aboutsummaryrefslogtreecommitdiff
path: root/src/video_core/gpu.cpp
Commit message (Collapse)AuthorAgeFilesLines
* Merge pull request #334 from snowstormxd/codex/upstream-mkw-testJonas29 hours1-1/+3
|\ | | | | MKW performance improvements
| * nx2: rearm clean read areas when the GPU writes them againsnowstormxd31 hours1-1/+3
| | | | | | | | | | | | | | Port the coherency changes from alarok/Eglantine 77fc365f2 without its diagnostic framework or shutdown changes. Only GPU-dirty CPU reads become permanently preflushable, clean areas can be trapped on the next write, non-core readers avoid indexing core arrays, and collected CPU writes precede readback. Assisted-by: Codex (source review, selective integration and verification) (cherry picked from commit b64e211957180aa218ec6d9a3bcb020332279c3f)
* | nx2: map GPU memory before later command lists are readSssnowyyy32 hours1-0/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 837133792d queues the GMMU maps on the GPU thread (GPU::RunInOrder) so the service thread never waits for the texture cache. With the GPU split (Low GPU accuracy) the GPU thread defers queued functions to its back, but its front reads the command lists: it read the lists submitted after a map through GPU addresses that weren't mapped yet. Pokopia stopped after its first frame. Add GPU::MapInOrder for the GMMU maps and the sparse tile remaps. It runs in order like RunInOrder, and with the split it runs on the front after draining the back, before the front reads further command lists. The service thread still never takes the texture cache lock, and the back is idle while the map changes the caches. Unmaps and the other queued work still run on the back. Assisted-by: Claude Opus 5.5 (bisect, fix, testing and commit message)
* | nx2: don't stall presents on a current-frame wait the GPU can't finishSssnowyyy32 hours1-0/+4
|/ | | | | | | | | | | | | | | | | 0f14696044 made PresentClientBuffer also wait for the current frame's fence whenever the previous frame's wasn't reached. Mario Kart World's GPU waits on semaphores its CPU releases only after the present returns, so in that state the current fence can't be reached until the present returns, and each such present stalled up to the 4 s deadline. Wait for the current frame by default, but give up after a few ms once a GPU channel was seen suspended on a semaphore acquire and wait for the previous frame instead. EDEN_NX2_FRAMES_AHEAD=1 always waits for the previous frame (with 0f14696044's current-frame wait when the GPU is behind), =0 always for the current one. The scheduler gains a lock-free flag so the present thread can see suspended channels. Assisted-by: Claude Sonnet 5.5 (implementation and commit message)
* buffer_cache: no buffer downloads at every fence for Switch 2 titlesTakoroka5 days1-0/+7
| | | | | | | | | | | | | Eden copied every GPU-written buffer range back to guest memory at the next fence. Title 1 signals a fence per compute dispatch, so that was 250-330 MB/s in 13,000-20,000 ranges a second, with the fencing thread holding the buffer cache while it wrote them. Measured: the native MME's DMA reads never touched those pages (they read query words written on the CPU side), and with reactive flushing the guest CPU read GPU-written memory at most 64 times in 200 s. Without them, for Switch 2 titles (native tiling): fences ~1.1 us instead of ~5, draw setup 6.1 us instead of 6.4-6.7, macro calls 160 ns instead of 167-170, GPU-thread lock spinning 1.8% (was 5.5%). Regions stay marked GPU written and are still flushed on demand. EDEN_NX2_BUFFER_DOWNLOADS=0/1 forces either way. Reactive flushing reads of GPU-written memory are now counted in the log.
* Fix the P6 scratch save in R2UR/MATCH/HFMA2.RELU; GPU memory dump toolTakoroka6 days1-0/+4
| | | | | | | | | | | | | | | | | The colour cast on title 1 (magenta/cyan, then red/green, changing per course) came from the sky/atmosphere volume shader (a pixel shader drawing a 32^3 RGBA16F table slice by slice), which starts with an unguarded R2UR. R2UR, MATCH and HFMA2.RELU use P6 as scratch and saved/restored it with the uniform-guard mask, 1 << (6 - 7) for an unguarded instruction: undefined, in practice bit 31, which shifted into the opcode bits and turned the save into another instruction. They now always save and restore P6. Debugging aids used to find it: - EDEN_NX2_DUMP=<present>:<addr>:<size>,...: flush GPU memory ranges at a present (on the GPU thread) and save them next to the frame captures; nx2work/vol3d.py untiles Switch 2 3D textures from such dumps. - GPU::RunOnGpuThread. - Logs of Switch 2 3D images and descriptors.
