aboutsummaryrefslogtreecommitdiff
path: root/src/video_core/shader_cache.h
Commit message (Collapse)AuthorAgeFilesLines
* shader_cache: batch memory-write invalidations without skipping rangessnowstormxd35 hours1-0/+4
| | | | | | | Adapt the batching from ymatrix1/Eglantine PR #237, retaining all CPU invalidations. Acquire the lock once and remove pending shaders once per write batch. Compare real cache contents with sequential invalidation across randomized overlapping writes and pending removals. Assisted-by: Codex (implementation, review, commit text and automated validation) (cherry picked from commit b65f818d7789ffadfb58fe4543b18060fd88ce53)
* nx2: capture and lower native shader programs oncekraibse2 days1-0/+2
| | | | | | | | | | | | | Native programs were captured and lowered when their ShaderInfo was made and again for every environment built for their pipelines. The ShaderInfo now keeps the lowered code and the environments adopt it. Capture also reads a 4 KiB page at a time instead of flushing caches per instruction, and cached native programs fold bank 1 reads like live ones. Ported from Wael3rd/Eglantine-MKW (0ae093058e, 65a6aa877a, df076133c4). Assisted-by: Claude Sonnet 5.5 (porting the fixes) (cherry picked from commit 3af11ac401ad3061c183281d8c98c16c1d6da456)
* video_core: draw Switch 2 task/mesh pipelines with VK_EXT_mesh_shaderTakoroka3 days1-0/+2
| | | | | | | | | | | | | While SET_MESH_CONTROL is on, a native draw binds a task program in the vertex slot, its mesh program in the tessellation slot and a header-only pass-through geometry program naming the per-primitive outputs. Translate the first two as task and mesh programs with the mesh registers (SET_MESH_SHADER_A/B, SET_MESH_INIT_SHADER) and that header, key their pipelines by them, and build pipelines with task/mesh stages and no vertex input, input assembly or tessellation state. The draw's vertex count is the number of task workgroups: record vkCmdDrawMeshTasksEXT. Enable VK_EXT_mesh_shader (task and mesh only) when the host has it; draws are skipped without it, as they were before.
* nx2: memo for compute macros; one lock per inline upload; cheaper shader ↵Takoroka6 days1-0/+4
| | | | | | | | | | | | | | | invalidation - Compute engine macros replay repeated calls like 3D ones (title 1's compute macro path fell from ~27% to ~14% of the GPU thread). - AccelerateInlineToMemory takes the buffer and texture cache mutexes together instead of one after the other. - ShaderCache::InvalidatePagesInRegion looks at the rejected programs only when the write can reach one (bounds kept as they are added); it ran a locked scan for every inline upload and pre-draw invalidation. Inline uploads went from ~10% to ~4% of the GPU thread.
* nx2: measure the TSC on AMD, no rethrows for rejected shaders, debug hooks ↵Takoroka6 days1-0/+2
| | | | | | | | | | | | | | | off the hot path - AMD CPUs have no CPUID leaf 0x15, so the TSC frequency stayed 0 and every guest CNTPCT read went through QueryPerformanceCounter (~10% of the game's main core in title 1). Measure it like Intel CPUs without a crystal frequency. - A rejected native shader was rethrown (two C++ throws) every time it was bound; return false instead (~1-2% of the GPU thread). - The semaphore watch, pushbuffer entry hashing, first-write trace and per-macro debug logs ran on every command word or macro call; they now cost a flag check unless their environment variable is set (~4% of the GPU thread). Also starts the macro engine's profile zone.
* Jump-table programs end at their self-branch; aligned wide zero storesTakoroka6 days1-0/+5
| | | | | | | | | | | - With a BRX the jump targets live in a constant buffer, so capture and lowering keep going past EXITs to the compiler's closing branch-to-self (a title 1 compute program jumped to 0xae0, just past the first EXIT). - 128-bit stores of RZ use R248-R251 (Eden wants 4-aligned data registers). - The rejected-program cache covers environment construction, where native programs are lowered (it never applied before), and logs the stage slot. Title 1 reaches its title screen with the ground drawn.
