Without the CONFIG option, find_package will perform Module search. On
at least Linux Mint 20 (I'm unable to reproduce this on CentOS and Arch
Linux), my guess is that this causes CMake to find "dirty" modules that
modify the configuration state despite the Boost version being too
low/absent.
Use CONFIG to put CMake into pure Config mode and avoid Module search.
Building SDL2 from externals is incompatible with Conan's version of
libiconv, a requirement of Conan's Boost package. Solution is to use the
same Boost package in use by the linux-fresh container. This tells CMake
to download boost_1_75_0.tar.xz from yuzu-emu/ext-linux-bin at CMake's
configuration step, much the same way Qt and FFmpeg are downloaded for
Windows.
Also makes DownloadExternals.cmake cross-platform. Although the CMake
code is not entirely specific to Linux, only Linux has Boost libraries
available at ext-linux-bin, whereas there is no equivalent Boost package
for Windows at ext-windows-bin. caveat emptor
If SDL2 is not found, the error is handled by falling back to externals.
No need spill the full warning at the find_package if it's going to be
handled later, so add QUIET to it.
Sets find_package(FFmpeg) to QUIET instead of REQUIRED. This allows
using the FFmpeg external in cases where there is no suitable installed
version of FFmpeg.
Also fixes a bug where multiple CMake configures causes FFmpeg_LIBRARIES
to concatenate on itself, producing cyclical dependencies. Unsets the
variable before building it in the foreach loop.
Fixes FFmpeg_INCLUDE_DIR not including the headers generated at run
time.
Forces using SDL 2.0.14. Upgrades the SDL external to that version. Adds
a message when switching to the external.
Fixes an error where input_common only links to SDL when SDL2_FOUND is
set, but externals/CMakeLists cannot set that variable to the required
scope. Switch to using ENABLE_SDL2, which we can use since we now
include the SDL source.
Since Bintray is (soon to be) no more, there needs to be a way to
acquire SDL2. Since 20.04's version is older than our minimum required
version (2.0.12), add it as an external.
- Bintray will be deprecated on May 1st 2021 (https://bintray.com/)
- We were previously using this for Qt (non-Windows) and SDL.
- I've moved to bundled SDL on Windows.
We had used conan for opus before, but there was a bug in the AVX detection.
However we still had the Findopus.cmake file within the repository, but not used.
This patch reenables the Findopus helper and prefer the system wide installation of opus.
Sets YUZU_USE_BUNDLED_FFMPEG as a CMake dependent option that is OFF on
Linux and ON for WIN32 targets. If FFmpeg is not found when
YUZU_USE_BUNDLED_FFMPEG is OFF, the bundled module/binaries are used
instead.
Reverts earlier changes to FindFFmpeg a bit, mostly to keep parity with
it's Citra version a bit. Now _FFmpeg_ALL_COMPONENTS lists all
components. We overwrite FFmpeg_LIBRARIES and FFmpeg_INCLUDE_DIR after
using the module.
Tells CMake to look for either nasm or yasm as it is required to build
FFmpeg. Avoids a compile-time error by checking for it during
configuration.
Adds a workaround for Ubuntu Bionic's old version of make not
communicating jobserver details properly.
Also renames related CMake variables to match both the Find*FFmpeg* and
variables defined within the file. Fixes odd errors produced by the old
FindFFmpeg.
Citra's FindFFmpeg is slightly modified here: adds Citra's copyright at
the beginning, renames FFmpeg_INCLUDES to FFmpeg_INCLUDE_DIR, disables a
few components in _FFmpeg_ALL_COMPONENTS, and adds the missing avutil
component to the comment above.
For Linux, instructs CMake to use the FFmpeg submodule in externals.
This is HEAVILY based on our usage of the late Unicorn. Minimal change
to MSVC as it uses the yuzu-emu/ext-windows-bin. MinGW now targets the
same ext-windows-bin libraries as MSVC for FFmpeg. Adds FFMPEG_LIBRARIES
to WIN32 and simplifies video_core/CMakeLists.txt a bit.
Update conan package version used for building.
A couple of new joystick-related functions might pose interest to yuzu's input system. Some sort of LED management have been added, but it doesn't seem to support leds used for player number indication JoyCons/ProCons use.
Boxcat is a web service but is still enabled if ENABLE_WEB_SERVICE is
disabled during the CMake stage, which causes compilation issues with
either missing headers or missing libraries.
