diff --git a/Utilities/SSemaphore.cpp b/Utilities/SSemaphore.cpp
deleted file mode 100644
index fd1aad9fbb..0000000000
--- a/Utilities/SSemaphore.cpp
+++ /dev/null
@@ -1,85 +0,0 @@
-#include "stdafx.h"
-#include "Utilities/SSemaphore.h"
-#include "Emu/System.h"
-
-void SSemaphore::wait()
-{
-	u32 order;
-	{
-		std::lock_guard<std::mutex> lock(m_mutex);
-		if (m_count && m_out_order == m_in_order)
-		{
-			m_count--;
-			return;
-		}
-		order = m_in_order++;
-	}
-
-	std::unique_lock<std::mutex> cv_lock(m_cv_mutex);
-	
-	while (true)
-	{
-		CHECK_EMU_STATUS;
-
-		m_cond.wait_for(cv_lock, std::chrono::milliseconds(1));
-		
-		{
-			std::lock_guard<std::mutex> lock(m_mutex);
-			if (m_count)
-			{
-				if (m_out_order == order)
-				{
-					m_count--;
-					m_out_order++;
-					return;
-				}
-				else
-				{
-					m_cond.notify_one();
-				}
-			}
-		}
-	}
-}
-
-bool SSemaphore::try_wait()
-{
-	std::lock_guard<std::mutex> lock(m_mutex);
-
-	if (m_count && m_in_order == m_out_order)
-	{
-		m_count--;
-		return true;
-	}
-	else
-	{
-		return false;
-	}
-}
-
-void SSemaphore::post()
-{
-	std::lock_guard<std::mutex> lock(m_mutex);
-
-	if (m_count < m_max)
-	{
-		m_count++;
-	}
-	else
-	{
-		return;
-	}
-
-	m_cond.notify_one();
-}
-
-bool SSemaphore::post_and_wait()
-{
-	// TODO: merge these functions? Probably has a race condition.
-	if (try_wait()) return false;
-
-	post();
-	wait();
-
-	return true;
-}
diff --git a/Utilities/SSemaphore.h b/Utilities/SSemaphore.h
deleted file mode 100644
index 2e61262959..0000000000
--- a/Utilities/SSemaphore.h
+++ /dev/null
@@ -1,41 +0,0 @@
-#pragma once
-
-class SSemaphore
-{
-	const u32 m_max;
-	u32 m_count;
-	u32 m_in_order;
-	u32 m_out_order;
-	std::mutex m_cv_mutex;
-	std::mutex m_mutex;
-	std::condition_variable m_cond;
-
-public:
-	SSemaphore(u32 value, u32 max = 1)
-		: m_max(max > 0 ? max : 0xffffffff)
-		, m_count(value > m_max ? m_max : value)
-		, m_in_order(0)
-		, m_out_order(0)
-	{
-	}
-
-	SSemaphore()
-		: m_max(0xffffffff)
-		, m_count(0)
-		, m_in_order(0)
-		, m_out_order(0)
-	{
-	}
-
-	~SSemaphore()
-	{
-	}
-
-	void wait();
-
-	bool try_wait();
-
-	void post();
-
-	bool post_and_wait();
-};
\ No newline at end of file
diff --git a/Utilities/Semaphore.cpp b/Utilities/Semaphore.cpp
new file mode 100644
index 0000000000..ed3d419e12
--- /dev/null
+++ b/Utilities/Semaphore.cpp
@@ -0,0 +1,120 @@
+#include "stdafx.h"
+#include "Utilities/Semaphore.h"
+
+bool semaphore_t::try_wait()
+{
+	// check m_value without interlocked op
+	if (m_var.load().value == 0)
+	{
+		return false;