* Cache rejected native shaders and log each rejection onceTakoroka6 days1-0/+6
| | | | | | | A rejected program was lowered again (and logged) on every draw or dispatch: 1.7M log lines in 13 minutes of title 1. The shader cache keeps the first exception per address until guest memory there is written, and the error logs are deduplicated.
* [common] Switch to boost::unordered_flat containers (#4326)crueter2026-08-311-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Replaces all instances of ankerl's unordered map/set with boost's `unordered_flat_*` classes. This uses std::hash since boost::hash is actually a lot slower. Also adds an abstraction layer in `Common` so future changes are quicker and easier. Other implementation details: - ankerl provided hash specializations for tuple and pair, so those were ported here - std::erase_if doesn't work on boost, so just used the ADL'd erase_if This should be about equal or superior performance as unordered_dense for everything except iteration. Signed-off-by: crueter <crueter@eden-emu.dev> - [x] I have read and followed the [Contribution Guidelines](https://git.eden-emu.dev/eden-emu/eden/src/branch/master/CONTRIBUTING.md#code-contributions). - [x] I have read and followed the [AI Policy](https://git.eden-emu.dev/eden-emu/eden/src/branch/master/docs/policies/AI.md) - [x] I have read and followed the [Coding Guidelines](https://git.eden-emu.dev/eden-emu/eden/src/branch/master/docs/policies/Coding.md) to the best of my ability. ------------------- Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/4326 Reviewed-by: Lizzie and Samuel <lizzie@eden-emu.dev> Reviewed-by: MaranBr <maranbr@eden-emu.dev>
* [*] change all std::unordered_map and std::unordered_set into ↵lizzie2026-02-101-4/+4
| | | | | | | | | | | | | | | | | | | ankerl::unordered_dense::map/set variants (#3442) mainly doing this to reduce memory footprint; we all know how nice ankerl::unordered_dense is in theory 4x faster - in practice these maps arent that "hot" anyways so not likely to have much perf gained i just want to reduce mem fragmentation to ease my porting process, plus it helps other platforms as well (ahem weak Mediatek devices) :) Signed-off-by: lizzie <lizzie@eden-emu.dev> Co-authored-by: Caio Oliveira <caiooliveirafarias0@gmail.com> Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/3442 Reviewed-by: DraVee <dravee@eden-emu.dev> Reviewed-by: CamilleLaVey <camillelavey99@gmail.com> Co-authored-by: lizzie <lizzie@eden-emu.dev> Co-committed-by: lizzie <lizzie@eden-emu.dev>
* [compat] Debian stable gcc12/clang14 compilation fixes (#2763)lizzie2025-10-181-6/+0
| | | | | | | | | | | | | Mainly because - while we can just give out an AppImage and call it a day - building natively should be an option for all major distros. And "base" stable debian doesn't provide a new enough g++/clang++ so... we need to make some "fixups". Signed-off-by: lizzie <lizzie@eden-emu.dev> Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2763 Reviewed-by: crueter <crueter@eden-emu.dev> Reviewed-by: MaranBr <maranbr@eden-emu.dev> Co-authored-by: lizzie <lizzie@eden-emu.dev> Co-committed-by: lizzie <lizzie@eden-emu.dev>
* [common] remove ranges polyfill (#2546)lizzie2025-09-241-1/+4
| | | | | | | | | Signed-off-by: lizzie <lizzie@eden-emu.dev> Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/2546 Reviewed-by: MaranBr <maranbr@eden-emu.dev> Reviewed-by: crueter <crueter@eden-emu.dev> Co-authored-by: lizzie <lizzie@eden-emu.dev> Co-committed-by: lizzie <lizzie@eden-emu.dev>
* Core: Clang format and other small issues.Fernando Sahmkow2024-01-181-1/+1
