Requires qt6-static, obviously... at least for eden. eden-cli also can
be built fully static
Notable challenges n such:
1. VkMemAlloc conflicts with Qt, since it embeds vk_mem_alloc.h in
qrhivulkan; we can get around this by conditionally defining
VMA_IMPLEMENTATION; that is, define it in the SDL2 frontend and undef
it in the Qt frontend. It's not ideal, but I mean... it works, no?
2. find_library, pkgconfig, and some Config modules will always look for
a .dll, so we have to tell CMake to look for .a
3. In spite of this, some will end up using .dll.a (implib) as their
link targets; this is, well, bad, so we create a find_library hook
that rejects dll.a
4. Some libraries have specific configs (boost lol)
5. Some libraries use _static targets (zstd, mbedtls)
6. Some extra libraries need to be linked, i.e. jbig, lzma, etc
7. QuaZip is sad
Needs testing on all platforms, and for both frontends on desktop, to
ensure Vulkan still works as expected.
(also: CI). Resulting executables are:
- 71MB for eden.exe
- 39MB for eden-cli.exe
Considering the entire libicudt is included (thanks Qt), that's a great size all things considered. No need to bundle all those plugins and translation files too.
Theoretically, this lays the groundwork towards fully static executables for other platforms too; though Linux doesn't have a huge benefit since AppImages are needed regardless. eden-room though maybe?
Fixes comp for clangarm64 because -msse4.1
Also allows macOS to build with qt6-static. macOS can't build static executables, but with these changes it ONLY relies on system libraries like libc and frameworks. So in theory we don't even need macdeployqt.
Signed-off-by: crueter <crueter@eden-emu.dev>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2994
Patience is key when building from source on an Intel Atom N455.
Of course, no SSE4.1 is in an atom... which is so unfortunate :(
We only get SSSE3 - but CI handles for building better codegen doesn't it?
Signed-off-by: lizzie <lizzie@eden-emu.dev>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2872
Reviewed-by: Caio Oliveira <caiooliveirafarias0@gmail.com>
Reviewed-by: crueter <crueter@eden-emu.dev>
Co-authored-by: lizzie <lizzie@eden-emu.dev>
Co-committed-by: lizzie <lizzie@eden-emu.dev>
This is part of a series of PRs made in preparation for the QML rewrite. this PR specifically moves a bunch of utility functions from main.cpp into qt_common, with the biggest benefit being that QML can reuse the exact same code through ctx passthrough.
Also, QtCommon::Frontend is an abstraction layer over several previously Widgets-specific stuff like QMessageBox that gets used everywhere. The idea is that once QML is implemented, these functions can have a Quick version implemented for systems that don't work well with Widgets (sun) or for those on Plasma 6+ (reduces memory usage w/o Widgets linkage) although Quick from C++ is actually anal, but whatever.
Other than that this should also just kinda reduce the size of main.cpp which is a 6000-line behemoth rn, and clangd straight up gives up with it for me (likely caused by the massive amount of headers, which this DOES reduce).
In the future, I probably want to create a common strings lookup table that both Qt and QML can reference--though I'm not sure how much linguist likes that--which should give us a way to keep language consistent (use frozen-map).
TODO: Docs for Qt stuff
Co-authored-by: MaranBr <maranbr@outlook.com>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/94
Reviewed-by: MaranBr <maranbr@eden-emu.dev>
Reviewed-by: Shinmegumi <shinmegumi@eden-emu.dev>
Compilation and CMake fixes for both Windows on ARM and clang-cl, meaning Windows can now be built on both MSVC and clang on both amd64 and aarch64.
Compiling on clang is *dramatically* faster so this should be useful for CI.
