"Merge Tagged PR 1012"

"Merge Tagged PR 1340"
"Merge Tagged PR 1703"
2019-10-02 23:35:05 +00:00 · 2019-10-02 23:35:04 +00:00 · 2019-10-02 23:35:04 +00:00 · 2019-10-02 23:35:02 +00:00 · 2019-10-02 23:35:01 +00:00 · 2019-10-02 23:35:01 +00:00
68 changed files with 1500 additions and 483 deletions
@@ -0,0 +1,45 @@
+# Download all pull requests as patches that match a specific label
+# Usage: python download-patches-by-label.py <Label to Match> <Root Path Folder to DL to>
+
+import requests, sys, json, shutil, subprocess, os, traceback
+
+org = os.getenv("PRIVATEMERGEORG", "yuzu-emu")
+repo = os.getenv("PRIVATEMERGEREPO", "yuzu-private")
+tagline = sys.argv[3]
+user = sys.argv[1]
+
+dl_list = {}
+
+TAG_NAME = sys.argv[2]
+
+def check_individual(repo_id, pr_id):
+    url = 'https://%sdev.azure.com/%s/%s/_apis/git/repositories/%s/pullRequests/%s/labels?api-version=5.1-preview.1' % (user, org, repo, repo_id, pr_id)
+    response = requests.get(url)
+    if (response.ok):
+        try:
+            js = response.json()
+            return any(tag.get('name') == TAG_NAME for tag in js['value'])
+        except:
+            return False
+    return False
+
+def merge_pr(pn, ref):
+    print("Matched PR# %s" % pn)
+    print(subprocess.check_output(["git", "fetch", "https://%sdev.azure.com/%s/_git/%s" % (user, org, repo), ref, "-f"]))
+    print(subprocess.check_output(["git", "merge", "--squash", 'origin/' + ref.replace('refs/heads/','')]))
+    print(subprocess.check_output(["git", "commit", "-m\"Merge %s PR %s\"" % (tagline, pn)]))
+
+def main():
+    url = 'https://%sdev.azure.com/%s/%s/_apis/git/pullrequests?api-version=5.1' % (user, org, repo)
+    response = requests.get(url)
+    if (response.ok):
+        js = response.json()
+        tagged_prs = filter(lambda pr: check_individual(pr['repository']['id'], pr['pullRequestId']), js['value'])
+        map(lambda pr: merge_pr(pr['pullRequestId'], pr['sourceRefName']), tagged_prs)
+
+if __name__ == '__main__':
+    try:
+        main()
+    except:
+        traceback.print_exc(file=sys.stdout)
+        sys.exit(-1)
@@ -1,7 +1,9 @@
 # Download all pull requests as patches that match a specific label
 # Usage: python download-patches-by-label.py <Label to Match> <Root Path Folder to DL to>

-import requests, sys, json, urllib3.request, shutil, subprocess
+import requests, sys, json, urllib3.request, shutil, subprocess, os
+
+tagline = sys.argv[2]

 http = urllib3.PoolManager()
 dl_list = {}
@@ -12,17 +14,23 @@ def check_individual(labels):
            return True
    return False

-try:
-    url = 'https://api.github.com/repos/yuzu-emu/yuzu/pulls'
+def do_page(page):
+    url = 'https://api.github.com/repos/yuzu-emu/yuzu/pulls?page=%s' % page
    response = requests.get(url)
    if (response.ok):
        j = json.loads(response.content)
+        if j == []:
+            return
        for pr in j:
            if (check_individual(pr["labels"])):
                pn = pr["number"]
                print("Matched PR# %s" % pn)
                print(subprocess.check_output(["git", "fetch", "https://github.com/yuzu-emu/yuzu.git", "pull/%s/head:pr-%s" % (pn, pn), "-f"]))
                print(subprocess.check_output(["git", "merge", "--squash", "pr-%s" % pn]))
-                print(subprocess.check_output(["git", "commit", "-m\"Merge PR %s\"" % pn]))
+                print(subprocess.check_output(["git", "commit", "-m\"Merge %s PR %s\"" % (tagline, pn)]))
+
+try:
+    for i in range(1,30):
+        do_page(i)
 except:
    sys.exit(-1)
@@ -0,0 +1,32 @@
+$GITDATE = $(git show -s --date=short --format='%ad') -replace "-",""
+$GITREV = $(git show -s --format='%h')
+$RELEASE_DIST = "yuzu-windows-msvc"
+
+$MSVC_BUILD_ZIP = "yuzu-windows-msvc-$GITDATE-$GITREV.zip" -replace " ", ""
+$MSVC_BUILD_PDB = "yuzu-windows-msvc-$GITDATE-$GITREV-debugsymbols.zip" -replace " ", ""
+$MSVC_SEVENZIP = "yuzu-windows-msvc-$GITDATE-$GITREV.7z" -replace " ", ""
+
+$env:BUILD_ZIP = $MSVC_BUILD_ZIP
+$env:BUILD_SYMBOLS = $MSVC_BUILD_PDB
+$env:BUILD_UPDATE = $MSVC_SEVENZIP
+
+$BUILD_DIR = ".\build\bin\Release"
+
+mkdir pdb
+Get-ChildItem "$BUILD_DIR\" -Recurse -Filter "*.pdb" | Copy-Item -destination .\pdb
+7z a -tzip $MSVC_BUILD_PDB .\pdb\*.pdb
+rm "$BUILD_DIR\*.pdb"
+mkdir $RELEASE_DIST
+mkdir "artifacts"
+
+Copy-Item "$BUILD_DIR\*" -Destination $RELEASE_DIST -Recurse
+rm "$RELEASE_DIST\*.exe"
+Get-ChildItem "$BUILD_DIR" -Recurse -Filter "yuzu*.exe" | Copy-Item -destination $RELEASE_DIST
+Get-ChildItem "$BUILD_DIR" -Recurse -Filter "QtWebEngineProcess*.exe" | Copy-Item -destination $RELEASE_DIST
+Copy-Item .\license.txt -Destination $RELEASE_DIST
+Copy-Item .\README.md -Destination $RELEASE_DIST
+7z a -tzip $MSVC_BUILD_ZIP $RELEASE_DIST\*
+7z a $MSVC_SEVENZIP $RELEASE_DIST
+
+Get-ChildItem . -Filter "*.zip" | Copy-Item -destination "artifacts"
+Get-ChildItem . -Filter "*.7z" | Copy-Item -destination "artifacts"
@@ -0,0 +1,21 @@
+parameters:
+  artifactSource: 'true'
+  cache: 'false'
+
+steps:
+- script: mkdir build && cd build && set DATE=`date '+%Y.%m.%d'` && set CI=true && set AZURE_REPO_NAME=yuzu-emu/yuzu-$(BuildName) && set AZURE_REPO_TAG=$(BuildName)-$DATE && cmake -G "Visual Studio 15 2017 Win64" --config Release -DYUZU_USE_BUNDLED_QT=1 -DYUZU_USE_BUNDLED_SDL2=1 -DYUZU_USE_BUNDLED_UNICORN=1 -DYUZU_USE_QT_WEB_ENGINE=ON -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON -DYUZU_ENABLE_COMPATIBILITY_REPORTING=${COMPAT} -DUSE_DISCORD_PRESENCE=ON .. && cd ..
+  displayName: 'Configure CMake'
+- task: MSBuild@1
+  displayName: 'Build'
+  inputs:
+    solution: 'build/yuzu.sln'
+    maximumCpuCount: true
+    configuration: release
+- task: PowerShell@2
+  displayName: 'Package Artifacts'
+  inputs:
+    targetType: 'filePath'
+    filePath: './.ci/scripts/windows/upload.ps1'
+- publish: artifacts
+  artifact: 'yuzu-$(BuildName)-windows-msvc'
+  displayName: 'Upload Artifacts'
@@ -13,7 +13,7 @@ steps:
    key: yuzu-v1-$(BuildName)-$(BuildSuffix)-$(CacheSuffix)
    path: $(System.DefaultWorkingDirectory)/ccache
    cacheHitVar: CACHE_RESTORED
- script: chmod a+x ./.ci/scripts/$(ScriptFolder)/exec.sh && ./.ci/scripts/$(ScriptFolder)/exec.sh
+- script: export DATE=`date '+%Y.%m.%d'` && export CI=true && export AZURE_REPO_NAME=yuzu-emu/yuzu-$(BuildName) && export AZURE_REPO_TAG=$(BuildName)-$DATE && chmod a+x ./.ci/scripts/$(ScriptFolder)/exec.sh && ./.ci/scripts/$(ScriptFolder)/exec.sh
  displayName: 'Build'
 - script: chmod a+x ./.ci/scripts/$(ScriptFolder)/upload.sh && RELEASE_NAME=$(BuildName) ./.ci/scripts/$(ScriptFolder)/upload.sh
  displayName: 'Package Artifacts'
@@ -0,0 +1,47 @@
+jobs:
+- job: merge
+  displayName: 'pull requests'
+  steps:
+  - checkout: self
+    submodules: recursive
+  - template: ./mergebot-private.yml
+    parameters:
+      matchLabel: '$(BuildName)-merge'
+      matchLabelPublic: '$(PublicBuildName)-merge'
+  - task: ArchiveFiles@2
+    displayName: 'Package Source'
+    inputs:
+      rootFolderOrFile: '$(System.DefaultWorkingDirectory)'
+      includeRootFolder: false
+      archiveType: '7z'
+      archiveFile: '$(Build.ArtifactStagingDirectory)/yuzu-$(BuildName)-source.7z'
+  - task: PublishPipelineArtifact@1
+    displayName: 'Upload Artifacts'
+    inputs:
+      targetPath: '$(Build.ArtifactStagingDirectory)/yuzu-$(BuildName)-source.7z'
+      artifact: 'yuzu-$(BuildName)-source'
+      replaceExistingArchive: true
+- job: upload_source
+  displayName: 'upload'
+  dependsOn: merge
+  steps:
+  - template: ./sync-source.yml
+    parameters:
+      artifactSource: 'true'
+      needSubmodules: 'true'
+  - script: chmod a+x $(System.DefaultWorkingDirectory)/.ci/scripts/merge/yuzubot-git-config.sh && $(System.DefaultWorkingDirectory)/.ci/scripts/merge/yuzubot-git-config.sh
+    displayName: 'Apply Git Configuration'
+  - script: git tag -a $(BuildName)-$(Build.BuildId) -m "yuzu $(BuildName) $(Build.BuildNumber) $(Build.DefinitionName)"
+    displayName: 'Tag Source'
+  - script: git remote add other $(GitRepoPushChangesURL)
+    displayName: 'Register Repository'
+  - script: git push --follow-tags --force other HEAD:$(GitPushBranch)
+    displayName: 'Update Code'
+  - script: git rev-list -n 1 $(BuildName)-$(Build.BuildId) > $(Build.ArtifactStagingDirectory)/tag-commit.sha
+    displayName: 'Calculate Release Point'
+  - task: PublishPipelineArtifact@1
+    displayName: 'Upload Release Point'
+    inputs:
+      targetPath: '$(Build.ArtifactStagingDirectory)/tag-commit.sha'
+      artifact: 'yuzu-$(BuildName)-release-point'
+      replaceExistingArchive: true
@@ -0,0 +1,30 @@
+parameters:
+  matchLabel: 'dummy-merge'
+  matchLabelPublic: 'dummy-merge'
+
+steps:
+  - script: mkdir $(System.DefaultWorkingDirectory)/patches && pip install requests urllib3
+    displayName: 'Prepare Environment'
+  - script: chmod a+x $(System.DefaultWorkingDirectory)/.ci/scripts/merge/yuzubot-git-config.sh && $(System.DefaultWorkingDirectory)/.ci/scripts/merge/yuzubot-git-config.sh
+    displayName: 'Apply Git Configuration'
+  - task: PythonScript@0
+    displayName: 'Discover, Download, and Apply Patches (Mainline)'
+    inputs:
+      scriptSource: 'filePath'
+      scriptPath: '.ci/scripts/merge/apply-patches-by-label.py'
+      arguments: '${{ parameters.matchLabelPublic }} $(MergeTaglinePublic) patches-public'
+      workingDirectory: '$(System.DefaultWorkingDirectory)'
+  - task: PythonScript@0
+    displayName: 'Discover, Download, and Apply Patches (Patreon Public)'
+    inputs:
+      scriptSource: 'filePath'
+      scriptPath: '.ci/scripts/merge/apply-patches-by-label.py'
+      arguments: '${{ parameters.matchLabel }} "$(MergeTaglinePrivate) Public" patches-mixed-public'
+      workingDirectory: '$(System.DefaultWorkingDirectory)'
+  - task: PythonScript@0
+    displayName: 'Discover, Download, and Apply Patches (Patreon Private)'
+    inputs:
+      scriptSource: 'filePath'
+      scriptPath: '.ci/scripts/merge/apply-patches-by-label-private.py'
+      arguments: '$(PrivateMergeUser) ${{ parameters.matchLabel }} "$(MergeTaglinePrivate) Private" patches-private'
+      workingDirectory: '$(System.DefaultWorkingDirectory)'
@@ -11,5 +11,5 @@ steps:
    inputs:
      scriptSource: 'filePath'
      scriptPath: '.ci/scripts/merge/apply-patches-by-label.py'
-      arguments: '${{ parameters.matchLabel }} patches'
+      arguments: '${{ parameters.matchLabel }} Tagged patches'
      workingDirectory: '$(System.DefaultWorkingDirectory)'
@@ -2,7 +2,7 @@ steps:
  - task: DownloadPipelineArtifact@2
    displayName: 'Download Windows Release'
    inputs:
-      artifactName: 'yuzu-$(BuildName)-windows-mingw'
+      artifactName: 'yuzu-$(BuildName)-windows-msvc'
      buildType: 'current'
      targetPath: '$(Build.ArtifactStagingDirectory)'
  - task: DownloadPipelineArtifact@2
@@ -15,12 +15,47 @@ stages:
  dependsOn: format
  displayName: 'build'
  jobs:
-  - template: ./templates/build-standard.yml
-    parameters:
-      cache: 'true'
+  - job: build
+    displayName: 'standard'
+    pool:
+      vmImage: ubuntu-latest
+    strategy:
+      maxParallel: 10
+      matrix:
+        linux:
+          BuildSuffix: 'linux'
+          ScriptFolder: 'linux'
+    steps:
+    - template: ./templates/sync-source.yml
+      parameters:
+        artifactSource: $(parameters.artifactSource)
+        needSubmodules: 'true'
+    - template: ./templates/build-single.yml
+      parameters:
+        artifactSource: 'false'
+        cache: 'true'
+- stage: build_win
+  dependsOn: format
+  displayName: 'build-windows'
+  jobs:
+  - job: build
+    displayName: 'msvc'
+    pool:
+      vmImage: vs2017-win2016
+    steps:
+    - template: ./templates/sync-source.yml
+      parameters:
+        artifactSource: $(parameters.artifactSource)
+        needSubmodules: 'true'
+    - template: ./templates/build-msvc.yml
+      parameters:
+        artifactSource: 'false'
+        cache: 'true'
 - stage: release
  displayName: 'Release'
-  dependsOn: build
+  dependsOn:
+  - build
+  - build_win
  jobs:
  - job: github
    displayName: 'GitHub Release'
@@ -0,0 +1,8 @@
+trigger:
+- master
+
+stages:
+- stage: merge
+  displayName: 'merge'
+  jobs:
+  - template: ./templates/merge-private.yml
@@ -0,0 +1,30 @@
+trigger:
+- master
+
+stages:
+- stage: format
+  displayName: 'format'
+  jobs:
+  - job: format
+    displayName: 'clang'
+    pool:
+      vmImage: ubuntu-latest
+    steps:
+    - template: ./templates/format-check.yml
+- stage: build
+  dependsOn: format
+  displayName: 'build'
+  jobs:
+  - job: build
+    displayName: 'windows-msvc'
+    pool:
+      vmImage: vs2017-win2016
+    steps:
+    - template: ./templates/sync-source.yml
+      parameters:
+        artifactSource: $(parameters.artifactSource)
+        needSubmodules: 'true'
+    - template: ./templates/build-msvc.yml
+      parameters:
+        artifactSource: 'false'
+        cache: $(parameters.cache)
@@ -1,19 +0,0 @@
-# Starter pipeline
-# Start with a minimal pipeline that you can customize to build and deploy your code.
-# Add steps that build, run tests, deploy, and more:
-# https://aka.ms/yaml
-
-trigger:
- master
-
-pool:
-  vmImage: 'ubuntu-latest'
-
-steps:
- script: echo Hello, world!
-  displayName: 'Run a one-line script'
-
- script: |
-    echo Add other tasks to build, test, and deploy your project.
-    echo See https://aka.ms/yaml
-  displayName: 'Run a multi-line script'
@@ -1,7 +1,6 @@
 yuzu emulator
 =============
 [![Travis CI Build Status](https://travis-ci.org/yuzu-emu/yuzu.svg?branch=master)](https://travis-ci.org/yuzu-emu/yuzu)
-[![AppVeyor CI Build Status](https://ci.appveyor.com/api/projects/status/77k97svb2usreu68?svg=true)](https://ci.appveyor.com/project/bunnei/yuzu)
 [![Azure Mainline CI Build Status](https://dev.azure.com/yuzu-emu/yuzu/_apis/build/status/yuzu%20mainline?branchName=master)](https://dev.azure.com/yuzu-emu/yuzu/)

 yuzu is an experimental open-source emulator for the Nintendo Switch from the creators of [Citra](https://citra-emu.org/).
@@ -10,6 +10,9 @@ if (DEFINED ENV{CI})
  elseif(DEFINED ENV{APPVEYOR})
    set(BUILD_REPOSITORY $ENV{APPVEYOR_REPO_NAME})
    set(BUILD_TAG $ENV{APPVEYOR_REPO_TAG_NAME})
+  elseif(DEFINED ENV{AZURE})
+    set(BUILD_REPOSITORY $ENV{AZURE_REPO_NAME})
+    set(BUILD_TAG $ENV{AZURE_REPO_TAG})
  endif()
 endif()
 add_custom_command(OUTPUT scm_rev.cpp
@@ -28,18 +28,14 @@ __declspec(noinline, noreturn)
 }