This disables YUZU_ENABLE_BOXCAT regardless of the input if
ENABLE_WEB_SERVICE is disabled.
Addresses an issue with the two competing versions of Conan's Boost
package that are currently floating around.
Adds the Boost::context target only if it's recognized by CMake as a
target.
Fixes regression by 761206cf81, causing
yuzu to not build on Linux with any version of Boost except a cached
1.73 Conan version from before about a day ago.
Moves the Boost requirement out of the `REQUIRED_LIBS` psuedo-2D-array
for Conan to instead be manually configured, using Conan as a fallback
solution if the system does not meet our requirements.
Requires any update from the linux-fresh container in order to build.
**DO NOT MERGE** until someone with the MSVC toolchain can verify this
works there, too.
Unicorn long-since lost most of its use, due to dynarmic gaining support
for handling most instructions. At this point any further issues
encountered should be used to make dynarmic better.
This also allows us to remove our dependency on Python.
This commit aims to implement the NVDEC (Nvidia Decoder) functionality, with video frame decoding being handled by the FFmpeg library.
The process begins with Ioctl commands being sent to the NVDEC and VIC (Video Image Composer) emulated devices. These allocate the necessary GPU buffers for the frame data, along with providing information on the incoming video data. A Submit command then signals the GPU to process and decode the frame data.
To decode the frame, the respective codec's header must be manually composed from the information provided by NVDEC, then sent with the raw frame data to the ffmpeg library.
Currently, H264 and VP9 are supported, with VP9 having some minor artifacting issues related mainly to the reference frame composition in its uncompressed header.
Async GPU is not properly implemented at the moment.
Co-Authored-By: David <25727384+ogniK5377@users.noreply.github.com>
Provides the buildbot with one builder that is always tracking the
latest version of the C++ standard, allowing us to progressively rectify
our code and amend any differences between standards over time instead
of waiting for a complete standard change, potentially breaking a lot of
code all at once.
Keeps the package up to date with the latest major release of fmt.
This version brings in quite a bit of improvements, such as code size
reduction, etc.
In file included from src/input_common/gcadapter/gc_adapter.cpp:8:
src/./input_common/gcadapter/gc_adapter.h:11:10: fatal error: 'libusb.h' file not found
#include <libusb.h>
^~~~~~~~~~
CMake Error at src/yuzu/CMakeLists.txt:7 (add_executable):
Target "yuzu" links to target "Qt5::WebEngineCore" but the target was not
found. Perhaps a find_package() call is missing for an IMPORTED target, or
an ALIAS target is missing?
* externals: Revert to libressl, as build is broken with find_package(OpenSLL).
* fixup! externals: Revert to libressl, as build is broken with find_package(OpenSLL).
* fixup! externals: Revert to libressl, as build is broken with find_package(OpenSLL).
* Remove git submodules that will be loaded through conan
* Move custom Find modules to their own folder
* Use conan for downloading missing external dependencies
* CI: Change the yuzu source folder user to the user that the containers run on
* Attempt to remove dirty mingw build hack
* Install conan on the msvc build
* Only set release build type when using not using multi config generator
* Re-add qt bundled to workaround an issue with conan qt not downloading prebuilt binaries
* Add workaround for submodules that use legacy CMAKE variables
* Re-add USE_BUNDLED_QT on the msvc build bot
An implementation of the cemuhook motion/touch protocol, this adds the
ability for users to connect several different devices to citra to send
direct motion and touch data to citra.
Co-Authored-By: jroweboy <jroweboy@gmail.com>
This fixes the early-access builds on Windows (tested on EA 58). Cmake
was previously looking for git-related files that were stripped out of
the early access builds and failing; check if those exist before reading
them.
* CMake: Get Git submodule dependencies via CMake
* CMakeLists: Fixed unintentional line break
* travis: Bring parity between linux-mingw and linux build script
* CMakeLists: Fixed typo in error message
VS 2019 is binary compatible with VS 2017, so we can safely use
the prebuilt libraries for VS 2017 with VS 2019. This makes it less
annoying to build yuzu with the most up to date toolchain.
We generally shouldn't be hijacking CMAKE_CXX_FLAGS, etc as a means to
append flags to the targets, since this adds the compilation flags to
everything, including our externals, which can result in weird issues
and makes the build hierarchy fragile.
Instead, we want to just apply these compilation flags to our targets,
and let those managing external libraries to properly specify their
compilation flags.
This also results in us not getting as many warnings, as we don't raise
the warning level on every external target.
