The syncpoint maximum value represents the maximum possible syncpt value at a given time, however due to PBs being submitted before max was incremented, for a brief moment of time this is not the case which could lead to invalid behaviour if a game waits on the fence at that specific time.
* Implement syscall handlers using a source generator
* Copy FlushProcessDataCache implementation to Syscall since it was only implemented on Syscall32
* Fix wrong argument order in some syscalls
* Delete old Reflection.Emit based syscall handling code
* Improvements to the code generation
* ControlCodeMemory address and size is always 64-bit
* Refactor CPU interface
* Use IExecutionContext interface on SVC handler, change how CPU interrupts invokes the handlers
* Make CpuEngine take a ITickSource rather than returning one
The previous implementation had the scenario where the CPU engine had to implement the tick source in mind, like for example, when we have a hypervisor and the game can read CNTPCT on the host directly. However given that we need to do conversion due to different frequencies anyway, it's not worth it. It's better to just let the user pass the tick source and redirect any reads to CNTPCT to the user tick source
* XML docs for the public interfaces
* PPTC invalidation due to NativeInterface function name changes
* Fix build of the CPU tests
* PR feedback
* Enable JIT service LLE
* Force disable PPTC when using the JIT service
PPTC does not support multiple guest processes
* Fix build
* Make SM service registration per emulation context rather than global
* Address PR feedback
* Fix shared memory leak on Windows
* Fix memory leak caused by RO session disposal not decrementing the memory manager ref count
* Fix UnmapViewInternal deadlock
* Was not supposed to add those back
* Back to the origins: Make memory manager take guest PA rather than host address once again
* Direct mapping with alias support on Windows
* Fixes and remove more of the emulated shared memory
* Linux support
* Make shared and transfer memory not depend on SharedMemoryStorage
* More efficient view mapping on Windows (no more restricted to 4KB pages at a time)
* Handle potential access violations caused by partial unmap
* Implement host mapping using shared memory on Linux
* Add new GetPhysicalAddressChecked method, used to ensure the virtual address is mapped before address translation
Also align GetRef behaviour with software memory manager
* We don't need a mirrorable memory block for software memory manager mode
* Disable memory aliasing tests while we don't have shared memory support on Mac
* Shared memory & SIGBUS handler for macOS
* Fix typo + nits + re-enable memory tests
* Set MAP_JIT_DARWIN on x86 Mac too
* Add back the address space mirror
* Only set MAP_JIT_DARWIN if we are mapping as executable
* Disable aliasing tests again (still fails on Mac)
* Fix UnmapView4KB (by not casting size to int)
* Use ref counting on memory blocks to delay closing the shared memory handle until all blocks using it are disposed
* Address PR feedback
* Make RO hold a reference to the guest process memory manager to avoid early disposal
Co-authored-by: nastys <nastys@users.noreply.github.com>
* hle: Some cleanup
This PR cleaned up a bit the HLE folder and the VirtualFileSystem one, since we use LibHac, we can use some class of it directly instead of duplicate things. The "Content" of VFS folder is removed since it should be handled in the NCM service directly.
A larger cleanup should be done later since there is still be duplicated code here and there.
* Fix Headless.SDL2
* Addresses gdkchan feedback
* De-tile GOB when DMA copying from block linear to pitch kind memory regions
* XML docs + nits
* Remove using
* No flush for regular buffer copies
* Add back ulong casts, fix regression due to oversight
* oslc: Fix condition in GetSaveDataBackupSetting
This PR fixes a condition previously implemented in #3190 where ACNH can't be booted without an existing savedata.
Closes#3206
* Addresses gdkchan feedback
* ntc: Implement IEnsureNetworkClockAvailabilityService
This PR implement a basic `IEnsureNetworkClockAvailabilityService` checked by RE. It's needed by Splatoon 2 with Guest Internet Access enabled. Game is now playable with this setting.
* Update Ryujinx.HLE/HOS/Services/Nim/Ntc/StaticService/IEnsureNetworkClockAvailabilityService.cs
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
* Update IGeneralService.cs
Fix IPV4 local ip related frame drop in fire emblem by rewriting [CommandHipc(12)]
* Fix IPV4 Local IP Slowdown & Style Fixes
fix a missing space
* Remove unnecessary line
* Fix for hardcoding which index to use
* Replace argument with empty string.
By sending an empty string to Dns.GetHostAddresses("") you get back localhost info only.
* Add caching, undo change in GetCurrentIpAddress
Implement caching and revert the GetCurrentIP() function, speed improvements still present.
