Initial PR was made by Avan for Ryubing.
This allows for Trails in the Sky 1st Chapter (and by extension Trails through Daybreak, Trails through Daybreak II, Trails Beyond the Horizon, and the future Trails in the Sky 2nd Chapter – tested via demo – to be fully playable.
Initial PR caused issues on macOS devices, specifically crashing in certain games (such as Mario Kart 8 Deluxe and the aforementioned Trails games). It was reverted on the original Ryubing project (at the time of writing this description) for causing small rendering issues on The Legend of Zelda: Breath of the Wild.
The rendering issues do not appear on macOS.
This commit improves the original PR by eliminating said macOS crash.
Trails in the Sky 1st writes its exposure/brightness value through an R32G32Float 1x1 texture and later reads the raw bits back through an R32Uint 2x1 texture mapped to the same guest memory. The two texture formats/dimensions are fully incompatible as texture views, so they end
up as separate host textures, and nothing kept their contents coherent - the reader saw stale zeroed data, making the exposure calculation (and the rendered image) too dark. (This is the initial fix for the game)
TextureGroup already had a raw-copy dependency mechanism intended to
handle this class of alias (TextureDependency, TextureGroupHandle
raw-copy plumbing), gated by CanCreateRawCopyDependency's strict same-guest-memory checks. This commit fixes three bugs that kept that mechanism from working correctly (and thus crashing Trails and other games, such as Mario Kart 8 Deluxe):
(1) TextureGroup.InitializeOverlaps() and TextureGroup.RegisterIncompatibleOverlap() only forwarded overlaps to CreateCopyDependency() when compatibility was LayoutIncompatible or better, silently excluding the fully Incompatible case the raw-copy path exists for. Widened both guards to let Incompatible overlaps through so CanCreateRawCopyDependency actually gets a chance to run.
(2) Once Incompatible overlaps were allowed through, CreateCopyDependency(TextureGroup, ...) could still fall back to a regular, non-raw textureCopy for such pairs, since ViewLayoutCompatible/CopySizeMatches only check byte size and were never meant to reason about fully incompatible pairs (e.g. a depth format aliasing a color format with the same byte size). On Vulkan running through MoltenVK on macOS, that non-raw copy path requires a pixel-format-reinterpreting texture view, and MoltenVK/Metal refuses to create any view onto a depth-format texture ("not castable"), aborting emulation. textureCopy is now forced false for Incompatible pairs, leaving raw copy (which already excludes depth/stencil formats) as the only route for that severity level.
(3) TextureGroupHandle.Inherit() copied a handle's pending DeferredCopy to the new handle when a view was recreated, but did not copy DeferredCopyRaw alongside it. A handle that inherited a pending raw copy would then execute it through the regular (non-raw) CopyTo path once acknowledged, reinterpreting the source bytes with the wrong row layout and corrupting the image - visible as vertical flickering stripes whenever a view happened to be recreated with a raw copy still pending. DeferredCopyRaw is now carried over together with DeferredCopy.
(As mentioned earlier, vertical stripes aren't present on macOS, so (3) doesn't negatively have an effect on macOS in any way).
It has been confirmed that Divinity: Original Sin 2 may describe the same or overlapping guest GPU memory using BCn-compressed textures with different logical dimensions, such as 104×104 and 102×102.
BCn formats store texture data in blocks, with each BC block covering 4×4 texels. After rounding the dimensions up to complete blocks, both 104×104 and 102×102 textures require a 26×26 grid of BC blocks. As a result, they have the same block footprint at the base mip level. Ryujinx originally determined size compatibility in TextureCompatibility.ViewSizeMatches() primarily from this block footprint, which could cause these textures to be classified as Full view-compatible and allowed to share the same Vulkan VkImage directly.