 #define ASSERT(_a_)                                                                                \
-    do                                                                                             \
-        if (!(_a_)) {                                                                              \
-            assert_noinline_call([] { LOG_CRITICAL(Debug, "Assertion Failed!"); });                \
-        }                                                                                          \
-    while (0)
+    if (!(_a_)) {                                                                                  \
+        LOG_CRITICAL(Debug, "Assertion Failed!");                                                  \
+    }

 #define ASSERT_MSG(_a_, ...)                                                                       \
-    do                                                                                             \
-        if (!(_a_)) {                                                                              \
-            assert_noinline_call([&] { LOG_CRITICAL(Debug, "Assertion Failed!\n" __VA_ARGS__); }); \
-        }                                                                                          \
-    while (0)
+    if (!(_a_)) {                                                                                  \
+        LOG_CRITICAL(Debug, "Assertion Failed! " __VA_ARGS__);                                     \
+    }

 #define UNREACHABLE() ASSERT_MSG(false, "Unreachable code!")
 #define UNREACHABLE_MSG(...) ASSERT_MSG(false, __VA_ARGS__)
@@ -252,6 +252,8 @@ struct System::Impl {
        is_powered_on = false;
        exit_lock = false;

+        gpu_core->WaitIdle();
+
        // Shutdown emulation session
        renderer.reset();
        GDBStub::Shutdown();
@@ -23,6 +23,8 @@ SharedPtr<TransferMemory> TransferMemory::Create(KernelCore& kernel, VAddr base_
    transfer_memory->owner_permissions = permissions;
    transfer_memory->owner_process = kernel.CurrentProcess();

+    transfer_memory->MapMemory(base_address, size, permissions);
+
    return transfer_memory;
 }

@@ -48,17 +48,8 @@ SharedPtr<Thread> WaitObject::GetHighestPriorityReadyThread() const {
        if (ShouldWait(thread.get()))
            continue;

-        // A thread is ready to run if it's either in ThreadStatus::WaitSynch
-        // and the rest of the objects it is waiting on are ready.
-        bool ready_to_run = true;
-        if (thread_status == ThreadStatus::WaitSynch) {
-            ready_to_run = thread->AllWaitObjectsReady();
-        }
-
-        if (ready_to_run) {
-            candidate = thread.get();
-            candidate_priority = thread->GetPriority();
-        }
+        candidate = thread.get();
+        candidate_priority = thread->GetPriority();
    }

    return candidate;
@@ -841,17 +841,16 @@ private:
    void PopInteractiveOutData(Kernel::HLERequestContext& ctx) {
        LOG_DEBUG(Service_AM, "called");

-        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
-
        const auto storage = applet->GetBroker().PopInteractiveDataToGame();
        if (storage == nullptr) {
            LOG_ERROR(Service_AM,
                      "storage is a nullptr. There is no data in the current interactive channel");
-
+            IPC::ResponseBuilder rb{ctx, 2};
            rb.Push(ERR_NO_DATA_IN_CHANNEL);
            return;
        }

+        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IStorage>(std::move(*storage));
    }
@@ -27,9 +27,9 @@ AppletDataBroker::AppletDataBroker(Kernel::KernelCore& kernel) {
    state_changed_event = Kernel::WritableEvent::CreateEventPair(
        kernel, Kernel::ResetType::Manual, "ILibraryAppletAccessor:StateChangedEvent");
    pop_out_data_event = Kernel::WritableEvent::CreateEventPair(
-        kernel, Kernel::ResetType::Manual, "ILibraryAppletAccessor:PopDataOutEvent");
+        kernel, Kernel::ResetType::Automatic, "ILibraryAppletAccessor:PopDataOutEvent");
    pop_interactive_out_data_event = Kernel::WritableEvent::CreateEventPair(
-        kernel, Kernel::ResetType::Manual, "ILibraryAppletAccessor:PopInteractiveDataOutEvent");
+        kernel, Kernel::ResetType::Automatic, "ILibraryAppletAccessor:PopInteractiveDataOutEvent");
 }

 AppletDataBroker::~AppletDataBroker() = default;
@@ -91,6 +91,7 @@ void SoftwareKeyboard::ExecuteInteractive() {

    if (status == INTERACTIVE_STATUS_OK) {
        complete = true;
+        broker.SignalStateChanged();
    } else {
        std::array<char16_t, SWKBD_OUTPUT_INTERACTIVE_BUFFER_SIZE / 2 - 2> string;
        std::memcpy(string.data(), data.data() + 4, string.size() * 2);
@@ -40,7 +40,10 @@ static FileSys::VirtualDir GetDirectoryRelativeWrapped(FileSys::VirtualDir base,
    if (dir_name.empty() || dir_name == "." || dir_name == "/" || dir_name == "\\")
        return base;

-    return base->GetDirectoryRelative(dir_name);
+    const auto res = base->GetDirectoryRelative(dir_name);
+    if (res == nullptr)
+        return base->CreateDirectoryRelative(dir_name);
+    return res;
 }

 VfsDirectoryServiceWrapper::VfsDirectoryServiceWrapper(FileSys::VirtualDir backing_)
@@ -165,13 +165,14 @@ void Controller_NPad::InitNewlyAddedControler(std::size_t controller_idx) {
    controller.battery_level[0] = BATTERY_FULL;
    controller.battery_level[1] = BATTERY_FULL;
    controller.battery_level[2] = BATTERY_FULL;
+    styleset_changed_events[controller_idx].writable->Signal();
 }

 void Controller_NPad::OnInit() {
    auto& kernel = system.Kernel();
    for (std::size_t i = 0; i < styleset_changed_events.size(); i++) {
        styleset_changed_events[i] = Kernel::WritableEvent::CreateEventPair(
-            kernel, Kernel::ResetType::Automatic, fmt::format("npad:NpadStyleSetChanged_{}", i));
+            kernel, Kernel::ResetType::Manual, fmt::format("npad:NpadStyleSetChanged_{}", i));
    }

    if (!IsControllerActivated()) {
@@ -433,7 +434,6 @@ void Controller_NPad::SetSupportedNPadIdTypes(u8* data, std::size_t length) {
    supported_npad_id_types.clear();
    supported_npad_id_types.resize(length / sizeof(u32));
    std::memcpy(supported_npad_id_types.data(), data, length);
-    bool had_controller_update = false;
    for (std::size_t i = 0; i < connected_controllers.size(); i++) {
        auto& controller = connected_controllers[i];
        if (!controller.is_connected) {
@@ -452,10 +452,6 @@ void Controller_NPad::SetSupportedNPadIdTypes(u8* data, std::size_t length) {
                controller.type = requested_controller;
                InitNewlyAddedControler(i);
            }
-            had_controller_update = true;
-        }
-        if (had_controller_update) {
-            styleset_changed_events[i].writable->Signal();
        }
    }
 }
@@ -481,7 +477,6 @@ void Controller_NPad::SetNpadMode(u32 npad_id, NPadAssignments assignment_mode)
    const std::size_t npad_index = NPadIdToIndex(npad_id);
    ASSERT(npad_index < shared_memory_entries.size());
    if (shared_memory_entries[npad_index].pad_assignment != assignment_mode) {
-        styleset_changed_events[npad_index].writable->Signal();
        shared_memory_entries[npad_index].pad_assignment = assignment_mode;
    }
 }
@@ -507,7 +502,6 @@ Kernel::SharedPtr<Kernel::ReadableEvent> Controller_NPad::GetStyleSetChangedEven
    // TODO(ogniK): Figure out the best time to signal this event. This event seems that it should
    // be signalled at least once, and signaled after a new controller is connected?
    const auto& styleset_event = styleset_changed_events[NPadIdToIndex(npad_id)];
-    styleset_event.writable->Signal();
    return styleset_event.readable;
 }

@@ -201,13 +201,13 @@ Hid::Hid(Core::System& system) : ServiceFramework("hid"), system(system) {
        {120, &Hid::SetNpadJoyHoldType, "SetNpadJoyHoldType"},
        {121, &Hid::GetNpadJoyHoldType, "GetNpadJoyHoldType"},
        {122, &Hid::SetNpadJoyAssignmentModeSingleByDefault, "SetNpadJoyAssignmentModeSingleByDefault"},
-        {123, nullptr, "SetNpadJoyAssignmentModeSingleByDefault"},
+        {123, &Hid::SetNpadJoyAssignmentModeSingle, "SetNpadJoyAssignmentModeSingle"},
        {124, &Hid::SetNpadJoyAssignmentModeDual, "SetNpadJoyAssignmentModeDual"},
        {125, &Hid::MergeSingleJoyAsDualJoy, "MergeSingleJoyAsDualJoy"},
        {126, &Hid::StartLrAssignmentMode, "StartLrAssignmentMode"},
        {127, &Hid::StopLrAssignmentMode, "StopLrAssignmentMode"},
        {128, &Hid::SetNpadHandheldActivationMode, "SetNpadHandheldActivationMode"},
-        {129, nullptr, "GetNpadHandheldActivationMode"},
+        {129, &Hid::GetNpadHandheldActivationMode, "GetNpadHandheldActivationMode"},
        {130, &Hid::SwapNpadAssignment, "SwapNpadAssignment"},
        {131, nullptr, "IsUnintendedHomeButtonInputProtectionEnabled"},
        {132, nullptr, "EnableUnintendedHomeButtonInputProtection"},
@@ -544,10 +544,126 @@ void Hid::SetNpadJoyAssignmentModeSingleByDefault(Kernel::HLERequestContext& ctx
    LOG_WARNING(Service_HID, "(STUBBED) called, npad_id={}, applet_resource_user_id={}", npad_id,
                applet_resource_user_id);

+    auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
+    controller.SetNpadMode(npad_id, Controller_NPad::NPadAssignments::Single);
+
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
 }

+void Hid::SetNpadJoyAssignmentModeSingle(Kernel::HLERequestContext& ctx) {
+    // TODO: Check the differences between this and SetNpadJoyAssignmentModeSingleByDefault
+    IPC::RequestParser rp{ctx};
+    const auto npad_id{rp.Pop<u32>()};
+    const auto applet_resource_user_id{rp.Pop<u64>()};
+    const auto npad_joy_device_type{rp.Pop<u64>()};
+
+    LOG_WARNING(Service_HID,
+                "(STUBBED) called, npad_id={}, applet_resource_user_id={}, npad_joy_device_type={}",
+                npad_id, applet_resource_user_id, npad_joy_device_type);
+
+    auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
+    controller.SetNpadMode(npad_id, Controller_NPad::NPadAssignments::Single);
+
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
+
+void Hid::SetNpadJoyAssignmentModeDual(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto npad_id{rp.Pop<u32>()};
+    const auto applet_resource_user_id{rp.Pop<u64>()};
+
+    LOG_DEBUG(Service_HID, "called, npad_id={}, applet_resource_user_id={}", npad_id,
+              applet_resource_user_id);
+
+    auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
+    controller.SetNpadMode(npad_id, Controller_NPad::NPadAssignments::Dual);
+
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
+
+void Hid::MergeSingleJoyAsDualJoy(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto unknown_1{rp.Pop<u32>()};
+    const auto unknown_2{rp.Pop<u32>()};
+    const auto applet_resource_user_id{rp.Pop<u64>()};
+
+    LOG_WARNING(Service_HID,
+                "(STUBBED) called, unknown_1={}, unknown_2={}, applet_resource_user_id={}",
+                unknown_1, unknown_2, applet_resource_user_id);
+
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
+
+void Hid::StartLrAssignmentMode(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto applet_resource_user_id{rp.Pop<u64>()};
+
+    LOG_DEBUG(Service_HID, "called, applet_resource_user_id={}", applet_resource_user_id);
+    auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
+    controller.StartLRAssignmentMode();
+
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
+
+void Hid::StopLrAssignmentMode(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto applet_resource_user_id{rp.Pop<u64>()};
+
+    LOG_DEBUG(Service_HID, "called, applet_resource_user_id={}", applet_resource_user_id);
+    auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
+    controller.StopLRAssignmentMode();
+
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
+
+void Hid::SetNpadHandheldActivationMode(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto applet_resource_user_id{rp.Pop<u64>()};
+    const auto mode{rp.Pop<u64>()};
+
+    LOG_WARNING(Service_HID, "(STUBBED) called, applet_resource_user_id={}, mode={}",
+                applet_resource_user_id, mode);
+
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
+
+void Hid::GetNpadHandheldActivationMode(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto applet_resource_user_id{rp.Pop<u64>()};
+
+    LOG_WARNING(Service_HID, "(STUBBED) called, applet_resource_user_id={}",
+                applet_resource_user_id);
+
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
+
+void Hid::SwapNpadAssignment(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto npad_1{rp.Pop<u32>()};
+    const auto npad_2{rp.Pop<u32>()};
+    const auto applet_resource_user_id{rp.Pop<u64>()};
+
+    LOG_DEBUG(Service_HID, "called, applet_resource_user_id={}, npad_1={}, npad_2={}",
+              applet_resource_user_id, npad_1, npad_2);
+
+    auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
+    IPC::ResponseBuilder rb{ctx, 2};
+    if (controller.SwapNpadAssignment(npad_1, npad_2)) {
+        rb.Push(RESULT_SUCCESS);
+    } else {
+        LOG_ERROR(Service_HID, "Npads are not connected!");
+        rb.Push(ERR_NPAD_NOT_CONNECTED);
+    }
+}
+
 void Hid::BeginPermitVibrationSession(Kernel::HLERequestContext& ctx) {
    IPC::RequestParser rp{ctx};
    const auto applet_resource_user_id{rp.Pop<u64>()};
@@ -622,47 +738,6 @@ void Hid::GetActualVibrationValue(Kernel::HLERequestContext& ctx) {
        applet_resource->GetController<Controller_NPad>(HidController::NPad).GetLastVibration());
 }

-void Hid::SetNpadJoyAssignmentModeDual(Kernel::HLERequestContext& ctx) {
-    IPC::RequestParser rp{ctx};
-    const auto npad_id{rp.Pop<u32>()};
-    const auto applet_resource_user_id{rp.Pop<u64>()};
-
-    LOG_DEBUG(Service_HID, "called, npad_id={}, applet_resource_user_id={}", npad_id,
-              applet_resource_user_id);
-
-    auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
-    controller.SetNpadMode(npad_id, Controller_NPad::NPadAssignments::Dual);
-
-    IPC::ResponseBuilder rb{ctx, 2};
-    rb.Push(RESULT_SUCCESS);
-}
-
-void Hid::MergeSingleJoyAsDualJoy(Kernel::HLERequestContext& ctx) {
-    IPC::RequestParser rp{ctx};
-    const auto unknown_1{rp.Pop<u32>()};
-    const auto unknown_2{rp.Pop<u32>()};
-    const auto applet_resource_user_id{rp.Pop<u64>()};
-
-    LOG_WARNING(Service_HID,
-                "(STUBBED) called, unknown_1={}, unknown_2={}, applet_resource_user_id={}",
-                unknown_1, unknown_2, applet_resource_user_id);
-
-    IPC::ResponseBuilder rb{ctx, 2};
-    rb.Push(RESULT_SUCCESS);
-}
-
-void Hid::SetNpadHandheldActivationMode(Kernel::HLERequestContext& ctx) {
-    IPC::RequestParser rp{ctx};
-    const auto applet_resource_user_id{rp.Pop<u64>()};
-    const auto mode{rp.Pop<u64>()};
-
-    LOG_WARNING(Service_HID, "(STUBBED) called, applet_resource_user_id={}, mode={}",
-                applet_resource_user_id, mode);
-
-    IPC::ResponseBuilder rb{ctx, 2};
-    rb.Push(RESULT_SUCCESS);
-}
-
 void Hid::GetVibrationDeviceInfo(Kernel::HLERequestContext& ctx) {
    LOG_DEBUG(Service_HID, "called");

@@ -756,49 +831,6 @@ void Hid::SetPalmaBoostMode(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
 }

