* Implement faster address translation and write tracking on the MMU
* Rename MemoryAlloc to MemoryManagement, and other nits
* Support multi-level page tables
* Fix typo
* Reword comment a bit
* Support scalar vector loads/stores on the memory fast path, and minor fixes
* Add missing cast
* Alignment
* Fix VirtualFree function signature
* Change MemoryProtection enum to uint aswell for consistency
* Implement ConvertScalingMode properly
* Fix up the naming
* Only values 2 and 4 are allowed
* Return a nullable enum from ConvetScalingMode
* Fix typo on method name
* Use convertedScalingMode
* Implement ARM exclusive load/store with compare exchange insts, and enable multicore by default
* Fix comment typo
* Support Linux and OSX on MemoryAlloc and CompareExchange128, some cleanup
* Use intel syntax on assembly code
* Adjust identation
* Add CPUID check and fix exclusive reservation granule size
* Update schema multicore scheduling default value
* Make the cpu id check code lower case aswell
* Make it possibles to load hb-loader and hb-menu
One issue remains with hb-menu homebrew icons because of SIMD issues
(libjpeg-turbo related) and netloader doesn't work.
* Implement GetApplicationControlData
* Fix shared fonts for NSO/NRO
* Add homebrew NRO romfs support
This readd the NRO support by parsing the ASET header
* Address comments about HomebrewRomFs
* override Dispose in homebrew romfs stream
* Use a struct for file timestamp
* Simplify positional increments in GetApplicationControlData
* Address comments
* improve readability of the memory permission check in SetProcessMemoryPermission
* Fix previous broken check
* Add address space checks in SetProcessMemoryPermission
* Logging: Refactor log targets into Ryujinx.Common
* Logger: Implement JSON Log Target
* Logger: Optimize Console/File logging targets
Implement a simple ObjectPool to pool up StringBuilders to avoid causing excessive GCing of gen1/2 items when large amounts of log entries are being generated.
We can also pre-determine the async overflow action at initialization time, allowing for an easy optimization in the message enqueue function, avoiding a number of comparisons.
* Logger: Implement LogFormatters
* Config: Refactor configuration file and loading
* Config: Rename to .jsonc to avoid highlighting issues in VSC and GitHub
* Resolve style nits
* Config: Resolve incorrect default key binding
* Config: Also update key binding default in schema
* Tidy up namespace imports
* Config: Update CONFIG.md to reflect new Config file
* Implement speculative translation on the cpu, and change the way how branches to unknown or untranslated addresses works
* Port t0opt changes and other cleanups
* Change namespace from translation related classes to ChocolArm64.Translation, other minor tweaks
* Fix typo
* Translate higher quality code for indirect jumps aswell, and on some cases that were missed when lower quality (tier 0) code was available
* Remove debug print
* Remove direct argument passing optimization, and enable tail calls for BR instructions
* Call delegates directly with Callvirt rather than calling Execute, do not emit calls for tier 0 code
* Remove unused property
* Rename argument on ArmSubroutine delegate
* Implement ARM32 memory instructions: LDM, LDR, LDRB, LDRD, LDRH, LDRSB, LDRSH, STM, STR, STRB, STRD, STRH (immediate and register + immediate variants), implement CMP (immediate and register shifted by immediate variants)
* Rename some opcode classes and flag masks for consistency
* Fix a few suboptimal ARM32 codegen issues, only loads should be considered on decoder when checking if Rt == PC, and only NZCV flags should be considered for comparison optimizations
* Take into account Rt2 for LDRD instructions aswell when checking if the instruction changes PC
* Re-align arm32 instructions on the opcode table
* Remove ARM32 interpreter and add ARM32 support on the translator
* Nits.