+	}
+
+	// try to decrement m_value atomically
+	const auto old = m_var.atomic_op([](sync_var_t& var)
+	{
+		if (var.value)
+		{
+			var.value--;
+		}
+	});
+
+	// recheck atomic result
+	if (old.value == 0)
+	{
+		return false;
+	}
+
+	return true;
+}
+
+bool semaphore_t::try_post()
+{
+	// check m_value without interlocked op
+	if (m_var.load().value >= max_value)
+	{
+		return false;
+	}
+
+	// try to increment m_value atomically
+	const auto old = m_var.atomic_op([&](sync_var_t& var)
+	{
+		if (var.value < max_value)
+		{
+			var.value++;
+		}
+	});
+
+	// recheck atomic result
+	if (old.value >= max_value)
+	{
+		return false;
+	}
+
+	if (old.waiters)
+	{
+		// notify waiting thread
+		std::lock_guard<std::mutex> lock(m_mutex);
+
+		m_cv.notify_one();
+	}
+
+	return true;
+}
+
+void semaphore_t::wait()
+{
+	if (m_var.atomic_op([](sync_var_t& var) -> bool
+	{
+		if (var.value)
+		{
+			var.value--;
+
+			return true;
+		}
+		else
+		{
+			//var.waiters++;
+
+			return false;
+		}
+	}))
+	{
+		return;
+	}
+
+	std::unique_lock<std::mutex> lock(m_mutex);
+
+	m_var.atomic_op([](sync_var_t& var)
+	{
+		var.waiters++;
+	});
+
+	while (!m_var.atomic_op([](sync_var_t& var) -> bool
+	{
+		if (var.value)
+		{
+			var.value--;
+			var.waiters--;
+
+			return true;
+		}
+		else
+		{
+			return false;
+		}
+	}))
+	{
+		m_cv.wait(lock);
+	}
+}
+
+bool semaphore_t::post_and_wait()
+{
+	// TODO: merge these functions? Probably has a race condition.
+	if (try_wait()) return false;
+
+	try_post();
+	wait();
+
+	return true;
+}
diff --git a/Utilities/Semaphore.h b/Utilities/Semaphore.h
new file mode 100644
index 0000000000..03d3a77168
--- /dev/null
+++ b/Utilities/Semaphore.h
@@ -0,0 +1,37 @@
+#pragma once
+
+class semaphore_t
+{
+	// semaphore mutex
+	std::mutex m_mutex;
+
+	// semaphore condition variable
+	std::condition_variable m_cv;
+
+	struct sync_var_t
+	{
+		u32 value; // current semaphore value
+		u32 waiters; // current amount of waiters
+	};
+
+	// current semaphore value
+	atomic_t<sync_var_t> m_var;
+
+public:
+	// max semaphore value
+	const u32 max_value;
+
+	semaphore_t(u32 max_value = 1, u32 value = 0)
+		: m_var({ value, 0 })
+		, max_value(max_value)
+	{
+	}
+
+	bool try_wait();
+
+	bool try_post();
+
+	void wait();
+
+	bool post_and_wait();
+};
\ No newline at end of file
diff --git a/rpcs3/Emu/RSX/GCM.h b/rpcs3/Emu/RSX/GCM.h
index 375788fdb7..71aa10ee82 100644
--- a/rpcs3/Emu/RSX/GCM.h
+++ b/rpcs3/Emu/RSX/GCM.h
@@ -224,9 +224,9 @@ typedef s32(CellGcmContextCallback)(vm::ptr<CellGcmContextData>, u32);
 