However, because the two textures have different logical dimensions, their mip chains are also different. For example, at mip level 3, a 104×104 texture is reduced to 13×13, while a 102×102 texture is reduced to 12×12. If a 104×104 child texture view shares the backing image of a 102×102 parent texture, a subsequent full mip upload may attempt to write 13×13 compressed data into a Vulkan image subresource whose actual dimensions are only 12×12.
This produces a compressed buffer-to-image copy that does not match the geometry of the backing image. After the invalid command is submitted to the GPU, the Vulkan driver may asynchronously report VK_ERROR_DEVICE_LOST during command buffer submission, waiting, or presentation. In Ryujinx, this eventually appears as: VulkanException: Unexpected API error "ErrorDeviceLost".
The fix adds a logical-dimension check to TextureCompatibility.ViewSizeMatches(). If either the parent or child uses a compressed texture format, and the actual width or height of the corresponding parent mip does not match the logical width or height of the child, the relationship is no longer classified as Full view-compatible. Instead, it is downgraded to CopyOnly.
CopyOnly indicates that the two textures may still describe the same or overlapping guest memory and that their contents must remain synchronized, but they cannot directly share a single fixed-size Vulkan image. Ryujinx instead allows each texture to retain a host texture matching its own logical dimensions and synchronizes their contents through the existing copy-dependency mechanism.
As a result, the 13×13 mip of the 104×104 texture is uploaded to an actual 13×13 destination mip, while the 12×12 mip of the 102×102 texture is uploaded to an actual 12×12 destination mip.
Fix an "Invalid texture format 0x25A5A (sRGB: True)" error that occurs when running OCTOPATH TRAVELER 0.
The game may leave texture descriptor heap entries filled with 0x5A. Interpreting such uninitialized entries as valid texture descriptors produces an invalid format and causes texture creation to proceed with garbage descriptor data.
Validate the texture address before decoding the descriptor. Descriptors with a zero or unmapped address are marked as invalid and skipped. The invalid state is cleared when the corresponding texture pool entry is modified, allowing the descriptor to be evaluated again.
Trails in the Sky 1st uses texture descriptors with different formats and dimensions that map to the same region of guest GPU memory. For example, the game interprets the same underlying data as both an R32Uint 2x1 texture and an R32G32Float 1x1 texture. Both textures have an 8-byte logical payload, but their texel formats and dimensions are different.
Ryubing cannot represent these two descriptors as normally compatible texture views, so it creates separate host texture objects for them. Since these host textures represent the same guest memory, a modification made to one of them should become visible to the other before the other texture is read. However, the existing synchronization mechanism does not fully handle this situation.
The game uses this pattern for its exposure or brightness history. A compute shader first writes the updated exposure data to the R32G32Float 1x1 host texture. A later operation then reads the raw bits of the same data through the R32Uint 2x1 host texture. The original Ryubing implementation did not synchronize the contents of these two separate host textures. As a result, the reading texture received stale zero values, which caused the shader to calculate an incorrectly low exposure value and made the rendered image appear too dark.
TextureGroup already has a guest-memory synchronization mechanism for incompatible overlaps. When _flushIncompatibleOverlaps is enabled, the most recent texture contents are written back to guest memory, after which the other texture reloads the data. However, this mechanism was primarily enabled through IsFormatHostIncompatible, which checks whether an individual texture format can be correctly supported by the host GPU.
In this case, both R32Uint and R32G32Float are individually supported by the host GPU. Therefore, IsFormatHostIncompatible does not enable this synchronization path. The issue is not that either format is unsupported. The issue is that two individually supported aliases, represented by separate host textures, do not maintain coherent contents.
In theory, the latest contents could be flushed to guest memory immediately after every GPU write to a host texture, allowing other textures that map to the same guest memory to reload the updated data. However, doing so could frequently trigger expensive operations such as GPU waits, data readbacks, guest texture layout conversions, and memory-tracking updates. This would be particularly costly for textures that are updated every frame or during every compute dispatch.
