GS/HW: Refactor StretchRect() to have single entry to renderer/reduce duplication.

We currently have a barrier so let's do date_one in a single pass, plus this avoids any issues with copies and stencil issues.
On dx copies are slower so we can only use the optimization if we have barriers already present.

.github/workflows/lint_gamedb.yml

@@ -23,7 +23,7 @@ jobs:
       - name: Install Packages
         run: |
           npm install -g ajv-cli prettier
-          sudo apt-get -y install yamllint
+          pip install yamllint
 
       - name: Validate YAML
         run: |

bin/resources/game_controller_db.txt

@@ -175,7 +175,6 @@
 030000001a1c00000001000000000000,Datel Arcade Joystick,a:b1,b:b2,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b4,lefttrigger:b6,leftx:a0,lefty:a1,rightshoulder:b5,righttrigger:b7,rightx:a2,righty:a5,start:b9,x:b0,y:b3,platform:Windows,
 03000000451300000830000000000000,Defender Game Racer X7,a:b0,b:b1,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b4,leftstick:b10,lefttrigger:b6,leftx:a0,lefty:a1,rightshoulder:b5,rightstick:b11,righttrigger:b7,rightx:a2,righty:a3,start:b9,x:b2,y:b3,platform:Windows,
 03000000791d00000103000000000000,Dual Box Wii,a:b2,b:b1,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,guide:b10,leftshoulder:b6,lefttrigger:b4,leftx:a0,lefty:a1,rightshoulder:b7,righttrigger:b5,rightx:a2,righty:a3,start:b9,x:b3,y:b0,platform:Windows,
-03000000c0160000e105000000000000,Dual Controller,a:b1,b:b2,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b4,leftstick:b10,lefttrigger:b6,leftx:a0,lefty:a1,rightshoulder:b5,rightstick:b11,righttrigger:b7,rightx:a3,righty:a4,start:b9,x:b0,y:b3,platform:Windows,
 030000004f040000070f000000000000,Dual Power,a:b8,b:b9,back:b4,dpdown:b1,dpleft:b2,dpright:b3,dpup:b0,leftshoulder:b13,leftstick:b6,lefttrigger:b14,leftx:a0,lefty:a1,rightshoulder:b12,rightstick:b7,righttrigger:b15,start:b5,x:b10,y:b11,platform:Windows,
 030000004f04000012b3000000000000,Dual Power 3,a:b0,b:b2,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b4,leftstick:b10,lefttrigger:b5,leftx:a0,lefty:a1,rightshoulder:b6,rightstick:b11,righttrigger:b7,rightx:a2,righty:a3,start:b9,x:b1,y:b3,platform:Windows,
 030000004f04000020b3000000000000,Dual Trigger,a:b0,b:b2,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b4,leftstick:b10,lefttrigger:b5,leftx:a0,lefty:a1,rightshoulder:b6,rightstick:b11,righttrigger:b7,rightx:a2,righty:a3,start:b9,x:b1,y:b3,platform:Windows,
@@ -839,6 +838,7 @@
 03000000172700004431000000000000,Xiaomi Controller,a:b0,b:b1,back:b10,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,guide:b20,leftshoulder:b6,leftstick:b13,lefttrigger:b8,leftx:a0,lefty:a1,rightshoulder:b7,rightstick:b14,righttrigger:a7,rightx:a2,righty:a5,start:b11,x:b3,y:b4,platform:Windows,
 03000000172700003350000000000000,Xiaomi XMGP01YM,a:b0,b:b1,back:b10,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,guide:b12,leftshoulder:b6,leftstick:b13,leftx:a0,lefty:a1,rightshoulder:b7,rightstick:b14,rightx:a3,righty:a4,start:b11,x:b3,y:b4,platform:Windows,
 03000000bc2000005060000000000000,Xiaomi XMGP01YM,+lefty:+a2,+righty:+a5,-lefty:-a1,-righty:-a4,a:b0,b:b1,back:b10,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,guide:b12,leftshoulder:b6,leftstick:b13,lefttrigger:b8,leftx:a0,rightshoulder:b7,rightstick:b14,righttrigger:b9,rightx:a3,start:b11,x:b3,y:b4,platform:Windows,
+03000000c0160000e105000000000000,XinMo Dual Arcade,a:b1,b:b2,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b4,lefttrigger:b6,leftx:a0,lefty:a1,rightshoulder:b5,righttrigger:b7,start:b9,x:b0,y:b3,platform:Windows,
 xinput,XInput Controller,a:b0,b:b1,back:b6,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,guide:b10,leftshoulder:b4,leftstick:b8,lefttrigger:a2,leftx:a0,lefty:a1,rightshoulder:b5,rightstick:b9,righttrigger:a5,rightx:a3,righty:a4,start:b7,x:b2,y:b3,platform:Windows,
 030000007d0400000340000000000000,Xterminator Digital Gamepad,a:b0,b:b1,back:b10,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:-a4,lefttrigger:+a4,leftx:a0,lefty:a1,paddle1:b7,paddle2:b6,rightshoulder:b5,rightstick:b9,righttrigger:b2,rightx:a3,righty:a5,start:b8,x:b3,y:b4,platform:Windows,
 030000002c3600000100000000000000,Yawman Arrow,+rightx:h0.2,+righty:h0.4,-rightx:h0.8,-righty:h0.1,a:b4,b:b5,back:b6,dpdown:b15,dpleft:b14,dpright:b16,dpup:b13,leftshoulder:b10,leftstick:b0,lefttrigger:-a4,leftx:a0,lefty:a1,paddle1:b11,paddle2:b12,rightshoulder:b8,rightstick:b9,righttrigger:+a4,start:b3,x:b1,y:b2,platform:Windows,
@@ -1119,6 +1119,7 @@ xinput,XInput Controller,a:b0,b:b1,back:b6,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,
 030000005f140000c501000000020000,Trust Gamepad,a:b2,b:b1,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,guide:b12,leftshoulder:b4,leftstick:b10,lefttrigger:b6,leftx:a0,lefty:a1,rightshoulder:b5,rightstick:b11,righttrigger:b7,rightx:a2,righty:a3,start:b9,x:b3,y:b0,platform:Mac OS X,
 03000000100800000100000000000000,Twin USB Joystick,a:b4,b:b2,back:b16,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b12,leftstick:b20,lefttrigger:b8,leftx:a0,lefty:a2,rightshoulder:b14,rightstick:b22,righttrigger:b10,rightx:a6,righty:a4,start:b18,x:b6,y:b0,platform:Mac OS X,
 03000000632500002605000000010000,Uberwith Controller,a:b0,b:b1,back:b10,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b6,leftstick:b13,lefttrigger:a5,leftx:a0,lefty:a1,rightshoulder:b7,rightstick:b14,righttrigger:a4,rightx:a2,righty:a3,start:b11,x:b3,y:b4,platform:Mac OS X,
+03000000c0160000e105000000040000,Ultimate Atari Fight Stick,a:b2,b:b4,back:b18,dpdown:+a2,dpleft:-a0,dpright:+a0,dpup:-a2,rightshoulder:b8,righttrigger:b10,start:b16,x:b0,y:b6,platform:Mac OS X,
 03000000151900005678000010010000,Uniplay U6,a:b3,b:b6,back:b25,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b17,leftstick:b31,lefttrigger:b21,leftx:a1,lefty:a3,rightshoulder:b19,rightstick:b33,righttrigger:b23,rightx:a4,righty:a5,start:b27,x:b11,y:b13,platform:Mac OS X,
 030000006f0e00000302000025040000,Victrix PS4 Pro Fightstick,a:b1,b:b2,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,guide:b12,leftshoulder:b4,lefttrigger:b6,rightshoulder:b5,righttrigger:b7,start:b9,touchpad:b13,x:b0,y:b3,platform:Mac OS X,
 030000006f0e00000702000003060000,Victrix PS4 Pro Fightstick,a:b1,b:b2,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,guide:b12,leftshoulder:b4,leftstick:b10,lefttrigger:b6,rightshoulder:b5,righttrigger:b7,start:b9,touchpad:b13,x:b0,y:b3,platform:Mac OS X,
@@ -1774,6 +1775,7 @@ xinput,XInput Controller,a:b0,b:b1,back:b6,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,
 030000005f140000c501000010010000,Trust Gamepad,a:b2,b:b1,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,guide:b12,leftshoulder:b4,leftstick:b10,lefttrigger:b6,leftx:a0,lefty:a1,rightshoulder:b5,rightstick:b11,righttrigger:b7,rightx:a2,righty:a3,start:b9,x:b3,y:b0,platform:Linux,
 06000000f51000000870000003010000,Turtle Beach Recon,a:b0,b:b1,back:b6,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,guide:b8,leftshoulder:b4,leftstick:b9,lefttrigger:a2,leftx:a0,lefty:a1,rightshoulder:b5,rightstick:b10,righttrigger:a5,rightx:a3,righty:a4,start:b7,x:b2,y:b3,platform:Linux,
 03000000100800000100000010010000,Twin PS2 Adapter,a:b2,b:b1,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b6,leftstick:b10,lefttrigger:b4,leftx:a0,lefty:a1,rightshoulder:b7,rightstick:b11,righttrigger:b5,rightx:a3,righty:a2,start:b9,x:b3,y:b0,platform:Linux,
+03000000c0160000e105000010010000,Ultimate Atari Fight Stick,a:b1,b:b2,back:b9,dpdown:+a1,dpleft:-a0,dpright:+a0,dpup:-a1,rightshoulder:b4,righttrigger:b5,start:b8,x:b0,y:b3,platform:Linux,
 03000000151900005678000010010000,Uniplay U6,a:b0,b:b1,back:b10,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b6,leftstick:b13,lefttrigger:a5,leftx:a0,lefty:a1,rightshoulder:b7,rightstick:b14,righttrigger:a4,rightx:a2,righty:a3,start:b11,x:b3,y:b4,platform:Linux,
 03000000100800000300000010010000,USB Gamepad,a:b2,b:b1,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b6,leftstick:b10,lefttrigger:b4,leftx:a0,lefty:a1,rightshoulder:b7,rightstick:b11,righttrigger:b5,rightx:a3,righty:a2,start:b9,x:b3,y:b0,platform:Linux,
 03000000790000000600000007010000,USB gamepad,a:b2,b:b1,back:b8,dpdown:h0.4,dpleft:h0.8,dpright:h0.2,dpup:h0.1,leftshoulder:b4,leftstick:b10,lefttrigger:b6,leftx:a0,lefty:a1,rightshoulder:b5,rightstick:b11,righttrigger:b7,rightx:a3,righty:a4,start:b9,x:b3,y:b0,platform:Linux,

pcsx2-qt/Debugger/ThreadModel.cpp

@@ -13,7 +13,7 @@ ThreadModel::ThreadModel(DebugInterface& cpu, QObject* parent)
 
 int ThreadModel::rowCount(const QModelIndex&) const
 {
-	return m_cpu.GetThreadList().size();
+	return static_cast<int>(m_threads.size());
 }
 
 int ThreadModel::columnCount(const QModelIndex&) const

pcsx2-qt/MainWindow.cpp

@@ -1099,7 +1099,7 @@ bool MainWindow::shouldMouseLock() const
 	if (!Host::GetBoolSettingValue("EmuCore", "EnableMouseLock", false))
 		return false;
 
