* slightly fix gambatte peeking
* fix gambatte layer masking, move layer masking to sync settings (these settings affect sync, they shouldn't be normal settings)
* block toggle layer hotkeys when movie recording
* refactor Gambatte's RTC
* add a way to set an inital RTC state during a movie recording
* prevent Gambatte from saving RTC data when recording a movie, needed to create save-anchored movies
The goal of the separate stacks was to allow this, but I never quite finished the job. Now, when a SEH exception (generally a Rust panic in a guest syscall handler, or a C# Exception in a callback) tries to unwind through guest code, it works. Note that we don't actually unwind the guest stack, as there's nothing useful to be gained from that; When an emulator core throws an exception like this, it should be considered completely hosed. Throw it out and get a new one.
There were two bugs stopping this from working.
First of all, we had custom thunks that lacked sufficient unwind information for RtlUnwind to get through. For the sysv <-> msabi adapter, this was fixed by making it regular Rust code instead of hand assembled junkus. So the compiler generates valid unwind information for all of that. Then we just JIT a small stub on top in the MsHostSysVGuest code, which needs no unwind information because it won't throw an exception itself and transparently passes execution to something with valid unwind information without invalidating that information. (NB: Clr JIT stubs use the same strategy.) For the host <-> guest stack transition code, a small hand generated unwind stub was added to interop.s that is registered with `RtlAddFunctionTable`. I've seen the unwind work successfully without this second set of unwind information, but better safe than sorry.
Secondly, our misuse of SubSystemTib caught up with us. It's an old field, allegedly from OS/2, that we repurposed to hold TLS information needed for the waterbox stack transitions. Most people think nothing uses it any more, but in fact if it's set to a non-NULL value, but doesn't contain valid information, `KERNELBASE!GetModuleFileNameW` will crash when it tries to get a module name from there. The fix here was to simply tighten up our usage of SubSystemTib: We were already nulling it out when returning from guest code, but not when calling back to host code in guest code.
Fixes#2487. Unwinding of this sort has never worked well in waterbox; the reason why that issue is more recent is that the particular reproducing case of firmware didn't cause an exception in a callback in older code; the exception happened in pure managed code.
* post-build step moves .exes up a level, to output
* also fixed OpenTK.dll.config location
* also removed redundant step from packaging scripts
* future executables need to be capable of resolving assemblies in dll
* Move .so libraries to dll dir, update some build scripts
* Move OpenTK.dll.config with OpenTK.dll
* Keep EmuHawkMono.sh in Windows-built artifacts
* Add Package.sh to match QuickTestBuildAndPackage.bat
used as `Dist/BuildRelease.sh && Dist/Package.sh`
* Update GitLab CI to use Package.sh
A2600 standard controller and A7800 standard controller were using the same
keybinds, now only Atari2600Hawk uses those and A7800Hawk's are blank unless
configured manually or configured on config regen
Waterbox supports threads now, but they're not real threads on the host side because that's complicated and can be nondeterministic. Instead, everything is scheduled to share one host thread. This means that scheduling is actually cooperative and certain patterns of spinlocks and other nonsense can fail to work at all, but "regular" code probably will.
With this, add DobieStation PS2 core. This core was selected because it has threads and is otherwise simple to port; easy to build and a good core/frontend separation. It's not a wonderful core however, with low speed (made abysmally lower by our lack of real threads) and low compatibility, so it remains a curiosity for now.
The description in the previous commit is accurate, but the problem runs deeper and was on the whole a complete failure for me to appreciate the difference between active and swapped in on memoryblocks. Bleeecch.
This was broken by 175556529e, with two related issues: When we allowed for some operations to happen even when the block is inactive, we didn't account for the fact that in swapin, the block technically is not active yet (the lock is not on the self), and similarly in swapout, the lock has already been moved out of self. The former caused all memory areas to revert to RWX at the host OS level after a swap, so no dirty detection was done. After the former was fixed, the latter caused saved costacks to still get missed.
At the same time we ran into a perfect storm with costacks on Windows; if a stack page is not yet dirty, but we hit a fault for something else, Windows will not call our VEH handler unless the TIB stack extents are satisfactory, since it needs userspace to fix up the TIB extents via VEH or SEH handler, but there's already an exception pending.