 struct CellGcmContextData
 {
-	be_t<u32> begin;
-	be_t<u32> end;
-	be_t<u32> current;
+	vm::bptr<u32> begin;
+	vm::bptr<u32> end;
+	vm::bptr<u32> current;
 	vm::bptr<CellGcmContextCallback> callback;
 };
 
diff --git a/rpcs3/Emu/RSX/GSRender.cpp b/rpcs3/Emu/RSX/GSRender.cpp
index 371db32257..cebc2d08a8 100644
--- a/rpcs3/Emu/RSX/GSRender.cpp
+++ b/rpcs3/Emu/RSX/GSRender.cpp
@@ -12,7 +12,6 @@ GSLock::GSLock(GSRender& renderer, GSLockType type)
 	switch (m_type)
 	{
 	case GS_LOCK_NOT_WAIT: m_renderer.m_cs_main.lock(); break;
-	case GS_LOCK_WAIT_FLUSH: m_renderer.m_sem_flush.wait(); break;
 	case GS_LOCK_WAIT_FLIP: m_renderer.m_sem_flip.wait(); break;
 	}
 }
@@ -22,8 +21,7 @@ GSLock::~GSLock()
 	switch (m_type)
 	{
 	case GS_LOCK_NOT_WAIT: m_renderer.m_cs_main.unlock(); break;
-	case GS_LOCK_WAIT_FLUSH: m_renderer.m_sem_flush.post(); break;
-	case GS_LOCK_WAIT_FLIP: m_renderer.m_sem_flip.post(); break;
+	case GS_LOCK_WAIT_FLIP: m_renderer.m_sem_flip.try_post(); break;
 	}
 }
 
diff --git a/rpcs3/Emu/RSX/GSRender.h b/rpcs3/Emu/RSX/GSRender.h
index 4b376e5187..d0755e5659 100644
--- a/rpcs3/Emu/RSX/GSRender.h
+++ b/rpcs3/Emu/RSX/GSRender.h
@@ -17,7 +17,6 @@ struct GSRender : public RSXThread
 enum GSLockType
 {
 	GS_LOCK_NOT_WAIT,
-	GS_LOCK_WAIT_FLUSH,
 	GS_LOCK_WAIT_FLIP,
 };
 
diff --git a/rpcs3/Emu/RSX/RSXThread.cpp b/rpcs3/Emu/RSX/RSXThread.cpp
index 8363adc6a5..3fb31f9c95 100644
--- a/rpcs3/Emu/RSX/RSXThread.cpp
+++ b/rpcs3/Emu/RSX/RSXThread.cpp
@@ -258,6 +258,8 @@ void RSXThread::DoCmd(const u32 fcmd, const u32 cmd, const u32 args_addr, const
 			});
 		}
 
+		m_sem_flip.post_and_wait();
+
 		auto sync = [&]()
 		{
 			double limit;
@@ -2499,14 +2501,6 @@ void RSXThread::Task()
 
 		if (put == get || !Emu.IsRunning())
 		{
-			if (put == get)
-			{
-				if (m_flip_status == 0)
-					m_sem_flip.post_and_wait();
-
-				m_sem_flush.post_and_wait();
-			}
-
 			std::this_thread::sleep_for(std::chrono::milliseconds(1)); // hack
 			continue;
 		}
diff --git a/rpcs3/Emu/RSX/RSXThread.h b/rpcs3/Emu/RSX/RSXThread.h
index ce0d92c819..6cacabc967 100644
--- a/rpcs3/Emu/RSX/RSXThread.h
+++ b/rpcs3/Emu/RSX/RSXThread.h
@@ -5,7 +5,7 @@
 #include "RSXFragmentProgram.h"
 
 #include <stack>
-#include "Utilities/SSemaphore.h"
+#include "Utilities/Semaphore.h"
 #include "Utilities/Thread.h"
 #include "Utilities/Timer.h"
 
@@ -155,8 +155,7 @@ public:
 
 public:
 	std::mutex m_cs_main;
-	SSemaphore m_sem_flush;
-	SSemaphore m_sem_flip;
+	semaphore_t m_sem_flip;
 	u64 m_last_flip_time;
 	vm::ptr<void(u32)> m_flip_handler;
 	vm::ptr<void(u32)> m_user_handler;
diff --git a/rpcs3/Emu/SysCalls/Modules/cellGcmSys.cpp b/rpcs3/Emu/SysCalls/Modules/cellGcmSys.cpp
index 3c3d292c91..e4b4cf1ffc 100644
--- a/rpcs3/Emu/SysCalls/Modules/cellGcmSys.cpp
+++ b/rpcs3/Emu/SysCalls/Modules/cellGcmSys.cpp
@@ -320,9 +320,9 @@ void _cellGcmFunc15(vm::ptr<CellGcmContextData> context)
 u32 g_defaultCommandBufferBegin, g_defaultCommandBufferFragmentCount;
 