-	if (m_display_created == false || m_display_surface == nullptr && !isRenderingToMain())
+	if (m_display_created == false || m_display_surface == nullptr)
 		return false;
 
 	bool windowsHidden = (!g_debugger_window || g_debugger_window->isHidden()) &&
@@ -2570,11 +2570,21 @@ void MainWindow::checkMousePosition(int x, int y)
 		const QSize logicalSize = displayWindow->size();
 		const QPoint logicalPosition = displayWindow->position() + displayWindow->parent()->position();
 
+		// The offset to the origin of the current screen is in device-independent pixels while the origin itself is native!
+		// The logicalPosition is the sum of these two values, so we need to separate them and only scale the offset
+		const QScreen* screen = displayWindow->screen();
+
+		// If we fail to get the screen associated with the window, avoid mouse locking as it's probably in an unexpected position.
+		if (!screen)
+			return;
+
+		const QPoint screenPosition = screen->geometry().topLeft();
+
 		// physical frame rect
 		const qreal scale = displayWindow->devicePixelRatio();
 		const QRectF physicalBounds(
-			logicalPosition.x() * scale,
-			logicalPosition.y() * scale,
+			screenPosition.x() + (logicalPosition.x() - screenPosition.x()) * scale,
+			screenPosition.y() + (logicalPosition.y() - screenPosition.y()) * scale,
 			logicalSize.width() * scale,
 			logicalSize.height() * scale);
 

pcsx2-qt/Settings/AchievementLoginDialog.cpp

@@ -127,7 +127,25 @@ void AchievementLoginDialog::processLoginResult(bool result, const QString& mess
 		}
 	}
 
-	done(0);
+	// Show success messagebox
+	const std::string username = Host::GetBaseStringSettingValue("Achievements", "Username");
+	QMessageBox::information(
+		this, tr("Login Successful"),
+		tr("Successfully logged in to RetroAchievements as %1.").arg(QString::fromStdString(username)));
+	
+	m_ui.status->setText(tr("Successfully logged in as %1.").arg(QString::fromStdString(username)));
+	m_ui.status->setStyleSheet("color: green; font-weight: bold;");
+	
+	disconnect(m_ui.buttonBox, &QDialogButtonBox::accepted, this, &AchievementLoginDialog::loginClicked);
+	
+	m_login->setVisible(false);
+	QPushButton* dismissButton = m_ui.buttonBox->addButton(tr("&Dismiss"), QDialogButtonBox::AcceptRole);
+	dismissButton->setDefault(true);
+	dismissButton->setFocus();
+	
+	connect(dismissButton, &QPushButton::clicked, this, [this]() { done(0); });
+	
+	enableUI(false);
 }
 
 void AchievementLoginDialog::connectUi()

pcsx2-qt/Settings/InterfaceSettingsWidget.cpp

@@ -209,7 +209,7 @@ InterfaceSettingsWidget::InterfaceSettingsWidget(SettingsWindow* settings_dialog
 		tr("Shows the game you are currently playing as part of your profile in Discord."));
 	dialog()->registerWidgetHelp(
 		m_ui.mouseLock, tr("Enable Mouse Lock"), tr("Unchecked"),
-		tr("Locks the mouse cursor to the windows when PCSX2 is in focus and all other windows are closed.<br><b>Unavailable on Linux Wayland.</b><br><b>Requires accessibility permissions on macOS.</b><br><b>Limited support for mixed-resolution with non-100% DPI configurations.</b>"));
+		tr("Locks the mouse cursor to the windows when PCSX2 is in focus and all other windows are closed.<br><b>Unavailable on Linux Wayland.</b><br><b>Requires accessibility permissions on macOS.</b>"));
 	dialog()->registerWidgetHelp(
 		m_ui.doubleClickTogglesFullscreen, tr("Double-Click Toggles Fullscreen"), tr("Checked"),
 		tr("Allows switching in and out of fullscreen mode by double-clicking the game window."));

pcsx2/Counters.cpp

@@ -252,7 +252,7 @@ static void vSyncInfoCalc(vSyncTimingInfo* info, double framesPerSecond, u32 sca
 	const u64 accumilatedHRenderError = (hRender % 10000) + (hBlank % 10000);
 	const u64 accumilatedHFractional = accumilatedHRenderError % 10000;
 	info->hRender += (u32)(accumilatedHRenderError / 10000);
-	info->hSyncError = (accumilatedHFractional * (scansPerFrame / (IsInterlacedVideoMode() ? 2 : 1))) / 10000;
+	info->hSyncError = (u32)((accumilatedHFractional * (scansPerFrame / (IsInterlacedVideoMode() ? 2 : 1))) / 10000);
 
 	// Note: In NTSC modes there is some small rounding error in the vsync too,
 	// however it would take thousands of frames for it to amount to anything and

pcsx2/DEV9/InternalServers/DNS_Server.cpp

@@ -160,7 +160,7 @@ namespace InternalServers
 			//Counts
 			ret->questions = dns.questions;
 
-			DNS_State* state = new DNS_State(reqs.size(), reqs, ret, payload->sourcePort);
+			DNS_State* state = new DNS_State(static_cast<int>(reqs.size()), reqs, ret, payload->sourcePort);
 			outstandingQueries++;
 
 			for (size_t i = 0; i < reqs.size(); i++)

pcsx2/DebugTools/BiosDebugData.cpp

@@ -8,22 +8,19 @@
 
 std::vector<std::unique_ptr<BiosThread>> getEEThreads()
 {
-	std::vector<std::unique_ptr<BiosThread>> threads;
-
-	if (CurrentBiosInformation.eeThreadListAddr <= 0)
-		return threads;
+	if (!VMManager::HasValidVM() || CurrentBiosInformation.eeThreadListAddr == 0)
+		return {};
 
 	const u32 start = CurrentBiosInformation.eeThreadListAddr & 0x3fffff;
 
+	std::vector<std::unique_ptr<BiosThread>> threads;
 	for (int tid = 0; tid < 256; tid++)
 	{
+		EEInternalThread* internal = static_cast<EEInternalThread*>(
+			PSM(start + tid * sizeof(EEInternalThread)));
 
-		EEInternalThread* internal = static_cast<EEInternalThread*>(PSM(start + tid * sizeof(EEInternalThread)));
-		if (internal->status != (int)ThreadStatus::THS_BAD)
-		{
-			auto thread = std::make_unique<EEThread>(tid, *internal);
-			threads.push_back(std::move(thread));
-		}
+		if (internal && internal->status != (int)ThreadStatus::THS_BAD)
+			threads.emplace_back(std::make_unique<EEThread>(tid, *internal));
 	}
 
 	return threads;
@@ -63,8 +60,7 @@ std::vector<std::unique_ptr<BiosThread>> getIOPThreads()
 
 		data.PC = iopMemRead32(data.SavedSP + 0x8c);
 
-		auto thread = std::make_unique<IOPThread>(data);
-		threads.push_back(std::move(thread));
+		threads.emplace_back(std::make_unique<IOPThread>(data));
 
 		item = iopMemRead32(item + 0x24);
 	}
@@ -93,7 +89,7 @@ std::vector<IopMod> getIOPModules()
 		if (i > 1000)
 			return {};
 
-		IopMod mod;
+		IopMod& mod = modlist.emplace_back();
 
 		u32 nstr = iopMemRead32(maddr + 4);
 		if (nstr)
@@ -113,8 +109,6 @@ std::vector<IopMod> getIOPModules()
 		mod.data_size = iopMemRead32(maddr + 0x20);
 		mod.bss_size = iopMemRead32(maddr + 0x24);
 
-		modlist.push_back(mod);
-
 		maddr = iopMemRead32(maddr);
 	}
 

pcsx2/GS/GS.cpp

@@ -674,11 +674,11 @@ void GSgetStats(SmallStringBase& info)
 		}
 		else if (pps >= _1kb)
 		{
-			pps /= _1kb; // Kpps
-			prefix = 'K';
+			pps /= _1kb; // kpps
+			prefix = 'k';
 		}
 
