* stencil fixes hope it fixes driveclub
* revert image copy to the one that had driveclub worked
* reverted texture cache change
* some more fixes and reverts
* added logging for overlap again
* fixed image copy to ignore stencil aspect
* Added logging for "Encountered unresolvable image overlap with equal memory address."
* fixed overlap issues with different pitch , added more detailed logging for rest of overlap issues
* improved log error
* maybe mipmaps ?
* array layers or different mip map range
* rewrote case new image has fewer mip levels than cached image
* array with 2 layers?
* last case
* improved
* no it didn't work
* video_core: Refactor render target bind to allow disabling MSAA
* video_core: Implement swapping of backing samples
* clang format
* video_core: Better implementation
Instead of downgrading to 1 sample, always try to match depth samples. This avoids needing to copy depth-stencil attachment and copying multisampled stencil is not possible on some vendors
* video_core: Small bugfixes
* image: Add null check
* vk_rasterizer: Swap backing samples on resolve dst
* vk_presenter: Reset backing samples before present
* video_core: Small refactor to make this implementation better
* reinterpret: Fix channel check for degamma
Seems this was simpler than I thought, hardware doesn't apply degamma on the W channel regardless of swizzle
* image: Add missing end rendering call
* blit_helper: Fix bug in old reinterpret path
* blit_helper: Remove unused layer vertex
Should be used in the future if copying many layers is needed
* vk_rasterizer: Apply suggestion
* vk_rasterizer: More bind refactor
* vk_instance: Re-enable extensions
* vk_rasterizer: Reorder image query in fast clear elimination
Fixes missing clears when a texture is being cleared using this method but never actually used for rendering purposes by ensuring the texture cache has at least a chance to register cmask
* shader_recompiler: Partial support for ANCILLARY_ENA
* pixel_format: Add number conversion of BC6 srgb format
* texture_cache: Support aliases of 3D and 2D array images
Used be UE to render its post processing LUT
* pixel_format: Test BC6 srgb as unorm
Still not sure what is up with snorm/unorm can be useful to have both actions to compare for now
* video_core: Use attachment feedback layout instead of general if possible
UE games often do mipgen passes where the previous mip of the image being rendered to is bound for reading. This appears to cause corruption issues so use attachment feedback loop extension to ensure correct output
* renderer_vulkan: Improve feedback loop code
* Set proper usage flag for feedback loop usage
* Add dynamic state extension and enable it for color aspect when necessary
* Check if image is bound instead of force_general for better code consistency
* shader_recompiler: More proper depth export implementation
* shader_recompiler: Fix bug in output modifiers
* shader_recompiler: Fix sampling from MSAA images
This is not allowed by any graphics API but seems hardware supports it somehow and it can be encountered. To avoid glitched output translate to to a texelFetch call on sample 0
* clang format
* image: Add back missing code
* shader_recompiler: Better ancillary implementation
Now is implemented with a custom attribute that is constant propagated depending on which parts of it are extracted. It will assert if an unknown part is used or if the attribute itself is not removed by dead code elim
* copy_shader: Ignore not enabled export channels
* constant_propagation: Invalidate ancillary after successful elimination
* spirv: Fix f11/f10 conversion to f32
---------
Co-authored-by: georgemoralis <giorgosmrls@gmail.com>
* video_core: Rework detiling
* video_core: Support tiling and macrotile detiling
* clang format
* image_info: Cleanups
* resource: Revert some changes
* texture_cache: Fix small error
* image_info: Set depth flag on depth promote
* buffer_cache: Remove level check
* tile_manager: Handle case of staging buffer causing flush
* image_info: Add 2D thick array mode
* image_info: Add slices to mip size
* tile_manager: Set bank swizzle
* buffer_cache: Support image copies from DmaData
* vk_rasterizer: Accelerate trivial render target copies with compute
Before tiling PR compute image copies were done with the following sequence
vkCmdCopyImageToBuffer (in SynchronizeBufferFromImage) -> vkCmdDispatch (copy) -> vkCmdCopyBufferToImage (in RefreshImage)
With the tiling PR it added extra tiling/detiling steps
vkCmdCopyImageToBuffer -> vkCmdDispatch (tiling) -> vkCmdDispatch (copy) -> vkCmdDispatch (detiling) -> vkCmdCopyBufferToImage
This is quite a bit of overhead for a simple image copy. This commit tries to detect trivial image copies i.e cs shaders that copy the full source image to all of the destination.