 // Called by cellGcmInit
-s32 _cellGcmInitBody(vm::ptr<CellGcmContextData> context, u32 cmdSize, u32 ioSize, u32 ioAddress)
+s32 _cellGcmInitBody(vm::pptr<CellGcmContextData> context, u32 cmdSize, u32 ioSize, u32 ioAddress)
 {
-	cellGcmSys.Warning("_cellGcmInitBody(context=*0x%x, cmdSize=0x%x, ioSize=0x%x, ioAddress=0x%x)", context, cmdSize, ioSize, ioAddress);
+	cellGcmSys.Warning("_cellGcmInitBody(context=**0x%x, cmdSize=0x%x, ioSize=0x%x, ioAddress=0x%x)", context, cmdSize, ioSize, ioAddress);
 
 	if(!local_size && !local_addr)
 	{
@@ -365,8 +365,8 @@ s32 _cellGcmInitBody(vm::ptr<CellGcmContextData> context, u32 cmdSize, u32 ioSiz
 	g_defaultCommandBufferBegin = ioAddress;
 	g_defaultCommandBufferFragmentCount = cmdSize / (32 * 1024);
 
-	current_context.begin = g_defaultCommandBufferBegin + 4096; // 4 kb reserved at the beginning
-	current_context.end = g_defaultCommandBufferBegin + 32 * 1024 - 4; // 4b at the end for jump
+	current_context.begin.set(g_defaultCommandBufferBegin + 4096); // 4 kb reserved at the beginning
+	current_context.end.set(g_defaultCommandBufferBegin + 32 * 1024 - 4); // 4b at the end for jump
 	current_context.current = current_context.begin;
 	current_context.callback.set(Emu.GetRSXCallback() - 4);
 
@@ -376,7 +376,7 @@ s32 _cellGcmInitBody(vm::ptr<CellGcmContextData> context, u32 cmdSize, u32 ioSiz
 	gcm_info.label_addr = vm::alloc(0x1000, vm::main); // ???
 
 	vm::get_ref<CellGcmContextData>(gcm_info.context_addr) = current_context;
-	vm::write32(context.addr(), gcm_info.context_addr);
+	context->set(gcm_info.context_addr);
 
 	auto& ctrl = vm::get_ref<CellGcmControl>(gcm_info.control_addr);
 	ctrl.put.store(0);
@@ -480,59 +480,46 @@ void cellGcmSetFlipStatus()
 	Emu.GetGSManager().GetRender().m_flip_status = 0;
 }
 
-s32 cellGcmSetPrepareFlip(PPUThread& CPU, vm::ptr<CellGcmContextData> ctxt, u32 id)
+s32 cellGcmSetPrepareFlip(PPUThread& ppu, vm::ptr<CellGcmContextData> ctxt, u32 id)
 {
-	cellGcmSys.Log("cellGcmSetPrepareFlip(ctx=0x%x, id=0x%x)", ctxt.addr(), id);
+	cellGcmSys.Log("cellGcmSetPrepareFlip(ctx=*0x%x, id=0x%x)", ctxt, id);
 
-	if(id > 7)
+	if (id > 7)
 	{
 		cellGcmSys.Error("cellGcmSetPrepareFlip : CELL_GCM_ERROR_FAILURE");
 		return CELL_GCM_ERROR_FAILURE;
 	}
 