-		info.format("{} SW | {} SP | {} P | {} D | {:.2f} S | {:.2f} U | {:.2f} {}pps",
+		info.format("{} SW | {} SYNP | {} PRIM | {} DRW | {:.2f} SWIZ | {:.2f} UNSWIZ | {:.2f} {}pps",
 			api_name,
 			(int)pm.Get(GSPerfMon::SyncPoint),
 			(int)pm.Get(GSPerfMon::Prim),
@@ -693,7 +693,7 @@ void GSgetStats(SmallStringBase& info)
 	}
 	else
 	{
-		info.format("{} HW | {} P | {} D | {} DC | {} B | {} RP | {} RB | {} TC | {} TU",
+		info.format("{} HW | {} PRIM | {} DRW | {} DRWC | {} BAR | {} RP | {} RB | {} TC | {} TU",
 			api_name,
 			(int)pm.Get(GSPerfMon::Prim),
 			(int)pm.Get(GSPerfMon::Draw),
@@ -733,7 +733,7 @@ void GSgetMemoryStats(SmallStringBase& info)
 	{
 		const double hashcache_MB = get_MB(static_cast<double>(g_texture_cache->GetHashCacheMemoryUsage()));
 		const double total_MB = targets_MB + sources_MB + hashcache_MB + pool_MB;
-		info.format("VRAM: {} MB | T: {} MB | S: {} MB | H: {} MB | P: {} MB",
+		info.format("VRAM: {} MB | TGT: {} MB | SRC: {} MB | HC: {} MB | PL: {} MB",
 			format_precision(total_MB),
 			format_precision(targets_MB),
 			format_precision(sources_MB),
@@ -743,7 +743,7 @@ void GSgetMemoryStats(SmallStringBase& info)
 	else
 	{
 		const double total_MB = targets_MB + sources_MB + pool_MB;
-		info.format("VRAM: {} MB | T: {} MB | S: {} MB | P: {} MB",
+		info.format("VRAM: {} MB | TGT: {} MB | SRC: {} MB | PL: {} MB",
 			format_precision(total_MB),
 			format_precision(targets_MB),
 			format_precision(sources_MB),

pcsx2/GS/GSPerfMon.h

@@ -36,7 +36,7 @@ public:
 protected:
 	double m_counters[CounterLast] = {};
 	double m_stats[CounterLast] = {};
-	u64 m_frame = 0;
+	int m_frame = 0;
 	clock_t m_lastframe = 0;
 	int m_count = 0;
 	int m_disp_fb_sprite_blits = 0;
@@ -46,8 +46,8 @@ public:
 
 	void Reset();
 
-	void SetFrame(u64 frame) { m_frame = frame; }
-	u64 GetFrame() { return m_frame; }
+	void SetFrame(int frame) { m_frame = frame; }
+	int GetFrame() { return m_frame; }
 	void EndFrame(bool frame_only);
 
 	void Put(counter_t c, double val) { m_counters[c] += val; }
@@ -68,4 +68,4 @@ public:
 	void Dump(const std::string& filename, bool hw);
 };
 
-extern GSPerfMon g_perfmon;
+extern GSPerfMon g_perfmon;

pcsx2/GS/Renderers/Common/GSDevice.cpp

@@ -723,6 +723,39 @@ GSTexture* GSDevice::CreateTexture(int w, int h, int mipmap_levels, GSTexture::F
 	return FetchSurface(GSTexture::Type::Texture, w, h, levels, format, false, m_features.prefer_new_textures && !prefer_reuse);
 }
 
+void GSDevice::DoStretchRectWithAssertions(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+	GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear)
+{
+	pxAssert((dTex && dTex->IsDepthStencil()) == HasDepthOutput(shader));
+	pxAssert(linear ? SupportsBilinear(shader) : SupportsNearest(shader));
+	GL_INS("StretchRect(%d) {%d,%d} %dx%d -> {%d,%d) %dx%d", shader, int(sRect.left), int(sRect.top),
+		int(sRect.right - sRect.left), int(sRect.bottom - sRect.top), int(dRect.left), int(dRect.top),
+		int(dRect.right - dRect.left), int(dRect.bottom - dRect.top));
+	DoStretchRect(sTex, sRect, dTex, dRect, cms, shader, linear);
+}
+
+void GSDevice::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+	bool red, bool green, bool blue, bool alpha, ShaderConvert shader)
+{
+	GSHWDrawConfig::ColorMaskSelector cms;
+
+	cms.wr = red;
+	cms.wg = green;
+	cms.wb = blue;
+	cms.wa = alpha;
+
+	pxAssert(HasVariableWriteMask(shader));
+	GL_INS("ColorCopy Red:%d Green:%d Blue:%d Alpha:%d", cms.wr, cms.wg, cms.wb, cms.wa);
+
+	DoStretchRectWithAssertions(sTex, sRect, dTex, dRect, cms, shader, false);
+}
+
+void GSDevice::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+	ShaderConvert shader, bool linear)
+{
+	DoStretchRectWithAssertions(sTex, sRect, dTex, dRect, GSHWDrawConfig::ColorMaskSelector(ShaderConvertWriteMask(shader)), shader, linear);
+}
+
 void GSDevice::StretchRect(GSTexture* sTex, GSTexture* dTex, const GSVector4& dRect, ShaderConvert shader, bool linear)
 {
 	StretchRect(sTex, GSVector4(0, 0, 1, 1), dTex, dRect, shader, linear);
@@ -734,11 +767,11 @@ void GSDevice::DrawMultiStretchRects(
 	for (u32 i = 0; i < num_rects; i++)
 	{
 		const MultiStretchRect& sr = rects[i];
-		pxAssert(shader == ShaderConvert::COPY || shader == ShaderConvert::RTA_CORRECTION || rects[0].wmask.wrgba == 0xf);
+		pxAssert(HasVariableWriteMask(shader) || rects[0].wmask.wrgba == 0xf);
 		if (rects[0].wmask.wrgba != 0xf)
 		{
 			g_gs_device->StretchRect(sr.src, sr.src_rect, dTex, sr.dst_rect, rects[0].wmask.wr,
-				rects[0].wmask.wg, rects[0].wmask.wb, rects[0].wmask.wa);
+				rects[0].wmask.wg, rects[0].wmask.wb, rects[0].wmask.wa, shader);
 		}
 		else
 		{

pcsx2/GS/Renderers/Common/GSDevice.h

@@ -69,6 +69,27 @@ enum class ShaderInterlace
 	Count
 };
 
+static inline bool HasVariableWriteMask(ShaderConvert shader)
+{
+	switch (shader)
+	{
+		case ShaderConvert::COPY:
+		case ShaderConvert::RTA_CORRECTION:
+			return true;
+		default:
+			return false;
+	}
+}
+
+static inline int GetShaderIndexForMask(ShaderConvert shader, int mask)
+{
+	pxAssert(HasVariableWriteMask(shader));
+	int index = mask;
+	if (shader == ShaderConvert::RTA_CORRECTION)
+		index |= 1 << 4;
+	return index;
+}
+
 static inline bool HasDepthOutput(ShaderConvert shader)
 {
 	switch (shader)
@@ -549,6 +570,7 @@ struct alignas(16) GSHWDrawConfig
 		constexpr ColorMaskSelector(): key(0xF) {}
 		constexpr ColorMaskSelector(u8 c): key(0) { wrgba = c; }
 	};
+
 #pragma pack(pop)
 	struct alignas(16) VSConstantBuffer
 	{
@@ -922,6 +944,12 @@ protected:
 	/// Perform texture operations for ImGui
 	void UpdateImGuiTextures();
 
+protected:
+	// Entry point to the renderer-specific StretchRect code.
+	virtual void DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+		GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear) = 0;
+	void DoStretchRectWithAssertions(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear);
+
 public:
 	GSDevice();
 	virtual ~GSDevice();
@@ -1037,9 +1065,9 @@ public:
 	virtual std::unique_ptr<GSDownloadTexture> CreateDownloadTexture(u32 width, u32 height, GSTexture::Format format) = 0;
 
 	virtual void CopyRect(GSTexture* sTex, GSTexture* dTex, const GSVector4i& r, u32 destX, u32 destY) = 0;
-	virtual void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ShaderConvert shader = ShaderConvert::COPY, bool linear = true) = 0;
-	virtual void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red, bool green, bool blue, bool alpha, ShaderConvert shader = ShaderConvert::COPY) = 0;
 
+	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red, bool green, bool blue, bool alpha, ShaderConvert shader = ShaderConvert::COPY);
+	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ShaderConvert shader = ShaderConvert::COPY, bool linear = true);
 	void StretchRect(GSTexture* sTex, GSTexture* dTex, const GSVector4& dRect, ShaderConvert shader = ShaderConvert::COPY, bool linear = true);
 
 	/// Performs a screen blit for display. If dTex is null, it assumes you are writing to the system framebuffer/swap chain.

pcsx2/GS/Renderers/DX11/GSDevice11.cpp

@@ -1274,28 +1274,18 @@ void GSDevice11::CopyRect(GSTexture* sTex, GSTexture* dTex, const GSVector4i& r,
 	dTex->SetState(GSTexture::State::Dirty);
 }
 
-void GSDevice11::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ShaderConvert shader, bool linear)
+void GSDevice11::DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+	GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear)
 {
-	pxAssert(dTex->IsDepthStencil() == HasDepthOutput(shader));
-	pxAssert(linear ? SupportsBilinear(shader) : SupportsNearest(shader));
-	StretchRect(sTex, sRect, dTex, dRect, m_convert.ps[static_cast<int>(shader)].get(), nullptr,
-		m_convert.bs[ShaderConvertWriteMask(shader)].get(), linear);
+	DoStretchRect(sTex, sRect, dTex, dRect, m_convert.ps[static_cast<int>(shader)].get(), nullptr, m_convert.bs[cms.wrgba].get(), linear);
 }
 
-void GSDevice11::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ID3D11PixelShader* ps, ID3D11Buffer* ps_cb, bool linear)
+void GSDevice11::DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ID3D11PixelShader* ps, ID3D11Buffer* ps_cb, bool linear)
 {
-	StretchRect(sTex, sRect, dTex, dRect, ps, ps_cb, m_convert.bs[D3D11_COLOR_WRITE_ENABLE_ALL].get(), linear);
+	DoStretchRect(sTex, sRect, dTex, dRect, ps, ps_cb, m_convert.bs[D3D11_COLOR_WRITE_ENABLE_ALL].get(), linear);
 }
 
-void GSDevice11::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red, bool green, bool blue, bool alpha, ShaderConvert shader)
-{
-	const u8 index = static_cast<u8>(red) | (static_cast<u8>(green) << 1) | (static_cast<u8>(blue) << 2) |
-	                 (static_cast<u8>(alpha) << 3);
-	StretchRect(sTex, sRect, dTex, dRect, m_convert.ps[static_cast<int>(shader)].get(), nullptr,
-		m_convert.bs[index].get(), false);
-}
-
-void GSDevice11::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ID3D11PixelShader* ps, ID3D11Buffer* ps_cb, ID3D11BlendState* bs, bool linear)
+void GSDevice11::DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ID3D11PixelShader* ps, ID3D11Buffer* ps_cb, ID3D11BlendState* bs, bool linear)
 {
 	CommitClear(sTex);
 