* Remove unnecessary using
* Syntax fixes and removing extra lines
Requested changes by AcK77
* Properly unsubscribe from event handler
Adds an unsubscribe in the dispose section of IGeneralService
* olsc: Implement GetSaveDataBackupSetting
This PR implement GetSaveDataBackupSetting of OLSC service which is now needed by ACNH 2.0.5. The game is playable as usual if you use the same user profile as the original save file (I don't know if it was the case before), everything is checked by RE.
* addresses gdkchan feedback
* Collapse AsSpan().Slice(..) calls into AsSpan(..)
Less code and a bit faster
* Collapse an Array.Clear(array, 0, array.Length) call to Array.Clear(array)
This should prevent filesystem services from blocking other services that don't have their own ServerBase. May improve filesystem related stutters in certain titles.
Improves button advanced cutscenes such as Miqol's Request in Xenoblade: DE when the game is on a network share (used to stutter when voice lines played).
Should probably be tested to make sure no mysterious bugs have been unearthed, and to see if any other filesystem related perf issues are improved.
* Implement/Stub mnpp:app service and some hid calls
This PR Implement/Stub the `mnpp:app` service (closes#3107) accordingly to RE, which seems to do some telemetry for China region only, so everything is stubbed.
This PR fixes some inconsistencies in the hid service too and stub EnableSixAxisSensorUnalteredPassthrough, IsSixAxisSensorUnalteredPassthroughEnabled, LoadSixAxisSensorCalibrationParameter, GetSixAxisSensorIcInformation calls (closes#3123 and closes#3124).
* Addresses Thog review
* added trace log level
* use trace log level instead of debug ( #1547)
* alignment #1547
* moved trace logs toggle at the bottom #1547
* bumped config file version #3096
* added migration step #3096
* setting moved to the dev section #1547
* performance warning displayed when trace is enabled #1547
- Run the extra data fix in FixExtraData on non-system saves that have no owner ID.
- Set the owner ID in the dummy application control property if an application doesn't have a proper one available.
* Start refactoring kernel a bit and import some changes from kernel decoupling PR
* kernel: Put output always at the start in Syscall functions
* kernel: Rewrite GetThreadContext3 to use a structure and to be accurate
* kernel: make KernelTransfer use generic types and simplify
* Fix some warning and do not use getters on MemoryInfo
* Address gdkchan's comment
* GetThreadContext3: use correct pause flag
* kernel: Fix deadlock when pinning in interrupt handler
This fix a deadlock on DoDonPachi Resurrection when starting a new game
* Address gdkchan's comment
* implement certain servicessl functions
* ssl: Implement more of SSL connection and abstract it
This adds support to non blocking SSL operations and unlink the SSL
implementation from the IPC logic.
* Rename SslDefaultSocketConnection to SslManagedSocketConnection
* Fix regression on Pokemon TV
* Address gdkchan's comment
* Simplify value read from previous commit
* ssl: some changes
- Implement builtin certificates parsing and retrieving
- Fix issues with SSL version handling
- Improve managed SSL socket error handling
- Ensure to only return a certificate on DoHandshake when actually requested
* Add missing BuiltInCertificateManager initialization call
* Address gdkchan's comment
* Address Ack's comment
Co-authored-by: InvoxiPlayGames <webmaster@invoxiplaygames.uk>
* bsd: Revamp API and make socket abstract
This part of the code was really ancient and needed some love.
As such this commit aims at separating the socket core logic from the IClient class and make it uses more modern APIs to read/write/parse data.
* Address gdkchan's comment
* Move TryConvertSocketOption to WinSockHelper
* Allow reusing old fds and add missing locks around SocketInternal and ShutdownAllSockets
* bsd: ton of changes
- Make sockets per process
- Implement eventfds
- Rework Poll for support of eventfds
- Handle protocol auto selection by type (used by gRPC)
- Handle IPv6 socket creation
* Address most of gdkchan comments
* Fix inverted read logic for BSD socket read
* bsd: Make Poll abstract via IBsdSocketPollManager
* bsd: Improve naming of everything
* Fix build issue from last commit (missed to save on VC)
* Switch BsdContext registry to a concurrent dictionary
* bsd: Implement socket creation flags logic and the non blocking flag
* Remove unused enum from previous commit
* bsd: Fix poll logic when 0 fds are present for a given poll manager and when timeout is very small (or 0)
* Address gdkchan's comment
* friend: Stub IsFriendListCacheAvailable and EnsureFriendListAvailable
This PR stubs IsFriendListCacheAvailable and EnsureFriendListAvailable call of friend service which close#2896.