-	GSLockCurrent gslock(GS_LOCK_WAIT_FLUSH);
-
-	u32 current = ctxt->current;
-
-	if (current + 8 == ctxt->begin)
+	if (ctxt->current + 2 >= ctxt->end)
 	{
-		cellGcmSys.Error("cellGcmSetPrepareFlip : queue is full");
-		return CELL_GCM_ERROR_FAILURE;
-	}
-
-	if (current + 8 >= ctxt->end)
-	{
-		cellGcmSys.Error("Bad flip!");
-		if (s32 res = ctxt->callback(CPU, ctxt, 8 /* ??? */))
+		if (s32 res = ctxt->callback(ppu, ctxt, 8 /* ??? */))
 		{
 			cellGcmSys.Error("cellGcmSetPrepareFlip : callback failed (0x%08x)", res);
 			return res;
 		}
 	}
 
-	current = ctxt->current;
-	vm::write32(current, 0x3fead | (1 << 18));
-	vm::write32(current + 4, id);
-	ctxt->current += 8;
+	*ctxt->current++ = 0x3fead | (1 << 18);
+	*ctxt->current++ = id;
 
-	if(ctxt.addr() == gcm_info.context_addr)
+	if (ctxt.addr() == gcm_info.context_addr)
 	{
-		auto& ctrl = vm::get_ref<CellGcmControl>(gcm_info.control_addr);
-		ctrl.put.atomic_op([](be_t<u32>& value)
-		{
-			value += 8;
-		});
+		vm::get_ref<CellGcmControl>(gcm_info.control_addr).put += 8;
 	}
 
 	return id;
 }
 
-s32 cellGcmSetFlip(PPUThread& CPU, vm::ptr<CellGcmContextData> ctxt, u32 id)
+s32 cellGcmSetFlip(PPUThread& ppu, vm::ptr<CellGcmContextData> ctxt, u32 id)
 {
-	cellGcmSys.Log("cellGcmSetFlip(ctx=0x%x, id=0x%x)", ctxt.addr(), id);
+	cellGcmSys.Log("cellGcmSetFlip(ctxt=*0x%x, id=0x%x)", ctxt, id);
 
-	s32 res = cellGcmSetPrepareFlip(CPU, ctxt, id);
-	return res < 0 ? CELL_GCM_ERROR_FAILURE : CELL_OK;
+	if (s32 res = cellGcmSetPrepareFlip(ppu, ctxt, id))
+	{
+		if (res < 0) return CELL_GCM_ERROR_FAILURE;
+	}
+
+	return CELL_OK;
 }
 
 s32 cellGcmSetSecondVFrequency(u32 freq)
@@ -610,9 +597,10 @@ void cellGcmSetVBlankHandler(vm::ptr<void(u32)> handler)
 
 s32 cellGcmSetWaitFlip(vm::ptr<CellGcmContextData> ctxt)
 {
-	cellGcmSys.Log("cellGcmSetWaitFlip(ctx=*0x%x)", ctxt);
+	cellGcmSys.Warning("cellGcmSetWaitFlip(ctx=*0x%x)", ctxt);
+
+	// TODO: emit RSX command for "wait flip" operation
 
-	GSLockCurrent lock(GS_LOCK_WAIT_FLIP);
 	return CELL_OK;
 }
 
@@ -1101,18 +1089,18 @@ void cellGcmSetDefaultCommandBuffer()
 // Other
 //------------------------------------------------------------------------
 
-s32 _cellGcmSetFlipCommand(PPUThread& CPU, vm::ptr<CellGcmContextData> ctx, u32 id)
+s32 _cellGcmSetFlipCommand(PPUThread& ppu, vm::ptr<CellGcmContextData> ctx, u32 id)
 {
 	cellGcmSys.Log("cellGcmSetFlipCommand(ctx=*0x%x, id=0x%x)", ctx, id);
 
-	return cellGcmSetPrepareFlip(CPU, ctx, id);
+	return cellGcmSetPrepareFlip(ppu, ctx, id);
 }
 