@@ -1439,7 +1429,7 @@ void GSDevice11::UpdateCLUTTexture(GSTexture* sTex, float sScale, u32 offsetX, u
 
 	const GSVector4 dRect(0, 0, dSize, 1);
 	const ShaderConvert shader = (dSize == 16) ? ShaderConvert::CLUT_4 : ShaderConvert::CLUT_8;
-	StretchRect(sTex, GSVector4::zero(), dTex, dRect, m_convert.ps[static_cast<int>(shader)].get(), m_merge.cb.get(), nullptr, false);
+	DoStretchRect(sTex, GSVector4::zero(), dTex, dRect, m_convert.ps[static_cast<int>(shader)].get(), m_merge.cb.get(), nullptr, false);
 }
 
 void GSDevice11::ConvertToIndexedTexture(GSTexture* sTex, float sScale, u32 offsetX, u32 offsetY, u32 SBW, u32 SPSM, GSTexture* dTex, u32 DBW, u32 DPSM)
@@ -1457,7 +1447,7 @@ void GSDevice11::ConvertToIndexedTexture(GSTexture* sTex, float sScale, u32 offs
 
 	const GSVector4 dRect(0, 0, dTex->GetWidth(), dTex->GetHeight());
 	const ShaderConvert shader = ((SPSM & 0xE) == 0) ? ShaderConvert::RGBA_TO_8I : ShaderConvert::RGB5A1_TO_8I;
-	StretchRect(sTex, GSVector4::zero(), dTex, dRect, m_convert.ps[static_cast<int>(shader)].get(), m_merge.cb.get(), nullptr, false);
+	DoStretchRect(sTex, GSVector4::zero(), dTex, dRect, m_convert.ps[static_cast<int>(shader)].get(), m_merge.cb.get(), nullptr, false);
 }
 
 void GSDevice11::FilteredDownsampleTexture(GSTexture* sTex, GSTexture* dTex, u32 downsample_factor, const GSVector2i& clamp_min, const GSVector4& dRect)
@@ -1477,7 +1467,7 @@ void GSDevice11::FilteredDownsampleTexture(GSTexture* sTex, GSTexture* dTex, u32
 	m_ctx->UpdateSubresource(m_merge.cb.get(), 0, nullptr, &cb, 0, 0);
 
 	const ShaderConvert shader = ShaderConvert::DOWNSAMPLE_COPY;
-	StretchRect(sTex, GSVector4::zero(), dTex, dRect, m_convert.ps[static_cast<int>(shader)].get(), m_merge.cb.get(), nullptr, false);
+	DoStretchRect(sTex, GSVector4::zero(), dTex, dRect, m_convert.ps[static_cast<int>(shader)].get(), m_merge.cb.get(), nullptr, false);
 }
 
 void GSDevice11::DrawMultiStretchRects(const MultiStretchRect* rects, u32 num_rects, GSTexture* dTex, ShaderConvert shader)
@@ -1596,7 +1586,7 @@ void GSDevice11::DoMerge(GSTexture* sTex[3], GSVector4* sRect, GSTexture* dTex,
 	// Save 2nd output
 	if (feedback_write_2)
 	{
-		StretchRect(dTex, full_r, sTex[2], dRect[2], m_convert.ps[static_cast<int>(ShaderConvert::YUV)].get(),
+		DoStretchRect(dTex, full_r, sTex[2], dRect[2], m_convert.ps[static_cast<int>(ShaderConvert::YUV)].get(),
 			m_merge.cb.get(), nullptr, linear);
 	}
 
@@ -1607,12 +1597,12 @@ void GSDevice11::DoMerge(GSTexture* sTex[3], GSVector4* sRect, GSTexture* dTex,
 	if (sTex[0])
 	{
 		// 1st output is enabled. It must be blended
-		StretchRect(sTex[0], sRect[0], dTex, dRect[0], m_merge.ps[PMODE.MMOD].get(), m_merge.cb.get(), m_merge.bs.get(), linear);
+		DoStretchRect(sTex[0], sRect[0], dTex, dRect[0], m_merge.ps[PMODE.MMOD].get(), m_merge.cb.get(), m_merge.bs.get(), linear);
 	}
 
 	if (feedback_write_1)
 	{
-		StretchRect(dTex, full_r, sTex[2], dRect[2], m_convert.ps[static_cast<int>(ShaderConvert::YUV)].get(),
+		DoStretchRect(dTex, full_r, sTex[2], dRect[2], m_convert.ps[static_cast<int>(ShaderConvert::YUV)].get(),
 			m_merge.cb.get(), nullptr, linear);
 	}
 }
@@ -1621,7 +1611,7 @@ void GSDevice11::DoInterlace(GSTexture* sTex, const GSVector4& sRect, GSTexture*
 {
 	m_ctx->UpdateSubresource(m_interlace.cb.get(), 0, nullptr, &cb, 0, 0);
 
-	StretchRect(sTex, sRect, dTex, dRect, m_interlace.ps[static_cast<int>(shader)].get(), m_interlace.cb.get(), linear);
+	DoStretchRect(sTex, sRect, dTex, dRect, m_interlace.ps[static_cast<int>(shader)].get(), m_interlace.cb.get(), linear);
 }
 
 void GSDevice11::DoFXAA(GSTexture* sTex, GSTexture* dTex)
@@ -1647,7 +1637,7 @@ void GSDevice11::DoFXAA(GSTexture* sTex, GSTexture* dTex)
 			return;
 	}
 
-	StretchRect(sTex, sRect, dTex, dRect, m_fxaa_ps.get(), nullptr, true);
+	DoStretchRect(sTex, sRect, dTex, dRect, m_fxaa_ps.get(), nullptr, true);
 }
 
 void GSDevice11::DoShadeBoost(GSTexture* sTex, GSTexture* dTex, const float params[4])
@@ -1659,7 +1649,7 @@ void GSDevice11::DoShadeBoost(GSTexture* sTex, GSTexture* dTex, const float para
 
 	m_ctx->UpdateSubresource(m_shadeboost.cb.get(), 0, nullptr, params, 0, 0);
 
-	StretchRect(sTex, sRect, dTex, dRect, m_shadeboost.ps.get(), m_shadeboost.cb.get(), false);
+	DoStretchRect(sTex, sRect, dTex, dRect, m_shadeboost.ps.get(), m_shadeboost.cb.get(), false);
 }
 
 void GSDevice11::SetupVS(VSSelector sel, const GSHWDrawConfig::VSConstantBuffer* cb)
@@ -2635,6 +2625,8 @@ void GSDevice11::RenderHW(GSHWDrawConfig& config)
 		}
 	}
 
+	// Destination Alpha Setup
+	const bool multidraw_fb_copy = m_features.multidraw_fb_copy && (config.require_one_barrier || config.require_full_barrier);
 	GSTexture* primid_texture = nullptr;
 	if (config.destination_alpha == GSHWDrawConfig::DestinationAlphaMode::PrimIDTracking)
 	{
@@ -2645,11 +2637,11 @@ void GSDevice11::RenderHW(GSHWDrawConfig& config)
 			return;
 		}
 
-		StretchRect(colclip_rt ? colclip_rt : config.rt, GSVector4(config.drawarea) / GSVector4(rtsize).xyxy(),
+		DoStretchRect(colclip_rt ? colclip_rt : config.rt, GSVector4(config.drawarea) / GSVector4(rtsize).xyxy(),
 			primid_texture, GSVector4(config.drawarea), m_date.primid_init_ps[static_cast<u8>(config.datm)].get(), nullptr, false);
 	}
 	else if (config.destination_alpha == GSHWDrawConfig::DestinationAlphaMode::Stencil ||
-			 config.destination_alpha == GSHWDrawConfig::DestinationAlphaMode::StencilOne)
+			 (config.destination_alpha == GSHWDrawConfig::DestinationAlphaMode::StencilOne && !multidraw_fb_copy))
 		SetupDATE(colclip_rt ? colclip_rt : config.rt, config.ds, config.datm, config.drawarea);
 
 	if (config.vs.expand != GSHWDrawConfig::VSExpand::None)
@@ -2764,6 +2756,11 @@ void GSDevice11::RenderHW(GSHWDrawConfig& config)
 
 	OMSetRenderTargets(draw_rt, draw_ds, &config.scissor, read_only_dsv);
 	SetupOM(config.depth, OMBlendSelector(config.colormask, config.blend), config.blend.constant);
+
+	// Clear stencil as close as possible to the RT bind, to avoid framebuffer swaps.
+	if (config.destination_alpha == GSHWDrawConfig::DestinationAlphaMode::StencilOne && multidraw_fb_copy)
+		m_ctx->ClearDepthStencilView(*static_cast<GSTexture11*>(draw_ds), D3D11_CLEAR_STENCIL, 0.0f, 1);
+
 	SendHWDraw(config, draw_rt_clone, draw_rt, config.require_one_barrier, config.require_full_barrier, false);
 
 	if (config.blend_multi_pass.enable)

pcsx2/GS/Renderers/DX11/GSDevice11.h

@@ -253,6 +253,10 @@ private:
 	D3D11ShaderCache m_shader_cache;
 	std::string m_tfx_source;
 
+protected:
+	virtual void DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+		GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear) override;
+
 public:
 	GSDevice11();
 	~GSDevice11() override;
@@ -298,10 +302,8 @@ public:
 
 	void CopyRect(GSTexture* sTex, GSTexture* dTex, const GSVector4i& r, u32 destX, u32 destY) override;
 
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ShaderConvert shader = ShaderConvert::COPY, bool linear = true) override;
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ID3D11PixelShader* ps, ID3D11Buffer* ps_cb, bool linear = true);
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red, bool green, bool blue, bool alpha, ShaderConvert shader = ShaderConvert::COPY) override;
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ID3D11PixelShader* ps, ID3D11Buffer* ps_cb, ID3D11BlendState* bs, bool linear = true);
+	void DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ID3D11PixelShader* ps, ID3D11Buffer* ps_cb, bool linear);
+	void DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ID3D11PixelShader* ps, ID3D11Buffer* ps_cb, ID3D11BlendState* bs, bool linear);
 	void PresentRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, PresentShader shader, float shaderTime, bool linear) override;
 	void UpdateCLUTTexture(GSTexture* sTex, float sScale, u32 offsetX, u32 offsetY, GSTexture* dTex, u32 dOffset, u32 dSize) override;
 	void ConvertToIndexedTexture(GSTexture* sTex, float sScale, u32 offsetX, u32 offsetY, u32 SBW, u32 SPSM, GSTexture* dTex, u32 DBW, u32 DPSM) override;

pcsx2/GS/Renderers/DX12/GSDevice12.cpp

@@ -1425,30 +1425,17 @@ void GSDevice12::CopyRect(GSTexture* sTex, GSTexture* dTex, const GSVector4i& r,
 	dTex12->SetState(GSTexture::State::Dirty);
 }
 