* nx2: unmap GPU memory in submission orderTakoroka7 days1-0/+8
| | | | | | | | | | | NormalUnmapGpuMemory4K and destroying GPU memory are jobs the console's graphics server completes behind the work already submitted. Run the GMMU unmaps on the GPU thread after the command lists queued before them (GPU::RunInOrder) and signal the job's event only then, instead of unmapping and signalling at once while Eden's GPU thread may still be behind. Also EDEN_NX2_FRAMES_AHEAD (experimental): wait for the previous frame's fence at present instead of this one's.
* [video_core, settings] readback for nce invalidations (#4490) optimization ↵xbzk8 days1-3/+3
| | | | | | | | | | | | | | | | | | and nuke toggles (#4502) (#4531) - [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. ------------------- Follow up for #4490: Originally i was racing locks between dirty and clean invalidations, but +94% were clean so it was causing useless locks. Solutioin was to share locks between both. P.S.: THIS IS INTENDED TO FIX RANDOM CRASHES IN UE, OBJECTIVELY REPRODUCIBLE IN ENDER MAGNOLIA IN ANDROID. Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/4531 Reviewed-by: CamilleLaVey <camillelavey99@gmail.com> Reviewed-by: Shinmegumi <shinmegumi@eden-emu.dev>
* [settings] add PR4490 & PR4492 toggles (#4502)xbzk2026-10-051-9/+11
| | | | | | | | | | | | | | - [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. ------------------- TEMPORARY! Pushed under Camille's instruction. Will be dropped as soon as they get optimized. Aimed to avoid reported perf impact from PRs 4490 and 4492 on 30-sep-2026 nightly, Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/4502 Reviewed-by: MaranBr <maranbr@eden-emu.dev> Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
* [video_core] buffer cache readback for nce invalidations (#4490)xbzk2026-09-301-3/+13
| | | | | | | | | | | | | | | | | | - [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. ------------------- This is a follow up of 4473, a fix for the next Ender Magnolia known crash, the shackled beast's vaper crash, but specific for NCE. Imma try to be short: NCE invalidation was dooming whole guest pages when it shouldn't. If the GPU buffer cache owns some bytes on a page, invalidating first discards their ownership, and the retrying CPU can then observe stale data. On Android, this was corrupting some indirect draw packets and causing huge invalid index count reach staging memory allocation. Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/4490 Reviewed-by: CamilleLaVey <camillelavey99@gmail.com> Reviewed-by: lizzie <lizzie@eden-emu.dev>
* [gpu] Fix infinite hangup/freezes at shutdown/reset (#4395)lizzie2026-09-111-4/+5
| | | | | | | | | | | | | | | | | | | | | Thread may be destroyed at dtor(), but it hasn't fully shut down, so before notify shutdown, request immediate stop (effective immediately). Like the issue was that the thread didn't want to stop running, thus it would hang, it would also reference objects which were being destroyed without waiting for them to actually be destroyed So it would reference invalid data This PR tells the thread: "hey, STOP now, and DESTROY yourself" So all of that should be avoided Signed-off-by: lizzie <lizzie@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/4395 Reviewed-by: Maufeat <sahyno1996@gmail.com> Reviewed-by: MaranBr <maranbr@eden-emu.dev>
* [gpu] Destroy GPU thread after all non-trivial members had been destroyed ↵lizzie2026-09-081-3/+6
| | | | | | | | | | | | | | | | | | (#4368) Should help fix any game hangs after closing the emulator, or sporadic hangs when restarting. Signed-off-by: lizzie <lizzie@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/4368 Reviewed-by: CamilleLaVey <camillelavey99@gmail.com> Reviewed-by: Maufeat <sahyno1996@gmail.com>
* [common] Switch to boost::unordered_flat containers (#4326)crueter2026-08-311-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | 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>