-s32 _cellGcmSetFlipCommandWithWaitLabel(PPUThread& CPU, vm::ptr<CellGcmContextData> ctx, u32 id, u32 label_index, u32 label_value)
+s32 _cellGcmSetFlipCommandWithWaitLabel(PPUThread& ppu, vm::ptr<CellGcmContextData> ctx, u32 id, u32 label_index, u32 label_value)
 {
 	cellGcmSys.Log("cellGcmSetFlipCommandWithWaitLabel(ctx=*0x%x, id=0x%x, label_index=0x%x, label_value=0x%x)", ctx, id, label_index, label_value);
 
-	s32 res = cellGcmSetPrepareFlip(CPU, ctx, id);
+	s32 res = cellGcmSetPrepareFlip(ppu, ctx, id);
 	vm::write32(gcm_info.label_addr + 0x10 * label_index, label_value);
 	return res < 0 ? CELL_GCM_ERROR_FAILURE : CELL_OK;
 }
@@ -1200,9 +1188,7 @@ static bool isInCommandBufferExcept(u32 getPos, u32 bufferBegin, u32 bufferEnd)
 	return true;
 }
 
-// TODO: This function was originally located in lv2/SC_GCM and appears in RPCS3 as a lv2 syscall with id 1023,
-//       which according to lv2 dumps isn't the case. So, is this a proper place for this function?
-
+// TODO: Avoid using syscall 1023 for calling this function
 s32 cellGcmCallback(vm::ptr<CellGcmContextData> context, u32 count)
 {
 	cellGcmSys.Log("cellGcmCallback(context=*0x%x, count=0x%x)", context, count);
@@ -1210,16 +1196,16 @@ s32 cellGcmCallback(vm::ptr<CellGcmContextData> context, u32 count)
 	auto& ctrl = vm::get_ref<CellGcmControl>(gcm_info.control_addr);
 	const std::chrono::time_point<std::chrono::system_clock> enterWait = std::chrono::system_clock::now();
 	// Flush command buffer (ie allow RSX to read up to context->current)
-	ctrl.put.exchange(getOffsetFromAddress(context->current));
+	ctrl.put.exchange(getOffsetFromAddress(context->current.addr()));
 
-	std::pair<u32, u32> newCommandBuffer = getNextCommandBufferBeginEnd(context->current);
+	std::pair<u32, u32> newCommandBuffer = getNextCommandBufferBeginEnd(context->current.addr());
 	u32 offset = getOffsetFromAddress(newCommandBuffer.first);
 	// Write jump instruction
-	vm::write32(context->current, CELL_GCM_METHOD_FLAG_JUMP | offset);
+	*context->current = CELL_GCM_METHOD_FLAG_JUMP | offset;
 	// Update current command buffer
-	context->begin = newCommandBuffer.first;
-	context->current = newCommandBuffer.first;
-	context->end = newCommandBuffer.second;
+	context->begin.set(newCommandBuffer.first);
+	context->current.set(newCommandBuffer.first);
+	context->end.set(newCommandBuffer.second);
 
 	// Wait for rsx to "release" the new command buffer
 	while (!Emu.IsStopped())
@@ -1235,55 +1221,6 @@ s32 cellGcmCallback(vm::ptr<CellGcmContextData> context, u32 count)
 	}
 
