Opus support was enabled in 2ddfd51810. This commit adds the Opus
library and configures FFmpeg to be compiled with Opus support.
NOTE: It may be required to run `cmake -U '*FFMPEG_LIBRARIES*' .` in
your Blender build directory in order to refresh the `FFMPEG_LIBRARIES`
setting and add libopus.
Problem was twofold
1) `GENERATOR_IS_MULTI_CONFIG` is a property not a variable so
the test for it would always be false, unless you set a custom
CMAKE_INSTALL_PREFIX (like the buildbot does) the unit tests
would have a wrong working directory and complain about missing
dlls or blender executable
2) Tests added outside of `/test` (like libmv) would have no working
folder set since the variable would not be visible for them.
consulted @sergey who voiced the opinion that duplicating the code
to the test macro was slightly less evil than moving it to the main
CMakeLists.txt
Use cmake TIMESTAMP for BUILD_DATE+TIME
this simplifies code a lot
and even makes it more portable to other platforms
TIMESTAMP is available since cmake-2.8.11 ; blender already requires
cmake>=3.5 so that is fine.
Note that with CMake>=3.8, if defined, the SOURCE_DATE_EPOCH envvar
will be used by CMake here.
Reviewers: mont29, campbellbarton
Reviewed By: mont29, campbellbarton
Differential Revision: https://developer.blender.org/D5760
This change switches the debug symbol format from /Zi to /ZI for
debug builds of blender, allowing Edit and Continue to work.
This allows limited [1] code changes in the debugger without
having to stop the process and recompile a new binary leading
to improved developer productivity.
All MSVC versions we support support this flag, Clang on
windows does not mind the /ZI flag, but doesn't currently
emit the required information to have this feature work.
[1] https://docs.microsoft.com/en-us/visualstudio/debugger/supported-code-changes-cpp
The installer always upgraded the last version installed and did not allow for two versions to be installed side by side.
The reworked installer will allow side by side installs
install order:
```
2.81 -> 2.81a -> 2.82 : Allowed , will result in both 2.82 and 2.81a being installed
2.82 -> 2.81 -> 2.81a : Allowed , will result in both 2.82 and 2.81a being installed
2.82 -> 2.81a : Allowed , will result in both 2.82 and 2.81a being installed
2.82 -> 2.81a -> 2.81 : Not Allowed, 2.81 will only install if you manually remove 2.81a first.
```
Do note though that this will not apply to any previously issued installers and even for 2.80a this is not something we can fix.
This patch is for landing in 2.81 *only* and should be excluded from any possible 2.80a release.
Second change is a change to the compression level, building the MSI takes 30 minutes, which is crazy, perhaps worth it if the compression actually pays of.
```
MSI - none 1:35 247.0 MB (260,025,634 bytes)
MSI - mszip 2:02 89.6 MB ( 94,022,946 bytes)
MSI - low 2:35 81.6 MB ( 85,646,626 bytes)
MSI - medium 4:11 77.3 MB ( 81,136,930 bytes)
MSI - high 28:01 74.7 MB ( 78,384,418 bytes)
zip 1:32 93.2 MB ( 97,732,293 bytes)
7Z 2:22 65.0 MB ( 68,171,614 bytes)
```
It didn't, so I lowered it to medium, seemed reasonable.
Differential Revision: https://developer.blender.org/D5494
Reviewers: brecht, jesterking
bf_intern_openvdb makes a significant number of template instantiations
causing it go over the maximum number of sections (int16) in a coff file
when doing a debug build.
This change switches the compiler to use the extended coff format which
has this field extended (int32) all linkers post msvc2005 can process
this format so there's no reason not to turn this on globally.
Clang on windows does not need this change since clang switches implicitly
to the extended format when required. [1]
[1] https://reviews.llvm.org/rL217812
This change removes 32 bit support from the helper make.bat scripts
as we are dropping official 32 bit support, you can still build for
32 bit by configuring your build yourself using cmake and pointing
the LIBDIR cmake variable to your own 32 bit library folder.
This has faster builds, error checks and tests. The number of cmake options
for this type of thing has grown over the years and it's convenient to be
able to point new developers to a single target.
Previously the combination of all these options did not work correctly, now
all tests should pass.
The easiest way to use this is with the make wrapper, for example:
make full developer debug
Or set it manually with CMake:
cmake -C ../blender/build_files/cmake/config/blender_developer.cmake .
Differential Revision: https://developer.blender.org/D5149
gflags emits a few unused variable warnings since the main
CMakeLists.txt raised the warning from w4 down to w3. This
restores it back to w4 in the remove_strict_flags macro.
Precompiled headers were sharing the PCH file between debug and
release builds which is 'bad'. Adding the configuration to the
path fixes the issue.
Reported on chat by @mano-wii
Ninja was unable to see the dependency between the cpp
that generated the pch and the compile units that used
it. Explicitly managing this now makes precompiled headers
work with both msvc and clang, with both msbuild and ninja
based generators.
Ninja has issues detecting the implicit dependency on the
precompiled header output for freestyle. Disabled ninja
support for now until a proper solution can be found.
This brings down the build time for freestyle with MSVC from a
minute to 10-20 seconds.
vs2019 bf_freestyle debug before: 60464 ms after: 11028 ms
vs2019 bf_freestyle release before: 56984 ms after: 20526 ms
Differential Revision: https://developer.blender.org/D2606
Reviewed By: brecht , sergey
On GPUs that support it, we now present OpenGL contents via CAMetalLayer. This
fixes frame skipping issues found in T60043. If the system does not have a Metal
capable GPU, NSOpenGLView will continue to be used.
Patch by Tomoaki Kawada, with some changes by Brecht Van Lommel.
Differential Revision: https://developer.blender.org/D4619
This is in preparation of an the upcoming fix where we need to use a Metal
layer to avoid performance issue when drawing with OpenGL. Note that we already
only officially support 10.12+, the difference with this change is that Blender
will not start at all on 10.9 and 10.10.
This marks the headers in the LIBDIR as system headers
and changes the warn to /W0 on msvc versions that support it.
This resolves some warnings we would had to completely
repress otherwise.
Many modern computers support a lot of threads (parrallel building
jobs), but are somewhat restricted in memory, when some building jobs
can require several GB each.
Ninja builder has pools, which extend the usual `-j X` make
parallelizing option, by allowing to specify different numbers of
parallel jobs for different targets.
This commit defines three pools, one for linking, one for usual compile,
and one for compiling some 'heavy' cpp libs, when a single file can
require GB of RAM in full debug builds.
Simply enabling WITH_NINJA_POOL_JOBS will try to set default sensible
values for those three pools based on your machine specifications, you
can then tweak further the values of NINJA_MAX_NUM_PARALLEL_ settings,
if you like.
On my system (8 cores, 16GB RAM), it allows to build a full debug with
all ASAN options build with roughly 7GB of RAM used at most, pretty much
as quickly as without that option (which would require up to 11GB of
available RAM at some points).
Review task: D4780.