* [core, hle, video_core, memory] Improve multi-process, display layers, ↵Maufeat2026-08-291-0/+10
| | | | | | | | | | | | | | | | | | | | | dynamic shader cache and rework overlay (#4238) - [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. ------------------- - Dynamic shader cache reloading capability for qlaunch - Multi-process improvements (thanks to @frank1734), instead of just using the main application we now respect caller process (also did it for HID devices while I was at it) - Layer stack masks & shared buffer screenshot for video core, added different masks (screenshot, recording, etc.) this was discovered as an issue due to how in qlaunch the transition was not right between applications. May not be perfect but also fixes screenshots while using qlaunch - Reworked overlay display management (input and visibility) - instead of random numbers as I've previously did, I decided to add an AppletZIndex enum for better readability. Also split capability of input by touch and gamepad. - etc. Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/4238 Reviewed-by: Samuel <lizzie@eden-emu.dev> Reviewed-by: CamilleLaVey <camillelavey99@gmail.com> Reviewed-by: MaranBr <maranbr@eden-emu.dev>
* [core, gpu, threads] Multithreading refactor (#4254)CamilleLaVey2026-08-081-29/+57
| | | | | | This insufferable work tries to cover some holes on previous threading implementation from yuzu's team, starting with Windows and Linux reordering of priorities (NICE), reworks previous Android's threading and cpu affinity with adpf, adjust emulated clocks/gpu for better "accuracy" with their work, bumps android minSDK for all flavors, legacy will now work with AP 29 to cover A10 - A12, standard will reach A13 as base and finally the optimized build will come with API 35, mostly targeted on devices with A15 support and newer, NDK and AGP wasn't upgraded yet. The performance cost efficiency have been improved based on device power configuration; preventing overheating if certain devices tended to fall into NICE0 (not allocated threads priority, all task ran with higher priority, 11 tasks running within the limited 2 - 7 threads available on the most common configuration 1x3x4 or 1x4x3). Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/4254
* [video_core, hle] remove redundant parent references in system structs (#3908)lizzie2026-06-231-124/+58
| | | | | | | | | | reworked a bit to remove references of parent objects and instead pass as arguments to methods to prevent useless reloads Signed-off-by: lizzie <lizzie@eden-emu.dev> Co-authored-by: maufeat <sahyno1996@gmail.com> Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/3908 Reviewed-by: Maufeat <sahyno1996@gmail.com> Reviewed-by: crueter <crueter@eden-emu.dev>
* [hle/service{nvdrv,apm}] fixes for TetrisSwitch (#3983)lizzie2026-05-241-18/+17
| | | | | | | | | | | | - testriswitch submits buffers with a fence id of -1, just skip them instead of trying to process them? - apm:u, which is removed, but hey, backwards compat never hurted - another instance of shared_memory crashing NPad Signed-off-by: lizzie <lizzie@eden-emu.dev> Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/3983 Reviewed-by: MaranBr <maranbr@eden-emu.dev> Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>
* [host1x] fix ffmpeg not having va-api on freebsd, inline nvenc (#3878)lizzie2026-04-291-1/+1
| | | | | | | | | | | | | | - fix va-api not being used on freebsd small thingies dont affect a lot: - removes some pointer indirection (why save pointer to GMMU if its accesible via host1x) - use std::variant<> for decoder - miscelly vp9/v8/h264 opts Signed-off-by: lizzie <lizzie@eden-emu.dev> Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/3878 Reviewed-by: crueter <crueter@eden-emu.dev> Reviewed-by: MaranBr <maranbr@eden-emu.dev>
* [*] change all std::unordered_map and std::unordered_set into ↵lizzie2026-02-101-2/+2
| | | | | | | | | | | | | | | | | | | 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>
* [settings, frontend] Reorganize graphics/CPU settings, saner defaults (#3233)crueter2025-12-301-5/+4
