These changes modernize the CMake build system to (hopefully) match newer best practices
* Library linking is simpler and more automatic because of using imported targets
* Multi-configuration builds should be supported (Ninja Multi-Config, Visual Studio, etc. generators)
* Clean up build options using cmake_dependent_option
* Let CMake do its job in more cases, like finding the math/dl libraries and detecting and enabling LTO support
* Remove platform-specific kludges like the Fedora/flatpak LTO workaround and a bunch of Windows stuff
* Simplify Windows static builds
* Consistent formatting
also including:
* getting rid of shitty strings
* all new, cleaner ROM handling code
* base for DSi savestates
* GBA slot addons (for now, memory cart)
* make timers usable for measurement shito without being assfuckingly unreliable
* bürp
* Arisotura can you ever clean up your goddamn code
also regroup the timer code instead of having it split weirdly
* make the set-timing functions a tad less hacky
* congrats Arisotura you made an ass-enum
* add timing region tables, and separate timings for ARM9 DMA (exempt of 3c penalty)
* temp work on DMA timings, not finished
also, did you know? 'increment/reload' is also a thing for the source address
* begin work
* add some of the GBA slot/wifi timings
* complete it, I guess
* make some progress
* getting somewhere
* sdsdfs
* see, Arisotura, was it that hard? blarg.
* Add PoroCYon's DSP code.
* Remove some teakra iles that we dont need.
* make some requested changes.
* move DataMemoryOffset into namespace.
* use deault param.
* ad the switch change
* <Generic> forget about the default parameter
* Draft GPU3D renderer modularization
* Update sources C++ standard to C++17
The top-level `CMakeLists.txt` is already using the C++17 standard.
* Move GLCompositor into class type
Some other misc fixes to push towards better modularity
* Make renderer-implementation types move-only
These types are going to be holding onto handles
of GPU-side resources and shouldn't ever be copied around.
* Fix OSX: Remove 'register' storage class specifier
`register` has been removed in C++17...
But this keyword hasn't done anything in years anyways.
OSX builds consider this "warning" an error and it
stops the whole build.
* Add RestartFrame to Renderer3D interface
* Move Accelerated property to Renderer3D interface
There are points in the code base where we do:
`renderer != 0` to know if we are feeding
an openGL renderer. Rather than that we can instead just have this be
a property of the renderer itself.
With this pattern a renderer can just say how it wants its data to come
in rather than have everyone know that they're talking to an OpenGL
renderer.
* Remove Accelerated flag from GPU
* Move 2D_Soft interface in separate header
Also make the current 2D engine an "owned" unique_ptr.
* Update alignment attribute to standard alignas
Uses standardized `alignas` rather than compiler-specific
attributes.
https://en.cppreference.com/w/cpp/language/alignas
* Fix Clang: alignas specifier
Alignment must be specified before the array to align the entire array.
https://en.cppreference.com/w/cpp/language/alignas
* Converted Renderer3D Accelerated to variable
This flag is checked a lot during scanline rasterization. So rather
than having an expensive vtable-lookup call during mainline rendering
code, it is now a public constant bool type that is written to only once
during Renderer3D initialization.
* Use libepoxy to load in OpenGL functions
Prevents having to load them in manually
* Install libepoxy in the CI
* Do not link OpenGL libraries, libepoxy opens them itself
* Add libepoxy to build instructions
* Add FreeBSD support
* Fix indentation
* Fix Linux not finding OpenGL
* Link POSIX Realtime Extensions library
* Link POSIX Realtime Extensions when OpenGL is enabled too
* Fail if shm_open memory exists and also check for errors
* fix the last commit
* (try to) Setup FreeBSD CI
* Fix some issues with FreeBSD CI
* Make with all cores
* Remove FreeBSD CI
It doesn't want to work for some reason
* Offload NDS SRAM writing to separate thread, debounce writes to two seconds after last flush or DeInit.
* Fixed printf messages.
* Fixes after CR.
* Fixed potential portability issue with time_t
* use shm_open() instead of memfd_create() on macOS
malloc.h isn't a header on macOS
* Change OpenGL headers + create ifdef for DO_PROCLIST
macOS seems to already have the OpenGL functions defined, without the ifdef, it gives "ambiguous references" errors.
* macOS doesn't have ->gregs in uc_mcontext
and it doesn't have REG_RIP either
https://github.com/gperftools/gperftools/blob/master/m4/pc_from_ucontext.m4
* use getpid() to make memory file name unique
* #ifndef __APPLE__ for AF_PACKET and linux/if_packet.h
* Add include and link directories for macOS and link the OpenGL framework
* Add macOS CI
* Use newly added libslirp package from Homebrew
https://github.com/Homebrew/homebrew-core/pull/63412
* Use Apple's Clang instead of GNU GCC on macOS
* Add macOS build instructions to README
* Try to fix macOS undefined symbol
* snprintf doesn't take null terminator into account
* Map new memory on macOS for JIT
* Only use gcc-ar if using GNU Compiler
* re-add fastmem code - whoops!
* Fix style issue - use camelCase not snake_case
* Set Minimum macOS version
* Switch Minimum OS X version to 10.9
* Add macOS libpcap library name
* fix memory leak
* Fix binding keys in macOS
* Allow getting MAC address on macOS
melonDS on Linux uses AF_PACKET, which doesn't exist on macOS. Instead, this commit uses AF_LINK on macOS to get the MAC address.
* Remove unneeded macOS CI dependencies
* Build melonDS app bundle on macOS
Now it is no longer required to install the libraries on macOS, they come with the app bundle.
* fix macOS CI not being able to find macdeployqt
* copy melonDS.app with recursive because it's a folder
* Disable fastmem checkbox on macOS
* Disable fastmem by default in config
* forgot a semicolon
* Don't bundle libraries, causes issues on macOS <10.15
* Update README + allow finding version in Finder on macOS
* Make sure fastmem checkbox stays uncheckable