-void Hid::StartLrAssignmentMode(Kernel::HLERequestContext& ctx) {
-    IPC::RequestParser rp{ctx};
-    const auto applet_resource_user_id{rp.Pop<u64>()};
-
-    LOG_DEBUG(Service_HID, "called, applet_resource_user_id={}", applet_resource_user_id);
-    auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
-    controller.StartLRAssignmentMode();
-
-    IPC::ResponseBuilder rb{ctx, 2};
-    rb.Push(RESULT_SUCCESS);
-}
-
-void Hid::StopLrAssignmentMode(Kernel::HLERequestContext& ctx) {
-    IPC::RequestParser rp{ctx};
-    const auto applet_resource_user_id{rp.Pop<u64>()};
-
-    LOG_DEBUG(Service_HID, "called, applet_resource_user_id={}", applet_resource_user_id);
-    auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
-    controller.StopLRAssignmentMode();
-
-    IPC::ResponseBuilder rb{ctx, 2};
-    rb.Push(RESULT_SUCCESS);
-}
-
-void Hid::SwapNpadAssignment(Kernel::HLERequestContext& ctx) {
-    IPC::RequestParser rp{ctx};
-    const auto npad_1{rp.Pop<u32>()};
-    const auto npad_2{rp.Pop<u32>()};
-    const auto applet_resource_user_id{rp.Pop<u64>()};
-
-    LOG_DEBUG(Service_HID, "called, applet_resource_user_id={}, npad_1={}, npad_2={}",
-              applet_resource_user_id, npad_1, npad_2);
-
-    auto& controller = applet_resource->GetController<Controller_NPad>(HidController::NPad);
-    IPC::ResponseBuilder rb{ctx, 2};
-    if (controller.SwapNpadAssignment(npad_1, npad_2)) {
-        rb.Push(RESULT_SUCCESS);
-    } else {
-        LOG_ERROR(Service_HID, "Npads are not connected!");
-        rb.Push(ERR_NPAD_NOT_CONNECTED);
-    }
-}
-
 class HidDbg final : public ServiceFramework<HidDbg> {
 public:
    explicit HidDbg() : ServiceFramework{"hid:dbg"} {
@@ -105,14 +105,19 @@ private:
    void SetNpadJoyHoldType(Kernel::HLERequestContext& ctx);
    void GetNpadJoyHoldType(Kernel::HLERequestContext& ctx);
    void SetNpadJoyAssignmentModeSingleByDefault(Kernel::HLERequestContext& ctx);
+    void SetNpadJoyAssignmentModeSingle(Kernel::HLERequestContext& ctx);
+    void SetNpadJoyAssignmentModeDual(Kernel::HLERequestContext& ctx);
+    void MergeSingleJoyAsDualJoy(Kernel::HLERequestContext& ctx);
+    void StartLrAssignmentMode(Kernel::HLERequestContext& ctx);
+    void StopLrAssignmentMode(Kernel::HLERequestContext& ctx);
+    void SetNpadHandheldActivationMode(Kernel::HLERequestContext& ctx);
+    void GetNpadHandheldActivationMode(Kernel::HLERequestContext& ctx);
+    void SwapNpadAssignment(Kernel::HLERequestContext& ctx);
    void BeginPermitVibrationSession(Kernel::HLERequestContext& ctx);
    void EndPermitVibrationSession(Kernel::HLERequestContext& ctx);
    void SendVibrationValue(Kernel::HLERequestContext& ctx);
    void SendVibrationValues(Kernel::HLERequestContext& ctx);
    void GetActualVibrationValue(Kernel::HLERequestContext& ctx);
-    void SetNpadJoyAssignmentModeDual(Kernel::HLERequestContext& ctx);
-    void MergeSingleJoyAsDualJoy(Kernel::HLERequestContext& ctx);
-    void SetNpadHandheldActivationMode(Kernel::HLERequestContext& ctx);
    void GetVibrationDeviceInfo(Kernel::HLERequestContext& ctx);
    void CreateActiveVibrationDeviceList(Kernel::HLERequestContext& ctx);
    void PermitVibration(Kernel::HLERequestContext& ctx);
@@ -122,9 +127,6 @@ private:
    void StopSixAxisSensor(Kernel::HLERequestContext& ctx);
    void SetIsPalmaAllConnectable(Kernel::HLERequestContext& ctx);
    void SetPalmaBoostMode(Kernel::HLERequestContext& ctx);
-    void StartLrAssignmentMode(Kernel::HLERequestContext& ctx);
-    void StopLrAssignmentMode(Kernel::HLERequestContext& ctx);
-    void SwapNpadAssignment(Kernel::HLERequestContext& ctx);

    std::shared_ptr<IAppletResource> applet_resource;
    Core::System& system;
@@ -10,6 +10,8 @@
 #include "core/hle/service/lbl/lbl.h"
 #include "core/hle/service/service.h"
 #include "core/hle/service/sm/sm.h"
+#include "core/settings.h"
+#include "video_core/renderer_base.h"

 namespace Service::LBL {

@@ -18,21 +20,21 @@ public:
    explicit LBL() : ServiceFramework{"lbl"} {
        // clang-format off
        static const FunctionInfo functions[] = {
-            {0, nullptr, "SaveCurrentSetting"},
-            {1, nullptr, "LoadCurrentSetting"},
-            {2, nullptr, "SetCurrentBrightnessSetting"},
-            {3, nullptr, "GetCurrentBrightnessSetting"},
-            {4, nullptr, "ApplyCurrentBrightnessSettingToBacklight"},
-            {5, nullptr, "GetBrightnessSettingAppliedToBacklight"},
-            {6, nullptr, "SwitchBacklightOn"},
-            {7, nullptr, "SwitchBacklightOff"},
-            {8, nullptr, "GetBacklightSwitchStatus"},
-            {9, nullptr, "EnableDimming"},
-            {10, nullptr, "DisableDimming"},
-            {11, nullptr, "IsDimmingEnabled"},
-            {12, nullptr, "EnableAutoBrightnessControl"},
-            {13, nullptr, "DisableAutoBrightnessControl"},
-            {14, nullptr, "IsAutoBrightnessControlEnabled"},
+            {0, &LBL::SaveCurrentSetting, "SaveCurrentSetting"},
+            {1, &LBL::LoadCurrentSetting, "LoadCurrentSetting"},
+            {2, &LBL::SetCurrentBrightnessSetting, "SetCurrentBrightnessSetting"},
+            {3, &LBL::GetCurrentBrightnessSetting, "GetCurrentBrightnessSetting"},
+            {4, &LBL::ApplyCurrentBrightnessSettingToBacklight, "ApplyCurrentBrightnessSettingToBacklight"},
+            {5, &LBL::GetBrightnessSettingAppliedToBacklight, "GetBrightnessSettingAppliedToBacklight"},
+            {6, &LBL::SwitchBacklightOn, "SwitchBacklightOn"},
+            {7, &LBL::SwitchBacklightOff, "SwitchBacklightOff"},
+            {8, &LBL::GetBacklightSwitchStatus, "GetBacklightSwitchStatus"},
+            {9, &LBL::EnableDimming, "EnableDimming"},
+            {10, &LBL::DisableDimming, "DisableDimming"},
+            {11, &LBL::IsDimmingEnabled, "IsDimmingEnabled"},
+            {12, &LBL::EnableAutoBrightnessControl, "EnableAutoBrightnessControl"},
+            {13, &LBL::DisableAutoBrightnessControl, "DisableAutoBrightnessControl"},
+            {14, &LBL::IsAutoBrightnessControlEnabled, "IsAutoBrightnessControlEnabled"},
            {15, nullptr, "SetAmbientLightSensorValue"},
            {16, nullptr, "GetAmbientLightSensorValue"},
            {17, nullptr, "SetBrightnessReflectionDelayLevel"},
@@ -42,8 +44,8 @@ public:
            {21, nullptr, "SetCurrentAmbientLightSensorMapping"},
            {22, nullptr, "GetCurrentAmbientLightSensorMapping"},
            {23, nullptr, "IsAmbientLightSensorAvailable"},
-            {24, nullptr, "SetCurrentBrightnessSettingForVrMode"},
-            {25, nullptr, "GetCurrentBrightnessSettingForVrMode"},
+            {24, &LBL::SetCurrentBrightnessSettingForVrMode, "SetCurrentBrightnessSettingForVrMode"},
+            {25, &LBL::GetCurrentBrightnessSettingForVrMode, "GetCurrentBrightnessSettingForVrMode"},
            {26, &LBL::EnableVrMode, "EnableVrMode"},
            {27, &LBL::DisableVrMode, "DisableVrMode"},
            {28, &LBL::IsVrModeEnabled, "IsVrModeEnabled"},
@@ -53,13 +55,209 @@ public:
        RegisterHandlers(functions);
    }

+    void LoadFromSettings() {
+        current_brightness = Settings::values.backlight_brightness;
+        current_vr_mode_brightness = Settings::values.backlight_brightness;
+
+        if (auto_brightness_enabled) {
+            return;
+        }
+
+        if (vr_mode_enabled) {
+            Renderer().SetCurrentBrightness(current_vr_mode_brightness);
+        } else {
+            Renderer().SetCurrentBrightness(current_brightness);
+        }
+    }
+
 private:
+    f32 GetAutoBrightnessValue() const {
+        return 0.5f;
+    }
+
+    VideoCore::RendererBase& Renderer() {
+        return Core::System::GetInstance().Renderer();
+    }
+
+    void SaveCurrentSetting(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        Settings::values.backlight_brightness = current_brightness;
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
+    }
+
+    void LoadCurrentSetting(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        LoadFromSettings();
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
+    }
+
+    void SetCurrentBrightnessSetting(Kernel::HLERequestContext& ctx) {
+        IPC::RequestParser rp{ctx};
+        const auto value = rp.PopRaw<f32>();
+
+        LOG_DEBUG(Service_LBL, "called, value={:.3f}", value);
+
+        current_brightness = std::clamp(value, 0.0f, 1.0f);
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
+    }
+
+    void GetCurrentBrightnessSetting(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        IPC::ResponseBuilder rb{ctx, 3};
+        rb.Push(RESULT_SUCCESS);
+        rb.Push(current_brightness);
+    }
+
+    void ApplyCurrentBrightnessSettingToBacklight(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        if (!auto_brightness_enabled) {
+            Renderer().SetCurrentBrightness(vr_mode_enabled ? current_vr_mode_brightness
+                                                            : current_brightness);
+        }
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
+    }
+
+    void GetBrightnessSettingAppliedToBacklight(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        IPC::ResponseBuilder rb{ctx, 3};
+        rb.Push(RESULT_SUCCESS);
+        rb.Push(Renderer().GetCurrentResultantBrightness());
+    }
+
+    void SwitchBacklightOn(Kernel::HLERequestContext& ctx) {
+        IPC::RequestParser rp{ctx};
+        const auto fade_time = rp.PopRaw<u64>();
+
+        LOG_DEBUG(Service_LBL, "called, fade_time={:016X}", fade_time);
+
+        Renderer().SetBacklightStatus(true, fade_time);
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
+    }
+
+    void SwitchBacklightOff(Kernel::HLERequestContext& ctx) {
+        IPC::RequestParser rp{ctx};
+        const auto fade_time = rp.PopRaw<u64>();
+
+        LOG_DEBUG(Service_LBL, "called, fade_time={:016X}", fade_time);
+
+        Renderer().SetBacklightStatus(false, fade_time);
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
+    }
+
+    void GetBacklightSwitchStatus(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        IPC::ResponseBuilder rb{ctx, 3};
+        rb.Push(RESULT_SUCCESS);
+        rb.Push<u8>(Renderer().GetBacklightStatus());
+    }
+
+    void EnableDimming(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        dimming_enabled = true;
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
+    }
+
+    void DisableDimming(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "callled");
+
+        dimming_enabled = false;
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
+    }
+
+    void IsDimmingEnabled(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        IPC::ResponseBuilder rb{ctx, 3};
+        rb.Push(RESULT_SUCCESS);
+        rb.Push<u8>(dimming_enabled);
+    }
+
+    void EnableAutoBrightnessControl(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        auto_brightness_enabled = true;
+        Renderer().SetCurrentBrightness(GetAutoBrightnessValue());
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
+    }
+
+    void DisableAutoBrightnessControl(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        auto_brightness_enabled = false;
+        Renderer().SetCurrentBrightness(current_brightness);
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
+    }
+
+    void IsAutoBrightnessControlEnabled(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        IPC::ResponseBuilder rb{ctx, 3};
+        rb.Push(RESULT_SUCCESS);
+        rb.Push<u8>(auto_brightness_enabled);
+    }
+
+    void SetCurrentBrightnessSettingForVrMode(Kernel::HLERequestContext& ctx) {
+        IPC::RequestParser rp{ctx};
+        const auto value = rp.PopRaw<f32>();
+
+        LOG_DEBUG(Service_LBL, "called, value={:.3f}", value);
+
+        current_vr_mode_brightness = std::clamp(value, 0.0f, 1.0f);
+
+        if (vr_mode_enabled && !auto_brightness_enabled) {
+            Renderer().SetCurrentBrightness(value);
+        }
+
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(RESULT_SUCCESS);
+    }
+
+    void GetCurrentBrightnessSettingForVrMode(Kernel::HLERequestContext& ctx) {
+        LOG_DEBUG(Service_LBL, "called");
+
+        IPC::ResponseBuilder rb{ctx, 3};
+        rb.Push(RESULT_SUCCESS);
+        rb.Push(current_vr_mode_brightness);
+    }
+
    void EnableVrMode(Kernel::HLERequestContext& ctx) {
        LOG_DEBUG(Service_LBL, "called");

        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);

+        if (!vr_mode_enabled && !auto_brightness_enabled &&
+            current_brightness != current_vr_mode_brightness) {
+            Renderer().SetCurrentBrightness(current_vr_mode_brightness);
+        }
+
        vr_mode_enabled = true;
    }

@@ -69,6 +267,11 @@ private:
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);

+        if (vr_mode_enabled && !auto_brightness_enabled &&
+            current_brightness != current_vr_mode_brightness) {
+            Renderer().SetCurrentBrightness(current_brightness);
+        }
+
        vr_mode_enabled = false;
    }

@@ -80,9 +283,27 @@ private:
        rb.Push(vr_mode_enabled);
    }

+    bool auto_brightness_enabled = false;
+    bool dimming_enabled = true;
+
+    f32 current_brightness = GetAutoBrightnessValue();
+    f32 current_vr_mode_brightness = GetAutoBrightnessValue();
+
    bool vr_mode_enabled = false;
 };

+void RequestLoadCurrentSetting(SM::ServiceManager& sm) {
+    if (&sm == nullptr) {
+        return;
+    }
+
+    const auto lbl = sm.GetService<LBL>("lbl");
+
+    if (lbl) {
+        lbl->LoadFromSettings();
+    }
+}
+
 void InstallInterfaces(SM::ServiceManager& sm) {
    std::make_shared<LBL>()->InstallAsService(sm);
 }
@@ -10,6 +10,9 @@ class ServiceManager;

 namespace Service::LBL {

+// Requests the LBL service passed to load brightness values from Settings
+void RequestLoadCurrentSetting(SM::ServiceManager& sm);
+
 void InstallInterfaces(SM::ServiceManager& sm);

 } // namespace Service::LBL
@@ -5,6 +5,7 @@
 #include "common/assert.h"
 #include "common/logging/log.h"
 #include "core/core.h"
+#include "core/core_timing.h"
 #include "core/hle/service/nvdrv/devices/nvdisp_disp0.h"
 #include "core/hle/service/nvdrv/devices/nvmap.h"
 #include "core/perf_stats.h"
@@ -38,7 +39,10 @@ void nvdisp_disp0::flip(u32 buffer_handle, u32 offset, u32 format, u32 width, u3
        transform, crop_rect};

    system.GetPerfStats().EndGameFrame();
+    system.GetPerfStats().EndSystemFrame();
    system.GPU().SwapBuffers(&framebuffer);
+    system.FrameLimiter().DoFrameLimiting(system.CoreTiming().GetGlobalTimeUs());
+    system.GetPerfStats().BeginSystemFrame();
 }

 } // namespace Service::Nvidia::Devices
@@ -62,16 +62,26 @@ u32 nvhost_ctrl::IocCtrlEventWait(const std::vector<u8>& input, std::vector<u8>&
        return NvResult::BadParameter;
    }

+    u32 event_id = params.value & 0x00FF;
+
+    if (event_id >= MaxNvEvents) {
+        std::memcpy(output.data(), &params, sizeof(params));
+        return NvResult::BadParameter;
+    }
+
+    auto event = events_interface.events[event_id];
    auto& gpu = system.GPU();
    // This is mostly to take into account unimplemented features. As synced
    // gpu is always synced.
    if (!gpu.IsAsync()) {
+        event.writable->Signal();
        return NvResult::Success;
    }
    auto lock = gpu.LockSync();
    const u32 current_syncpoint_value = gpu.GetSyncpointValue(params.syncpt_id);
    const s32 diff = current_syncpoint_value - params.threshold;
    if (diff >= 0) {
+        event.writable->Signal();
        params.value = current_syncpoint_value;
        std::memcpy(output.data(), &params, sizeof(params));
        return NvResult::Success;
@@ -87,27 +97,6 @@ u32 nvhost_ctrl::IocCtrlEventWait(const std::vector<u8>& input, std::vector<u8>&
        return NvResult::Timeout;
    }

-    u32 event_id;
-    if (is_async) {
-        event_id = params.value & 0x00FF;
-        if (event_id >= MaxNvEvents) {
-            std::memcpy(output.data(), &params, sizeof(params));
-            return NvResult::BadParameter;
-        }
-    } else {
-        if (ctrl.fresh_call) {
-            const auto result = events_interface.GetFreeEvent();
-            if (result) {
-                event_id = *result;
-            } else {
-                LOG_CRITICAL(Service_NVDRV, "No Free Events available!");
-                event_id = params.value & 0x00FF;
-            }
-        } else {
-            event_id = ctrl.event_id;
-        }
-    }
-
    EventState status = events_interface.status[event_id];
    if (event_id < MaxNvEvents || status == EventState::Free || status == EventState::Registered) {
        events_interface.SetEventStatus(event_id, EventState::Waiting);
@@ -119,7 +108,7 @@ u32 nvhost_ctrl::IocCtrlEventWait(const std::vector<u8>& input, std::vector<u8>&
            params.value = ((params.syncpt_id & 0xfff) << 16) | 0x10000000;
        }
        params.value |= event_id;
-        events_interface.events[event_id].writable->Clear();
+        event.writable->Clear();
        gpu.RegisterSyncptInterrupt(params.syncpt_id, target_value);
        if (!is_async && ctrl.fresh_call) {
            ctrl.must_delay = true;
@@ -22,6 +22,18 @@ u32 nvhost_nvdec::ioctl(Ioctl command, const std::vector<u8>& input, const std::
    switch (static_cast<IoctlCommand>(command.raw)) {
    case IoctlCommand::IocSetNVMAPfdCommand:
        return SetNVMAPfd(input, output);
+    case IoctlCommand::IocSubmit:
+        return Submit(input, output);
+    case IoctlCommand::IocGetSyncpoint:
+        return GetSyncpoint(input, output);
+    case IoctlCommand::IocGetWaitbase:
+        return GetWaitbase(input, output);
+    case IoctlCommand::IocMapBuffer:
+        return MapBuffer(input, output);
+    case IoctlCommand::IocMapBufferEx:
+        return MapBufferEx(input, output);
+    case IoctlCommand::IocUnmapBufferEx:
+        return UnmapBufferEx(input, output);
    }