|
* SMMU: Initial adaptation to video_core.Fernando Sahmkow2024-01-181-2/+3
|
* Fix shader dumps with nvdisasmKelebek12023-08-031-1/+1
| | | | | skip fragment shaders when rasterizer is disabled initialize env_ptrs
* Memory Tracking: Optimize tracking to only use atomic writes when contested ↵Fernando Sahmkow2023-06-281-1/+1
| | | | with the host GPU
* general: fix compile for Apple ClangLiam2022-11-221-0/+1
|
* VideoCore: implement channels on gpu caches.Fernando Sahmkow2022-10-061-8/+7
|
* code: dodge PAGE_SIZE #defineKyle Kienapfel2022-08-191-2/+2
| | | | | | | | | | | | | | | | | | | | | Some header files, specifically for OSX and Musl libc define PAGE_SIZE to be a number This is great except in yuzu we're using PAGE_SIZE as a variable Specific example `static constexpr u64 PAGE_SIZE = u64(1) << PAGE_BITS;` PAGE_SIZE PAGE_BITS PAGE_MASK are all similar variables. Simply deleted the underscores, and then added YUZU_ prefix Might be worth noting that there are multiple uses in different classes/namespaces This list may not be exhaustive Core::Memory 12 bits (4096) QueryCacheBase 12 bits ShaderCache 14 bits (16384) TextureCache 20 bits (1048576, or 1MB) Fixes #8779
* general: Convert source file copyright comments over to SPDXMorph2022-04-231-3/+2
| | | | | This formats all copyright comments according to SPDX formatting guidelines. Additionally, this resolves the remaining GPLv2 only licensed files by relicensing them to GPLv2.0-or-later.
* video_core/shader_cache: Remove unused algorithm includeLioncash2022-02-011-1/+0
|
* video_core/shader_cache: Take std::span in RemoveShadersFromStorage()Lioncash2022-02-011-1/+1
| | | | | Same behavior, but without the need to move into the function to avoid an allocation.
* shader: Initial OpenGL implementationReinUsesLisp2021-07-221-2/+21
|
* shader: Move pipeline cache logic to separate filesReinUsesLisp2021-07-221-148/+50
| | | | | | | | | Move code to separate files to be able to reuse it from OpenGL. This greatly simplifies the pipeline cache logic on Vulkan. Transform feedback state is not yet abstracted and it's still intrusively stored inside vk_pipeline_cache. It will be moved when needed on OpenGL.
* shader_cache: Make use of std::erase_ifLioncash2020-07-141-2/+2
| | | | | Now that we use C++20, we can also make use of std::erase_if instead of needing to do the erase-remove idiom.
* shader_cache: Fix use-after-free and orphan invalidation cache entriesReinUsesLisp2020-07-011-29/+41
| | | | | | | | | | | | | This fixes some cases where entries could have been removed multiple times reading freed memory. To address this issue this commit removes duplicates from entries marked for removal and sorts out the removal process to fix another use-after-free situation. Another issue fixed in this commit is orphan invalidation cache entries. Previously only the entries that were invalidated in the current operations had its entries removed. This led to more use-after-free situations when these entries were actually invalidated but referenced an object that didn't exist.
* Rename PAGE_SHIFT to PAGE_BITSMerryMage2020-06-181-5/+5
| | | | macOS header files #define PAGE_SHIFT
* shader_cache: Implement a generic shader cacheReinUsesLisp2020-06-071-0/+228
Implement a generic shader cache for fast lookups and invalidations. Invalidations are cheap but expensive when a shader is invalidated. Use two mutexes instead of one to avoid locking invalidations for lookups and vice versa. When a shader has to be removed, lookups are locked as expected.
generated by cgit v1.2.3 (git 2.39.1) at 2026-11-04 05:35:05 +0000