So now all this sequence is just a vkCmdCopyImage. How much it triggers depends on the guest
* texture_cache: Fix build
* image: Copy all subresources with buffer too
* pixel_format: Remove unused tables, refactor
* host_compatibilty: Cleanup and support uncompressed views of compressed formats
* texture_cache: Handle compressed views of uncompressed images
* tile_manager: Bump max supported mips to 16
Fixes a crash during start
* oops
* texture_cache: Fix order of format compat check
* host_shaders: support for includes
* video_core: add a simpler vulkan asserts
* video_core: refactored post processing pipeline to another file
* renderer_vulkan: add define param to compile shader utility
* video_core: fsr implementation
* devtools: show resolution & fsr state
* ir_passes: Add barrier at end of block too
* vk_platform: Always assign names to resources
* texture_cache: Better overlap handling
* liverpool: Avoid resuming ce_task when its finished
* spirv_quad_rect: Skip default attributes
Fixes some crashes
* memory: Improve buffer size clamping
* liverpool: Relax binary header validity check
* liverpool: Stub SetPredication with a warning
* Better than outright crash
* emit_spirv: Implement round to zero mode
* liverpool: queue::pop takes the front element
* image_info: Remove obsolete assert
The old code assumed the mip only had 1 layer thus a right overlap could not return mip 0. But with the new path we handle images that are both mip-mapped and multi-layer, thus this can happen
* tile_manager: Fix size calculation
* spirv_quad_rect: Skip default attributes
---------
Co-authored-by: poly <47796739+polybiusproxy@users.noreply.github.com>
Co-authored-by: squidbus <175574877+squidbus@users.noreply.github.com>
* shader_recompiler: Account for instruction array flag in image type.
* shader_recompiler: Check da flag for all mimg instructions.
* shader_recompiler: Convert cube images into 2D arrays.
* shader_recompiler: Move image resource functions into sharp type.
* shader_recompiler: Use native AMD cube instructions when possible.
* specialization: Fix buffer storage mistake.
* texture_cache: Improve support for stencil reads
* libraries: Supress some spammy logs
* core: Support loading font libraries
* texture_cache: Remove assert
* Fixed false-positive image reuploads
* Fixed userfaultfd path, removed dead code, simplified calculations
* oopsie
* track potentially dirty images and hash them
* untrack only first page of the image in case of head access
* rebase, initialize hash, fix bounds check
* include image tail in the calculations
* video_core: texture_cache: interface refactor and better overlap handling
* resources binding moved into vk_rasterizer
* remove `virtual` flag leftover
* video_core: texture: image subresources state tracking
* shader_recompiler: use one binding if the same image is read and written
* video_core: added rebinding of changed textures after overlap resolve
* don't use pointers; slight `FindTexture` refactoring
* video_core: buffer_cache: don't copy over the image size
* redundant barriers removed; fixes
* regression fixes
* texture_cache: 3d texture layers count fixup
* shader_recompiler: support for partially bound cubemaps
* added support for cubemap arrays
* don't bind unused color buffers
* fixed depth promotion to do not use stencil
* doors
* bonfire lit
* cubemap array index calculation
* final touches
* texture_cache: Use invalidate threshhold
* It's possible for shaders to bind huge buffers and only write to lower portion of it. This is a problem if upper parts of the buffer overlap with render targets. If the image is very far away from buffer base it's unlikely the shader will want to write it, so skip invalidation for it
* video_core: Allow using texture cache to validate texture buffers
* texture_cache: Use buffer cache in all cases for data source
* Allows to correctly handle compute written micro tiled textures
* texture_cache: Fix depth pitch
* kernel: Remove missed code
* clang format
* video_core: Adjust depth format
* buffer_cache: Do not cache buffer views
* thread_management: Do not call createMutex on unlock
* temp: Revert this when pr is done
* buffer_cache: Dont skip cpu uploads with image sync
* Sometimes image does not fully overlap with a region
* fix build
* video_core: Improve invalidate heuristic
* small fixes
* video_core: Hopefully fix some vertex explosions
* translator: Use templates for stronger type guarantees
* spirv: Define buffer offsets upfront
* Saves a lot of shader instructions
* buffer_cache: Use dynamic vertex input when available
* Fixes issues when games like dark souls rebind vertex buffers with different stride
* externals: Update boost
* spirv: Use runtime array for ssbos
* ssbos can be large and typically their size will vary, especially in generic copy/clear cs shaders
* fs: Lock when doing case insensitive search
* Dark Souls does fs lookups from different threads
* texture_cache: More precise invalidation from compute
* Fixes unrelated render targets being cleared
* texture_cache: Use hashes for protect gpu modified images from reupload
* translator: Treat V_CNDMASK as float
* Sometimes it can have input modifiers. Worst this will cause is some extra calls to uintBitsToFloat and opposite. But most often this is used as float anyway
* translator: Small optimization for V_SAD_U32
* Fix review
* clang format
* texture_cache: interface refactoring
* a bit of fixes and improvements
* texture_cache: macro tile extents for bpp 128
* texture_cache: detiler: prefer host memory for large buffers upload
* amdgpu: proper CB and DB sizes calculation; minor refactoring
* texture_cache: separate file for image_info
* texture_cache: image guest address moved into image info
* texture_cache: surface size calculation
* shader_recompiler: fixed sin/cos
Thanks to red_pring and gandalfthewhite0173
* initial preparations for subresources upload
* review comments