    UNIMPLEMENTED_MSG("Unimplemented ioctl");
@@ -30,11 +42,67 @@ u32 nvhost_nvdec::ioctl(Ioctl command, const std::vector<u8>& input, const std::

 u32 nvhost_nvdec::SetNVMAPfd(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlSetNvmapFD params{};
-    std::memcpy(&params, input.data(), input.size());
+    std::memcpy(&params, input.data(), sizeof(IoctlSetNvmapFD));
    LOG_DEBUG(Service_NVDRV, "called, fd={}", params.nvmap_fd);

    nvmap_fd = params.nvmap_fd;
    return 0;
 }

+u32 nvhost_nvdec::Submit(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlSubmit params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlSubmit));
+    LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+    std::memcpy(output.data(), &params, sizeof(IoctlSubmit));
+    return 0;
+}
+
+u32 nvhost_nvdec::GetSyncpoint(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlGetSyncpoint params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlGetSyncpoint));
+    LOG_INFO(Service_NVDRV, "called, unknown=0x{:X}", params.unknown);
+    params.value = 0; // Seems to be hard coded at 0
+    std::memcpy(output.data(), &params, sizeof(IoctlGetSyncpoint));
+    return 0;
+}
+
+u32 nvhost_nvdec::GetWaitbase(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlGetWaitbase params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlGetWaitbase));
+    LOG_INFO(Service_NVDRV, "called, unknown=0x{:X}", params.unknown);
+    params.value = 0; // Seems to be hard coded at 0
+    std::memcpy(output.data(), &params, sizeof(IoctlGetWaitbase));
+    return 0;
+}
+
+u32 nvhost_nvdec::MapBuffer(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlMapBuffer params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlMapBuffer));
+    LOG_WARNING(Service_NVDRV, "(STUBBED) called with address={:08X}{:08X}", params.address_2,
+                params.address_1);
+    params.address_1 = 0;
+    params.address_2 = 0;
+    std::memcpy(output.data(), &params, sizeof(IoctlMapBuffer));
+    return 0;
+}
+
+u32 nvhost_nvdec::MapBufferEx(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlMapBufferEx params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlMapBufferEx));
+    LOG_WARNING(Service_NVDRV, "(STUBBED) called with address={:08X}{:08X}", params.address_2,
+                params.address_1);
+    params.address_1 = 0;
+    params.address_2 = 0;
+    std::memcpy(output.data(), &params, sizeof(IoctlMapBufferEx));
+    return 0;
+}
+
+u32 nvhost_nvdec::UnmapBufferEx(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlUnmapBufferEx params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlUnmapBufferEx));
+    LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+    std::memcpy(output.data(), &params, sizeof(IoctlUnmapBufferEx));
+    return 0;
+}
+
 } // namespace Service::Nvidia::Devices
@@ -23,16 +23,66 @@ public:
 private:
    enum class IoctlCommand : u32_le {
        IocSetNVMAPfdCommand = 0x40044801,
+        IocSubmit = 0xC0400001,
+        IocGetSyncpoint = 0xC0080002,
+        IocGetWaitbase = 0xC0080003,
+        IocMapBuffer = 0xC01C0009,
+        IocMapBufferEx = 0xC0A40009,
+        IocUnmapBufferEx = 0xC0A4000A,
    };

    struct IoctlSetNvmapFD {
        u32_le nvmap_fd;
    };
-    static_assert(sizeof(IoctlSetNvmapFD) == 4, "IoctlSetNvmapFD is incorrect size");
+    static_assert(sizeof(IoctlSetNvmapFD) == 0x4, "IoctlSetNvmapFD is incorrect size");
+
+    struct IoctlSubmit {
+        INSERT_PADDING_BYTES(0x40); // TODO(DarkLordZach): RE this structure
+    };
+    static_assert(sizeof(IoctlSubmit) == 0x40, "IoctlSubmit has incorrect size");
+
+    struct IoctlGetSyncpoint {
+        u32 unknown; // seems to be ignored? Nintendo added this
+        u32 value;
+    };
+    static_assert(sizeof(IoctlGetSyncpoint) == 0x08, "IoctlGetSyncpoint has incorrect size");
+
+    struct IoctlGetWaitbase {
+        u32 unknown; // seems to be ignored? Nintendo added this
+        u32 value;
+    };
+    static_assert(sizeof(IoctlGetWaitbase) == 0x08, "IoctlGetWaitbase has incorrect size");
+
+    struct IoctlMapBuffer {
+        u32 unknown;
+        u32 address_1;
+        u32 address_2;
+        INSERT_PADDING_BYTES(0x10); // TODO(DarkLordZach): RE this structure
+    };
+    static_assert(sizeof(IoctlMapBuffer) == 0x1C, "IoctlMapBuffer is incorrect size");
+
+    struct IoctlMapBufferEx {
+        u32 unknown;
+        u32 address_1;
+        u32 address_2;
+        INSERT_PADDING_BYTES(0x98); // TODO(DarkLordZach): RE this structure
+    };
+    static_assert(sizeof(IoctlMapBufferEx) == 0xA4, "IoctlMapBufferEx has incorrect size");
+
+    struct IoctlUnmapBufferEx {
+        INSERT_PADDING_BYTES(0xA4); // TODO(DarkLordZach): RE this structure
+    };
+    static_assert(sizeof(IoctlUnmapBufferEx) == 0xA4, "IoctlUnmapBufferEx has incorrect size");

    u32_le nvmap_fd{};

    u32 SetNVMAPfd(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 Submit(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 GetSyncpoint(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 GetWaitbase(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 MapBuffer(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 MapBufferEx(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 UnmapBufferEx(const std::vector<u8>& input, std::vector<u8>& output);
 };

 } // namespace Service::Nvidia::Devices
@@ -22,6 +22,18 @@ u32 nvhost_vic::ioctl(Ioctl command, const std::vector<u8>& input, const std::ve
    switch (static_cast<IoctlCommand>(command.raw)) {
    case IoctlCommand::IocSetNVMAPfdCommand:
        return SetNVMAPfd(input, output);
+    case IoctlCommand::IocSubmit:
+        return Submit(input, output);
+    case IoctlCommand::IocGetSyncpoint:
+        return GetSyncpoint(input, output);
+    case IoctlCommand::IocGetWaitbase:
+        return GetWaitbase(input, output);
+    case IoctlCommand::IocMapBuffer:
+        return MapBuffer(input, output);
+    case IoctlCommand::IocMapBufferEx:
+        return MapBuffer(input, output);
+    case IoctlCommand::IocUnmapBufferEx:
+        return UnmapBufferEx(input, output);
    }

    UNIMPLEMENTED_MSG("Unimplemented ioctl");
@@ -30,11 +42,67 @@ u32 nvhost_vic::ioctl(Ioctl command, const std::vector<u8>& input, const std::ve

 u32 nvhost_vic::SetNVMAPfd(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlSetNvmapFD params{};
-    std::memcpy(&params, input.data(), input.size());
+    std::memcpy(&params, input.data(), sizeof(IoctlSetNvmapFD));
    LOG_DEBUG(Service_NVDRV, "called, fd={}", params.nvmap_fd);

    nvmap_fd = params.nvmap_fd;
    return 0;
 }

+u32 nvhost_vic::Submit(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlSubmit params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlSubmit));
+    LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+    std::memcpy(output.data(), &params, sizeof(IoctlSubmit));
+    return 0;
+}
+
+u32 nvhost_vic::GetSyncpoint(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlGetSyncpoint params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlGetSyncpoint));
+    LOG_INFO(Service_NVDRV, "called, unknown=0x{:X}", params.unknown);
+    params.value = 0; // Seems to be hard coded at 0
+    std::memcpy(output.data(), &params, sizeof(IoctlGetSyncpoint));
+    return 0;
+}
+
+u32 nvhost_vic::GetWaitbase(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlGetWaitbase params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlGetWaitbase));
+    LOG_INFO(Service_NVDRV, "called, unknown=0x{:X}", params.unknown);
+    params.value = 0; // Seems to be hard coded at 0
+    std::memcpy(output.data(), &params, sizeof(IoctlGetWaitbase));
+    return 0;
+}
+
+u32 nvhost_vic::MapBuffer(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlMapBuffer params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlMapBuffer));
+    LOG_WARNING(Service_NVDRV, "(STUBBED) called with address={:08X}{:08X}", params.address_2,
+                params.address_1);
+    params.address_1 = 0;
+    params.address_2 = 0;
+    std::memcpy(output.data(), &params, sizeof(IoctlMapBuffer));
+    return 0;
+}
+
+u32 nvhost_vic::MapBufferEx(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlMapBufferEx params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlMapBufferEx));
+    LOG_WARNING(Service_NVDRV, "(STUBBED) called with address={:08X}{:08X}", params.address_2,
+                params.address_1);
+    params.address_1 = 0;
+    params.address_2 = 0;
+    std::memcpy(output.data(), &params, sizeof(IoctlMapBufferEx));
+    return 0;
+}
+
+u32 nvhost_vic::UnmapBufferEx(const std::vector<u8>& input, std::vector<u8>& output) {
+    IoctlUnmapBufferEx params{};
+    std::memcpy(&params, input.data(), sizeof(IoctlUnmapBufferEx));
+    LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+    std::memcpy(output.data(), &params, sizeof(IoctlUnmapBufferEx));
+    return 0;
+}
+
 } // namespace Service::Nvidia::Devices
@@ -23,6 +23,12 @@ public:
 private:
    enum class IoctlCommand : u32_le {
        IocSetNVMAPfdCommand = 0x40044801,
+        IocSubmit = 0xC0400001,
+        IocGetSyncpoint = 0xC0080002,
+        IocGetWaitbase = 0xC0080003,
+        IocMapBuffer = 0xC01C0009,
+        IocMapBufferEx = 0xC03C0009,
+        IocUnmapBufferEx = 0xC03C000A,
    };

    struct IoctlSetNvmapFD {
@@ -30,9 +36,53 @@ private:
    };
    static_assert(sizeof(IoctlSetNvmapFD) == 4, "IoctlSetNvmapFD is incorrect size");

+    struct IoctlSubmit {
+        INSERT_PADDING_BYTES(0x40); // TODO(DarkLordZach): RE this structure
+    };
+    static_assert(sizeof(IoctlSubmit) == 0x40, "IoctlSubmit is incorrect size");
+
+    struct IoctlGetSyncpoint {
+        u32 unknown; // seems to be ignored? Nintendo added this
+        u32 value;
+    };
+    static_assert(sizeof(IoctlGetSyncpoint) == 0x8, "IoctlGetSyncpoint is incorrect size");
+
+    struct IoctlGetWaitbase {
+        u32 unknown; // seems to be ignored? Nintendo added this
+        u32 value;
+    };
+    static_assert(sizeof(IoctlGetWaitbase) == 0x8, "IoctlGetWaitbase is incorrect size");
+
+    struct IoctlMapBuffer {
+        u32 unknown;
+        u32 address_1;
+        u32 address_2;
+        INSERT_PADDING_BYTES(0x10); // TODO(DarkLordZach): RE this structure
+    };
+    static_assert(sizeof(IoctlMapBuffer) == 0x1C, "IoctlMapBuffer is incorrect size");
+
+    struct IoctlMapBufferEx {
+        u32 unknown;
+        u32 address_1;
+        u32 address_2;
+        INSERT_PADDING_BYTES(0x30); // TODO(DarkLordZach): RE this structure
+    };
+    static_assert(sizeof(IoctlMapBufferEx) == 0x3C, "IoctlMapBufferEx is incorrect size");
+
+    struct IoctlUnmapBufferEx {
+        INSERT_PADDING_BYTES(0x3C); // TODO(DarkLordZach): RE this structure
+    };
+    static_assert(sizeof(IoctlUnmapBufferEx) == 0x3C, "IoctlUnmapBufferEx is incorrect size");
+
    u32_le nvmap_fd{};

    u32 SetNVMAPfd(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 Submit(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 GetSyncpoint(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 GetWaitbase(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 MapBuffer(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 MapBufferEx(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 UnmapBufferEx(const std::vector<u8>& input, std::vector<u8>& output);
 };

 } // namespace Service::Nvidia::Devices
@@ -134,7 +134,9 @@ void NVDRV::QueryEvent(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 3, 1};
    rb.Push(RESULT_SUCCESS);
    if (event_id < MaxNvEvents) {
-        rb.PushCopyObjects(nvdrv->GetEvent(event_id));
+        auto event = nvdrv->GetEvent(event_id);
+        event->Clear();
+        rb.PushCopyObjects(event);
        rb.Push<u32>(NvResult::Success);
    } else {
        rb.Push<u32>(0);
@@ -40,8 +40,8 @@ Module::Module(Core::System& system) {
    auto& kernel = system.Kernel();
    for (u32 i = 0; i < MaxNvEvents; i++) {
        std::string event_label = fmt::format("NVDRV::NvEvent_{}", i);
-        events_interface.events[i] = Kernel::WritableEvent::CreateEventPair(
-            kernel, Kernel::ResetType::Automatic, event_label);
+        events_interface.events[i] =
+            Kernel::WritableEvent::CreateEventPair(kernel, Kernel::ResetType::Manual, event_label);
        events_interface.status[i] = EventState::Free;
        events_interface.registered[i] = false;
    }
@@ -187,14 +187,18 @@ void NVFlinger::Compose() {
        MicroProfileFlip();

        if (!buffer) {
-            // There was no queued buffer to draw, render previous frame
-            system.GetPerfStats().EndGameFrame();
-            system.GPU().SwapBuffers({});
            continue;
        }

        const auto& igbp_buffer = buffer->get().igbp_buffer;

+        const auto& gpu = system.GPU();
+        const auto& multi_fence = buffer->get().multi_fence;
+        for (u32 fence_id = 0; fence_id < multi_fence.num_fences; fence_id++) {
+            const auto& fence = multi_fence.fences[fence_id];
+            gpu.WaitFence(fence.id, fence.value);
+        }
+
        // Now send the buffer to the GPU for drawing.
        // TODO(Subv): Support more than just disp0. The display device selection is probably based
        // on which display we're drawing (Default, Internal, External, etc)
@@ -6,6 +6,8 @@
 #include "core/core.h"
 #include "core/gdbstub/gdbstub.h"
 #include "core/hle/service/hid/hid.h"
+#include "core/hle/service/lbl/lbl.h"
+#include "core/hle/service/sm/sm.h"
 #include "core/settings.h"
 #include "video_core/renderer_base.h"

@@ -70,6 +72,7 @@ void Apply() {
    auto& system_instance = Core::System::GetInstance();
    if (system_instance.IsPoweredOn()) {
        system_instance.Renderer().RefreshBaseSettings();
+        Service::LBL::RequestLoadCurrentSetting(system_instance.ServiceManager());
    }

    Service::HID::ReloadInputDevices();
@@ -428,6 +428,8 @@ struct Values {
    float bg_green;
    float bg_blue;

+    float backlight_brightness = 0.5f;
+
    std::string log_filter;

    bool use_dev_keys;
@@ -47,10 +47,20 @@ void Fermi2D::HandleSurfaceCopy() {
        src_blit_x2 = static_cast<u32>((regs.blit_src_x >> 32) + regs.blit_dst_width);
        src_blit_y2 = static_cast<u32>((regs.blit_src_y >> 32) + regs.blit_dst_height);
    }
-    const Common::Rectangle<u32> src_rect{src_blit_x1, src_blit_y1, src_blit_x2, src_blit_y2};
-    const Common::Rectangle<u32> dst_rect{regs.blit_dst_x, regs.blit_dst_y,
-                                          regs.blit_dst_x + regs.blit_dst_width,
-                                          regs.blit_dst_y + regs.blit_dst_height};
+    const u32 dst_blit_x2 = regs.blit_dst_x + regs.blit_dst_width;
+    const u32 dst_blit_y2 = regs.blit_dst_x + regs.blit_dst_height;
+    const u32 excess_src_x2 = std::max<s32>(0, dst_blit_x2 - regs.dst.width);
+    const u32 excess_src_y2 = std::max<s32>(0, dst_blit_y2 - regs.dst.height);
+    const u32 excess_dst_x2 = std::max<s32>(0, src_blit_x2 - regs.src.width);
+    const u32 excess_dst_y2 = std::max<s32>(0, src_blit_y2 - regs.src.height);
+
+    const Common::Rectangle<u32> src_rect{
+        src_blit_x1, src_blit_y1, std::min<u32>(regs.src.width, src_blit_x2) - excess_src_x2,
+        std::min<u32>(regs.src.height, src_blit_y2) - excess_src_y2};
+    const Common::Rectangle<u32> dst_rect{
+        regs.blit_dst_x, regs.blit_dst_y,
+        std::min<u32>(regs.dst.width, dst_blit_x2) - excess_dst_x2,
+        std::min<u32>(regs.dst.height, dst_blit_y2) - excess_dst_y2};
    Config copy_config;
    copy_config.operation = regs.operation;
    copy_config.filter = regs.blit_control.filter;
@@ -544,7 +544,7 @@ enum class VoteOperation : u64 {
    Eq = 2,  // allThreadsEqualNV
 };

-enum class ImageAtomicSize : u64 {
+enum class ImageAtomicOperationType : u64 {
    U32 = 0,
    S32 = 1,
    U64 = 2,
@@ -1432,11 +1432,11 @@ union Instruction {
            ASSERT(mode == SurfaceDataMode::D_BA);
            return store_data_layout;
        }
-    } sust;
+    } suldst;

    union {
        BitField<28, 1, u64> is_ba;
-        BitField<51, 3, ImageAtomicSize> size;
+        BitField<51, 3, ImageAtomicOperationType> operation_type;
        BitField<33, 3, ImageType> image_type;
        BitField<29, 4, ImageAtomicOperation> operation;
        BitField<49, 2, OutOfBoundsStore> out_of_bounds_store;
@@ -1595,6 +1595,7 @@ public:
        TMML_B, // Texture Mip Map Level
        TMML,   // Texture Mip Map Level
        SUST,   // Surface Store
+        SULD,   // Surface Load
        SUATOM, // Surface Atomic Operation
        EXIT,
        NOP,
@@ -1884,6 +1885,7 @@ private:
            INST("110111110110----", Id::TMML_B, Type::Texture, "TMML_B"),
            INST("1101111101011---", Id::TMML, Type::Texture, "TMML"),
            INST("11101011001-----", Id::SUST, Type::Image, "SUST"),
+            INST("11101011000-----", Id::SULD, Type::Image, "SULD"),
            INST("1110101000------", Id::SUATOM, Type::Image, "SUATOM_D"),
            INST("0101000010110---", Id::NOP, Type::Trivial, "NOP"),
            INST("11100000--------", Id::IPA, Type::Trivial, "IPA"),
@@ -3,6 +3,7 @@
 // Refer to the license.txt file included.