-void GSDevice12::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
-	ShaderConvert shader /* = ShaderConvert::COPY */, bool linear /* = true */)
+void GSDevice12::DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+	GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear)
 {
-	pxAssert(HasDepthOutput(shader) == (dTex && dTex->GetType() == GSTexture::Type::DepthStencil));
-
-	GL_INS("StretchRect(%d) {%d,%d} %dx%d -> {%d,%d) %dx%d", shader, int(sRect.left), int(sRect.top),
-		int(sRect.right - sRect.left), int(sRect.bottom - sRect.top), int(dRect.left), int(dRect.top),
-		int(dRect.right - dRect.left), int(dRect.bottom - dRect.top));
-
+	const bool allow_discard = (cms.wrgba == 0xf);
+	const ID3D12PipelineState* state;
+	if (HasVariableWriteMask(shader))
+		state = m_color_copy[GetShaderIndexForMask(shader, cms.wrgba)].get();
+	else
+		state = dTex ? m_convert[static_cast<int>(shader)].get() : m_present[static_cast<int>(shader)].get();
 	DoStretchRect(static_cast<GSTexture12*>(sTex), sRect, static_cast<GSTexture12*>(dTex), dRect,
-		dTex ? m_convert[static_cast<int>(shader)].get() : m_present[static_cast<int>(shader)].get(), linear,
-		ShaderConvertWriteMask(shader) == 0xf);
-}
-
-void GSDevice12::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red,
-	bool green, bool blue, bool alpha, ShaderConvert shader)
-{
-	GL_PUSH("ColorCopy Red:%d Green:%d Blue:%d Alpha:%d", red, green, blue, alpha);
-
-	const u32 index = (red ? 1 : 0) | (green ? 2 : 0) | (blue ? 4 : 0) | (alpha ? 8 : 0);
-	int rta_offset = (shader == ShaderConvert::RTA_CORRECTION) ? 16 : 0;
-	const bool allow_discard = (index == 0xf);
-	DoStretchRect(static_cast<GSTexture12*>(sTex), sRect, static_cast<GSTexture12*>(dTex), dRect,
-		m_color_copy[index + rta_offset].get(), false, allow_discard);
+		state, linear, allow_discard);
 }
 
 void GSDevice12::PresentRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
@@ -1639,10 +1626,10 @@ void GSDevice12::DoMultiStretchRects(
 	SetPrimitiveTopology(D3D_PRIMITIVE_TOPOLOGY_TRIANGLESTRIP);
 	SetUtilityTexture(rects[0].src, rects[0].linear ? m_linear_sampler_cpu : m_point_sampler_cpu);
 
-	pxAssert(shader == ShaderConvert::COPY || shader == ShaderConvert::RTA_CORRECTION || rects[0].wmask.wrgba == 0xf);
-	int rta_bit = (shader == ShaderConvert::RTA_CORRECTION) ? 16 : 0;
-	SetPipeline((rects[0].wmask.wrgba != 0xf) ? m_color_copy[rects[0].wmask.wrgba | rta_bit].get() :
-												m_convert[static_cast<int>(shader)].get());
+	pxAssert(HasVariableWriteMask(shader) || rects[0].wmask.wrgba == 0xf);
+	SetPipeline((rects[0].wmask.wrgba != 0xf) ?
+		m_color_copy[GetShaderIndexForMask(shader, rects[0].wmask.wrgba)].get() :
+		m_convert[static_cast<int>(shader)].get());
 
 	if (ApplyUtilityState())
 		DrawIndexedPrimitive();
@@ -3831,8 +3818,12 @@ GSTexture12* GSDevice12::SetupPrimitiveTrackingDATE(GSHWDrawConfig& config, Pipe
 void GSDevice12::RenderHW(GSHWDrawConfig& config)
 {
 	// Destination Alpha Setup
-	const bool stencil_DATE = (config.destination_alpha == GSHWDrawConfig::DestinationAlphaMode::Stencil ||
-							   config.destination_alpha == GSHWDrawConfig::DestinationAlphaMode::StencilOne);
+	const bool stencil_DATE_One = config.destination_alpha == GSHWDrawConfig::DestinationAlphaMode::StencilOne;
+	const bool stencil_DATE = (config.destination_alpha == GSHWDrawConfig::DestinationAlphaMode::Stencil || stencil_DATE_One);
+
+	// TODO: Backport from vk.
+	if (stencil_DATE_One)
+		config.ps.date = 0;
 
 	GSTexture12* colclip_rt = static_cast<GSTexture12*>(g_gs_device->GetColorClipTexture());
 	GSTexture12* draw_rt = static_cast<GSTexture12*>(config.rt);

pcsx2/GS/Renderers/DX12/GSDevice12.h

@@ -390,6 +390,10 @@ private:
 
 	void DestroyResources();
 
+protected:
+	virtual void DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+		GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear) override;
+
 public:
 	GSDevice12();
 	~GSDevice12() override;
@@ -428,10 +432,6 @@ public:
 
 	void CopyRect(GSTexture* sTex, GSTexture* dTex, const GSVector4i& r, u32 destX, u32 destY) override;
 
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
-		ShaderConvert shader = ShaderConvert::COPY, bool linear = true) override;
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red,
-		bool green, bool blue, bool alpha, ShaderConvert shader = ShaderConvert::COPY) override;
 	void PresentRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
 		PresentShader shader, float shaderTime, bool linear) override;
 	void UpdateCLUTTexture(

pcsx2/GS/Renderers/HW/GSRendererHW.cpp

@@ -7939,7 +7939,8 @@ __ri void GSRendererHW::DrawPrims(GSTextureCache::Target* rt, GSTextureCache::Ta
 	}
 	else if (DATE_one)
 	{
-		if (features.texture_barrier)
+		const bool multidraw_fb_copy = features.multidraw_fb_copy && (m_conf.require_one_barrier || m_conf.require_full_barrier);
+		if (features.texture_barrier || multidraw_fb_copy)
 		{
 			m_conf.require_one_barrier = true;
 			m_conf.ps.date = 5 + m_cached_ctx.TEST.DATM;

pcsx2/GS/Renderers/HW/GSTextureCache.cpp

@@ -2663,8 +2663,10 @@ GSTextureCache::Target* GSTextureCache::LookupTarget(GIFRegTEX0 TEX0, const GSVe
 			for (auto i = list.begin(); i != list.end(); ++i)
 			{
 				Target* t = *i;
+				const bool half_buffer_match = GSConfig.UserHacks_TextureInsideRt >= GSTextureInRtMode::InsideTargets && TEX0.TBW == t->m_TEX0.TBW && TEX0.PSM == t->m_TEX0.PSM &&
+												bp == GSLocalMemory::GetStartBlockAddress(t->m_TEX0.TBP0, t->m_TEX0.TBW, t->m_TEX0.PSM, GSVector4i(0, size.y, size.x, size.y + 1));
 				// Make sure the target is inside the texture
-				if (t->m_TEX0.TBP0 <= bp && bp <= t->m_end_block && t->Inside(bp, TEX0.TBW, TEX0.PSM, GSVector4i::loadh(size)))
+				if (t->m_TEX0.TBP0 <= bp && bp <= t->m_end_block && (half_buffer_match || t->Inside(bp, TEX0.TBW, TEX0.PSM, GSVector4i::loadh(size))))
 				{
 					if (dst && (GSState::s_n - dst->m_last_draw) < (GSState::s_n - t->m_last_draw))
 						continue;

pcsx2/GS/Renderers/Metal/GSDeviceMTL.h

@@ -407,8 +407,6 @@ public:
 	void DrawStretchRect(const GSVector4& sRect, const GSVector4& dRect, const GSVector2& ds);
 	/// Copy from a position in sTex to the same position in the currently active render encoder using the given fs pipeline and rect
 	void RenderCopy(GSTexture* sTex, id<MTLRenderPipelineState> pipeline, const GSVector4i& rect);
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ShaderConvert shader = ShaderConvert::COPY, bool linear = true) override;
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red, bool green, bool blue, bool alpha, ShaderConvert shader = ShaderConvert::COPY) override;
 	void PresentRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, PresentShader shader, float shaderTime, bool linear) override;
 	void DrawMultiStretchRects(const MultiStretchRect* rects, u32 num_rects, GSTexture* dTex, ShaderConvert shader) override;
 	void UpdateCLUTTexture(GSTexture* sTex, float sScale, u32 offsetX, u32 offsetY, GSTexture* dTex, u32 dOffset, u32 dSize) override;
@@ -452,6 +450,10 @@ public:
 
 	void RenderImGui(ImDrawData* data);
 	u32 FrameNo() const { return m_frame; }
+
+protected:
+	virtual void DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+		GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear) override;
 };
 
 static constexpr bool IsCommandBufferCompleted(MTLCommandBufferStatus status)

pcsx2/GS/Renderers/Metal/GSDeviceMTL.mm

@@ -1600,33 +1600,21 @@ void GSDeviceMTL::RenderCopy(GSTexture* sTex, id<MTLRenderPipelineState> pipelin
 	g_perfmon.Put(GSPerfMon::DrawCalls, 1);
 }
 
-void GSDeviceMTL::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ShaderConvert shader, bool linear)
+void GSDeviceMTL::DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+	GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear)
 { @autoreleasepool {
 
-	pxAssert(linear ? SupportsBilinear(shader) : SupportsNearest(shader));
-
-	id<MTLRenderPipelineState> pipeline = m_convert_pipeline[static_cast<int>(shader)];
+	const LoadAction load_action = (cms.wrgba == 0xf) ? LoadAction::DontCareIfFull : LoadAction::Load;
+	id<MTLRenderPipelineState> pipeline;
+	if (HasVariableWriteMask(shader))
+		pipeline = m_convert_pipeline_copy_mask[GetShaderIndexForMask(shader, cms.wrgba)];
+	else
+		pipeline = m_convert_pipeline[static_cast<int>(shader)];
 	pxAssertRel(pipeline, fmt::format("No pipeline for {}", shaderName(shader)).c_str());
 