Sadly, Super Bomberman R Online is still stuck on the loading screen and keep calling `TryPopFromFriendInvitationStorageChannel`, probably because another issue somewhere.
* Add FW version
* Apply suggestions from gdkchan
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
* Update IFriendService.cs
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
This PR stub `SetMediaPlaybackStateForApplication` of the am service. Accordingly to gdkchan it's needed by the Youtube app. This is checked by RE aswell.
* kernel: Implement Thread pinning support
This commit adds support for 8.x thread pinning changes and implement SynchronizePreemptionState syscall.
Based on kernel 13.x reverse.
* Address gdkchan's comment
* kernel: fix missing critical section leave in SetActivity
Fix Unity games
* Implement missing bits on the interrupt handler and inline update pinning function as it cannot be generic
* Fix some bugs in SetActivity and SetCoreAndAffinityMask
* Address gdkchan's comments
While I'm looking to the code, I've found some syntax issue, and a little inconsistencie between `ActivateNpad` and `ActivateNpadWithRevision`. Nothing more.
* Add the ability to toggle mute in the status bar.
* Add the ability to toggle mute in the status bar.
* Formatting fixes
* Add hotkey (F2) to mute
* Add default hotkey to config.json
* Add ability to change volume via slider.
* Fix Headless
* Fix SDL2 Problem : Credits to d3xMachina
* Remove unnecessary work
* Address gdk comments
* Toggling with Hotkey now properly restores volume to original level.
* Toggling with Hotkey now properly restores volume to original level.
* Update UI to show Volume % instead of Muted/Unmuted
* Clean up the volume ui a bit.
* Undo unintentionally committed code.
* Implement AudRen Support
* Restore intiial volume level in function definition.
* Finalize UI
* Finalize UI
* Use clamp for bounds check
* Use Math.Clamp for volume in soundio
* Address comments by gdkchan
* Address remaining comments
* Fix missing semicolon
* Address remaining gdkchan comment
* Fix comment
* Change /* to //
* Allow volume slider to change volume immediately.
Also force label text to cast to int to prevent decimals from showing in status bar
* Remove blank line
* Undo setting of volume level when "Cancel" is pressed.
* Fix allignment for settings window code
* kernel: Define InfoTYpe and make it less obscure when reading GetInfo
Also map ThreadTickCount to 25 instead of 0xF0000002 like 13.x kernel.
* kernel: Implement GetInfo IsApplication
* kernel: Implement GetInfo FreeThreadCount
The .NET 6 changes ended up causing a difference in the way we deserialise and would end up deserializing wrongly.
This commit fixes it by resoring previous behaviour while still fixing the illink warning that caused that change in the first place
* Don't blow up everything if a DLC file is missing
* change comment
* More correctly setting the "enabled" check box on dlc dialog for the add-on NSP based on the enabled state of all NCAs in the package.
* Update Ryujinx.HLE/HOS/ApplicationLoader.cs
Co-authored-by: Ac_K <Acoustik666@gmail.com>
Co-authored-by: Logan Stromberg <lostromb@microsoft.com>
Co-authored-by: Ac_K <Acoustik666@gmail.com>
* infra: Migrate to .NET 6
* Rollback version naming change
* Workaround .NET 6 ZipArchive API issues
* ci: Switch to VS 2022 for AppVeyor
CI is now ready for .NET 6
* Suppress WebClient warning in DoUpdateWithMultipleThreads
* Attempt to workaround System.Drawing.Common changes on 6.0.0
* Change keyboard rendering from System.Drawing to ImageSharp
* Make the software keyboard renderer multithreaded
* Bump ImageSharp version to 1.0.4 to fix a bug in Image.Load
* Add fallback fonts to the keyboard renderer
* Fix warnings
* Address caian's comment
* Clean up linux workaround as it's uneeded now
* Update readme
Co-authored-by: Caian Benedicto <caianbene@gmail.com>
* kernel: Fix sleep timing accuracy
This commit corrects some mistake while comparing reversing of kernel
13.x with our own.
WaitAndCheckScheduledObjects timing accuracy was also improved.
* Make KTimeManager.WaitAndCheckScheduledObjects spin wait for sub milliseconds
Fix performance regression on Pokemon Let's Go games and possibly
others.
* Address rip's comment
* kernel: Fix issues with timeout of -1 (0xFFFFFFFF)
Fixes possible hang on Pokemon DP and possibly others
Add basic support for the CFI value being passed in X18 since 11.0.0 by the official kernel.
We do not implement any random generator atm in the kernel and as such the KSystemControl.GenerateRandom function is stubbed
This PR implements/stubs some missing calls introduced in recent firmware (13.0.0).