| | | | | | | | | | | | | | - Fast GPU now defaults to 256, removed 128 since it's useless. - Completely reorganized graphics and CPU settings on both platforms. Also got rid of Eden's Veil - Merged some "use ..." settings that weren't really necessary. - Changed ExtendedDynamicState to be a combo box Signed-off-by: crueter <crueter@eden-emu.dev> Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/3233 Reviewed-by: MaranBr <maranbr@eden-emu.dev> Reviewed-by: DraVee <dravee@eden-emu.dev> Reviewed-by: Lizzie <lizzie@eden-emu.dev>
* [meta] remove MicroProfile (#185)crueter2025-08-061-1/+3
| | | | | | Signed-off-by: crueter <crueter@eden-emu.dev> Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/185 Reviewed-by: Lizzie <lizzie@eden-emu.dev>
* Fast CPU Time & Improved Fast GPU Time (#109)swurl2025-05-291-1/+4
| | | | | | | | | | | needs android setting Signed-off-by: swurl <swurl@swurl.xyz> Co-authored-by: Aleksandr Popovich <alekpopo@pm.me> Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/109 Co-authored-by: swurl <swurl@swurl.xyz> Co-committed-by: swurl <swurl@swurl.xyz>
* Port changes from Early Accessdarktux2025-04-041-33/+0
|
* video_core: make gpu context aware of rendering programLiam2024-02-261-4/+4
|
* nvnflinger/gpu: implement applet captureLiam2024-02-091-0/+15
|
* nvnflinger/gpu: implement layer stack compositionLiam2024-01-311-19/+10
|
* GPU-SMMU: Estimate game leak and preallocate device region.Fernando Sahmkow2024-01-181-2/+0
|
* GPU SMMU: Expand to 34 bitsFernando Sahmkow2024-01-181-0/+1
|
* SMMU: Initial adaptation to video_core.Fernando Sahmkow2024-01-181-13/+15
|
* NVDRV: Implement sessions and initial implementation of SMMUFernando Sahmkow2024-01-181-1/+1
|
* Query Cachge: Fully rework Vulkan's query cacheFernando Sahmkow2023-09-231-1/+3
|
* Memory Tracking: Optimize tracking to only use atomic writes when contested ↵Fernando Sahmkow2023-06-281-1/+9
| | | | with the host GPU
* MemoryTracking: Initial setup of atomic writes.Fernando Sahmkow2023-06-281-1/+3
|
* (wall, native)_clock: Add GetGPUTickMorph2023-06-071-8/+3
| | | | Allows us to directly calculate the GPU tick without double conversion to and from the host clock tick.
* core_timing: Use CNTPCT as the guest CPU tickMorph2023-06-071-7/+7
| | | | | Previously, we were mixing the raw CPU frequency and CNTFRQ. The raw CPU frequency (1020 MHz) should've never been used as CNTPCT (whose frequency is CNTFRQ) is the only counter available.
* core: frontend: Refactor GraphicsContext to its own module.bunnei2023-06-031-0/+1
|
* GPU: Add Reactive flushingFernando Sahmkow2023-05-071-0/+19
|
* core_timing: Use higher precision sleeps on WindowsMorph2023-03-051-1/+1
| | | | | | The precision of sleep_for and wait_for is limited to 1-1.5ms on Windows. Using SleepForOneTick() allows us to sleep for exactly one interval of the current timer resolution. This allows us to take advantage of systems that have a timer resolution of 0.5ms to reduce CPU overhead in the event loop.
* Remove OnCommandListEndCommandBehunin2023-02-071-1/+1
| | | | Call rasterizer->ReleaseFences() directly
* renderer_opengl: refactor context acquireLiam2022-12-131-2/+3
|
* General: address feedbackFernando Sahmkow2022-10-061-1/+1
|
* VideoCore: Refactor fencing system.Fernando Sahmkow2022-10-061-10/+7
|
* NVDRV: Further refactors and eliminate old code.Fernando Sahmkow2022-10-061-48/+0
|
* NVDRV: Refactor Host1xFernando Sahmkow2022-10-061-25/+2
|
* VideoCore: Refactor syncing.Fernando Sahmkow2022-10-061-95/+102
|
* Texture cache: Fix the remaining issues with memory mnagement and unmapping.Fernando Sahmkow2022-10-061-0/+8
|
* NVHOST_CTRl: Implement missing method and fix some stuffs.Fernando Sahmkow2022-10-061-0/+5
|
* VideoCore: implement channels on gpu caches.Fernando Sahmkow2022-10-061-387/+81
|
* 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: Replace lock_guard with scoped_lockMerry2022-04-071-3/+3
|
generated by cgit v1.2.3 (git 2.39.1) at 2026-11-04 04:02:11 +0000