* Implements hardware acceleration for SHA256 instructions.
* Adds SHA256 instructions introduced in ARMv8 to A32 frontend.
* Implements polyfill for processors that do not support hardware
accelerated SHA instructions.
Inlines implementation of exclusive instructions into JITted code,
improving performance of applications relying heavily on these
instructions.
We also fastmem these instructions for additional speed, with
support for appropriate recompilation on fastmem failure.
An unsafe optimization to disable the intercore global_monitor is also
provided, should one wish to rely solely on cmpxchg semantics for
safety.
See also: merryhime/dynarmic#664
* this resolves the todo items in the CMakeLists.txt
* a version requirement check for ffmpeg is added to catch issues early
* for future-proof reasons, nasm/yasm is now only required when build on
x86/AMD64 systems
On Linux, due to the way we include SDL2 as a submodule, it makes it
difficult for us to specify which SDL_config.h we intended to include.
Before, CMake would default to the dummy one included with SDL and
ignore the generated one.
This tells CMake to use the generated one. In addition, we define
USING_GENERATED_CONFIG_H to throw an error in case the dummy config is
used by accident. Fixes Vulkan not working on Linux yuzu-cmd.
yuzu requires CMake 3.15 yet find_program was using REQUIRED, which is
only available on 3.18 and later. Instead, we check for
"<VAR>-NOTFOUND".
In addition, check for additional requirements before building libusb or
FFmpeg with autotools. Otherwise, CMake configuration will pass yet
compilation will fail.
Fixes an issue where libusb.h wouldn't be found when building yuzu on
MSVC.
This only affects the "traditional" CMake pathway for linking libusb to
yuzu AKA MSVC. For autotools we still want to set these variables before
configuring SDL.
Delegates libusb external communication to externals/CMakeLists.txt
Ensures an interface library `usb` for every pathway
input_common just links to the `usb` library now
externals/libusb/CMakeLists.txt sets variables to override SDL2's libusb
finding
Other minor cleanup
Building libusb was also broken on GCC (and maybe Clang) on our
CMakeLists after upgrading to 1.0.24, but it was not being checked
because our 18.04 container had libusb installed on it.
This builds on the MinGW work from earlier and extends it to the rest of
the GNU toolchains. In addition we make use of pkg-config when present
to find libusb. pkg-config is preferrable because we can specify a
minimum required version.
Whatever those settings do breaks controller detection on Windows, at
least with the MinGW container. If-guard it against WIN32 and just let
SDL2 configure using its defaults, aside from static linking.
After updating to 1.0.24, MinGW fails to build libusb as a result of
numerous errors. So we build libusb their way and let them update the
nontrivial stuff.
This only applies to MinGW: the old path is still in use for Linux
toolchains as well as MSVC.
This will dynamically link libusb, since I hit build errors with the old
way we used to resolve the conflict with SDL2.
6b6b9e593d does not exist on FFmpeg master, and tag n4.3.1 requires
manually fetching all of FFmpeg's tags. `git` reports that the commit
does not exist initially and can be confusing as a result. Instead,
checkout the immediately previous commit from n4.3.1 on their master
branch.
Building it as a shared library causes issues distributing it to an
AppImage, since linuxdeploy expects the executable to only dynamically
link to system libraries. Additionally, simply dynamically linking to a
library in the binary directory is bound to cause issues.
Solution is to use SDL's CMake switches and build it statically. We also
alias `SDL2` to `SDL2-static` on the external submodule for
compatibility with the rest of the project.
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.
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.
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.
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.
* The web_service http request is now fixed on Windows (R) platform.
* The issue is due to a complicated race-condition in `httplib`, a detailed
explanation is available at https://github.com/yhirose/cpp-httplib/pull/701
* A pending Pull Request on `httplib` has been applied to remedy the
said race-condition.
* The socket availability check is removed due to a behavioral chice of
`httplib` that a socket will not be created before any actual request
is sent.
* A regression was in 39c8d18 and token verification function was
broken.
* The reason being `httplib` now requires OpenSSL 1.1+ API while
LibreSSL 2.x provided OpenSSL 1.0 compatible API.
* The bundled LibreSSL has been updated to 3.2.2 so it now provides
OpenSSL 1.1 compatible API now.
* Also the path hint has been added so that it will find the correct
path to the CA certs on *nix systems.
* An option is provided so that *nix system distributions/providers can
use their own SSL implementations when compiling Yuzu/Citra to
(hopefully) complies with their maintenance guidelines.
* LURLParse is also removed since `httplib` can handle
`scheme:host:port` string itself now.
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>
Makes our error coverage a little more consistent across the board by
applying it to Linux side of things as well. This also makes it more
consistent with the warning settings in other libraries in the project.
This also updates httplib to 0.7.9, as there are several warning
cleanups made that allow us to enable several warnings as errors.
Given we have two libraries that seem to use the same identifier, we can
alter one of them so that the variable is used in place, effectively
changing the used identifier, but without altering the source of
libusb.
I made a request on the Xbyak issue tracker to allow some constructors
to be constexpr in order to avoid static constructors from needing to
execute for some of our register constants.
This request was implemented, so this updates Xbyak so that we can make
use of it.
We can place the external in an inner folder and manage the custom files
necessary to integrate it with CMake directly. This allows us to
directly change how we use it with our build system, as opposed to
needing to change a fork.
5.95 contains a potentially backward-compatibility breaking change, so
we should be updating to this to ensure that our code remains
forward-compatible.
Two of the members of the MicroProfileThreadLog contains two std::atomic
instances. Given these aren't trivially-copyable types, we shouldn't be
memsetting the structure, given implementation details can contain other
members within it.
To avoid potential undefined behavior on platforms, we can use aggregate
initialization to zero out the members while still having well-defined
behavior.
While we're at it we can also silence some sign conversion warnings.
* 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
Neither core nor web_services use OpenSSL nor LibreSSL.
However they need to link them as it's a requirement of httplib.
So let's declare this within httplib instead of core and web_services.
Update Sirit and its usage in vk_shader_decompiler. Highlights:
- Implement tessellation shaders
- Implement geometry shaders
- Implement some missing features
- Use native half float instructions when available.