For this reason, the fix does not require an immediate flush after every write. Instead, it extends Ryubing's existing TextureDependency and TextureGroupHandle mechanisms. When two fully incompatible textures map to exactly the same guest memory and satisfy a strict set of raw-copy requirements, the program creates a raw-copy dependency between them.
When the source texture is modified, the target texture is only marked as requiring a raw copy. The latest logical raw bytes are copied from the source only when the target texture is actually about to be used. This keeps the contents of the separate host texture aliases coherent while avoiding unnecessary round trips through guest memory.
if a buffer is inherited, ignore it and remove it from the list.
fixes a crash when a buffer is inherited during a sync and the containing ranges have been modified to not fit in the old buffer.
General memory improvements to decrease GC pressure and frequency.
Pool big arrays and objects that are created and deleted often.
Skip data copies when they aren't needed.
Inline flag checks to skip unneeded allocations.
From my testing the performance is about the same, but the GC frequency is much lower and collection is faster causing less and smaller spikes.
Add support for sampler sRGB disable (#7312)
Change image format view handling to allow view incompatible formats (#7311)
* Allow creating texture aliases on texture pool
* Delete old image format override code
* New format incompatible alias
* Missing bounds check
* GetForBinding now takes FormatInfo
* Make FormatInfo struct more compact
* "Vulkan: Feedback loop detection and barriers (#7226)"
* "Change image format view handling to allow view incompatible formats (#7311)"
* "Add support for sampler sRGB disable (#7312)"
* These changes prevent slowdowns in several areas in Paper Mario: The Thousand Year Door, possibly other games.
* Add Texture Size Capacity and 8GB Dram Build
* Update AutoDeleteCache.cs
* Dynamic Texture Cache (WIP)
* Change to float Multiplier, in-case it needs fine-tuning.
* Delete src/src.sln
* Update AutoDeleteCache.cs
* Format
* Fix Formatting
* Add DefaultTextureSizeCapacity and MemoryScaleFactor
- Also remove redundant New Lines
* Fix 4GB dram crashing
* Format newline
* Refractor
- Added Initialize() function to TextureCache and AutoDeleteCache
- Removed GetMaxTextureCapacity() function and instead added _maxCacheMemoryUsage
- Added private const MaxTextureSizeCapacity to AutoDelete Cache
- Added TextureCache.Initialize() to MemoryManager in order to fetch MaxGpuMemory at the right time.
- Moved and Changed Logger.Info for Gpu Memory to Logger.Notice and Moved it to PrintGpuInformation function.
- Opted to use a ternary operator for the Initialize function, I think it looks cleaner than bunch of if statements.
* Update src/Ryujinx.Graphics.Gpu/Image/AutoDeleteCache.cs
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
* maxMemory to CacheMemory, use Clamp instead of Ternary. Changed MinTextureCapacity 1GiB to 512 MiB
* Update src/Ryujinx.Graphics.Gpu/Image/AutoDeleteCache.cs
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
* Format comment
* comment context
* Increase TextureSize capacity for OpenGL back to 1024
- Added a new const ulong for OpenGLTextureSizeCapacity
* Fix changes from last commit.
* Adjust last OpenGL changes.
