Create 3 ring buffers which rotates between buffers each frame to avoid GPU/CPU conflicts
BindMappedUniformBuffer first tries to allocate from the ring buffer and falls back to staging pool only if allocation is too large.
Note to testers:- please test the performance since it is primarily a performance optimization and also look for visual bugs.
Co-authored-by: wildcard <wildcard@eden-emu.dev>
Co-authored-by: MaranBr <maranbr@outlook.com>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2698
Reviewed-by: MaranBr <maranbr@eden-emu.dev>
Co-authored-by: Shinmegumi <shinmegumi@eden-emu.dev>
Co-committed-by: Shinmegumi <shinmegumi@eden-emu.dev>
The barrier before the CPU-upload copy was using VK_PIPELINE_STAGE_HOST_BIT. Validation rules only allow HOST as the source stage if you’re also specifying host-side access flags; inside a command buffer the GPU isn’t executing “host work,” so pairing that stage with the usual image layout transition is technically invalid.
Switching the source stage to VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT keeps the ordering guarantee we need and satisfies the spec, while the actual host visibility is still handled correctly by the preceding vmaFlushAllocation.
Co-authored-by: Ribbit <ribbit@placeholder.com>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2681
Reviewed-by: MaranBr <maranbr@eden-emu.dev>
Reviewed-by: Shinmegumi <shinmegumi@eden-emu.dev>
Co-authored-by: Ribbit <ribbit@eden-emu.dev>
Co-committed-by: Ribbit <ribbit@eden-emu.dev>
Adding the shader read barrier keeps every render/compute/transfer write visible before the image is sampled, so it prevents the “read-before-writes-finish” hazards. Without it you can get random stale frames, flickering post process passes, partially updated HUD textures, and corrupted depth-to-color conversions especially in scenes that render into an offscreen image and immediately feed that image to a shader (reflections, bloom, dynamic resolution, depth visualizers, etc.). This fix makes those R2T chains deterministic again across all Vulkan drivers.
Co-authored-by: Ribbit <ribbit@placeholder.com>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2671
Reviewed-by: MaranBr <maranbr@eden-emu.dev>
Reviewed-by: crueter <crueter@eden-emu.dev>
Co-authored-by: Ribbit <ribbit@eden-emu.dev>
Co-committed-by: Ribbit <ribbit@eden-emu.dev>
We copy pixels into a CPU-side staging buffer and then ask the GPU to read from it. On some systems those CPU writes aren’t automatically visible to the GPU unless explicitly flushed, so the GPU can sometimes read stale data. By calling buffer.Flush() immediately after writing, we force those CPU changes to become visible to the device, ensuring the GPU sees the latest frame. However, this is an emulator, so sometimes what spec says may not work cause reasons.
Co-authored-by: Ribbit <ribbit@placeholder.com>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2624
Reviewed-by: Shinmegumi <shinmegumi@eden-emu.dev>
Reviewed-by: MaranBr <maranbr@eden-emu.dev>
Co-authored-by: Ribbit <ribbit@eden-emu.dev>
Co-committed-by: Ribbit <ribbit@eden-emu.dev>
a bank could be (incorrectly) considered a superset if it had enough image buffer descriptors but not enough storage image descriptors, causing the allocator to pick a bank that can’t actually satisfy VK_DESCRIPTOR_TYPE_STORAGE_IMAGE demand resulting in sham allocations and creation of new pools.
Note to testers,
please look for any regressions in terms of visuals and most importantly please test the performance and ram usage.
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2564
Reviewed-by: MaranBr <maranbr@eden-emu.dev>
Reviewed-by: Shinmegumi <shinmegumi@eden-emu.dev>
Reviewed-by: Lizzie <lizzie@eden-emu.dev>
Reviewed-by: crueter <crueter@eden-emu.dev>
Co-authored-by: wildcard <wildcard@eden-emu.dev>
Co-committed-by: wildcard <wildcard@eden-emu.dev>
This improves the asynchronous shader building process.
Fixes a TOTK inventory bug that caused some icons to be missing under certain circumstances when using asynchronous shader building.
Fixes an issue in Kirby and the Forgotten Land where arriving at the checkpoint would cause a graphical bug in the building when asynchronous shader building was enabled.