This was only ever used by the now-removed memory_util functions. Also,
given we don't plan to support 32-bit architectures, this is just a
leftover from citra at this point.
This is more localized to what we want to enforce directory-wise with
the project. CMAKE_SOURCE_DIR indicates the root of the source tree, but
this would cause the wrong behavior if someone included yuzu as part of
a larger buildsystem (for whatever reason). Instead, we want to use the
directory where the "project(yuzu)" command was declared as the root
path reference.
Removes the annoying step when generating sln for MSVC where you have to
click an extra checkbox after the first generate fails by using a
conditional option. The USE_BUNDLED options will be off by default, but
if the enable_lib option is enabled and the toolset is msvc, they are
turned ON.
Checks to see if clang-format can be found, and if it is, sets up a
custom target that will run against the src dir and auto formats all
files. In MSVC, this is a project, and in Makefiles, its a make target
* Port citra #3352 to yuzu
This change allows non x86_64 architectures to compile yuzu by skipping the building of dynarmic
* Fixed clang-format errors
* fixes more clang-format errors
On MSVC if unicorn isn't found, fallback to bundled unicorn
On everything else, fallback to building unicorn in externals
Also fixes loading unicorn in msvc
Adds a cmake custom target that will build unicorn on first compile and
uses this in the build scripts as well. Updates Appveyor and Travis
build scripts to work with the new unicorn build, and updates the paths
to all of the different artifacts.
When compiling on a case-sensitive filesystem on OSX, cmake doesn't find
the FindUnicorn file, because it looks for Findunicorn.cmake. We should
uses the correct case to avoid this issue.
The mingw builds aren't submitting telemetry because the curl library
they are linked against is configured to use openSSL and openSSL looks
for the certificates in the users home folder. This keeps it from
contacting web services because it can't communicate over SSL.
This commit adds a download in mingw builds that will download a
precompiled curl for mingw linked against winssl and sspi.
Citra SDL2 doesn't have a launcher, and citra.desktop tries to execute
citra-qt which is N/A unless built with ENABLE_QT. Limiting installed
files to one of the options also makes it easier to split them into
separate non-conflicting packages downstream.
Most modern Unix environments use 64-bit off_t by default: OpenBSD,
FreeBSD, OS X, and Linux libc implementations such as Musl.
glibc is the lone exception; it can default to 32 bits but this is
configurable by setting _FILE_OFFSET_BITS.
Avoiding the stat64()/fstat64() interfaces is desirable because they
are nonstandard and not implemented on many systems (including
OpenBSD and FreeBSD), and using 64 bits for stat()/fstat() is either
the default or trivial to set up.
In QT 5.7 they added a new check for CXX features which appends a minimum
required standard to the CXX_FLAGS. Because we were writing the flag
directly previously, cmake assumed it needed to add a c++11 flag to the
build. This tells cmake to use c++14 on every build.
This option makes the generated binary crash with an illegal
instruction when the target CPU doesn’t support the SSE4.1 extension
(see #1968), with no noticeable performance increase compared to a
generic build.
This requires bumping up to a minimum of CMake 3.1. The benefit
of using the imported target is that you can switch to the -pthread
compiler flag on request, which may be necessary for some systems if
available.
citra: Remove GLFW, Add SDL2
FindSDL2: Do not CACHE SDL2_* variables if library is not found
EmuWindow_SDL2: Set minimal client area at initialisation time
EmuWindow_SDL2: Corrections
EmuWindow_SDL2: Fix no decorations on startup on OS X
cmake: windows_copy_files
The `option` commands have been moved to the top of the file, so that
the relevant options are registered in the CMake cache even if one of
the required libraries is not found. This solves an ergonomic problem
when using bundled libraries where you have to first download GLFW
before being able to select the option to also download Qt.
Using this variable is problematic is the user has several versions of
Qt installed on their system. There is no way to know ahead of time if
the Qt version pointed to by QTDIR matches the toolchain that is being
targeted.
The Qt installation path can still be easily specified if it's not found
by CMake by setting the Qt5_DIR cache variable after the initial
configuration run, so this shouldn't present an usability issue.
The main advantage of switching to glad from glLoadGen is that, apart
from being actively maintained, it supports a customizable entrypoint
loader function, which makes it possible to also support OpenGL ES.
Apparently /DEBUG implicitly disables linker optimizations. This
explicitly re-enables them, giving a 40% reduction in binary sizes and
a very slight runtime speed improvement.
