* SCFG_BIOS now selects between NDS and DSi BIOS
Allowing experimental direct boot.
* - making NWRAM Priorities work as in the HW
- fixed loading DSi stage2 Bootloader when in direct boot (should not be loaded) and might interfere with the image loaded into the (N)WRAM previously
* NWRAM and SCFG Registers are now write-guarded by the corresponding SCFG_EXT7/9 Bits
* removed debugging remainder
* Moved NDSHeader dependent SCFG initialization into the new DSi::SetupDirectBoot function called from NDS::SetupDirectBoot when ConsoleType is DSi
* removed redundant SCFG_BIOS bit checks
* Set of changes from RSDuck's review
* removed a forgotten comment
* - removed the guarded debug outputs for NWRAM
- NWRAM writes to bank and window registers now apply their write masks.
- added comment on an existing TODO within the code describing, why this is OK
- fixed initial NWRAM bank reset just accessing set A
* NWRAM not mapped to 03... range if bit25 in SCFG_EXT is cleared
* removed NWRAM write block on cleared bit25 in SCFG_EXT
* changed type of iterator for MapNWRAM_x functions from s8 to int
* - reduced calculations/comparisons on NWRAM write operations
- changed u8 to unsigned int for an iterator MapNWRAM_x
Co-authored-by: Tim Seidel <t.seidel@kts-systeme.de>
* Implement NO$GBA debug registers.
NO$GBA comes with 4 debug registers that allow a ROM to print text to
the emulator terminal and 2 other status registers, one with the name of the
emulator and the other with the clock cycles count. This commit
implements them for the ARMv5 processor.
Some small things to note:
- `NocashPrint` was changed and now it takes an address to _the string_ instead of the flags before it (those
don't do anything anyways).
- The "Emulation ID" register contains the string "melonDS " followed by version, _not_ "NO$GBA"
* Fix styling issue and improve comment regarding NO$GBA message flags
* Fixed division edge case: Div64/32 and Div64/64 set the remainder to 0 if dividend == INT64_MIN && divisor == -1
* Fixed CPU edge case where ARM9 ALU ops would switch to Thumb even when they shouldn't
* Only clear the lowest bit of the jump address in ALU ops with rd==15 (on recommendation of RSDuck)
* 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.
* Fix directory path when extracting from archive
* Don't create new dir in execution dir of melonds
* Create it beside the archive instead
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* ArchiveUtil : Use QT functions for I/O
* Make it more platform independent, cleaner
* Fixes permission related crash on linux
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* NDSCart : Abstract out common code in LoadROM()
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Extract nds roms to memory
* Some stuff is still broken in the frontend
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* GBACart : Abstract out common code in LoadROM()
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Extract gba roms to memory
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Integrate archive support with recent files
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* onClickRecentFile : Pause emu thread conditionally
* Don't pause at start of the function
* If user opens an archive and hits cancel, it won't pause
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Handle Resets when loading from archives
* Ask user to pick the rom(s) again (i.e. GBA & NDS)
when there are multiple files in the archive(s)
* Directly load if only 1 file
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Archive support for drag-n-drop
* Also recent files support for drag-n-drop
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* main : Allocate rombuffer objects on stack
* Less messy, decreases chances of memory leaks
* Underlying implementation of qbytearray uses heap (hopefully?)
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* GetSavestateName : Archive support
* Construct ssname from srampath (since rompath has archive name)
NOTE: In general, archive name != rom file name !!!!!!!!!!
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Add srl and dsi as "direct-load" formats
* Direct-load = anything not in an archive
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Don't use static functions
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Remove QT stuff from Util_ROM
* Also, during reset, directly load file from archive (no rom picker)
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Remove QT includes from FrontendUtil.h
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Util_ROM/LoadROM() : Use SetupDSiNAND()
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Util_ROM/Reset() : Use strrchr()
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Util_ROM : Put Archive stuff behind ifdefs
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* main: Set parent widget for archive dialog boxes
Signed-off-by: Madhav Kanbur <abcdjdj@gmail.com>
* Revert "Util_ROM/Reset() : Use strrchr()"
This reverts commit c8af6f066f.
* Implement "lag frame" flag and counter, and expose flag and both frame counters. BizHawk wants these.
* Track frame count and lag frames while the system isn't running.
* 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
- savestates used to read a four bytes from a single byte value
- a few unassigned variables
- some other things
- also make the ROR macro an inline function
- more reliable code invalidation detection
- blocks aren't stopped at any branch, but are being followed
if possible to get larger blocks
- idle loop recognition
- optimised literal loads, load/store cycle counting
and loads/stores from constant addresses
- more reliable code invalidation detection
- blocks aren't stopped at any branch, but are being followed
if possible to get larger blocks
- idle loop recognition
- optimised literal loads, load/store cycle counting
and loads/stores from constant addresses
Previously, the GBA memory was 0xFF-filled
on both CPUs. However, GBATEK reports that
the deselected CPU is 0x00-filled, and that
some titles depend on this behavior to
function properly.
* immediately clear AES busy flag when the block count is zero (occurs when loading DSi cart games)
* implement NDMA start modes that have an old-DMA equivalent (except for GXFIFO mode)
now it boots DSi carts!
Arisotura