This broke any waterbox core that called in to native code in the same EnterExit() right after sealing. All nyma cores were broken, 32x was not, didn't check the rest. Regressed in 175556529e.
It worked fine in release mode, theoretically
Set up a second mirror of guest memory; easily accomplished because we were already using memfd_create / CreateFileMappingW.
This lets us simplify a lot of host code that has to access guest memory that may not be active right now, or might have been mprotect()ed to something weird. Activate is only needed now to run guest code, or when the C# side wants to peer into guest memory for memory domains and such (waterboxhost does not share the mirror address with the C# side).
Bizhawk never would hit this because it only ever runs waterboxes in one host thread, but an application that spun up many threads and ran waterboxes in each would leak 32 bytes of heap for each native thread destroyed, which is super duper not really meaningful at all
Waterbox guest code now runs on a stack inside the guest memory space. This removes some potential opportunities for nondeterminism and makes future porting of libco-enabled cores easier.
This replaces the old managed one. The only direct effect of this is to fix some hard to reproduce crashes in bsnes.
In the long run, we'll use this new code to help build more waterbox features.
gpgx upstream automatically gives 64KiB sram to every game of size 2MB or lower, unless specifically overrided. That sucks and we don't do it. But some games do need it; there's not enough information to autodetect otherwise. So add a flag that we can set in gameDB for it, and set it true for two known games that have problems otherwise.
* Add MelonDS.cs, support opening (but not really) .nds files.
* init MelonDS
* MelonDS: Load selected ROM.
* MelonDS: FrameAdvance and frame counter.
* MelonDS: IVideoProvider
* MelonDS: Add DLL files.
* MelonDS: IInputPollable
* MelonDS: IStatable (and add forgotten file MelonDS_InputPollable.cs)
* update libmelonDS.dll
* MelonDS: ISoundProvider
* Add NDS to Global.SystemInfo, and convert screen coords when running NDS.
* set up default NDS controller
* MelonDS: ISaveRam
* MelonDS: remove romlist.bin
* MelonDS: ISettable
* Create firmware folder if it doesn't exist on Windows; otherwise, an exception is thrown.
* Add database entries for NDS bios/firmware files.
* MelonDS: Use the bios/firmware files selected in BizHawk's "Firmwares" dialog.
* MelonDS: Re-work sync settings a bit.
* NDS's firmware file contains user settings; these are over-written by sync settings, so we shouldn't allow them to impact the hash
* MelonDS: Add (currently unused) bootToFirmware sync setting, and NDSSettings dialog.
* Update NDS firmware hash; it seems I had somehow corrupted mine.
* MelonDS: Use boot to firmware sync setting.
* MelonDS: Allow user to set some firmware user settings via the NDS settings dialog.
* MelonDS: Add singleInstance attribute to core.
* MelonDS: IMemoryDomains
* update libmelonDS.dll
* MelonDS: Set up default sync settings if none are provided.
* MelonDS: Allow user to reset settings to default.
* MelonDS: bios+firmware files are recommended
* libmelonDS.dll
* MelonDS: Don't use real time.
* MelonDS: Update to reflect new way of handling RTC in MelonDS.
* MelonDS: Notify if savestate load failed.
* update MelonDS.dll
* MelonDS: Allow user to set startup date/time in settings dialog.
* MelonDS: Create melon directory if it doesn't already exist.
* Don't include Designer's "fixes" in PR (partially reverts 56b474c00)
* Don't show a broken console window; alert user of need to restart instead.
This fixes an error related to MelonDS trying to use the broken stdout stream.
* update default NDS controls to match other updated controls
* Implement a system bus, using ARM9 read/writes.
* MelonDS: Allow BizHawk to change the contents of the frame buffer.
* update libmelonDS.dll
* fix stuff that was merged incorrectly, or was broken by merge
* update libmelonDS.dll
(includes memory leak fix)
* update libmelonDS.dll
(fixes memory leak and an occasional savestate crash)
* fix stuff that broke with the merge
* cleanups, remove stuff that is no longer needed by service interaces
* simplify DS MemoryDomains
* DS - fix order of controller buttons to be consistent with other consoles. This probably breaks any existing movies made on this core, but those would have been experiments, right?