 #include "common/assert.h"
+#include "common/microprofile.h"
 #include "core/core.h"
 #include "core/core_timing.h"
 #include "core/memory.h"
@@ -17,6 +18,8 @@

 namespace Tegra {

+MICROPROFILE_DEFINE(GPU_wait, "GPU", "Wait for the GPU", MP_RGB(128, 128, 192));
+
 GPU::GPU(Core::System& system, VideoCore::RendererBase& renderer, bool is_async)
    : system{system}, renderer{renderer}, is_async{is_async} {
    auto& rasterizer{renderer.Rasterizer()};
@@ -63,6 +66,16 @@ const DmaPusher& GPU::DmaPusher() const {
    return *dma_pusher;
 }

+void GPU::WaitFence(u32 syncpoint_id, u32 value) const {
+    // Synced GPU, is always in sync
+    if (!is_async) {
+        return;
+    }
+    MICROPROFILE_SCOPE(GPU_wait);
+    while (syncpoints[syncpoint_id].load() < value) {
+    }
+}
+
 void GPU::IncrementSyncPoint(const u32 syncpoint_id) {
    syncpoints[syncpoint_id]++;
    std::lock_guard lock{sync_mutex};
@@ -177,6 +177,12 @@ public:
    /// Returns a reference to the GPU DMA pusher.
    Tegra::DmaPusher& DmaPusher();

+    // Waits for the GPU to finish working
+    virtual void WaitIdle() const = 0;
+
+    /// Allows the CPU/NvFlinger to wait on the GPU before presenting a frame.
+    void WaitFence(u32 syncpoint_id, u32 value) const;
+
    void IncrementSyncPoint(u32 syncpoint_id);

    u32 GetSyncpointValue(u32 syncpoint_id) const;
@@ -44,4 +44,8 @@ void GPUAsynch::TriggerCpuInterrupt(const u32 syncpoint_id, const u32 value) con
    interrupt_manager.GPUInterruptSyncpt(syncpoint_id, value);
 }

+void GPUAsynch::WaitIdle() const {
+    gpu_thread.WaitIdle();
+}
+
 } // namespace VideoCommon
@@ -25,6 +25,7 @@ public:
    void FlushRegion(CacheAddr addr, u64 size) override;
    void InvalidateRegion(CacheAddr addr, u64 size) override;
    void FlushAndInvalidateRegion(CacheAddr addr, u64 size) override;
+    void WaitIdle() const override;

 protected:
    void TriggerCpuInterrupt(u32 syncpoint_id, u32 value) const override;
@@ -24,6 +24,7 @@ public:
    void FlushRegion(CacheAddr addr, u64 size) override;
    void InvalidateRegion(CacheAddr addr, u64 size) override;
    void FlushAndInvalidateRegion(CacheAddr addr, u64 size) override;
+    void WaitIdle() const override {}

 protected:
    void TriggerCpuInterrupt([[maybe_unused]] u32 syncpoint_id,
@@ -5,8 +5,6 @@
 #include "common/assert.h"
 #include "common/microprofile.h"
 #include "core/core.h"
-#include "core/core_timing.h"
-#include "core/core_timing_util.h"
 #include "core/frontend/scope_acquire_window_context.h"
 #include "video_core/dma_pusher.h"
 #include "video_core/gpu.h"
@@ -68,14 +66,10 @@ ThreadManager::~ThreadManager() {

 void ThreadManager::StartThread(VideoCore::RendererBase& renderer, Tegra::DmaPusher& dma_pusher) {
    thread = std::thread{RunThread, std::ref(renderer), std::ref(dma_pusher), std::ref(state)};
-    synchronization_event = system.CoreTiming().RegisterEvent(
-        "GPUThreadSynch", [this](u64 fence, s64) { state.WaitForSynchronization(fence); });
 }

 void ThreadManager::SubmitList(Tegra::CommandList&& entries) {
-    const u64 fence{PushCommand(SubmitListCommand(std::move(entries)))};
-    const s64 synchronization_ticks{Core::Timing::usToCycles(std::chrono::microseconds{9000})};
-    system.CoreTiming().ScheduleEvent(synchronization_ticks, synchronization_event, fence);
+    PushCommand(SubmitListCommand(std::move(entries)));
 }

 void ThreadManager::SwapBuffers(const Tegra::FramebufferConfig* framebuffer) {
@@ -96,16 +90,15 @@ void ThreadManager::FlushAndInvalidateRegion(CacheAddr addr, u64 size) {
    InvalidateRegion(addr, size);
 }

+void ThreadManager::WaitIdle() const {
+    while (state.last_fence > state.signaled_fence.load()) {
+    }
+}
+
 u64 ThreadManager::PushCommand(CommandData&& command_data) {
    const u64 fence{++state.last_fence};
    state.queue.Push(CommandDataContainer(std::move(command_data), fence));
    return fence;
 }

-MICROPROFILE_DEFINE(GPU_wait, "GPU", "Wait for the GPU", MP_RGB(128, 128, 192));
-void SynchState::WaitForSynchronization(u64 fence) {
-    while (signaled_fence.load() < fence)
-        ;
-}
-
 } // namespace VideoCommon::GPUThread
@@ -21,9 +21,6 @@ class DmaPusher;

 namespace Core {
 class System;
-namespace Timing {
-struct EventType;
-} // namespace Timing
 } // namespace Core

 namespace VideoCommon::GPUThread {
@@ -89,8 +86,6 @@ struct CommandDataContainer {
 struct SynchState final {
    std::atomic_bool is_running{true};

-    void WaitForSynchronization(u64 fence);
-
    using CommandQueue = Common::SPSCQueue<CommandDataContainer>;
    CommandQueue queue;
    u64 last_fence{};
@@ -121,6 +116,9 @@ public:
    /// Notify rasterizer that any caches of the specified region should be flushed and invalidated
    void FlushAndInvalidateRegion(CacheAddr addr, u64 size);

+    // Wait until the gpu thread is idle.
+    void WaitIdle() const;
+
 private:
    /// Pushes a command to be executed by the GPU thread
    u64 PushCommand(CommandData&& command_data);
@@ -128,7 +126,6 @@ private:
 private:
    SynchState state;
    Core::System& system;
-    Core::Timing::EventType* synchronization_event{};
    std::thread thread;
    std::thread::id thread_id;
 };
@@ -40,4 +40,35 @@ void RendererBase::RequestScreenshot(void* data, std::function<void()> callback,
    renderer_settings.screenshot_requested = true;
 }

+f32 RendererBase::GetCurrentResultantBrightness() const {
+    return renderer_settings.current_brightness / 2.0f;
+}
+
+void RendererBase::SetBacklightStatus(bool enabled, u64 fade_transition_time) {
+    if (fade_transition_time == 0) {
+        // Needed to ensure the renderer recognizes that a change must occur.
+        fade_transition_time = 1;
+    }
+
+    if (enabled && renderer_settings.current_brightness == 0) {
+        renderer_settings.current_brightness = current_brightness_backup;
+        renderer_settings.backlight_fade_time = fade_transition_time;
+    } else if (!enabled && renderer_settings.current_brightness != 0) {
+        current_brightness_backup = renderer_settings.current_brightness;
+        renderer_settings.current_brightness = 0;
+        renderer_settings.backlight_fade_time = fade_transition_time;
+    }
+}
+
+bool RendererBase::GetBacklightStatus() const {
+    return renderer_settings.current_brightness != 0;
+}
+
+void RendererBase::SetCurrentBrightness(f32 value) {
+    if (value != renderer_settings.current_brightness) {
+        renderer_settings.current_brightness = value * 2.0f;
+        renderer_settings.backlight_fade_time = 1;
+    }
+}
+
 } // namespace VideoCore
@@ -28,6 +28,10 @@ struct RendererSettings {
    void* screenshot_bits;
    std::function<void()> screenshot_complete_callback;
    Layout::FramebufferLayout screenshot_framebuffer_layout;
+
+    // Backlight & Brightness
+    std::atomic<f32> current_brightness{1.f};
+    std::atomic<u64> backlight_fade_time{0};
 };

 class RendererBase : NonCopyable {
@@ -86,6 +90,17 @@ public:
    void RequestScreenshot(void* data, std::function<void()> callback,
                           const Layout::FramebufferLayout& layout);

+    // Gets the current brightness, even if it has been changed from the set value. Most of the time
+    // for yuzu this will simply match what was returned, but implementations are free to change the
+    // value in settings.
+    f32 GetCurrentResultantBrightness() const;
+
+    void SetBacklightStatus(bool enabled, u64 fade_transition_time);
+
+    bool GetBacklightStatus() const;
+
+    void SetCurrentBrightness(f32 value);
+
 protected:
    Core::Frontend::EmuWindow& render_window; ///< Reference to the render window handle.
    std::unique_ptr<RasterizerInterface> rasterizer;
@@ -97,6 +112,9 @@ protected:
 private:
    /// Updates the framebuffer layout of the contained render window handle.
    void UpdateCurrentFramebufferLayout();
+
+    // Value of brightness before backlight switch used to preserve value.
+    f32 current_brightness_backup;
 };

 } // namespace VideoCore
@@ -2,8 +2,10 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.

+#include <algorithm>
 #include <array>
 #include <cstddef>
+#include <vector>
 #include <glad/glad.h>

 #include "common/logging/log.h"
@@ -30,9 +32,27 @@ bool TestProgram(const GLchar* glsl) {
    return link_status == GL_TRUE;
 }

+std::vector<std::string_view> GetExtensions() {
+    GLint num_extensions;
+    glGetIntegerv(GL_NUM_EXTENSIONS, &num_extensions);
+    std::vector<std::string_view> extensions;
+    extensions.reserve(num_extensions);
+    for (GLint index = 0; index < num_extensions; ++index) {
+        extensions.push_back(
+            reinterpret_cast<const char*>(glGetStringi(GL_EXTENSIONS, static_cast<GLuint>(index))));
+    }
+    return extensions;
+}
+
+bool HasExtension(const std::vector<std::string_view>& images, std::string_view extension) {
+    return std::find(images.begin(), images.end(), extension) != images.end();
+}
+
 } // Anonymous namespace

 Device::Device() {
+    const std::vector extensions = GetExtensions();
+
    uniform_buffer_alignment = GetInteger<std::size_t>(GL_UNIFORM_BUFFER_OFFSET_ALIGNMENT);
    shader_storage_alignment = GetInteger<std::size_t>(GL_SHADER_STORAGE_BUFFER_OFFSET_ALIGNMENT);
    max_vertex_attributes = GetInteger<u32>(GL_MAX_VERTEX_ATTRIBS);
@@ -40,6 +60,7 @@ Device::Device() {
    has_warp_intrinsics = GLAD_GL_NV_gpu_shader5 && GLAD_GL_NV_shader_thread_group &&
                          GLAD_GL_NV_shader_thread_shuffle;
    has_vertex_viewport_layer = GLAD_GL_ARB_shader_viewport_layer_array;
+    has_image_load_formatted = HasExtension(extensions, "GL_EXT_shader_image_load_formatted");
    has_variable_aoffi = TestVariableAoffi();
    has_component_indexing_bug = TestComponentIndexingBug();
    has_precise_bug = TestPreciseBug();
@@ -55,6 +76,7 @@ Device::Device(std::nullptr_t) {
    max_varyings = 15;
    has_warp_intrinsics = true;
    has_vertex_viewport_layer = true;
+    has_image_load_formatted = true;
    has_variable_aoffi = true;
    has_component_indexing_bug = false;
    has_precise_bug = false;
@@ -38,6 +38,10 @@ public:
        return has_vertex_viewport_layer;
    }

+    bool HasImageLoadFormatted() const {
+        return has_image_load_formatted;
+    }
+
    bool HasVariableAoffi() const {
        return has_variable_aoffi;
    }
@@ -61,6 +65,7 @@ private:
    u32 max_varyings{};
    bool has_warp_intrinsics{};
    bool has_vertex_viewport_layer{};
+    bool has_image_load_formatted{};
    bool has_variable_aoffi{};
    bool has_component_indexing_bug{};
    bool has_precise_bug{};
@@ -348,6 +348,7 @@ static constexpr auto RangeFromInterval(Map& map, const Interval& interval) {
 }

 void RasterizerOpenGL::UpdatePagesCachedCount(VAddr addr, u64 size, int delta) {
+    std::lock_guard lock{pages_mutex};
    const u64 page_start{addr >> Memory::PAGE_BITS};
    const u64 page_end{(addr + size + Memory::PAGE_SIZE - 1) >> Memory::PAGE_BITS};

@@ -1340,7 +1341,9 @@ void RasterizerOpenGL::SyncPolygonOffset() {
    state.polygon_offset.fill_enable = regs.polygon_offset_fill_enable != 0;
    state.polygon_offset.line_enable = regs.polygon_offset_line_enable != 0;
    state.polygon_offset.point_enable = regs.polygon_offset_point_enable != 0;
-    state.polygon_offset.units = regs.polygon_offset_units;
+
+    // Hardware divides polygon offset units by two
+    state.polygon_offset.units = regs.polygon_offset_units / 2.0f;
    state.polygon_offset.factor = regs.polygon_offset_factor;
    state.polygon_offset.clamp = regs.polygon_offset_clamp;

@@ -9,6 +9,7 @@
 #include <cstddef>
 #include <map>
 #include <memory>
+#include <mutex>
 #include <optional>
 #include <tuple>
 #include <utility>
@@ -230,6 +231,8 @@ private:

    using CachedPageMap = boost::icl::interval_map<u64, int>;
    CachedPageMap cached_pages;
+
+    std::mutex pages_mutex;
 };

 } // namespace OpenGL
@@ -211,14 +211,14 @@ CachedProgram SpecializeShader(const std::string& code, const GLShader::ShaderEn
    const auto primitive_mode{variant.primitive_mode};
    const auto texture_buffer_usage{variant.texture_buffer_usage};

-    std::string source = "#version 430 core\n"
-                         "#extension GL_ARB_separate_shader_objects : enable\n"
-                         "#extension GL_NV_gpu_shader5 : enable\n"
-                         "#extension GL_NV_shader_thread_group : enable\n"
-                         "#extension GL_NV_shader_thread_shuffle : enable\n";
-    if (entries.shader_viewport_layer_array) {
-        source += "#extension GL_ARB_shader_viewport_layer_array : enable\n";
-    }
+    std::string source = R"(#version 430 core
+#extension GL_ARB_separate_shader_objects : enable
+#extension GL_ARB_shader_viewport_layer_array : enable
+#extension GL_EXT_shader_image_load_formatted : enable
+#extension GL_NV_gpu_shader5 : enable
+#extension GL_NV_shader_thread_group : enable
+#extension GL_NV_shader_thread_shuffle : enable
+)";
    if (program_type == ProgramType::Compute) {
        source += "#extension GL_ARB_compute_variable_group_size : require\n";
    }
@@ -19,6 +19,7 @@
 #include "video_core/renderer_opengl/gl_device.h"
 #include "video_core/renderer_opengl/gl_rasterizer.h"
 #include "video_core/renderer_opengl/gl_shader_decompiler.h"
+#include "video_core/shader/node.h"
 #include "video_core/shader/shader_ir.h"

 namespace OpenGL::GLShader {
@@ -241,6 +242,26 @@ constexpr const char* GetTypeString(Type type) {
    }
 }