Co-authored-by: crueter <crueter@eden-emu.dev>
Co-authored-by: crueter <crueter@crueter.xyz>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/348
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
Reviewed-by: crueter <crueter@eden-emu.dev>
Co-authored-by: lizzie <lizzie@eden-emu.dev>
Co-committed-by: lizzie <lizzie@eden-emu.dev>
I promise I'm a UI developer
- mbedtls can now be used as a system package
- zycore can now be used as a system package
- cleaned up dynarmic externals
- fixed libusb incorrectly showing as bundled
- add version/tag formatting to JSON
- add custom GIT_HOST option for packages
- moved some of my repos to my new git
- slightly better version identification
- combined VUL/VH since they are codependent (using my combo vendor)
- fix cpmfile inclusion
- remove libusb submodule
This PR succeeds #383 since it includes it
Co-authored-by: SDK Chan <sdkchan@eden-emu.dev>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/392
Co-authored-by: crueter <crueter@crueter.xyz>
Co-committed-by: crueter <crueter@crueter.xyz>
- creates a CPMUtil.cmake module that makes my job 10x easier and removes boilerplate
- also lets us generate dependency names/versions at compiletime, thus letting the frontend display each dependency's versions.
Signed-off-by: crueter <crueter@eden-emu.dev>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/238
Transfers the majority of submodules and large externals to CPM, using source archives rather than full Git clones. Not only does this save massive amounts of clone and configure time, but dependencies are grabbed on-demand rather than being required by default. Additionally, CPM will (generally) automatically search for system dependencies, though certain dependencies have options to control this.
Testing shows gains ranging from 5x to 10x in terms of overall clone/configure time.
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/143
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
also, fixes CI not targeting the right flag settings.
this pr needs to add the room exe removal from lime3ds.
Signed-off-by: Aleksandr Popovich <alekpopo@pm.me>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/106
Co-authored-by: Aleksandr Popovich <alekpopo@pm.me>
Co-committed-by: Aleksandr Popovich <alekpopo@pm.me>
Includes citron, sudachi, yuzu
currently broken, because the eden dir is always made early?
Signed-off-by: swurl <swurl@swurl.xyz>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/91
Co-authored-by: swurl <swurl@swurl.xyz>
Co-committed-by: swurl <swurl@swurl.xyz>
Replaces every way of handling config for each frontend with SimpleIni. frontend_common's Config class is at the center where it saves and loads all of the cross-platform settings and provides a set of pure virtual functions for platform specific settings.
As a result of making config handling platform specific, several parts had to be moved to each platform's own config class or to other parts. Default keys were put in platform specific config classes and translatable strings for Qt were moved to shared_translation. Default hotkeys, default_theme, window geometry, and qt metatypes were moved to uisettings. Additionally, to reduce dependence on Qt, QStrings were converted to std::strings where applicable.
Note: For GCC there are still a huge number of `-Warray-bounds` warnings
coming from `externals/dynarmic`. I could have added a workaround in
`externals/CMakeLists.txt` similar to what this PR does for other
externals, but given Dynarmic's close affiliation with Yuzu, it would be
better to fix it upstream.
Besides that, on my machine, this makes the build warning-free except
for some warnings from glslangValidator and AutoMoc.
Details:
- Disable some warnings in externals.
- Disable `-Wnullability-completeness`, which is a Clang warning triggered
by the Vulkan SDK where if any pointers in the header are marked
_Nullable, it wants all pointers to be marked _Nullable or _Nonnull.
Most of them are, but some aren't. Who knows why.
- `src/web_service/verify_user_jwt.cpp`: Disable another warning when
including `jwt.hpp`.
- `src/input_common/input_poller.cpp`: Add missing `override` specifiers.
- src/common/swap.h: Remove redundant `operator&`. In general, this
file declares three overloads of each operator. Using `+` as an
example, the overloads are:
- a member function for `swapped_t + integer`
- a member function for `swapped_t + swapped_t`
- a free function for `integer + swapped_t`
But for `operator&`, there was an additional free function for
`swapped_t + integer`, which was redundant with the member function.
This caused a GCC warning saying "ISO C++ says that these are
ambiguous".
The latest version of MSVC STL brings C++23 standard library modules, which conflict with precompiled headers.
Disabling with /experimental:module- has no effect, so force C++20 in the meantime while we wait for module support in other compilers.
Enables shadow-uncaptured-locals and implicit-fallthrough for Clang.
implicit-fallthrough is not enabled by default in -Wall or -Wextra, and
shadow-uncaptured-local doesn't seem to be enabled by default by
-Wshadow, even though GCC has both of these by their respective cases.
crueter