- `acc:u0 InitializeApplicationInfoV2` needed by ACNH.
- `aoc:u NotifyMountAddOnContent/NotifyUnmountAddOnContent/CheckAddOnContentMountStatus` checked by RE. Needed by ACNH.
- `IPurchaseEventManager PopPurchasedProductInfo` needed by Dying Light.
Now ACNH 2.0 update is fully playable, and Dying Light can boot further:
* Remove comment from code.
* Added needed RegisterInfo for TODO.
* Forgot to add two words...
* Stop being tired and fixed the one issue.
* Removing suggested line from GDKchan.
* GDK seems to have been incorrect?
* Update Ryujinx.HLE/HOS/Services/Nfc/Nfp/VirtualAmiibo.cs
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
* Remove a bit of unsafety around
* Regenerate StructArrayHelpers with a max element value of 256
* hle: remove unsafe marker from all struct that had it
* hle: make SoftwareKeyboardRenderer.TryCopyTo safe
* hle: remove unsafety in NpadDevice and remove AllowUnsafeBlocks from csproj
* Revert "Regenerate StructArrayHelpers with a max element value of 256"
This reverts commit f32a6e5be0.
* Introduce ByteArray of various size and use that instead of ArrayXXX to avoid stackoverflow in .NET runtime type resolution
* Use ByteArray more
* Add some missing spaces on Pack = 1 for various structs
* Fix broken logic for TryCopyTo
* Address gdkchan's comment
* Address gdkchan's comment
* timezone: Make timezone implementation safe
* hle: Do not use TrimEnd to parse ASCII strings
This adds an util that handle reading an ASCII string in a safe way.
Previously it was possible to read malformed data that could cause
various undefined behaviours in multiple services.
* hid: Remove an useless unsafe modifier on keyboard update
* Address gdkchan's comment
* Address gdkchan's comment
* kernel: Clear pages allocated with SetHeapSize
Before this commit, all new pages allocated by SetHeapSize were not
cleared by the kernel.
This would cause undefined data to be pass to the userland and possibly
resulting in weird memory corruption.
This commit also add support for custom fill heap and ipc value (that is also
supported by the official kernel)
* Remove dots at the end of KPageTableBase.MapPages new documentation
* Remove unused _stackFillValue
* Fixup channel submit IOCTL syncpoint parameters
The current arguments worked by happenstance as games only ever submit
one syncpoint and request one fence back, if a game were to do something
other than this then the arguments would've been parsed entirely wrong.
* Address feedback
* kernel: Add resource limit related syscalls
This commit implements all resource limit related syscalls.
* Fix register mapping being wrong for SetResourceLimitLimitValue
* Address gdkchan's comment
* kernel: Implement SetMemoryPermission syscall
This commit implement the SetMemoryPermission syscall accurately.
This also fix KMemoryPermission not being an unsigned 32 bits type and
add the "DontCare" bit (used by shared memory, currently unused in
Ryujinx)
* Update MemoryPermission mask
* Address gdkchan's comments
* Fix a nit
* Address gdkchan's comment
* Fix race when EventWait is called and a wait is done on the CPU
* This is useless now
* Fix EventSignal
* Ensure the signal belongs to the current fence, to avoid stale signals
It seems that certain games (Link's Awakening, Xenoblade DE) had their fences reached already when posting framebuffers, so the signal that a frame was ready would go out _before_ the frame was enqueued, and the render loop would fail to dequeue anything and "skip" a frame.
This was resulting in their performance lowering dramatically after some loading transitions, as a frame signal would be consumed and presentation would be one frame behind.
It's possible this might have eventually caused deadlocks in these games or others, if it happened twice.
* spl: Implement IGeneralInterface GetConfig
This PR implement the GetConfig call of the spl service. This is currently needed for some homebrews which currently needs Ignore Missing Service to boot. Now it's fixed.
Implementation was done using Atmosphère code and REing too.
* Addresses gdkchan feedback
This PR stubs and implements some clkrst call because they are used to overclock the Switch hardware and it's pointless in our case as we emulate the system.
Everything was done checked by RE.
Fixes#2686
This PR stubs some irs service calls which are needed to get some games playable or at least bootable since we don't support IR data throught real JoyCon for now.
- Stubs `IIrSensorServer` `StopImageProcessor`, `RunMomentProcessor`, `RunClusteringProcessor`, `RunImageTransferProcessor`, `GetImageTransferProcessorState`, `RunTeraPluginProcessor`. All calls are a bit checked by RE.