* Rename Cond -> Condition
* Align code again
* Rename Data to Alu
* Enable ARM32 support and handle undefined instructions
* Use the IsThumb method to check if its a thumb opcode
* Remove another 32-bits check
* Implement some IPC related kernel SVCs properly
* Fix BLZ decompression when the segment also has a uncompressed chunck
* Set default cpu core on process start from ProgramLoader, remove debug message
* Load process capabilities properly on KIPs
* Fix a copy/paste error in UnmapPhysicalMemory64
* Implement smarter switching between old and new IPC system to support the old HLE services implementation without the manual switch
* Implement RegisterService on sm and AcceptSession (partial)
* Misc fixes and improvements on new IPC methods
* Move IPC related SVCs into a separate file, and logging on RegisterService (sm)
* Some small fixes related to receive list buffers and error cases
* Load NSOs using the correct pool partition
* Fix corner case on GetMaskFromMinMax where range is 64, doesn't happen in pratice however
* Fix send static buffer copy
* Session release, implement closing requests on client disconnect
* Implement ConnectToPort SVC
* KLightSession init
* Refactor Ryujinx.Common and HLE Stub Logging
* Resolve review comments
* Rename missed loop variable
* Optimize PrintStub logging function
* Pass the call-sites Thread ID through to the logger
* Remove superfluous lock from ConsoleLog
* Process logged data objects in the logger target
Pass the data object all the way to the output logger targets, to allow them to "serialize" this in whatever appropriate format they're logging in.
* Use existing StringBuilder to build the properties string
* Add a ServiceNotImplemented Exception
Useful for printing debug information about unimplemented service calls
* Resolve Style Nits
* Resolve Merge Issues
* Fix typo and align declarations
* Initial fixes for last release of libnx
For now, the framebuffer aren't okay but it will not crash/
* Improve code reaadability in NvFlinger parsing
* Make surfaces access more userfriendly
* Add ColorFormat
* Fix code style in ColorFormat.cs
* Add multiple framebuffer support in nvnflinger
This fix libnx console rendering
* Move ReadStruct/WriteStruct to Ryujinx.Common
* fix the last nit
* Fix inverted color for R5G6B5
Also add some other format that libnx might uses.
* Remove hardcoded BlockHeight in nvflinger
* Fix issues with compressed NSO without a section
Avoid trying to decompress something that doesn't exist
* don't use the lengths of the buffer directly
* Optimized memory modified check
This was initially in some cases more expensive than plainly sending the data. Now it should have way better performance.
* Small refactoring
* renamed InvalidAccessEventArgs
* Renamed PtPageBits
* Removed ValueRange(set)
They are currently unused and won't be likely to be used in the near future
* Initial nvdec implementation using FFmpeg
* Fix swapped channels on the video decoder and the G8R8 texture format
* Fix texture samplers not being set properly (regression)
* Rebased
* Remove unused code introduced on the rebase
* Add support for RGBA8 output format on the video image composer
* Correct spacing
* Some fixes for rebase and other tweaks
* Allow size mismatch on frame copy
* Get rid of GetHostAddress calls on VDec
* Initial implementation of KProcess
* Some improvements to the memory manager, implement back guest stack trace printing
* Better GetInfo implementation, improve checking in some places with information from process capabilities
* Allow the cpu to read/write from the correct memory locations for accesses crossing a page boundary
* Change long -> ulong for address/size on memory related methods to avoid unnecessary casts
* Attempt at implementing ldr:ro with new KProcess
* Allow BSS with size 0 on ldr:ro
* Add checking for memory block slab heap usage, return errors if full, exit gracefully
* Use KMemoryBlockSize const from KMemoryManager
* Allow all methods to read from non-contiguous locations
* Fix for TransactParcelAuto
* Address PR feedback, additionally fix some small issues related to the KIP loader and implement SVCs GetProcessId, GetProcessList, GetSystemInfo, CreatePort and ManageNamedPort
* Fix wrong check for source pages count from page list on MapPhysicalMemory
* Fix some issues with UnloadNro on ldr:ro
* Implement contentmanager and related services
* small changes
* read system firmware version from nand
* add pfs support, write directoryentry info for romfs files
* add file check in fsp-srv:8
* add support for open fs of internal files
* fix filename when accessing pfs
* use switch style paths for contentpath
* close nca after verifying type
* removed publishing profiles, align directory entry
* fix style
* lots of style fixes
* yasf(yet another style fix)
* yasf(yet another style fix) plus symbols
* enforce path check on every fs access
* change enum type to default
* fix typo
* Better implementation of the DMA pusher, misc fixes
* Remove some debug code
* Correct RGBX8 format
* Add support for linked Texture Sampler Control
* Attempt to fix upside down screen issue
* Started to implement the hwopus service
* Write outputs on decode method, some basic error handling
* Fix buffer size read from header and check
* Fix order of values
* Audio: Implement libsoundio as an alternative audio backend
libsoundio will be preferred over OpenAL if it is available on the machine. If neither are available, it will fallback to a dummy audio renderer that outputs no sound.