+constexpr const char* GetImageTypeDeclaration(Tegra::Shader::ImageType image_type) {
+    switch (image_type) {
+    case Tegra::Shader::ImageType::Texture1D:
+        return "1D";
+    case Tegra::Shader::ImageType::TextureBuffer:
+        return "Buffer";
+    case Tegra::Shader::ImageType::Texture1DArray:
+        return "1DArray";
+    case Tegra::Shader::ImageType::Texture2D:
+        return "2D";
+    case Tegra::Shader::ImageType::Texture2DArray:
+        return "2DArray";
+    case Tegra::Shader::ImageType::Texture3D:
+        return "3D";
+    default:
+        UNREACHABLE();
+        return "1D";
+    }
+}
+
 /// Generates code to use for a swizzle operation.
 constexpr const char* GetSwizzle(u32 element) {
    constexpr std::array swizzle = {".x", ".y", ".z", ".w"};
@@ -398,8 +419,6 @@ public:
                                                       usage.is_read, usage.is_written);
        }
        entries.clip_distances = ir.GetClipDistances();
-        entries.shader_viewport_layer_array =
-            IsVertexShader(stage) && (ir.UsesLayer() || ir.UsesViewportIndex());
        entries.shader_length = ir.GetLength();
        return entries;
    }
@@ -722,42 +741,6 @@ private:
    void DeclareImages() {
        const auto& images{ir.GetImages()};
        for (const auto& [offset, image] : images) {
-            const char* image_type = [&] {
-                switch (image.GetType()) {
-                case Tegra::Shader::ImageType::Texture1D:
-                    return "image1D";
-                case Tegra::Shader::ImageType::TextureBuffer:
-                    return "imageBuffer";
-                case Tegra::Shader::ImageType::Texture1DArray:
-                    return "image1DArray";
-                case Tegra::Shader::ImageType::Texture2D:
-                    return "image2D";
-                case Tegra::Shader::ImageType::Texture2DArray:
-                    return "image2DArray";
-                case Tegra::Shader::ImageType::Texture3D:
-                    return "image3D";
-                default:
-                    UNREACHABLE();
-                    return "image1D";
-                }
-            }();
-
-            const auto [type_prefix, format] = [&]() -> std::pair<const char*, const char*> {
-                if (!image.IsSizeKnown()) {
-                    return {"", ""};
-                }
-                switch (image.GetSize()) {
-                case Tegra::Shader::ImageAtomicSize::U32:
-                    return {"u", "r32ui, "};
-                case Tegra::Shader::ImageAtomicSize::S32:
-                    return {"i", "r32i, "};
-                default:
-                    UNIMPLEMENTED_MSG("Unimplemented atomic size={}",
-                                      static_cast<u32>(image.GetSize()));
-                    return {"", ""};
-                }
-            }();
-
            std::string qualifier = "coherent volatile";
            if (image.IsRead() && !image.IsWritten()) {
                qualifier += " readonly";
@@ -765,9 +748,10 @@ private:
                qualifier += " writeonly";
            }

-            code.AddLine("layout (binding = IMAGE_BINDING_{}) {} uniform "
-                         "{} {};",
-                         image.GetIndex(), qualifier, image_type, GetImage(image));
+            const char* format = image.IsAtomic() ? "r32ui, " : "";
+            const char* type_declaration = GetImageTypeDeclaration(image.GetType());
+            code.AddLine("layout ({}binding = IMAGE_BINDING_{}) {} uniform uimage{} {};", format,
+                         image.GetIndex(), qualifier, type_declaration, GetImage(image));
        }
        if (!images.empty()) {
            code.AddNewLine();
@@ -1234,28 +1218,13 @@ private:
    }

    std::string BuildImageValues(Operation operation) {
+        constexpr std::array constructors{"uint", "uvec2", "uvec3", "uvec4"};
        const auto meta{std::get<MetaImage>(operation.GetMeta())};
-        const auto [constructors, type] = [&]() -> std::pair<std::array<const char*, 4>, Type> {
-            constexpr std::array float_constructors{"float", "vec2", "vec3", "vec4"};
-            if (!meta.image.IsSizeKnown()) {
-                return {float_constructors, Type::Float};
-            }
-            switch (meta.image.GetSize()) {
-            case Tegra::Shader::ImageAtomicSize::U32:
-                return {{"uint", "uvec2", "uvec3", "uvec4"}, Type::Uint};
-            case Tegra::Shader::ImageAtomicSize::S32:
-                return {{"int", "ivec2", "ivec3", "ivec4"}, Type::Uint};
-            default:
-                UNIMPLEMENTED_MSG("Unimplemented image size={}",
-                                  static_cast<u32>(meta.image.GetSize()));
-                return {float_constructors, Type::Float};
-            }
-        }();

        const std::size_t values_count{meta.values.size()};
        std::string expr = fmt::format("{}(", constructors.at(values_count - 1));
        for (std::size_t i = 0; i < values_count; ++i) {
-            expr += Visit(meta.values.at(i)).As(type);
+            expr += Visit(meta.values.at(i)).AsUint();
            if (i + 1 < values_count) {
                expr += ", ";
            }
@@ -1264,29 +1233,6 @@ private:
        return expr;
    }

-    Expression AtomicImage(Operation operation, const char* opname) {
-        constexpr std::array constructors{"int(", "ivec2(", "ivec3(", "ivec4("};
-        const auto meta{std::get<MetaImage>(operation.GetMeta())};
-        ASSERT(meta.values.size() == 1);
-        ASSERT(meta.image.IsSizeKnown());
-
-        const auto type = [&]() {
-            switch (const auto size = meta.image.GetSize()) {
-            case Tegra::Shader::ImageAtomicSize::U32:
-                return Type::Uint;
-            case Tegra::Shader::ImageAtomicSize::S32:
-                return Type::Int;
-            default:
-                UNIMPLEMENTED_MSG("Unimplemented image size={}", static_cast<u32>(size));
-                return Type::Uint;
-            }
-        }();
-
-        return {fmt::format("{}({}, {}, {})", opname, GetImage(meta.image),
-                            BuildIntegerCoordinates(operation), Visit(meta.values[0]).As(type)),
-                type};
-    }
-
    Expression Assign(Operation operation) {
        const Node& dest = operation[0];
        const Node& src = operation[1];
@@ -1545,6 +1491,8 @@ private:
        case Tegra::Shader::HalfType::H1_H1:
            return {fmt::format("vec2({}[1])", operand.AsHalfFloat()), Type::HalfFloat};
        }
+        UNREACHABLE();
+        return {"0", Type::Int};
    }

    Expression HMergeF32(Operation operation) {
@@ -1809,6 +1757,19 @@ private:
        return {tmp, Type::Float};
    }

+    Expression ImageLoad(Operation operation) {
+        if (!device.HasImageLoadFormatted()) {
+            LOG_ERROR(Render_OpenGL,
+                      "Device lacks GL_EXT_shader_image_load_formatted, stubbing image load");
+            return {"0", Type::Int};
+        }
+
+        const auto meta{std::get<MetaImage>(operation.GetMeta())};
+        return {fmt::format("imageLoad({}, {}){}", GetImage(meta.image),
+                            BuildIntegerCoordinates(operation), GetSwizzle(meta.element)),
+                Type::Uint};
+    }
+
    Expression ImageStore(Operation operation) {
        const auto meta{std::get<MetaImage>(operation.GetMeta())};
        code.AddLine("imageStore({}, {}, {});", GetImage(meta.image),
@@ -1816,31 +1777,14 @@ private:
        return {};
    }

-    Expression AtomicImageAdd(Operation operation) {
-        return AtomicImage(operation, "imageAtomicAdd");
-    }
+    template <const std::string_view& opname>
+    Expression AtomicImage(Operation operation) {
+        const auto meta{std::get<MetaImage>(operation.GetMeta())};
+        ASSERT(meta.values.size() == 1);

-    Expression AtomicImageMin(Operation operation) {
-        return AtomicImage(operation, "imageAtomicMin");
-    }
-
-    Expression AtomicImageMax(Operation operation) {
-        return AtomicImage(operation, "imageAtomicMax");
-    }
-    Expression AtomicImageAnd(Operation operation) {
-        return AtomicImage(operation, "imageAtomicAnd");
-    }
-
-    Expression AtomicImageOr(Operation operation) {
-        return AtomicImage(operation, "imageAtomicOr");
-    }
-
-    Expression AtomicImageXor(Operation operation) {
-        return AtomicImage(operation, "imageAtomicXor");
-    }
-
-    Expression AtomicImageExchange(Operation operation) {
-        return AtomicImage(operation, "imageAtomicExchange");
+        return {fmt::format("imageAtomic{}({}, {}, {})", opname, GetImage(meta.image),
+                            BuildIntegerCoordinates(operation), Visit(meta.values[0]).AsUint()),
+                Type::Uint};
    }

    Expression Branch(Operation operation) {
@@ -2035,6 +1979,12 @@ private:
        Func() = delete;
        ~Func() = delete;

+        static constexpr std::string_view Add = "Add";
+        static constexpr std::string_view And = "And";
+        static constexpr std::string_view Or = "Or";
+        static constexpr std::string_view Xor = "Xor";
+        static constexpr std::string_view Exchange = "Exchange";
+
        static constexpr std::string_view ShuffleIndexed = "shuffleNV";
        static constexpr std::string_view ShuffleUp = "shuffleUpNV";
        static constexpr std::string_view ShuffleDown = "shuffleDownNV";
@@ -2172,14 +2122,14 @@ private:
        &GLSLDecompiler::TextureQueryLod,
        &GLSLDecompiler::TexelFetch,

+        &GLSLDecompiler::ImageLoad,
        &GLSLDecompiler::ImageStore,
-        &GLSLDecompiler::AtomicImageAdd,
-        &GLSLDecompiler::AtomicImageMin,
-        &GLSLDecompiler::AtomicImageMax,
-        &GLSLDecompiler::AtomicImageAnd,
-        &GLSLDecompiler::AtomicImageOr,
-        &GLSLDecompiler::AtomicImageXor,
-        &GLSLDecompiler::AtomicImageExchange,
+
+        &GLSLDecompiler::AtomicImage<Func::Add>,
+        &GLSLDecompiler::AtomicImage<Func::And>,
+        &GLSLDecompiler::AtomicImage<Func::Or>,
+        &GLSLDecompiler::AtomicImage<Func::Xor>,
+        &GLSLDecompiler::AtomicImage<Func::Exchange>,

        &GLSLDecompiler::Branch,
        &GLSLDecompiler::BranchIndirect,
@@ -90,7 +90,6 @@ struct ShaderEntries {
    std::vector<ImageEntry> images;
    std::vector<GlobalMemoryEntry> global_memory_entries;
    std::array<bool, Maxwell::NumClipDistances> clip_distances{};
-    bool shader_viewport_layer_array{};
    std::size_t shader_length{};
 };

@@ -343,20 +343,17 @@ std::optional<ShaderDiskCacheDecompiled> ShaderDiskCacheOpenGL::LoadDecompiledEn
        u8 is_bindless{};
        u8 is_written{};
        u8 is_read{};
-        u8 is_size_known{};
-        u32 size{};
+        u8 is_atomic{};
        if (!LoadObjectFromPrecompiled(offset) || !LoadObjectFromPrecompiled(index) ||
            !LoadObjectFromPrecompiled(type) || !LoadObjectFromPrecompiled(is_bindless) ||
            !LoadObjectFromPrecompiled(is_written) || !LoadObjectFromPrecompiled(is_read) ||
-            !LoadObjectFromPrecompiled(is_size_known) || !LoadObjectFromPrecompiled(size)) {
+            !LoadObjectFromPrecompiled(is_atomic)) {
            return {};
        }
        entry.entries.images.emplace_back(
            static_cast<std::size_t>(offset), static_cast<std::size_t>(index),
            static_cast<Tegra::Shader::ImageType>(type), is_bindless != 0, is_written != 0,
-            is_read != 0,
-            is_size_known ? std::make_optional(static_cast<Tegra::Shader::ImageAtomicSize>(size))
-                          : std::nullopt);
+            is_read != 0, is_atomic != 0);
    }

    u32 global_memory_count{};
@@ -382,12 +379,6 @@ std::optional<ShaderDiskCacheDecompiled> ShaderDiskCacheOpenGL::LoadDecompiledEn
        }
    }

-    bool shader_viewport_layer_array{};
-    if (!LoadObjectFromPrecompiled(shader_viewport_layer_array)) {
-        return {};
-    }
-    entry.entries.shader_viewport_layer_array = shader_viewport_layer_array;
-
    u64 shader_length{};
    if (!LoadObjectFromPrecompiled(shader_length)) {
        return {};
@@ -435,14 +426,13 @@ bool ShaderDiskCacheOpenGL::SaveDecompiledFile(u64 unique_identifier, const std:
        return false;
    }
    for (const auto& image : entries.images) {
-        const u32 size = image.IsSizeKnown() ? static_cast<u32>(image.GetSize()) : 0U;
        if (!SaveObjectToPrecompiled(static_cast<u64>(image.GetOffset())) ||
            !SaveObjectToPrecompiled(static_cast<u64>(image.GetIndex())) ||
            !SaveObjectToPrecompiled(static_cast<u32>(image.GetType())) ||
            !SaveObjectToPrecompiled(static_cast<u8>(image.IsBindless() ? 1 : 0)) ||
            !SaveObjectToPrecompiled(static_cast<u8>(image.IsWritten() ? 1 : 0)) ||
            !SaveObjectToPrecompiled(static_cast<u8>(image.IsRead() ? 1 : 0)) ||
-            !SaveObjectToPrecompiled(image.IsSizeKnown()) || !SaveObjectToPrecompiled(size)) {
+            !SaveObjectToPrecompiled(static_cast<u8>(image.IsAtomic() ? 1 : 0))) {
            return false;
        }
    }
@@ -464,10 +454,6 @@ bool ShaderDiskCacheOpenGL::SaveDecompiledFile(u64 unique_identifier, const std:
        }
    }

-    if (!SaveObjectToPrecompiled(entries.shader_viewport_layer_array)) {
-        return false;
-    }
-
    if (!SaveObjectToPrecompiled(static_cast<u64>(entries.shader_length))) {
        return false;
    }
@@ -54,11 +54,13 @@ in vec2 frag_tex_coord;
 out vec4 color;

 uniform sampler2D color_texture;
+uniform vec4 backlight;

 void main() {
    // Swap RGBA -> ABGR so we don't have to do this on the CPU. This needs to change if we have to
    // support more framebuffer pixel formats.
-    color = texture(color_texture, frag_tex_coord);
+    // Also multiply the color by the backlight multiplier supplied.
+    color = texture(color_texture, frag_tex_coord) * backlight;
 }
 )";

@@ -123,8 +125,13 @@ void RendererOpenGL::SwapBuffers(const Tegra::FramebufferConfig* framebuffer) {
        // Load the framebuffer from memory, draw it to the screen, and swap buffers
        LoadFBToScreenInfo(*framebuffer);

-        if (renderer_settings.screenshot_requested)
+        if (renderer_settings.screenshot_requested) {
            CaptureScreenshot();
+        }
+
+        if (renderer_settings.backlight_fade_time > 0) {
+            UpdateBacklight();
+        }

        DrawScreen(render_window.GetFramebufferLayout());

@@ -135,9 +142,6 @@ void RendererOpenGL::SwapBuffers(const Tegra::FramebufferConfig* framebuffer) {

    render_window.PollEvents();

-    system.FrameLimiter().DoFrameLimiting(system.CoreTiming().GetGlobalTimeUs());
-    system.GetPerfStats().BeginSystemFrame();
-
    // Restore the rasterizer state
    prev_state.AllDirty();
    prev_state.Apply();
@@ -210,9 +214,13 @@ void RendererOpenGL::InitOpenGLObjects() {
    state.Apply();
    uniform_modelview_matrix = glGetUniformLocation(shader.handle, "modelview_matrix");
    uniform_color_texture = glGetUniformLocation(shader.handle, "color_texture");
+    uniform_backlight = glGetUniformLocation(shader.handle, "backlight");
    attrib_position = glGetAttribLocation(shader.handle, "vert_position");
    attrib_tex_coord = glGetAttribLocation(shader.handle, "vert_tex_coord");

+    // Initialize backlight
+    glUniform4f(uniform_backlight, 1.f, 1.f, 1.f, 1.f);
+
    // Generate VBO handle for drawing
    vertex_buffer.Create();

@@ -421,6 +429,29 @@ void RendererOpenGL::CaptureScreenshot() {
    renderer_settings.screenshot_requested = false;
 }

+void RendererOpenGL::UpdateBacklight() {
+    constexpr u64 PER_FRAME_FADE_TIME = 1000000000.0f / 60;
+
+    const auto fade_time = renderer_settings.backlight_fade_time.load(std::memory_order_relaxed);
+    auto value = renderer_settings.current_brightness.load(std::memory_order_relaxed);
+    if (fade_time <= PER_FRAME_FADE_TIME) {
+        glUniform4f(uniform_backlight, value, value, value, value);
+        renderer_settings.backlight_fade_time = 0;
+        fade_time_max = 0;
+    } else {
+        if (fade_time_max == 0) {
+            fade_time_max = fade_time;
+            value_max = value;
+        }
+
+        value += (value_max - value) * PER_FRAME_FADE_TIME / fade_time_max;
+
+        glUniform4f(uniform_backlight, value, value, value, value);
+        renderer_settings.backlight_fade_time -= PER_FRAME_FADE_TIME;
+        renderer_settings.current_brightness = value;
+    }
+}
+
 static const char* GetSource(GLenum source) {
 #define RET(s)                                                                                     \
    case GL_DEBUG_SOURCE_##s:                                                                      \
@@ -70,6 +70,7 @@ private:
    void UpdateFramerate();

    void CaptureScreenshot();
+    void UpdateBacklight();