 	return CELL_OK;
-
-	//if (0)
-	//{
-	//	auto& ctrl = vm::get_ref<CellGcmControl>(gcm_info.control_addr);
-	//	be_t<u32> res = context->current - context->begin - ctrl.put.load();
-
-	//	if (res != 0)
-	//	{
-	//		GSLockCurrent gslock(GS_LOCK_WAIT_FLUSH);
-	//	}
-
-	//	memmove(vm::get_ptr<void>(context->begin), vm::get_ptr<void>(context->current - res), res);
-
-	//	context->current = context->begin + res;
-	//	ctrl.put.store(res);
-	//	ctrl.get.store(0);
-
-	//	return CELL_OK;
-	//}
-
-	//auto& ctrl = vm::get_ref<CellGcmControl>(gcm_info.control_addr);
-
-	// preparations for changing the place (for optimized FIFO mode)
-	//auto cmd = vm::ptr<u32>::make(context->current);
-	//cmd[0] = 0x41D6C;
-	//cmd[1] = 0x20;
-	//cmd[2] = 0x41D74;
-	//cmd[3] = 0; // some incrementing by module value
-	//context->current += 0x10;
-
-	//if (0)
-	//{
-	//	const u32 address = context->begin;
-	//	const u32 upper = offsetTable.ioAddress[address >> 20]; // 12 bits
-	//	assert(upper != 0xFFFF);
-	//	const u32 offset = (upper << 20) | (address & 0xFFFFF);
-	//	vm::write32(context->current, CELL_GCM_METHOD_FLAG_JUMP | offset); // set JUMP cmd
-
-	//	auto& ctrl = vm::get_ref<CellGcmControl>(gcm_info.control_addr);
-	//	ctrl.put.exchange(offset);
-	//}
-	//else
-	//{
-	//	vm::write32(context->current, CELL_GCM_METHOD_FLAG_JUMP | CELL_GCM_METHOD_FLAG_NON_INCREMENT | (0));
-	//}
-	
-	//context->current = context->begin; // rewind to the beginning
-	// TODO: something is missing
-	return CELL_OK;
 }
 
 //----------------------------------------------------------------------------
diff --git a/rpcs3/Emu/SysCalls/Modules/cellResc.cpp b/rpcs3/Emu/SysCalls/Modules/cellResc.cpp
index ac2b403f04..ab1c7bffd2 100644
--- a/rpcs3/Emu/SysCalls/Modules/cellResc.cpp
+++ b/rpcs3/Emu/SysCalls/Modules/cellResc.cpp
@@ -17,7 +17,7 @@ extern void cellGcmSetFlipHandler(vm::ptr<void(u32)> handler);
 extern void cellGcmSetVBlankHandler(vm::ptr<void(u32)> handler);
 extern s32 cellGcmAddressToOffset(u32 address, vm::ptr<u32> offset);
 extern s32 cellGcmSetDisplayBuffer(u32 id, u32 offset, u32 pitch, u32 width, u32 height);
-extern s32 cellGcmSetPrepareFlip(PPUThread& CPU, vm::ptr<CellGcmContextData> ctx, u32 id);
+extern s32 cellGcmSetPrepareFlip(PPUThread& ppu, vm::ptr<CellGcmContextData> ctx, u32 id);
 extern s32 cellGcmSetSecondVFrequency(u32 freq);
 extern u32 cellGcmGetLabelAddress(u8 index);
 extern u32 cellGcmGetTiledPitchSize(u32 size);
@@ -467,7 +467,6 @@ void InitMembers()
 void SetupRsxRenderingStates(vm::ptr<CellGcmContextData>& cntxt)
 {
 	//TODO: use cntxt
-	GSLockCurrent lock(GS_LOCK_WAIT_FLUSH);
 	GSRender& r = Emu.GetGSManager().GetRender();
 	r.m_set_color_mask = true; r.m_color_mask_a = r.m_color_mask_r = r.m_color_mask_g = r.m_color_mask_b = true;
 	r.m_set_depth_mask = true; r.m_depth_mask = 0;
@@ -514,7 +513,6 @@ void SetupRsxRenderingStates(vm::ptr<CellGcmContextData>& cntxt)
 
 void SetupVertexArrays(vm::ptr<CellGcmContextData>& cntxt)
 {
-	GSLockCurrent lock(GS_LOCK_WAIT_FLUSH);
 	GSRender& r = Emu.GetGSManager().GetRender();
 	