-	const LoadAction load_action = (ShaderConvertWriteMask(shader) == 0xf) ? LoadAction::DontCareIfFull : LoadAction::Load;
 	DoStretchRect(sTex, sRect, dTex, dRect, pipeline, linear, load_action, nullptr, 0);
 }}
 
-void GSDeviceMTL::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red, bool green, bool blue, bool alpha, ShaderConvert shader)
-{ @autoreleasepool {
-	int sel = 0;
-	if (red)   sel |= 1;
-	if (green) sel |= 2;
-	if (blue)  sel |= 4;
-	if (alpha) sel |= 8;
-	if (shader == ShaderConvert::RTA_CORRECTION) sel |= 16;
-	const int color_sel = sel & 15;
-
-	id<MTLRenderPipelineState> pipeline = m_convert_pipeline_copy_mask[sel];
-
-	DoStretchRect(sTex, sRect, dTex, dRect, pipeline, false, color_sel == 15 ? LoadAction::DontCareIfFull : LoadAction::Load, nullptr, 0);
-}}
-
 static_assert(sizeof(DisplayConstantBuffer) == sizeof(GSMTLPresentPSUniform));
 static_assert(offsetof(DisplayConstantBuffer, SourceRect)          == offsetof(GSMTLPresentPSUniform, source_rect));
 static_assert(offsetof(DisplayConstantBuffer, TargetRect)          == offsetof(GSMTLPresentPSUniform, target_rect));
@@ -1683,9 +1671,9 @@ void GSDeviceMTL::DrawMultiStretchRects(const MultiStretchRect* rects, u32 num_r
 		const u32 end = i * 4;
 		const u32 vertex_count = end - start;
 		const u32 index_count = vertex_count + (vertex_count >> 1); // 6 indices per 4 vertices
-		const int rta_bit = shader == ShaderConvert::RTA_CORRECTION ? 16 : 0;
+		pxAssert(HasVariableWriteMask(shader) || wmask == 0xf);
 		id<MTLRenderPipelineState> new_pipeline = wmask == 0xf ? m_convert_pipeline[static_cast<int>(shader)]
-		                                                       : m_convert_pipeline_copy_mask[wmask | rta_bit];
+		                                                       : m_convert_pipeline_copy_mask[GetShaderIndexForMask(shader, wmask)];
 		if (new_pipeline != pipeline)
 		{
 			pipeline = new_pipeline;

pcsx2/GS/Renderers/OpenGL/GSDeviceOGL.cpp

@@ -1253,7 +1253,7 @@ GSTexture* GSDeviceOGL::InitPrimDateTexture(GSTexture* rt, const GSVector4i& are
 		return nullptr;
 
 	GL_PUSH("PrimID Destination Alpha Clear");
-	StretchRect(rt, GSVector4(area) / GSVector4(rtsize).xyxy(), tex, GSVector4(area), m_date.primid_ps[static_cast<u8>(datm)], false);
+	DoStretchRect(rt, GSVector4(area) / GSVector4(rtsize).xyxy(), tex, GSVector4(area), m_date.primid_ps[static_cast<u8>(datm)], false);
 	return tex;
 }
 
@@ -1472,31 +1472,20 @@ void GSDeviceOGL::CopyRect(GSTexture* sTex, GSTexture* dTex, const GSVector4i& r
 	dTex->SetState(GSTexture::State::Dirty);
 }
 
-void GSDeviceOGL::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ShaderConvert shader, bool linear)
+void GSDeviceOGL::DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+	OMColorMaskSelector cms, ShaderConvert shader, bool linear)
 {
-	pxAssert(dTex->IsDepthStencil() == HasDepthOutput(shader));
-	pxAssert(linear ? SupportsBilinear(shader) : SupportsNearest(shader));
-	StretchRect(sTex, sRect, dTex, dRect, m_convert.ps[(int)shader], false, OMColorMaskSelector(ShaderConvertWriteMask(shader)), linear);
+	DoStretchRect(sTex, sRect, dTex, dRect, m_convert.ps[static_cast<int>(shader)], false, cms, linear);
 }
 
-void GSDeviceOGL::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, const GLProgram& ps, bool linear)
+void GSDeviceOGL::DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+	const GLProgram& ps, bool linear)
 {
-	StretchRect(sTex, sRect, dTex, dRect, ps, false, OMColorMaskSelector(), linear);
+	DoStretchRect(sTex, sRect, dTex, dRect, ps, false, OMColorMaskSelector(), linear);
 }
 
-void GSDeviceOGL::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red, bool green, bool blue, bool alpha, ShaderConvert shader)
-{
-	OMColorMaskSelector cms;
-
-	cms.wr = red;
-	cms.wg = green;
-	cms.wb = blue;
-	cms.wa = alpha;
-
-	StretchRect(sTex, sRect, dTex, dRect, m_convert.ps[(int)shader], false, cms, false);
-}
-
-void GSDeviceOGL::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, const GLProgram& ps, bool alpha_blend, OMColorMaskSelector cms, bool linear)
+void GSDeviceOGL::DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+	const GLProgram& ps, bool alpha_blend, OMColorMaskSelector cms, bool linear)
 {
 	CommitClear(sTex, true);
 
@@ -1816,12 +1805,12 @@ void GSDeviceOGL::DoMerge(GSTexture* sTex[3], GSVector4* sRect, GSTexture* dTex,
 			// Blend with a constant alpha
 			m_merge_obj.ps[1].Bind();
 			m_merge_obj.ps[1].Uniform4fv(0, GSVector4::unorm8(c).v);
-			StretchRect(sTex[0], sRect[0], dTex, dRect[0], m_merge_obj.ps[1], true, OMColorMaskSelector(), linear);
+			DoStretchRect(sTex[0], sRect[0], dTex, dRect[0], m_merge_obj.ps[1], true, OMColorMaskSelector(), linear);
 		}
 		else
 		{
 			// Blend with 2 * input alpha
-			StretchRect(sTex[0], sRect[0], dTex, dRect[0], m_merge_obj.ps[0], true, OMColorMaskSelector(), linear);
+			DoStretchRect(sTex[0], sRect[0], dTex, dRect[0], m_merge_obj.ps[0], true, OMColorMaskSelector(), linear);
 		}
 	}
 
@@ -1836,7 +1825,7 @@ void GSDeviceOGL::DoInterlace(GSTexture* sTex, const GSVector4& sRect, GSTexture
 	m_interlace.ps[static_cast<int>(shader)].Bind();
 	m_interlace.ps[static_cast<int>(shader)].Uniform4fv(0, cb.ZrH.F32);
 
-	StretchRect(sTex, sRect, dTex, dRect, m_interlace.ps[static_cast<int>(shader)], linear);
+	DoStretchRect(sTex, sRect, dTex, dRect, m_interlace.ps[static_cast<int>(shader)], linear);
 }
 
 bool GSDeviceOGL::CompileFXAAProgram()
@@ -1875,7 +1864,7 @@ void GSDeviceOGL::DoFXAA(GSTexture* sTex, GSTexture* dTex)
 	const GSVector4 sRect(0, 0, 1, 1);
 	const GSVector4 dRect(0, 0, s.x, s.y);
 
-	StretchRect(sTex, sRect, dTex, dRect, m_fxaa.ps, true);
+	DoStretchRect(sTex, sRect, dTex, dRect, m_fxaa.ps, true);
 }
 
 bool GSDeviceOGL::CompileShadeBoostProgram()
@@ -1909,7 +1898,7 @@ void GSDeviceOGL::DoShadeBoost(GSTexture* sTex, GSTexture* dTex, const float par
 	const GSVector4 sRect(0, 0, 1, 1);
 	const GSVector4 dRect(0, 0, s.x, s.y);
 
-	StretchRect(sTex, sRect, dTex, dRect, m_shadeboost.ps, false);
+	DoStretchRect(sTex, sRect, dTex, dRect, m_shadeboost.ps, false);
 }
 
 void GSDeviceOGL::SetupDATE(GSTexture* rt, GSTexture* ds, SetDATM datm, const GSVector4i& bbox)

pcsx2/GS/Renderers/OpenGL/GSDeviceOGL.h

@@ -265,6 +265,10 @@ private:
 
 	void DrawStretchRect(const GSVector4& sRect, const GSVector4& dRect, const GSVector2i& ds);
 
+protected:
+	virtual void DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+		GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear) override;
+
 public:
 	GSDeviceOGL();
 	virtual ~GSDeviceOGL();
@@ -317,10 +321,8 @@ public:
 	// BlitRect *does* mess with GL state, be sure to re-bind.
 	void BlitRect(GSTexture* sTex, const GSVector4i& r, const GSVector2i& dsize, bool at_origin, bool linear);
 
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, ShaderConvert shader = ShaderConvert::COPY, bool linear = true) override;
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, const GLProgram& ps, bool linear = true);
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red, bool green, bool blue, bool alpha, ShaderConvert shader = ShaderConvert::COPY) override;
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, const GLProgram& ps, bool alpha_blend, OMColorMaskSelector cms, bool linear = true);
+	void DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, const GLProgram& ps, bool linear);
+	void DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, const GLProgram& ps, bool alpha_blend, OMColorMaskSelector cms, bool linear);
 	void PresentRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, PresentShader shader, float shaderTime, bool linear) override;
 	void UpdateCLUTTexture(GSTexture* sTex, float sScale, u32 offsetX, u32 offsetY, GSTexture* dTex, u32 dOffset, u32 dSize) override;
 	void ConvertToIndexedTexture(GSTexture* sTex, float sScale, u32 offsetX, u32 offsetY, u32 SBW, u32 SPSM, GSTexture* dTex, u32 DBW, u32 DPSM) override;

pcsx2/GS/Renderers/SW/GSRasterizer.cpp

@@ -1549,7 +1549,7 @@ void GSRasterizerList::Queue(const GSRingHeap::SharedPtr<GSRasterizerData>& data
 	pxAssert(r.top >= 0 && r.top < 2048 && r.bottom >= 0 && r.bottom < 2048);
 
 	int top = r.top >> m_thread_height;
-	int bottom = std::min<int>((r.bottom + (1 << m_thread_height) - 1) >> m_thread_height, top + m_workers.size());
+	int bottom = std::min<int>((r.bottom + (1 << m_thread_height) - 1) >> m_thread_height, top + (int)m_workers.size());
 
 	while (top < bottom)
 	{

pcsx2/GS/Renderers/SW/GSRendererSW.cpp

@@ -1652,7 +1652,7 @@ void GSRendererSW::SharedData::UpdateSource()
 
 		std::string s;
 