* NDSSettings - make min value for day and month 0, whiel those aren't "valid" values they are the default values in the core for whatever reason, better to not crash on load and not show a value that isn't actually the setting. This can easily be reverted if the core changes to default to 1
Co-authored-by: YoshiRulz <OSSYoshiRulz@gmail.com>
Co-authored-by: adelikat <adelikat@tasvideos.org>
for the camhack to work we have to save a state, hack memory, advance twice to see the changes, then load the state to prevent desync. since we can omit the framebuffer in savestates, loading them can happen without updating the screen, so the hacked camera remains visible.
advancing 2 frames automatically is done like tastudio does it when it seeks to a frame, only from lua now.
and the most questionable part is "invisible emulation", which is how Gens calls this IIRC, when everything that can distract or slow us down is skipped: sound, video, tools updates.
new lua functions:
- client.invisibleemulation()
- client.seekframe()
* for a test, mGBA core uses fake video and audio buffers and renders to them when we want to "skip" rendering. proper setup would involve actually skipping rendering those inside the core.
* allow disabling video and audio updates for gpgx too (proper approach, no fake buffers involved)
* add the script for Sonic Advance
This doesn't compile because of Input.cs, didn't know what to do. Also search for Merge TODO for some commenting things that probably need to be deleted
# Conflicts:
# BizHawk.Client.EmuHawk/BizHawk.Client.EmuHawk.csproj
# BizHawk.Client.EmuHawk/CustomControls/InputRoll.Drawing.cs
# BizHawk.Client.EmuHawk/CustomControls/InputRoll/InputRoll.cs
# BizHawk.Client.EmuHawk/Program.cs
# BizHawk.Client.EmuHawk/tools/Lua/LuaConsole.cs
# BizHawk.Client.EmuHawk/tools/TAStudio/TAStudio.cs
# BizHawk.Client.EmuHawk/tools/ToolHelpers.cs
# BizHawk.Client.EmuHawk/tools/ToolManager.cs
# BizHawk.Client.EmuHawk/tools/TraceLogger.Designer.cs
# BizHawk.Client.EmuHawk/tools/TraceLogger.cs
# BizHawk.Client.EmuHawk/tools/Watch/RamSearch.Designer.cs
# BizHawk.Client.EmuHawk/tools/Watch/RamSearch.cs
# BizHawk.Common/BizInvoke/DynamicLibraryImportResolver.cs
* Move PlatformSpecificLinkedLibs and implementations to common and rename
* Specify file ext. at LoadPlatformSpecific call site
* Move Client.Common.Global.RunningOnUnix to PlatformLinkedLibSingleton
* Inline var Resolver
* Use PlatformLinkedLibManager internally
* Move plugin load check to LinkedLibManager, use LinkedLibManager
* Interpolate
* Return exit code from dlclose/FreeLibrary
* Skip all calls to externs in BlipBufDll when using mono
* Use PlatformLinkedLibManager in SevenZipLibraryManager
* Add expected return value to workaround (from testing on Win32)
* Remove ".dll" from DllImport attr, remove temporary workaround, see desc.
The library can be built by changing the output file name in
`.../blip_buf/Makefile` to `libblip_buf.so`, and running `make`. It will be
loaded if placed in the `.../output` folder.
* Remove unused code, add TODO (this class is req. for Waterbox.PeWrapper)
The TODO is to [rewrite with
C#](https://docs.microsoft.com/en-us/dotnet/standard/io/memory-mapped-files)
instead of importing from `kernel32.dll`.
* Update OpenTK again but better (for #1384)
* Add Mono run script
* Add libblip_buf.so (temporary)
Temporary because it should be a separate package which BizHawk depends on.
* Add distro detection, add "already running" and "unknown distro" messages
* Gray-out Lua Console on Unix
* Extract superclass from EmuLuaLibrary, add shell implementation for Unix
* Specify libdl version, Fedora doesn't have the versionless symlink
* Remove empty `ToolStripMenuItem`, null `Text` caused crash on Unix
* Transform OpenTK keyboard input into a `List<KeyEvent>` and read that
Also fixes crash on rebind
* Remove debug `using ...;`
Adds a small script which, when run, only allows input to P1's controller if both P1 and P2 holds down the specific button simultaneously. Also useful since it demonstrates joypad manipulation.
- use M64K Bus properly
- input display
- fceux font
- map and screen borders when camhack is on
- refactor out some functions
- manual lagtount tweaks
- use r24 alias instead of rl for reading addresses from RAM
YoshiRulz