* Audio: Fix SoundIoRingBuffer documentation
* Audio: Unroll and optimize the audio write callback
Copying one sample at a time is slow, this unrolls the most common audio channel layouts and manually copies the bytes between source and destination. This is over 2x faster than calling CopyBlockUnaligned every sample.
* Audio: Optimize the write callback further
This dramatically reduces the audio buffer copy time. When the sample size is one of handled sample sizes the buffer copy operation is almost 10x faster than CopyBlockAligned.
This works by copying full samples at a time, rather than the individual bytes that make up the sample. This allows for 2x or 4x faster copy operations depending on sample size.
* Audio: Fix typo in Stereo write callback
* Audio: Fix Surround (5.1) audio write callback
* Audio: Update Documentation
* Audio: Use built-in Unsafe.SizeOf<T>()
Built-in `SizeOf<T>()` is 10x faster than our `TypeSize<T>` helper. This also helps reduce code surface area.
* Audio: Keep fixed buffer style consistent
* Audio: Address styling nits
* Audio: More style nits
* Audio: Add additional documentation
* Audio: Move libsoundio bindings internal
As per discussion, moving the libsoundio native bindings into Ryujinx.Audio
* Audio: Bump Target Framework back up to .NET Core 2.1
* Audio: Remove voice mixing optimizations.
Leaves Saturation optimizations in place.
* Change naming convention for Ryujinx project
* Change naming convention for ChocolArm64 project
* Fix NaN
* Remove unneeded this. from Ryujinx project
* Adjust naming from new PRs
* Name changes based on feedback
* How did this get removed?
* Rebasing fix
* Change FP enum case
* Remove prefix from ChocolArm64 classes - Part 1
* Remove prefix from ChocolArm64 classes - Part 2
* Fix alignment from last commit's renaming
* Rename namespaces
* Rename stragglers
* Fix alignment
* Rename OpCode class
* Missed a few
* Adjust alignment
* Timing: Optimize Timestamp Aquisition
Currently, we make use of Environment.TickCount in a number of places. This has some downsides, mainly being that the TickCount is a signed 32-bit integer, and has an effective limit of ~25 days before overflowing and wrapping around. Due to the signed-ness of the value, this also caused issues with negative numbers. This resolves these issues by using a 64-bit tick count obtained from Performance Counters (via the Stopwatch class). This has a beneficial side effect of being significantly more accurate than the TickCount.
* Timing: Rename ElapsedTicks to ElapsedMilliseconds and expose TicksPerX
* Timing: Some style changes
* Timing: Align static variable initialization
This should provide accurate behaviours.
This implementation has been tested with ftpd and libtransistor bsd tests.
This implementation lacks OOB support.
* Print stack trace on invalid memory accesses
* Rebased, change code region base address for 39-bits address space, print stack trace on break and undefined instructions too
* Implement IRoInterface
This is required by Super Mario Party.
This commit also adds MapProcessCodeMemory and UnmapProcessCodeMemory functions in KMemoryManager. Those two calls might not reflect what the SVC of the same names do.