-		for (size_t i = 0; m_tex[i].t; i++)
+		for (u32 i = 0; m_tex[i].t; i++)
 		{
 			const GIFRegTEX0& TEX0 = g_gs_renderer->GetTex0Layer(i);
 

pcsx2/GS/Renderers/Vulkan/GSDeviceVK.cpp

@@ -2821,31 +2821,16 @@ void GSDeviceVK::CopyRect(GSTexture* sTex, GSTexture* dTex, const GSVector4i& r,
 	dTexVK->SetState(GSTexture::State::Dirty);
 }
 
-void GSDeviceVK::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
-	ShaderConvert shader /* = ShaderConvert::COPY */, bool linear /* = true */)
+void GSDeviceVK::DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+	GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear)
 {
-	pxAssert(HasDepthOutput(shader) == (dTex && dTex->GetType() == GSTexture::Type::DepthStencil));
-	pxAssert(linear ? SupportsBilinear(shader) : SupportsNearest(shader));
-
-	GL_INS("StretchRect(%d) {%d,%d} %dx%d -> {%d,%d) %dx%d", shader, int(sRect.left), int(sRect.top),
-		int(sRect.right - sRect.left), int(sRect.bottom - sRect.top), int(dRect.left), int(dRect.top),
-		int(dRect.right - dRect.left), int(dRect.bottom - dRect.top));
-
-	DoStretchRect(static_cast<GSTextureVK*>(sTex), sRect, static_cast<GSTextureVK*>(dTex), dRect,
-		dTex ? m_convert[static_cast<int>(shader)] : m_present[static_cast<int>(shader)], linear,
-		ShaderConvertWriteMask(shader) == 0xf);
-}
-
-void GSDeviceVK::StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red,
-	bool green, bool blue, bool alpha, ShaderConvert shader)
-{
-	GL_PUSH("ColorCopy Red:%d Green:%d Blue:%d Alpha:%d", red, green, blue, alpha);
-
-	const u32 index = (red ? 1 : 0) | (green ? 2 : 0) | (blue ? 4 : 0) | (alpha ? 8 : 0);
-	const bool allow_discard = (index == 0xf);
-	int rta_offset = (shader == ShaderConvert::RTA_CORRECTION) ? 16 : 0;
-	DoStretchRect(static_cast<GSTextureVK*>(sTex), sRect, static_cast<GSTextureVK*>(dTex), dRect, m_color_copy[index + rta_offset],
-		false, allow_discard);
+	const bool allow_discard = (cms.wrgba == 0xf);
+	VkPipeline state;
+	if (HasVariableWriteMask(shader))
+		state = m_color_copy[GetShaderIndexForMask(shader, cms.wrgba)];
+	else
+		state = dTex ? m_convert[static_cast<int>(shader)] : m_present[static_cast<int>(shader)];
+	DoStretchRect(static_cast<GSTextureVK*>(sTex), sRect, static_cast<GSTextureVK*>(dTex), dRect, state, linear, allow_discard);
 }
 
 void GSDeviceVK::PresentRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
@@ -2967,10 +2952,10 @@ void GSDeviceVK::DoMultiStretchRects(
 		BeginRenderPassForStretchRect(dTex, rc, rc, false);
 	SetUtilityTexture(rects[0].src, rects[0].linear ? m_linear_sampler : m_point_sampler);
 
-	pxAssert(shader == ShaderConvert::COPY || shader == ShaderConvert::RTA_CORRECTION || rects[0].wmask.wrgba == 0xf);
-	int rta_bit = (shader == ShaderConvert::RTA_CORRECTION) ? 16 : 0;
-	SetPipeline(
-		(rects[0].wmask.wrgba != 0xf) ? m_color_copy[rects[0].wmask.wrgba | rta_bit] : m_convert[static_cast<int>(shader)]);
+	pxAssert(HasVariableWriteMask(shader) || rects[0].wmask.wrgba == 0xf);
+	SetPipeline((rects[0].wmask.wrgba != 0xf) ?
+		m_color_copy[GetShaderIndexForMask(shader, rects[0].wmask.wrgba)] :
+		m_convert[static_cast<int>(shader)]);
 
 	if (ApplyUtilityState())
 		DrawIndexedPrimitive();

pcsx2/GS/Renderers/Vulkan/GSDeviceVK.h

@@ -470,6 +470,10 @@ private:
 
 	void DestroyResources();
 
+protected:
+	virtual void DoStretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
+		GSHWDrawConfig::ColorMaskSelector cms, ShaderConvert shader, bool linear) override;
+
 public:
 	GSDeviceVK();
 	~GSDeviceVK() override;
@@ -525,10 +529,6 @@ public:
 
 	void CopyRect(GSTexture* sTex, GSTexture* dTex, const GSVector4i& r, u32 destX, u32 destY) override;
 
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
-		ShaderConvert shader = ShaderConvert::COPY, bool linear = true) override;
-	void StretchRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect, bool red,
-		bool green, bool blue, bool alpha, ShaderConvert shader = ShaderConvert::COPY) override;
 	void PresentRect(GSTexture* sTex, const GSVector4& sRect, GSTexture* dTex, const GSVector4& dRect,
 		PresentShader shader, float shaderTime, bool linear) override;
 	void DrawMultiStretchRects(

pcsx2/Gif_Unit.h

@@ -521,7 +521,7 @@ struct Gif_Path
 	// GS Packets that MTGS hasn't yet processed
 	u32 GetPendingGSPackets()
 	{
-		return mtvu.gsPackQueue.size();
+		return (u32)mtvu.gsPackQueue.size();
 	}
 };
 

pcsx2/ImGui/FullscreenUI.cpp

@@ -6866,6 +6866,46 @@ void FullscreenUI::DrawPauseMenu(MainWindowType type)
 		BeginMenuButtons(submenu_item_count[static_cast<u32>(s_current_pause_submenu)], 1.0f, ImGuiFullscreen::LAYOUT_MENU_BUTTON_X_PADDING,
 			ImGuiFullscreen::LAYOUT_MENU_BUTTON_Y_PADDING, ImGuiFullscreen::LAYOUT_MENU_BUTTON_HEIGHT_NO_SUMMARY);
 
+		if (!ImGui::IsPopupOpen(0u, ImGuiPopupFlags_AnyPopup))
+		{
+			const bool up_pressed = ImGui::IsKeyPressed(ImGuiKey_GamepadDpadUp, ImGuiInputFlags_Repeat, ImGuiKeyOwner_NoOwner) ||
+									ImGui::IsKeyPressed(ImGuiKey_UpArrow, ImGuiInputFlags_Repeat, ImGuiKeyOwner_NoOwner);
+			const bool down_pressed = ImGui::IsKeyPressed(ImGuiKey_GamepadDpadDown, ImGuiInputFlags_Repeat, ImGuiKeyOwner_NoOwner) ||
+									  ImGui::IsKeyPressed(ImGuiKey_DownArrow, ImGuiInputFlags_Repeat, ImGuiKeyOwner_NoOwner);
+
+			if (up_pressed || down_pressed)
+			{
+				const ImGuiID current_focus_id = ImGui::GetFocusID();
+				ImGuiWindow* window = ImGui::GetCurrentWindow();
+				ImGuiID first_id = 0;
+				ImGuiID last_id = 0;
+
+				switch (s_current_pause_submenu)
+				{
+					case PauseSubMenu::None:
+						first_id = ImGui::GetID(FSUI_ICONSTR(ICON_FA_PLAY, "Resume Game"));
+						last_id = ImGui::GetID(FSUI_ICONSTR(ICON_FA_POWER_OFF, "Close Game"));
+						break;
+					case PauseSubMenu::Exit:
+						first_id = ImGui::GetID(FSUI_ICONSTR(ICON_PF_BACKWARD, "Back To Pause Menu"));
+						last_id = ImGui::GetID(FSUI_ICONSTR(ICON_FA_POWER_OFF, "Exit Without Saving"));
+						break;
+					case PauseSubMenu::Achievements:
+						first_id = ImGui::GetID(FSUI_ICONSTR(ICON_PF_BACKWARD, "Back To Pause Menu"));
+						last_id = ImGui::GetID(FSUI_ICONSTR(ICON_FA_STOPWATCH, "Leaderboards"));
+						break;
+				}
+
+				if (first_id != 0 && last_id != 0)
+				{
+					if (up_pressed && current_focus_id == first_id)
+						ImGui::SetFocusID(last_id, window);
+					else if (down_pressed && current_focus_id == last_id)
+						ImGui::SetFocusID(first_id, window);
+				}
+			}
+		}
+
 		switch (s_current_pause_submenu)
 		{
 			case PauseSubMenu::None:
@@ -10113,6 +10153,10 @@ TRANSLATE_NOOP("FullscreenUI", "Compression Method");
 TRANSLATE_NOOP("FullscreenUI", "Compression Level");
 TRANSLATE_NOOP("FullscreenUI", "Use Debug Device");
 TRANSLATE_NOOP("FullscreenUI", "Resume Game");
+TRANSLATE_NOOP("FullscreenUI", "Close Game");
+TRANSLATE_NOOP("FullscreenUI", "Back To Pause Menu");
+TRANSLATE_NOOP("FullscreenUI", "Exit Without Saving");
+TRANSLATE_NOOP("FullscreenUI", "Leaderboards");
 TRANSLATE_NOOP("FullscreenUI", "Toggle Frame Limit");
 TRANSLATE_NOOP("FullscreenUI", "Game Properties");
 TRANSLATE_NOOP("FullscreenUI", "Achievements");
@@ -10120,11 +10164,7 @@ TRANSLATE_NOOP("FullscreenUI", "Save Screenshot");
 TRANSLATE_NOOP("FullscreenUI", "Switch To Software Renderer");
 TRANSLATE_NOOP("FullscreenUI", "Switch To Hardware Renderer");
 TRANSLATE_NOOP("FullscreenUI", "Change Disc");
-TRANSLATE_NOOP("FullscreenUI", "Close Game");
-TRANSLATE_NOOP("FullscreenUI", "Exit Without Saving");
-TRANSLATE_NOOP("FullscreenUI", "Back To Pause Menu");
 TRANSLATE_NOOP("FullscreenUI", "Exit And Save State");
-TRANSLATE_NOOP("FullscreenUI", "Leaderboards");
 TRANSLATE_NOOP("FullscreenUI", "Delete Save");
 TRANSLATE_NOOP("FullscreenUI", "Close Menu");
 TRANSLATE_NOOP("FullscreenUI", "Default Boot");

pcsx2/ImGui/ImGuiOverlays.cpp

@@ -622,7 +622,7 @@ __ri void ImGuiManager::DrawSettingsOverlay(float scale, float margin, float spa
 		else
 			APPEND("IR={} ", static_cast<unsigned>(GSConfig.UpscaleMultiplier));
 