Closes#2267, #2248, #2126
Night Vision and SpyAlarm are now bootable (but still unplayable due to the lack of the IR data):
* vi: Unify resolutions values and accurate implementation of them.
To continue what was made in #2618, I've REd `vi` service a bit. Now values and checks related to displays are more accurate.
- `am` GetDefaultDisplayResolution / GetDefaultDisplayResolutionChangeEvent have more informations on what the service does.
- `vi:u/vi:m/vi:s` GetDisplayService are now accurate.
- `IApplicationDisplay` GetRelayService, GetSystemDisplayService, GetManagerDisplayService, GetIndirectDisplayTransactionService, ListDisplays, OpenDisplay, OpenDefaultDisplay, CloseDisplay, GetDisplayResolution are now properly implemented.
- Some other calls are cleaned or have extra checks accordingly to RE.
Additionnaly, `IFriendService` have some wrong aligned things, and `pm:info` service placeholder was missing.
* just use _openedDisplayInfo.Remove()
* use context.Memory.Fill()
* fix some casting
* remove unneeded comment
* cleanup
* uses TryAdd
* displayId > ulong
* GetDisplayResolution > ulong
* UL
* Disable Pause/Resume menu instead of trying to hide them
* Fix Resume menu being active before renderer starts
* Fix emulator not being able to close properly
* hos: Cleanup the project
Since a lot of changes has been done on the HOS project, there are some leftover here and there, or class just used in one service, things at wrong places, and more.
This PR fixes that, additionnally to that, I've realigned some vars because I though it make the code more readable.
* Address gdkchan feedback
* addresses Thog feedback
* Revert ElfSymbol
* Add a "Pause Emulation" option and hotkey
Closes Ryujinx#1604
* Refactoring how pause is handled
* Applied suggested changes from review
* Applied suggested fixes
* Pass correct suspend type to threads for suspend/resume
* Fix NRE after stoping emulation
* Removing SimulateWakeUpMessage call after resuming emulation
* Skip suspending non game process
* Pause the tickCounter in the ExecutionContext
* Refactoring tickCounter pause/resume as suggested
* Fix Config migration to add pause hotkey
* Fixed pausing only application threads
* Fix exiting emulator while paused
* Avoid pause/resume while already paused/resumed
* Cleanup unused code
* Avoid restarting audio if stopping emulation while in pause.
* Added suggested changes
* Fix ConfigurationState
* Initial Implementation
About as fast as nvidia GL multithreading, can be improved with faster command queuing.
* Struct based command list
Speeds up a bit. Still a lot of time lost to resource copy.
* Do shader init while the render thread is active.
* Introduce circular span pool V1
Ideally should be able to use structs instead of references for storing these spans on commands. Will try that next.
* Refactor SpanRef some more
Use a struct to represent SpanRef, rather than a reference.
* Flush buffers on background thread
* Use a span for UpdateRenderScale.
Much faster than copying the array.
* Calculate command size using reflection
* WIP parallel shaders
* Some minor optimisation
* Only 2 max refs per command now.
The command with 3 refs is gone. 😌
* Don't cast on the GPU side
* Remove redundant casts, force sync on window present
* Fix Shader Cache
* Fix host shader save.
* Fixup to work with new renderer stuff
* Make command Run static, use array of delegates as lookup
Profile says this takes less time than the previous way.
* Bring up to date
* Add settings toggle. Fix Muiltithreading Off mode.
* Fix warning.
* Release tracking lock for flushes
* Fix Conditional Render fast path with threaded gal
* Make handle iteration safe when releasing the lock
This is mostly temporary.
* Attempt to set backend threading on driver
Only really works on nvidia before launching a game.
* Fix race condition with BufferModifiedRangeList, exceptions in tracking actions
* Update buffer set commands
* Some cleanup
* Only use stutter workaround when using opengl renderer non-threaded
* Add host-conditional reservation of counter events
There has always been the possibility that conditional rendering could use a query object just as it is disposed by the counter queue. This change makes it so that when the host decides to use host conditional rendering, the query object is reserved so that it cannot be deleted. Counter events can optionally start reserved, as the threaded implementation can reserve them before the backend creates them, and there would otherwise be a short amount of time where the counter queue could dispose the event before a call to reserve it could be made.
* Address Feedback
* Make counter flush tracked again.
Hopefully does not cause any issues this time.
* Wait for FlushTo on the main queue thread.
Currently assumes only one thread will want to FlushTo (in this case, the GPU thread)
* Add SDL2 headless integration
* Add HLE macro commands.
Co-authored-by: Mary <mary@mary.zone>