* Fix some code style issues
* Use MakeError to clarify error code
* Add NRR and NRO constants
* Fix some codestyle issues
* Fix InvalidMemoryState error code
* Simplify render target bindings
* Implement multiple viewports
* Pack glViewportIndexed calls into a single glViewportArray
* Use ARB_viewport_array when available
* Cache framebuffer attachments
* Use get accessors in OGLExtension
* Address feedback
* Full 5.X stubbed IHidServer
Since we can't support all those Hid calls in the right way, we can stub them with more information as possible in the logs.
I have added all symbols in it to be more revelant as possible.
It's remove some Hid spam in few games too, because we assign some var as the game want.
* Fix issues
* Fix code according to review
* Remove unused tracing functionality from the CPU
* GetNsoExecutable -> GetExecutable
* Unsigned comparison
* Re-add cpu tracing
* Config change
* Remove cold methods from the translation cache on the cpu
* Replace lock with try lock, pass new ATranslatorCache instead of ATranslator
* Rebase fixups
* support reading control data
* show game info on titlebar
* use first language is default is not available
* use seperate language enums for titles
* fix hex display
* Started to rewrite the thread scheduler
* Add a single core-like scheduling mode, enabled by default
* Clear exclusive monitor on context switch
* Add SetThreadActivity, misc fixes
* Implement WaitForAddress and SignalToAddress svcs, misc fixes
* Misc fixes (on SetActivity and Arbiter), other tweaks
* Rebased
* Add missing null check
* Rename multicore key on config, fix UpdatePriorityInheritance
* Make scheduling data MLQs private
* nit: Ordering
* (Re)Implement format reinterpretation, other changes
* Implement writeback to guest memory, some refactoring
* More refactoring, implement reinterpretation the old way again
* Clean up
* Some fixes on M2MF (old Dma engine), added partial support for P2MF, fix conditional ssy, add Z24S8 zeta format, other fixes
* nit: Formatting
* Address PR feedback
* Rewrite the C++ Demangler
This new Demangler provides support to almost every possible mangled
symbols and should behaves like GNU c++filt.
It works on 98.9% of the sdk's symbols and 99.5%
of Puyo Puyo Tetris's symbols.
* Fix code style
* Fix noexcept enclosed expression parsing issues
* fix code style issues
* Initial Framerate limit implementation
* use seperate event for limiter
* check for vsync signal after queue up framebuffer
* removed ingame toggle
* fix nits
* Move GPU LLE emulation from HLE to Graphics
* Graphics: Move Gal/Texture to Texture
* Remove Engines/ directory and namespace
* Use tables for image formats
* Abstract OpCode decoding
* Simplify image table
* Do not leak Read* symbols in TextureReader
* Fixups
* Rename IGalFrameBuffer -> IGalRenderTarget
* Remove MaxBpp hardcoded value
* Change yet again texture data and add G8R8 flipping
* Rename GalFrameBufferFormat to GalSurfaceFormat
* Unident EnsureSetup in ImageHandler
* Add IsCompressed
* Address some feedback
* Update BSD service to handle libnx's 'smart IPC buffers' for address info
* Use existing "GetBufferType0x21" for certain BSD socket methods
* Parse address port as unsigned short
* Fix bounds check on reading the IPC buffer
* Implement Read, Write methods
* rebased and cleaned
* addressed nits
* remove unused swap method
* fixed alignments
- Manage end of the log packet correctly.
- Add drop count, time, and program name parsing.