-		APPEND("B={} PL={} ", static_cast<unsigned>(GSConfig.AccurateBlendingUnit), static_cast<unsigned>(GSConfig.TexturePreloading));
+		APPEND("BL={} TPL={} ", static_cast<unsigned>(GSConfig.AccurateBlendingUnit), static_cast<unsigned>(GSConfig.TexturePreloading));
 		if (GSConfig.GPUPaletteConversion)
 			APPEND("PLTX ");
 

pcsx2/IopBios.cpp

@@ -44,39 +44,39 @@
 
 typedef struct
 {
-	unsigned int mode;
-	unsigned int attr;
-	unsigned int size;
-	unsigned char ctime[8];
-	unsigned char atime[8];
-	unsigned char mtime[8];
-	unsigned int hisize;
+	u32 mode;
+	u32 attr;
+	u32 size;
+	u8 ctime[8];
+	u8 atime[8];
+	u8 mtime[8];
+	u32 hisize;
 } fio_stat_t;
 typedef struct
 {
 	fio_stat_t _fioStat;
 	/** Number of subs (main) / subpart number (sub) */
-	unsigned int private_0;
-	unsigned int private_1;
-	unsigned int private_2;
-	unsigned int private_3;
-	unsigned int private_4;
+	u32 private_0;
+	u32 private_1;
+	u32 private_2;
+	u32 private_3;
+	u32 private_4;
 	/** Sector start.  */
-	unsigned int private_5;
+	u32 private_5;
 } fxio_stat_t;
 
 typedef struct
 {
 	fio_stat_t stat;
 	char name[256];
-	unsigned int unknown;
+	u32 unknown;
 } fio_dirent_t;
 
 typedef struct
 {
 	fxio_stat_t stat;
 	char name[256];
-	unsigned int unknown;
+	u32 unknown;
 } fxio_dirent_t;
 
 static std::string hostRoot;
@@ -171,7 +171,7 @@ namespace R3000A
 		if (!FileSystem::StatFile(file_path.c_str(), &file_stats))
 			return -IOP_ENOENT;
 
-		host_stats->size = file_stats.st_size;
+		host_stats->size = (u32)file_stats.st_size;
 		host_stats->hisize = 0;
 
 		// Convert the mode.
@@ -315,13 +315,13 @@ namespace R3000A
 			switch (whence)
 			{
 				case IOP_SEEK_SET:
-					err = ::lseek(fd, offset, SEEK_SET);
+					err = static_cast<int>(::lseek(fd, offset, SEEK_SET));
 					break;
 				case IOP_SEEK_CUR:
-					err = ::lseek(fd, offset, SEEK_CUR);
+					err = static_cast<int>(::lseek(fd, offset, SEEK_CUR));
 					break;
 				case IOP_SEEK_END:
-					err = ::lseek(fd, offset, SEEK_END);
+					err = static_cast<int>(::lseek(fd, offset, SEEK_END));
 					break;
 				default:
 					return -IOP_EIO;
@@ -332,12 +332,12 @@ namespace R3000A
 
 		virtual int read(void* buf, u32 count) /* Flawfinder: ignore */
 		{
-			return translate_error(::read(fd, buf, count));
+			return translate_error(static_cast<int>(::read(fd, buf, count)));
 		}
 
 		virtual int write(void* buf, u32 count)
 		{
-			return translate_error(::write(fd, buf, count));
+			return translate_error(static_cast<int>(::write(fd, buf, count)));
 		}
 	};
 
@@ -762,7 +762,7 @@ namespace R3000A
 					v0 = host_stat(full_path, (fxio_stat_t*)&buf);
 
 					for (size_t i = 0; i < sizeof(fxio_stat_t); i++)
-						iopMemWrite8(data + i, buf[i]);
+						iopMemWrite8(static_cast<u32>(data + i), buf[i]);
 				}
 				else
 				{
@@ -770,7 +770,7 @@ namespace R3000A
 					v0 = host_stat(full_path, (fio_stat_t*)&buf);
 
 					for (size_t i = 0; i < sizeof(fio_stat_t); i++)
-						iopMemWrite8(data + i, buf[i]);
+						iopMemWrite8(static_cast<u32>(data + i), buf[i]);
 				}
 				pc = ra;
 				return 1;

pcsx2/MTVU.cpp

@@ -186,7 +186,7 @@ void VU_Thread::ExecuteRingBuffer()
 					break;
 				case MTVU_VIF_UNPACK:
 				{
-					u32 vif_copy_size = (uptr)&vif.StructEnd - (uptr)&vif.tag;
+					u32 vif_copy_size = static_cast<u32>((uptr)&vif.StructEnd - (uptr)&vif.tag);
 					Read(&vif.tag, vif_copy_size);
 					ReadRegs(&vifRegs);
 					u32 size = Read();
@@ -467,7 +467,7 @@ void VU_Thread::ExecuteVU(u32 vu_addr, u32 vif_top, u32 vif_itop, u32 fbrst)
 void VU_Thread::VifUnpack(vifStruct& _vif, VIFregisters& _vifRegs, const u8* data, u32 size)
 {
 	MTVU_LOG("MTVU - VifUnpack!");
-	u32 vif_copy_size = (uptr)&_vif.StructEnd - (uptr)&_vif.tag;
+	u32 vif_copy_size = (u32)((uptr)&_vif.StructEnd - (uptr)&_vif.tag);
 	ReserveSpace(1 + size_u32(vif_copy_size) + size_u32(sizeof(VIFregistersMTVU)) + 1 + size_u32(size));
 	Write(MTVU_VIF_UNPACK);
 	Write(&_vif.tag, vif_copy_size);

pcsx2/Pcsx2Config.cpp

@@ -1844,6 +1844,7 @@ bool Pcsx2Config::PadOptions::Port::operator!=(const PadOptions::Port& right) co
 
 Pcsx2Config::AchievementsOptions::AchievementsOptions()
 {
+	bitset = 0;
 	Enabled = false;
 	HardcoreMode = false;
 	EncoreMode = false;

pcsx2/Recording/InputRecordingFile.cpp

@@ -66,12 +66,12 @@ const std::string& InputRecordingFile::getFilename() const noexcept
 	return m_filename;
 }
 
-unsigned long InputRecordingFile::getTotalFrames() const noexcept
+u32 InputRecordingFile::getTotalFrames() const noexcept
 {
 	return m_totalFrames;
 }
 
-unsigned long InputRecordingFile::getUndoCount() const noexcept
+u32 InputRecordingFile::getUndoCount() const noexcept
 {
 	return m_undoCount;
 }
@@ -237,7 +237,7 @@ std::vector<PadData> InputRecordingFile::bulkReadPadData(u32 frameStart, u32 fra
 	}
 
 	// TODO - no multi-tap support
-	for (uint64_t currFrame = frameStart; currFrame < frameEnd; currFrame++)
+	for (u32 currFrame = frameStart; currFrame < frameEnd; currFrame++)
 	{
 		const auto padData = readPadData(currFrame, port, 0);
 		if (padData)

pcsx2/Recording/InputRecordingFile.h

@@ -68,8 +68,8 @@ public:
 
 	// Retrieve the input recording's filename (not the path)
 	const std::string& getFilename() const noexcept;
-	unsigned long getTotalFrames() const noexcept;
-	unsigned long getUndoCount() const noexcept;
+	u32 getTotalFrames() const noexcept;
+	u32 getUndoCount() const noexcept;
 
 	void logRecordingMetadata();
 	std::vector<PadData> bulkReadPadData(u32 frameStart, u32 frameEnd, const uint port);
@@ -92,8 +92,8 @@ private:
 	bool m_savestate = false;
 
 	// An signed 32-bit frame limit is equivalent to 1.13 years of continuous 60fps footage
-	unsigned long m_totalFrames = 0;
-	unsigned long m_undoCount = 0;
+	u32 m_totalFrames = 0;
+	u32 m_undoCount = 0;
 
 	// Calculates the position of the current frame in the input recording
 	size_t getRecordingBlockSeekPoint(const u32 frame) const noexcept;

pcsx2/ps2/BiosTools.cpp

@@ -177,7 +177,7 @@ static bool LoadBiosVersion(std::FILE* fp, u32& version, std::string& descriptio
 																  "",
 			serial.c_str());
 
-		version = strtol(vermaj, (char**)NULL, 0) << 8;
+		version = static_cast<u32>(strtol(vermaj, (char**)NULL, 0) << 8);
 		version |= strtol(vermin, (char**)NULL, 0);
 
 		Console.WriteLn("BIOS Found: %s", description.c_str());

pcsx2/vtlb.h

@@ -174,7 +174,7 @@ namespace vtlb_private
 		/// Assumes the entry is a handler, and gets the raw handler ID
 		u8 assumeHandlerGetID() const { return value; }
 		/// Assumes the entry is a handler, and gets the physical address
-		u32 assumeHandlerGetPAddr(u32 vaddr) const { return (value + vaddr - assumeHandlerGetID()) & ~POINTER_SIGN_BIT; }
+		u32 assumeHandlerGetPAddr(u32 vaddr) const { return static_cast<u32>((value + vaddr - assumeHandlerGetID()) & ~POINTER_SIGN_BIT); }
 		/// Assumes the entry is a handler, returning it as a void*
 		void *assumeHandlerGetRaw(int index, bool write) const;
 		/// Assumes the entry is a handler, returning it