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2560
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
Reviewed-by: crueter <crueter@eden-emu.dev>
Co-authored-by: MaranBr <maranbr@outlook.com>
Co-committed-by: MaranBr <maranbr@outlook.com>
temproary non-msaa images used for msaa uploads were being destroyed before the GPU finished using them, causing validation errors/crashes.
This keeps the temp image alive until the GPU finishes and switches the aspect-mask check to bitwise so the MSAA path only runs when needed.
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2550
Reviewed-by: MaranBr <maranbr@eden-emu.dev>
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
Co-authored-by: Maufeat <sahyno1996@gmail.com>
Co-committed-by: Maufeat <sahyno1996@gmail.com>
This fixes the crashes on game launch caused by MacOS not being present in host_manager.cpp and enables primitiveRestart for MoltenVK to suppress a bunch of errors given in the log about MoltenVK requiring primitiveRestart. Fixes an crash when switching kingdoms in Mario Odyssey as well
EDS is forced to 0, otherwise games do not show graphics
Note: For now only dynarmicc is working, performance will be slow
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/372
Reviewed-by: Lizzie <lizzie@eden-emu.dev>
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
Reviewed-by: MaranBr <maranbr@outlook.com>
Co-authored-by: innix <dev@innix.space>
Co-committed-by: innix <dev@innix.space>
Validation flagged writes to a VK_DESCRIPTOR_TYPE_STORAGE_IMAGE descriptor because the bound VkImageView had a non-identity component mapping and hence the vuid-00336 error, this fixes the said error.
------
This commit helps to fix some graphical issues on games like Trident's Tale, where game didn't render anything than just plain terrain, helps to stabilize Nier Automata graphical issues, meanwhile the most annoying glitches are gone, there's still remain other issues.
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/321
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
Co-authored-by: wildcard <nubieluv@gmail.com>
Co-committed-by: wildcard <nubieluv@gmail.com>
fixes this error [ 18.505526] Render.Vulkan <Info> video_core/vulkan_common/vulkan_debug_callback.cpp:DebugUtilCallback:59: Validation Information: [ UNASSIGNED-BestPractices-SemaphoreCount ] | MessageID = 0x6cfe18a5 | pSubmits[0].pWaitSemaphores is set, but pSubmits[0].waitSemaphoreCount is 0.
This patch is only corrective in nature and is trivial and should not fix or break anything just one of the best practices in vulkan. It nulls pWaitSemaphores / pWaitDstStageMask / pSignalSemaphores when the corresponding counts are zero.
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/285
Reviewed-by: Shinmegumi <shinmegumi@eden-emu.dev>
Co-authored-by: wildcard <wildcard@eden-emu.dev>
Co-committed-by: wildcard <wildcard@eden-emu.dev>
Maxwell format `VK_FORMAT_A8B8G8R8_SRGB_PACK32` does not support storage. However a `A8B8G8R8_UNORM` view is created for a image with that format which supports storage. The previous patch ignored image view format usage making it impossible for the pipeline to render to the texture.
This commit reverts the image usage override. However, there is still a mismatch between image format usage and image view format usage.
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/241
Reviewed-by: Shinmegumi <shinmegumi@eden-emu.dev>
Co-authored-by: weakboson <weakboson@quantum-field.net>
Co-committed-by: weakboson <weakboson@quantum-field.net>
> The shared_ptr<Image> capture ensures the temporary image outlives all queued GPU work (both the upload/download step and the MSAA compute copy). Without this, drivers read freed memory
> The temp image always has STORAGE & TRANSFER usage, matching what the compute MSAA path actually binds.
> Since this only a temp fix ontop of the previous commit, we only use the MSAA path for color; depth/stencil still use the original safe route.
> Should still be reworked though, as seen in the other MSAA commis that are open.
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/245
Reviewed-by: crueter <crueter@eden-emu.dev>
Reviewed-by: Shinmegumi <shinmegumi@eden-emu.dev>
Co-authored-by: Maufeat <sahyno1996@gmail.com>
Co-committed-by: Maufeat <sahyno1996@gmail.com>
Ribbit
unknown
lizzie
crueter
Shinmegumi
CamilleLaVey
wildcard
MaranBr
innix
wildcard
weakboson