- Use the correct buffer type. (0x21 not 0x9)
- Prefix unknown fields with "Field"
* Add multiple color outputs for fragment shaders
* Add registers and gal enums
* Use textures for framebuffers and split color and zeta framebuffers
* Abstract texture and framebuffer targets as an image
* Share images between framebuffers and textures
* Unstub formats
* Add some formats
* Disable multiple attachments
* Cache framebuffer attachments
* Handle format types
* Add some rendertarget formats
* Code cleanup
* Fixup half float types
* Address feedback
* Disable multiple attachments in shaders
* Add A4B4G4R4 image format
* Add reversed section for image enums
* Fix MapPhysicalMemory when block position is less than map position
* Only try to free pages that are actually mapped
* Only unmap after freeing the memory
* Some style fixes and nits on ITimeZoneService
* Remove some unneeded usings
* Remove the Ryujinx.HLE.OsHle.Handles namespace
* Remove hbmenu automatic load on process exit
* Rename Ns to Device, rename Os to System, rename SystemState to State
* Move Exceptions and Utilities out of OsHle
* Rename OsHle to HOS
* Rename OsHle folder to HOS
* IManagerDisplayService and ISystemDisplayService style fixes
* BsdError shouldn't be public
* Add a empty new line before using static
* Remove unused file
* Some style fixes on NPDM
* Exit gracefully when the application is closed
* Code style fixes on IGeneralService
* Add 0x prefix on values printed as hex
* Small improvements on finalization code
* Move ProcessId and ThreadId out of AThreadState
* Rename VFs to FileSystem
* FsAccessHeader shouldn't be public. Also fix file names casing
* More case changes on NPDM
* Remove unused files
* Move using to the correct place on NPDM
* Use properties on KernelAccessControlMmio
* Address PR feedback
* Keep track mapped buffers with fixed offsets
* Started rewriting the memory manager
* Initial support for MapPhysicalMemory and UnmapPhysicalMemory, other tweaks
* MapPhysicalMemory/UnmapPhysicalMemory support, other tweaks
* Rebased
* Optimize the map/unmap physical memory svcs
* Integrate shared font support
* Fix address space reserve alignment
* Some fixes related to gpu memory mapping
* Some cleanup
* Only try uploading const buffers that are really used
* Check if memory region is contiguous
* Rebased
* Add missing count increment on IsRegionModified
* Check for reads/writes outside of the address space, optimize translation with a tail call
* Implement Shared Fonts
This fully implements shared fonts.
This commit is provided without fonts.
This commit also add Size to HSharedMem.Positions to be able to add fonts to shared zones when RequestLoad is called.
* Require the user to provide fonts in RyuFS/system
* Use File.Exits instead of relying ona try/catch and change system resource exception format a bit
* Make sure that font sum doesn't exceed 17MB
Also rename font data dictionary for coherence.
* Skip repeated cache tests between same sync
* Skip some checks for regions where just one resource is resident
* Dehardcode residency page size
* Some cleanup
* Directly send host address to buffer data
* Cleanup OGLShader
* Directly copy vertex and index data too
* Revert shader bind "cache"
* Address feedback
* Attempt to support deswizzle of sparse tiled textures
* Use correct frame buffer and viewport sizes, started to clean up the copy engine
* Correct texture width alignment
* Use Scale/Translate registers to calculate viewport rect
* Allow texture copy between frame buffers
This commit implements the argv config key in Ryujinx (by creating a temporary copy of the homebrew executable in the sdmc VFS) to make it possible
to load libnx's "romfs" files.
This commit also call Os.Dispose in Ns.OnFinish to dispose all resources when exiting
* Partial voice implementation on audio renderer
* Implemented audren resampler (based on original impl)
* Fix BiquadFilter struct
* Pause audio playback on last stream buffer
* Split audren/audout files into separate folders, some minor cleanup
* Use AudioRendererParameter on GetWorkBufferSize aswell
* Bump audren version to REV4, name a few things, increase sample buffer size
* Remove useless new lines
* return correct size of ASTC 5x5, 6x6, 8x8, 10x10 and 12x12
* return correct Bytes Per Pixel
* Use method in order to get CompressedTextureSize
* Add Read16BptCompressedTexture method
* add Bpb integer argument
* Loop GLScreen with custom method
* Fix deadlocks
* Fix screen resizing
* Change event to bool
* Try to fix quitting error
* Set title from main thread
* Queue max 1 vsync, fix high FPS after a slowdown
* GetConfig should return 0x30006 in production mode
* GetConfig will now check settings only if nv!rmos_set_production_mode is set to "0"
* Code formatting, TryGetValue
* Slight fixup
* dont forget the setting
* Implemented non production mode setting grabbing
* format issue
* style changes