* Remove garbage VSC file
* Update src/Ryujinx.Graphics.Gpu/Image/AutoDeleteCache.cs
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
* Update src/Ryujinx.Graphics.Gpu/Image/AutoDeleteCache.cs
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
* Update src/Ryujinx.Graphics.Gpu/Image/AutoDeleteCache.cs
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
---------
Co-authored-by: gdkchan <gab.dark.100@gmail.com>
* refactor(perf): pass MemoryOwner<byte> around as itself rather than IMemoryOwner<byte>
* fix(perf): get span via MemoryOwner<byte>.Span property instead of through Memory property
* fix: incorrect comment change
* Allow creating texture aliases on texture pool
* Delete old image format override code
* New format incompatible alias
* Missing bounds check
* GetForBinding now takes FormatInfo
* Make FormatInfo struct more compact
* chore: replace `ByteMemoryPool` usage with `MemoryOwner<byte>`
* refactor: `PixelConverter.ConvertR4G4ToR4G4B4A4()` - rename old `outputSpan` to `outputSpanUInt16`, reuse same output `Span<byte>` as newly-freed name `outputSpan`
* eliminate temporary buffer allocations
* chore, perf: use MemoryOwner<byte> instead of IMemoryOwner<byte>
* Ensure descriptor sets are only re-used when all command buffers using it have completed
* Fix some SPIR-V capabilities
* Set update after bind flag if we exceed limits
* Simpler fix for Intel
* Format whitespace
* Make struct readonly
* Add barriers for extra set arrays too
* Report base and extra sets from the backend
* Pass texture set index everywhere
* Key textures using set and binding (rather than just binding)
* Start using extra sets for array textures
* Shader cache version bump
* Separate new commands, some PR feedback
* Introduce new manual descriptor set reservation method that prevents it from being used by something else while owned by an array
* Move bind extra sets logic to new method
* Should only use separate array is MaximumExtraSets is not zero
* Format whitespace
* GPU: Migrate buffers on GPU project, pre-emptively flush device local mappings
Essentially retreading #4540, but it's on the GPU project now instead of the backend. This allows us to have a lot more control + knowledge of where the buffer backing has been changed and allows us to pre-emptively flush pages to host memory for quicker readback. It will allow us to do other stuff in the future, but we'll get there when we get there.
Performance greatly improved in Hyrule Warriors: Age of Calamity. Performance notably improved in TOTK (average). Performance for BOTW restored to how it was before #4911, perhaps a bit better.
- Rewrites a bunch of buffer migration stuff. Might want to tighten up how dispose stuff works.
- Fixed an issue where the copy for texture pre-flush would happen _after_ the syncpoint.
TODO: remove a page from pre-flush if it isn't flushed after a certain number of copies.
* Add copy deactivation
* Fix dependent virtual buffers
* Remove logging
* Fix format issues (maybe)
* Vulkan: Remove backing swap
* Add explicit memory access types for most buffers
* Fix typo
* Add device local force expiry, change buffer inheritance behaviour
* General cleanup, OGL fix
* BufferPreFlush comments
* BufferBackingState comments
* Add an extra precaution to BufferMigration
This is very unlikely, but it's important to cover loose ends like this.
* Address some feedback
* Docs
* Add support for bindless textures from shader input (vertex buffer)
* Shader cache version bump
* Format whitespace
* Remove cache entries on pool removal, disable for OpenGL
* PR feedback
* rebase
* add methods Ryyjinx.Common EmbeddedResources and SteamUtils
* GAL changes - change SetData() methods and ThreadedTexture commands to use IMemoryOwner<byte> instead of SpanOrArray<byte>
* Ryujinx.Graphics.Texture: change texture conversion methods to return IMemoryOwner<byte> and allocate from ByteMemoryPool
* Ryujinx.Graphics.OpenGL: update ITexture and Texture-like types with SetData() methods to take IMemoryOwner<byte> instead of SpanOrArray<byte>
* Ryujinx.Graphics.Vulkan: update ITexture and Texture-like types with SetData() methods to take IMemoryOwner<byte> instead of SpanOrArray<byte>
* Ryujinx.Graphics.Gpu: update ITexture and Texture-like types with SetData() methods to take IMemoryOwner<byte> instead of SpanOrArray<byte>
* Remove now-unused SpanOrArray<T>
* post-rebase cleanup
* PixelConverter: remove unsafe modifier on safe methods, and remove one unnecessary cast
* use ByteMemoryPool.Rent() in GetWritableRegion() impls
* fix formatting, rename `ReadRentedMemory()` to `ReadFileToRentedMemory()``
* Texture.ConvertToHostCompatibleFormat(): dispose of `result` in Astc decode branch
Neo