
- Remove obsolete patches
- Bump dependent ports' revisions
Approved by: pawel / wg (mentors)
- Switched to automake 1.11.6, see CVE-2012-3386.
- #14669: Fixed extraction of CC from gmp.h.
- Fixed case of intermediate zero real or imaginary part in mpc_fma,
found by hydra with GMP_CHECK_RANDOMIZE=1346362345.
This is on top of the following changes from version 1.0
- Licence change towards LGPLv3+ for the code and GFDLv1.3+ (with no
invariant sections) for the documentation.
- 100% of all lines are covered by tests
- Renamed functions
. mpc_mul_2exp to mpc_mul_2ui
. mpc_div_2exp to mpc_div_2ui
- 0^0, which returned (NaN,NaN) previously, now returns (1,+0).
- Removed compatibility with K&R compilers, which was untestable due
to lack of such compilers.
- New functions
. mpc_log10
. mpc_mul_2si, mpc_div_2si
- Speed-ups
. mpc_fma
- Bug fixes
. mpc_div and mpc_norm now return a value indicating the effective
rounding direction, as the other functions.
. mpc_mul, mpc_sqr and mpc_norm now return correct results even if
there are over- or underflows during the computation.
. mpc_asin, mpc_proj, mpc_sqr: Wrong result when input variable has
infinite part and equals output variable is corrected.
. mpc_fr_sub: Wrong return value for imaginary part is corrected.
Convert to the new LIB_DEPENDS standard and remove hard-coded
.so versions from a couple of dependent ports.
Bump PORTREVISIONS of all dependent ports.
PR: 183141
Approved by: portmgr (bdrewery)
Right now this is a noop in the former case and a noop in the latter
case unless lang/gcc44 has been installed explicitly.
This puts a bit more emphasis on standardizing on a canonical version
"current" GCC and makes it easier to update that canonical version
by changing the default in Mk/bsd.gcc.mk and updating the lang/gcc port.
That is, USE_GCC=yes means "use a decent/modern version of GCC" without
having to worry about details.
Approved by: portmgr (bdrewery)
brings support for several new cameras, and some minor bugfixes
- Remove no longer needed patching (const gint -> gint)
- Relax types of cache counters (to unbreak with GCC/i386)
white balance settings for several cameras, etc.
Note that some translations were removed: development team decided to exclude
those which are not properly finished from releases.