    // Loads framebuffer from emulated memory into the display information structure
    void LoadFBToScreenInfo(const Tegra::FramebufferConfig& framebuffer);
@@ -97,6 +98,7 @@ private:
    // Shader uniform location indices
    GLuint uniform_modelview_matrix;
    GLuint uniform_color_texture;
+    GLuint uniform_backlight;

    // Shader attribute input indices
    GLuint attrib_position;
@@ -105,6 +107,10 @@ private:
    /// Used for transforming the framebuffer orientation
    Tegra::FramebufferConfig::TransformFlags framebuffer_transform_flags;
    Common::Rectangle<int> framebuffer_crop_rect;
+
+    // Used for backlight transitions
+    u64 fade_time_max = 0;
+    f32 value_max = 0;
 };

 } // namespace OpenGL
@@ -19,6 +19,7 @@
 #include "video_core/engines/shader_header.h"
 #include "video_core/renderer_vulkan/vk_device.h"
 #include "video_core/renderer_vulkan/vk_shader_decompiler.h"
+#include "video_core/shader/node.h"
 #include "video_core/shader/shader_ir.h"

 namespace Vulkan::VKShader {
@@ -939,6 +940,11 @@ private:
        return {};
    }

+    Id ImageLoad(Operation operation) {
+        UNIMPLEMENTED();
+        return {};
+    }
+
    Id ImageStore(Operation operation) {
        UNIMPLEMENTED();
        return {};
@@ -949,16 +955,6 @@ private:
        return {};
    }

-    Id AtomicImageMin(Operation operation) {
-        UNIMPLEMENTED();
-        return {};
-    }
-
-    Id AtomicImageMax(Operation operation) {
-        UNIMPLEMENTED();
-        return {};
-    }
-
    Id AtomicImageAnd(Operation operation) {
        UNIMPLEMENTED();
        return {};
@@ -1440,10 +1436,9 @@ private:
        &SPIRVDecompiler::TextureQueryLod,
        &SPIRVDecompiler::TexelFetch,

+        &SPIRVDecompiler::ImageLoad,
        &SPIRVDecompiler::ImageStore,
        &SPIRVDecompiler::AtomicImageAdd,
-        &SPIRVDecompiler::AtomicImageMin,
-        &SPIRVDecompiler::AtomicImageMax,
        &SPIRVDecompiler::AtomicImageAnd,
        &SPIRVDecompiler::AtomicImageOr,
        &SPIRVDecompiler::AtomicImageXor,
@@ -41,11 +41,46 @@ u32 ShaderIR::DecodeImage(NodeBlock& bb, u32 pc) {
    const Instruction instr = {program_code[pc]};
    const auto opcode = OpCode::Decode(instr);

+    const auto GetCoordinates = [this, instr](Tegra::Shader::ImageType image_type) {
+        std::vector<Node> coords;
+        const std::size_t num_coords{GetImageTypeNumCoordinates(image_type)};
+        coords.reserve(num_coords);
+        for (std::size_t i = 0; i < num_coords; ++i) {
+            coords.push_back(GetRegister(instr.gpr8.Value() + i));
+        }
+        return coords;
+    };
+
    switch (opcode->get().GetId()) {
+    case OpCode::Id::SULD: {
+        UNIMPLEMENTED_IF(instr.suldst.mode != Tegra::Shader::SurfaceDataMode::P);
+        UNIMPLEMENTED_IF(instr.suldst.out_of_bounds_store !=
+                         Tegra::Shader::OutOfBoundsStore::Ignore);
+
+        const auto type{instr.suldst.image_type};
+        auto& image{instr.suldst.is_immediate ? GetImage(instr.image, type)
+                                              : GetBindlessImage(instr.gpr39, type)};
+        image.MarkRead();
+
+        u32 indexer = 0;
+        for (u32 element = 0; element < 4; ++element) {
+            if (!instr.suldst.IsComponentEnabled(element)) {
+                continue;
+            }
+            MetaImage meta{image, {}, element};
+            Node value = Operation(OperationCode::ImageLoad, meta, GetCoordinates(type));
+            SetTemporary(bb, indexer++, std::move(value));
+        }
+        for (u32 i = 0; i < indexer; ++i) {
+            SetRegister(bb, instr.gpr0.Value() + i, GetTemporary(i));
+        }
+        break;
+    }
    case OpCode::Id::SUST: {
-        UNIMPLEMENTED_IF(instr.sust.mode != Tegra::Shader::SurfaceDataMode::P);
-        UNIMPLEMENTED_IF(instr.sust.out_of_bounds_store != Tegra::Shader::OutOfBoundsStore::Ignore);
-        UNIMPLEMENTED_IF(instr.sust.component_mask_selector != 0xf); // Ensure we have an RGBA store
+        UNIMPLEMENTED_IF(instr.suldst.mode != Tegra::Shader::SurfaceDataMode::P);
+        UNIMPLEMENTED_IF(instr.suldst.out_of_bounds_store !=
+                         Tegra::Shader::OutOfBoundsStore::Ignore);
+        UNIMPLEMENTED_IF(instr.suldst.component_mask_selector != 0xf); // Ensure we have RGBA

        std::vector<Node> values;
        constexpr std::size_t hardcoded_size{4};
@@ -53,58 +88,51 @@ u32 ShaderIR::DecodeImage(NodeBlock& bb, u32 pc) {
            values.push_back(GetRegister(instr.gpr0.Value() + i));
        }

-        std::vector<Node> coords;
-        const std::size_t num_coords{GetImageTypeNumCoordinates(instr.sust.image_type)};
-        for (std::size_t i = 0; i < num_coords; ++i) {
-            coords.push_back(GetRegister(instr.gpr8.Value() + i));
-        }
-
-        const auto type{instr.sust.image_type};
-        auto& image{instr.sust.is_immediate ? GetImage(instr.image, type)
-                                            : GetBindlessImage(instr.gpr39, type)};
+        const auto type{instr.suldst.image_type};
+        auto& image{instr.suldst.is_immediate ? GetImage(instr.image, type)
+                                              : GetBindlessImage(instr.gpr39, type)};
        image.MarkWrite();

-        MetaImage meta{image, values};
-        bb.push_back(Operation(OperationCode::ImageStore, meta, std::move(coords)));
+        MetaImage meta{image, std::move(values)};
+        bb.push_back(Operation(OperationCode::ImageStore, meta, GetCoordinates(type)));
        break;
    }
    case OpCode::Id::SUATOM: {
        UNIMPLEMENTED_IF(instr.suatom_d.is_ba != 0);

-        Node value = GetRegister(instr.gpr0);
-
-        std::vector<Node> coords;
-        const std::size_t num_coords{GetImageTypeNumCoordinates(instr.sust.image_type)};
-        for (std::size_t i = 0; i < num_coords; ++i) {
-            coords.push_back(GetRegister(instr.gpr8.Value() + i));
-        }
-
        const OperationCode operation_code = [instr] {
-            switch (instr.suatom_d.operation) {
-            case Tegra::Shader::ImageAtomicOperation::Add:
-                return OperationCode::AtomicImageAdd;
-            case Tegra::Shader::ImageAtomicOperation::Min:
-                return OperationCode::AtomicImageMin;
-            case Tegra::Shader::ImageAtomicOperation::Max:
-                return OperationCode::AtomicImageMax;
-            case Tegra::Shader::ImageAtomicOperation::And:
-                return OperationCode::AtomicImageAnd;
-            case Tegra::Shader::ImageAtomicOperation::Or:
-                return OperationCode::AtomicImageOr;
-            case Tegra::Shader::ImageAtomicOperation::Xor:
-                return OperationCode::AtomicImageXor;
-            case Tegra::Shader::ImageAtomicOperation::Exch:
-                return OperationCode::AtomicImageExchange;
+            switch (instr.suatom_d.operation_type) {
+            case Tegra::Shader::ImageAtomicOperationType::S32:
+            case Tegra::Shader::ImageAtomicOperationType::U32:
+                switch (instr.suatom_d.operation) {
+                case Tegra::Shader::ImageAtomicOperation::Add:
+                    return OperationCode::AtomicImageAdd;
+                case Tegra::Shader::ImageAtomicOperation::And:
+                    return OperationCode::AtomicImageAnd;
+                case Tegra::Shader::ImageAtomicOperation::Or:
+                    return OperationCode::AtomicImageOr;
+                case Tegra::Shader::ImageAtomicOperation::Xor:
+                    return OperationCode::AtomicImageXor;
+                case Tegra::Shader::ImageAtomicOperation::Exch:
+                    return OperationCode::AtomicImageExchange;
+                }
            default:
-                UNIMPLEMENTED_MSG("Unimplemented operation={}",
-                                  static_cast<u32>(instr.suatom_d.operation.Value()));
-                return OperationCode::AtomicImageAdd;
+                break;
            }
+            UNIMPLEMENTED_MSG("Unimplemented operation={} type={}",
+                              static_cast<u64>(instr.suatom_d.operation.Value()),
+                              static_cast<u64>(instr.suatom_d.operation_type.Value()));
+            return OperationCode::AtomicImageAdd;
        }();

-        const auto& image{GetImage(instr.image, instr.suatom_d.image_type, instr.suatom_d.size)};
+        Node value = GetRegister(instr.gpr0);
+
+        const auto type = instr.suatom_d.image_type;
+        auto& image = GetImage(instr.image, type);
+        image.MarkAtomic();
+
        MetaImage meta{image, {std::move(value)}};
-        SetRegister(bb, instr.gpr0, Operation(operation_code, meta, std::move(coords)));
+        SetRegister(bb, instr.gpr0, Operation(operation_code, meta, GetCoordinates(type)));
        break;
    }
    default:
@@ -114,35 +142,32 @@ u32 ShaderIR::DecodeImage(NodeBlock& bb, u32 pc) {
    return pc;
 }

-Image& ShaderIR::GetImage(Tegra::Shader::Image image, Tegra::Shader::ImageType type,
-                          std::optional<Tegra::Shader::ImageAtomicSize> size) {
+Image& ShaderIR::GetImage(Tegra::Shader::Image image, Tegra::Shader::ImageType type) {
    const auto offset{static_cast<std::size_t>(image.index.Value())};
-    if (const auto image = TryUseExistingImage(offset, type, size)) {
+    if (const auto image = TryUseExistingImage(offset, type)) {
        return *image;
    }

    const std::size_t next_index{used_images.size()};
-    return used_images.emplace(offset, Image{offset, next_index, type, size}).first->second;
+    return used_images.emplace(offset, Image{offset, next_index, type}).first->second;
 }

-Image& ShaderIR::GetBindlessImage(Tegra::Shader::Register reg, Tegra::Shader::ImageType type,
-                                  std::optional<Tegra::Shader::ImageAtomicSize> size) {
+Image& ShaderIR::GetBindlessImage(Tegra::Shader::Register reg, Tegra::Shader::ImageType type) {
    const Node image_register{GetRegister(reg)};
    const auto [base_image, cbuf_index, cbuf_offset]{
        TrackCbuf(image_register, global_code, static_cast<s64>(global_code.size()))};
    const auto cbuf_key{(static_cast<u64>(cbuf_index) << 32) | static_cast<u64>(cbuf_offset)};

-    if (const auto image = TryUseExistingImage(cbuf_key, type, size)) {
+    if (const auto image = TryUseExistingImage(cbuf_key, type)) {
        return *image;
    }

    const std::size_t next_index{used_images.size()};
-    return used_images.emplace(cbuf_key, Image{cbuf_index, cbuf_offset, next_index, type, size})
+    return used_images.emplace(cbuf_key, Image{cbuf_index, cbuf_offset, next_index, type})
        .first->second;
 }

-Image* ShaderIR::TryUseExistingImage(u64 offset, Tegra::Shader::ImageType type,
-                                     std::optional<Tegra::Shader::ImageAtomicSize> size) {
+Image* ShaderIR::TryUseExistingImage(u64 offset, Tegra::Shader::ImageType type) {
    auto it = used_images.find(offset);
    if (it == used_images.end()) {
        return nullptr;
@@ -150,14 +175,6 @@ Image* ShaderIR::TryUseExistingImage(u64 offset, Tegra::Shader::ImageType type,
    auto& image = it->second;
    ASSERT(image.GetType() == type);

-    if (size) {
-        // We know the size, if it's known it has to be the same as before, otherwise we can set it.
-        if (image.IsSizeKnown()) {
-            ASSERT(image.GetSize() == size);
-        } else {
-            image.SetSize(*size);
-        }
-    }
    return &image;
 }

@@ -149,10 +149,10 @@ enum class OperationCode {
    TextureQueryLod,        /// (MetaTexture, float[N] coords) -> float4
    TexelFetch,             /// (MetaTexture, int[N], int) -> float4

-    ImageStore,          /// (MetaImage, int[N] values) -> void
+    ImageLoad,  /// (MetaImage, int[N] coords) -> void
+    ImageStore, /// (MetaImage, int[N] coords) -> void
+
    AtomicImageAdd,      /// (MetaImage, int[N] coords) -> void
-    AtomicImageMin,      /// (MetaImage, int[N] coords) -> void
-    AtomicImageMax,      /// (MetaImage, int[N] coords) -> void
    AtomicImageAnd,      /// (MetaImage, int[N] coords) -> void
    AtomicImageOr,       /// (MetaImage, int[N] coords) -> void
    AtomicImageXor,      /// (MetaImage, int[N] coords) -> void
@@ -294,21 +294,18 @@ private:

 class Image final {
 public:
-    constexpr explicit Image(std::size_t offset, std::size_t index, Tegra::Shader::ImageType type,
-                             std::optional<Tegra::Shader::ImageAtomicSize> size)
-        : offset{offset}, index{index}, type{type}, is_bindless{false}, size{size} {}
+    constexpr explicit Image(std::size_t offset, std::size_t index, Tegra::Shader::ImageType type)
+        : offset{offset}, index{index}, type{type}, is_bindless{false} {}

    constexpr explicit Image(u32 cbuf_index, u32 cbuf_offset, std::size_t index,
-                             Tegra::Shader::ImageType type,
-                             std::optional<Tegra::Shader::ImageAtomicSize> size)
+                             Tegra::Shader::ImageType type)
        : offset{(static_cast<u64>(cbuf_index) << 32) | cbuf_offset}, index{index}, type{type},
-          is_bindless{true}, size{size} {}
+          is_bindless{true} {}

    constexpr explicit Image(std::size_t offset, std::size_t index, Tegra::Shader::ImageType type,
-                             bool is_bindless, bool is_written, bool is_read,
-                             std::optional<Tegra::Shader::ImageAtomicSize> size)
+                             bool is_bindless, bool is_written, bool is_read, bool is_atomic)
        : offset{offset}, index{index}, type{type}, is_bindless{is_bindless},
-          is_written{is_written}, is_read{is_read}, size{size} {}
+          is_written{is_written}, is_read{is_read}, is_atomic{is_atomic} {}

    void MarkWrite() {
        is_written = true;
@@ -318,8 +315,10 @@ public:
        is_read = true;
    }

-    void SetSize(Tegra::Shader::ImageAtomicSize size_) {
-        size = size_;
+    void MarkAtomic() {
+        MarkWrite();
+        MarkRead();
+        is_atomic = true;
    }

    constexpr std::size_t GetOffset() const {
@@ -346,21 +345,17 @@ public:
        return is_read;
    }

+    constexpr bool IsAtomic() const {
+        return is_atomic;
+    }
+
    constexpr std::pair<u32, u32> GetBindlessCBuf() const {
        return {static_cast<u32>(offset >> 32), static_cast<u32>(offset)};
    }

-    constexpr bool IsSizeKnown() const {
-        return size.has_value();
-    }
-
-    constexpr Tegra::Shader::ImageAtomicSize GetSize() const {
-        return size.value();
-    }
-
    constexpr bool operator<(const Image& rhs) const {
-        return std::tie(offset, index, type, size, is_bindless) <
-               std::tie(rhs.offset, rhs.index, rhs.type, rhs.size, rhs.is_bindless);
+        return std::tie(offset, index, type, is_bindless) <
+               std::tie(rhs.offset, rhs.index, rhs.type, rhs.is_bindless);
    }

 private:
@@ -370,7 +365,7 @@ private:
    bool is_bindless{};
    bool is_written{};
    bool is_read{};
-    std::optional<Tegra::Shader::ImageAtomicSize> size{};
+    bool is_atomic{};
 };

 struct GlobalMemoryBase {
@@ -402,6 +397,7 @@ struct MetaTexture {
 struct MetaImage {
    const Image& image;
    std::vector<Node> values;
+    u32 element{};
 };

 /// Parameters that modify an operation but are not part of any particular operand
@@ -284,16 +284,13 @@ private:
                                      bool is_shadow);

    /// Accesses an image.
-    Image& GetImage(Tegra::Shader::Image image, Tegra::Shader::ImageType type,
-                    std::optional<Tegra::Shader::ImageAtomicSize> size = {});
+    Image& GetImage(Tegra::Shader::Image image, Tegra::Shader::ImageType type);

    /// Access a bindless image sampler.
-    Image& GetBindlessImage(Tegra::Shader::Register reg, Tegra::Shader::ImageType type,
-                            std::optional<Tegra::Shader::ImageAtomicSize> size = {});
+    Image& GetBindlessImage(Tegra::Shader::Register reg, Tegra::Shader::ImageType type);

    /// Tries to access an existing image, updating it's state as needed
-    Image* TryUseExistingImage(u64 offset, Tegra::Shader::ImageType type,
-                               std::optional<Tegra::Shader::ImageAtomicSize> size);
+    Image* TryUseExistingImage(u64 offset, Tegra::Shader::ImageType type);

    /// Extracts a sequence of bits from a node
    Node BitfieldExtract(Node value, u32 offset, u32 bits);
@@ -224,8 +224,13 @@ public:
                     const Tegra::Engines::Fermi2D::Regs::Surface& dst_config,
                     const Tegra::Engines::Fermi2D::Config& copy_config) {
        std::lock_guard lock{mutex};
-        std::pair<TSurface, TView> dst_surface = GetFermiSurface(dst_config);
-        std::pair<TSurface, TView> src_surface = GetFermiSurface(src_config);
+        SurfaceParams src_params = SurfaceParams::CreateForFermiCopySurface(src_config);
+        SurfaceParams dst_params = SurfaceParams::CreateForFermiCopySurface(dst_config);
+        const GPUVAddr src_gpu_addr = src_config.Address();
+        const GPUVAddr dst_gpu_addr = dst_config.Address();
+        DeduceBestBlit(src_params, dst_params, src_gpu_addr, dst_gpu_addr);
+        std::pair<TSurface, TView> dst_surface = GetSurface(dst_gpu_addr, dst_params, true, false);
+        std::pair<TSurface, TView> src_surface = GetSurface(src_gpu_addr, src_params, true, false);
        ImageBlit(src_surface.second, dst_surface.second, copy_config);
        dst_surface.first->MarkAsModified(true, Tick());
    }
@@ -357,6 +362,29 @@ private:
        BufferCopy = 3,
    };

+    enum class DeductionType : u32 {
+        DeductionComplete,
+        DeductionIncomplete,
+        DeductionFailed,
+    };
+
+    struct Deduction {
+        DeductionType type{DeductionType::DeductionFailed};
+        TSurface surface{};
+
+        bool Failed() const {
+            return type == DeductionType::DeductionFailed;
+        }
+
+        bool Incomplete() const {
+            return type == DeductionType::DeductionIncomplete;
+        }
+
+        bool IsDepth() const {
+            return surface->GetSurfaceParams().IsPixelFormatZeta();
+        }
+    };
+
    /**
     * `PickStrategy` takes care of selecting a proper strategy to deal with a texture recycle.
     * @param overlaps, the overlapping surfaces registered in the cache.
@@ -691,6 +719,118 @@ private:
                              MatchTopologyResult::FullMatch);
    }

+    /**
+     * `DeduceSurface` gets the starting address and parameters of a candidate surface and tries
+     * to find a matching surface within the cache that's similar to it. If there are many textures
+     * or the texture found if entirely incompatible, it will fail. If no texture is found, the
+     * blit will be unsuccessful.
+     * @param gpu_addr, the starting address of the candidate surface.
+     * @param params, the paremeters on the candidate surface.
+     **/
+    Deduction DeduceSurface(const GPUVAddr gpu_addr, const SurfaceParams& params) {
+        const auto host_ptr{system.GPU().MemoryManager().GetPointer(gpu_addr)};
+        const auto cache_addr{ToCacheAddr(host_ptr)};
+
+        if (!cache_addr) {
+            Deduction result{};
+            result.type = DeductionType::DeductionFailed;
+            return result;
+        }
+
+        if (const auto iter = l1_cache.find(cache_addr); iter != l1_cache.end()) {
+            TSurface& current_surface = iter->second;
+            const auto topological_result = current_surface->MatchesTopology(params);
+            if (topological_result != MatchTopologyResult::FullMatch) {
+                Deduction result{};
+                result.type = DeductionType::DeductionFailed;
+                return result;
+            }
+            const auto struct_result = current_surface->MatchesStructure(params);
+            if (struct_result != MatchStructureResult::None &&
+                current_surface->MatchTarget(params.target)) {
+                Deduction result{};
+                result.type = DeductionType::DeductionComplete;
+                result.surface = current_surface;
+                return result;
+            }
+        }
+
+        const std::size_t candidate_size = params.GetGuestSizeInBytes();
+        auto overlaps{GetSurfacesInRegion(cache_addr, candidate_size)};
+
+        if (overlaps.empty()) {
+            Deduction result{};
+            result.type = DeductionType::DeductionIncomplete;
+            return result;
+        }
+
+        if (overlaps.size() > 1) {
+            Deduction result{};
+            result.type = DeductionType::DeductionFailed;
+            return result;
+        } else {
+            Deduction result{};
+            result.type = DeductionType::DeductionComplete;
+            result.surface = overlaps[0];
+            return result;
+        }
+    }
+
+    /**
+     * `DeduceBestBlit` gets the a source and destination starting address and parameters,
+     * and tries to deduce if they are supposed to be depth textures. If so, their
+     * parameters are modified and fixed into so.
+     * @param gpu_addr, the starting address of the candidate surface.
+     * @param params, the parameters on the candidate surface.
+     **/
+    void DeduceBestBlit(SurfaceParams& src_params, SurfaceParams& dst_params,
+                        const GPUVAddr src_gpu_addr, const GPUVAddr dst_gpu_addr) {
+        auto deduc_src = DeduceSurface(src_gpu_addr, src_params);
+        auto deduc_dst = DeduceSurface(src_gpu_addr, src_params);
+        if (deduc_src.Failed() || deduc_dst.Failed()) {
+            return;
+        }
+
+        const bool incomplete_src = deduc_src.Incomplete();
+        const bool incomplete_dst = deduc_dst.Incomplete();
+
+        if (incomplete_src && incomplete_dst) {
+            return;
+        }
+
+        const bool any_incomplete = incomplete_src || incomplete_dst;
+
+        if (!any_incomplete && !(deduc_src.IsDepth() && deduc_dst.IsDepth())) {
+            return;
+        }
+
+        if (incomplete_src && !(deduc_dst.IsDepth())) {
+            return;
+        }
+
+        if (incomplete_dst && !(deduc_src.IsDepth())) {
+            return;
+        }
+
+        const auto inherit_format = ([](SurfaceParams& to, TSurface from) {
+            const SurfaceParams& params = from->GetSurfaceParams();
+            to.pixel_format = params.pixel_format;
+            to.component_type = params.component_type;
+            to.type = params.type;
+        });
+        // Now we got the cases where one or both is Depth and the other is not known
+        if (!incomplete_src) {
+            inherit_format(src_params, deduc_src.surface);
+        } else {
+            inherit_format(src_params, deduc_dst.surface);
+        }
+        if (!incomplete_dst) {
+            inherit_format(dst_params, deduc_dst.surface);
+        } else {
+            inherit_format(dst_params, deduc_src.surface);
+        }
+    }
+
    std::pair<TSurface, TView> InitializeSurface(GPUVAddr gpu_addr, const SurfaceParams& params,
                                                 bool preserve_contents) {
        auto new_surface{GetUncachedSurface(gpu_addr, params)};
@@ -62,6 +62,8 @@ ConfigureGraphics::ConfigureGraphics(QWidget* parent)
        }
        UpdateBackgroundColorButton(new_bg_color);
    });
+    connect(ui->brightness_reset, &QPushButton::pressed, this,
+            [this] { ui->brightness_slider->setValue(100); });
 }

 ConfigureGraphics::~ConfigureGraphics() = default;
@@ -80,6 +82,7 @@ void ConfigureGraphics::SetConfiguration() {
    ui->force_30fps_mode->setChecked(Settings::values.force_30fps_mode);
    UpdateBackgroundColorButton(QColor::fromRgbF(Settings::values.bg_red, Settings::values.bg_green,
                                                 Settings::values.bg_blue));
+    ui->brightness_slider->setValue(Settings::values.backlight_brightness * 100 + 50);
 }

 void ConfigureGraphics::ApplyConfiguration() {
@@ -93,6 +96,7 @@ void ConfigureGraphics::ApplyConfiguration() {
    Settings::values.bg_red = static_cast<float>(bg_color.redF());
    Settings::values.bg_green = static_cast<float>(bg_color.greenF());
    Settings::values.bg_blue = static_cast<float>(bg_color.blueF());
+    Settings::values.backlight_brightness = (ui->brightness_slider->value() - 50.0f) / 100.0f;
 }

 void ConfigureGraphics::changeEvent(QEvent* event) {
@@ -111,6 +111,68 @@
          </item>
         </layout>
        </item>
+        <item>
+         <layout class="QHBoxLayout" name="horizontalLayout_3">
+          <item>
+           <widget class="QLabel" name="label_2">
+            <property name="text">
+             <string>Brightness</string>
+            </property>
+           </widget>
+          </item>
+          <item>
+           <spacer name="horizontalSpacer">
+            <property name="orientation">
+             <enum>Qt::Horizontal</enum>
+            </property>
+            <property name="sizeHint" stdset="0">
+             <size>
+              <width>40</width>
+              <height>20</height>
+             </size>
+            </property>
+           </spacer>
+          </item>
+          <item>
+           <widget class="QSlider" name="brightness_slider">
+            <property name="minimum">
+             <number>50</number>
+            </property>
+            <property name="maximum">
+             <number>150</number>
+            </property>
+            <property name="singleStep">
+             <number>10</number>
+            </property>
+            <property name="pageStep">
+             <number>20</number>
+            </property>
+            <property name="value">
+             <number>100</number>
+            </property>
+            <property name="orientation">
+             <enum>Qt::Horizontal</enum>
+            </property>
+            <property name="tickPosition">
+             <enum>QSlider::NoTicks</enum>
+            </property>
+           </widget>
+          </item>
+          <item>
+           <widget class="QPushButton" name="brightness_reset">
+            <property name="sizePolicy">
+             <sizepolicy hsizetype="Fixed" vsizetype="Fixed">
+              <horstretch>0</horstretch>
+              <verstretch>0</verstretch>
+             </sizepolicy>
+            </property>
+            <property name="text">
+             <string>Reset</string>
+            </property>
+           </widget>
+          </item>
+         </layout>
+        </item>
       </layout>
      </widget>
     </item>
Author	SHA1	Message	Date
yuzubot	8da8f4bfaf	"Merge Tagged PR 1012"	2019-10-02 23:35:05 +00:00
yuzubot	717741698e	"Merge Tagged PR 1340"	2019-10-02 23:35:04 +00:00
yuzubot	8297770368	"Merge Tagged PR 1703"	2019-10-02 23:35:04 +00:00
yuzubot	75d3bab835	"Merge Tagged PR 2542"	2019-10-02 23:35:02 +00:00
yuzubot	07b795b909	"Merge Tagged PR 2710"	2019-10-02 23:35:01 +00:00
yuzubot	49d1830b1e	"Merge Tagged PR 2859"	2019-10-02 23:35:01 +00:00
yuzubot	948485ff07	"Merge Tagged PR 2912"	2019-10-02 23:35:00 +00:00
yuzubot	d007e48c24	"Merge Tagged PR 2914"	2019-10-02 23:34:59 +00:00
yuzubot	f913507685	"Merge Tagged PR 2917"	2019-10-02 23:34:58 +00:00
yuzubot	3ab2a12ed3	"Merge Tagged PR 2927"	2019-10-02 23:34:58 +00:00
bunnei	6bfabdedfd	Merge pull request #2937 from DarkLordZach/azure-msvc ci: Add windows MSVC builds to patreon and mainline pipelines	2019-10-02 19:28:03 -04:00
Zach Hilman	a86e52a375	ci: Correct mainline release dependency	2019-10-02 18:54:05 -04:00
Zach Hilman	53be058e74	ci: Add Mainline tagline	2019-10-02 18:51:21 -04:00
Zach Hilman	d648cd562a	ci: Use MSVC windows for patreon	2019-10-02 18:23:09 -04:00
Zach Hilman	bfa60e2d4e	ci: Use MSVC windows for mainline	2019-10-02 17:58:52 -04:00
Zach Hilman	514b74a098	ci: Add MSVC build template	2019-10-02 17:58:33 -04:00
Zach Hilman	49344111cc	ci: Add Windows MSVC package script	2019-10-02 17:53:53 -04:00
Zach Hilman	c56822a405	ci: Fix unset environment variable bug	2019-10-02 17:53:37 -04:00
bunnei	f8ce672b67	Merge pull request #2932 from DarkLordZach/azure-gitpages ci: Iterate through GitHub pages for pull requests	2019-10-01 23:22:49 -04:00
Zach Hilman	0e58bfedfd	ci: Check additional pages on GitHub	2019-10-01 21:54:26 -04:00
Zach Hilman	7d46416a16	ci: Iterate through GitHub pages for pull requests	2019-10-01 21:38:09 -04:00
bunnei	5e677a3178	Merge pull request #2904 from ogniK5377/better-signal-hid Signal styleset changes at a better time	2019-10-01 10:58:02 -04:00
bunnei	c26e9c4cd1	Merge pull request #2924 from MysticExile/readme-2 Remove AppVeyor status from the readme.md	2019-10-01 10:57:19 -04:00
bunnei	80d6abc08b	Merge pull request #2929 from DarkLordZach/azure-patreon-merge-extra ci: Merge patreon tagged PRs from public repository during merge step.	2019-10-01 10:56:50 -04:00
Zach Hilman	4ea425d6cf	ci: Correct arguments for mergebot script	2019-09-30 13:43:34 -04:00
Zach Hilman	e6f3aad84e	ci: Populate patreon PRs on public repository	2019-09-30 09:29:28 -04:00
bunnei	e11afeb34d	Merge pull request #2923 from yuzu-emu/revert-2574-dynarmic-jit-nullptr Revert "arm_dynarmic: Check if jit is nullptr when preparing reschedule"	2019-09-29 21:54:36 -04:00
bunnei	dc29919bbe	Revert "arm_dynarmic: Check if jit is nullptr when preparing reschedule"	2019-09-29 21:54:19 -04:00
bunnei	28538bba9c	Merge pull request #2574 from DarkLordZach/dynarmic-jit-nullptr arm_dynarmic: Check if jit is nullptr when preparing reschedule	2019-09-29 21:44:10 -04:00
Ethan	4bd6196b0a	Remove AppVeyor status	2019-09-29 22:32:44 +02:00
Rodrigo Locatti	8a2e96de2e	Merge pull request #2909 from MysticExile/fmt-2 externals: Update fmt to master	2019-09-27 14:09:32 -03:00
Rodrigo Locatti	c4a99944f2	Merge pull request #2911 from DarkLordZach/azure-patreon ci: Add azure patreon pipeline	2019-09-26 19:30:33 -03:00
Zach Hilman	867e1db287	ci: Add two step patreon build pipeline	2019-09-26 09:42:22 -04:00
Zach Hilman	0ba7055873	ci: Add private mergebot CI templates	2019-09-26 09:29:33 -04:00
Zach Hilman	3f52bb5677	ci: Optionally prepend commit messages with environment variable	2019-09-26 09:29:15 -04:00
Zach Hilman	ab8d122384	ci: Add private mergebot script Uses Azure instead of GitHub	2019-09-26 09:28:59 -04:00
Ethan	3e5790de79	externals: Update fmt to master	2019-09-26 00:34:37 +02:00
bunnei	da07373599	Merge pull request #2864 from ReinUsesLisp/fix-clang-decompiler gl_shader_decompiler: Fix clang build issues	2019-09-25 17:46:02 -04:00
David Marcec	77fbf29047	Signal styleset changes at a better time We should signal when a net controller is added and our event should be manual, not automatic.	2019-09-24 17:24:37 +10:00
ReinUsesLisp	f926230ab1	gl_shader_decompiler: Add tailing return for HUnpack2	2019-09-24 01:03:59 -03:00
ReinUsesLisp	25bfaffdff	gl_shader_decompiler: Fix clang build issues	2019-09-24 01:03:27 -03:00
bunnei	376f1a4432	Merge pull request #2869 from ReinUsesLisp/suld shader/image: Implement SULD and fix SUATOM	2019-09-23 21:47:03 -04:00
James Rowe	2f5545f8de	Merge pull request #2901 from DarkLordZach/mainline-title-bar ci: Properly parse Azure CI details in title bar	2019-09-22 18:14:08 -06:00
Zach Hilman	921dae5e83	ci: Correct GitHub Release name	2019-09-22 19:06:48 -04:00
Zach Hilman	15805f4e01	ci: Determine build date via bash	2019-09-22 18:39:29 -04:00
Zach Hilman	851c5d67ae	ci: Update to use date as build number	2019-09-22 17:23:47 -04:00
Zach Hilman	14248685af	cmake: Add SCM detection for Azure	2019-09-22 17:23:10 -04:00
ReinUsesLisp	44000971e2	gl_shader_decompiler: Use uint for images and fix SUATOM In the process remove implementation of SUATOM.MIN and SUATOM.MAX as these require a distinction between U32 and S32. These have to be implemented with imageCompSwap loop.	2019-09-21 17:33:52 -03:00
ReinUsesLisp	675f23aedc	shader/image: Implement SULD and remove irrelevant code * Implement SULD as float. * Remove conditional declaration of GL_ARB_shader_viewport_layer_array.	2019-09-21 17:32:48 -03:00
ReinUsesLisp	4de0f1e1c8	shader_bytecode: Add SULD encoding	2019-09-21 17:31:46 -03:00
Zach Hilman	425cdf946c	arm_dynarmic: Check if jit is nullptr when preparing reschedule Prevents crash with multiprocess loading.	2019-06-10 00:14:25 -04:00