 	//TODO
@@ -538,7 +536,6 @@ void SetupSurfaces(vm::ptr<CellGcmContextData>& cntxt)
 		dstOffset1 = s_rescInternalInstance->m_dstOffsets[s_rescInternalInstance->m_bufIdPalMidNow];
 	}
 
-	GSLockCurrent lock(GS_LOCK_WAIT_FLUSH);
 	GSRender& r = Emu.GetGSManager().GetRender();
 
 	r.m_surface_type = CELL_GCM_SURFACE_PITCH;
@@ -1023,8 +1020,9 @@ s32 cellRescSetConvertAndFlip(PPUThread& CPU, vm::ptr<CellGcmContextData> cntxt,
 
 s32 cellRescSetWaitFlip()
 {
-	cellResc.Log("cellRescSetWaitFlip()");
-	GSLockCurrent lock(GS_LOCK_WAIT_FLIP);
+	cellResc.Warning("cellRescSetWaitFlip()");
+	
+	// TODO: emit RSX command for "wait flip" operation
 
 	return CELL_OK;
 }
diff --git a/rpcs3/emucore.vcxproj b/rpcs3/emucore.vcxproj
index b0b9837289..2542c6a4e7 100644
--- a/rpcs3/emucore.vcxproj
+++ b/rpcs3/emucore.vcxproj
@@ -30,7 +30,7 @@
     <ClCompile Include="..\Utilities\rPlatform.cpp" />
     <ClCompile Include="..\Utilities\rTime.cpp" />
     <ClCompile Include="..\Utilities\rXml.cpp" />
-    <ClCompile Include="..\Utilities\SSemaphore.cpp" />
+    <ClCompile Include="..\Utilities\Semaphore.cpp" />
     <ClCompile Include="..\Utilities\StrFmt.cpp" />
     <ClCompile Include="..\Utilities\Thread.cpp" />
     <ClCompile Include="Emu\Cell\PPUInterpreter.cpp" />
@@ -318,9 +318,9 @@
     <ClInclude Include="..\Utilities\rPlatform.h" />
     <ClInclude Include="..\Utilities\rTime.h" />
     <ClInclude Include="..\Utilities\rXml.h" />
+    <ClInclude Include="..\Utilities\Semaphore.h" />
     <ClInclude Include="..\Utilities\simpleini\ConvertUTF.h" />
     <ClInclude Include="..\Utilities\simpleini\SimpleIni.h" />
-    <ClInclude Include="..\Utilities\SSemaphore.h" />
     <ClInclude Include="..\Utilities\StrFmt.h" />
     <ClInclude Include="..\Utilities\Thread.h" />
     <ClInclude Include="..\Utilities\Timer.h" />
diff --git a/rpcs3/emucore.vcxproj.filters b/rpcs3/emucore.vcxproj.filters
index f69dd07f16..f00fc35ec6 100644
--- a/rpcs3/emucore.vcxproj.filters
+++ b/rpcs3/emucore.vcxproj.filters
@@ -476,9 +476,6 @@
     <ClCompile Include="Emu\SysCalls\Modules\sys_http.cpp">
       <Filter>Emu\SysCalls\currently_unused</Filter>
     </ClCompile>
-    <ClCompile Include="..\Utilities\SSemaphore.cpp">
-      <Filter>Utilities</Filter>
-    </ClCompile>
     <ClCompile Include="Emu\SysCalls\lv2\sys_process.cpp">
       <Filter>Emu\SysCalls\lv2</Filter>
     </ClCompile>
@@ -884,6 +881,9 @@
     <ClCompile Include="Emu\SysCalls\Modules\sceNpUtil.cpp">
       <Filter>Emu\SysCalls\Modules</Filter>
     </ClCompile>
+    <ClCompile Include="..\Utilities\Semaphore.cpp">
+      <Filter>Utilities</Filter>
+    </ClCompile>
   </ItemGroup>
   <ItemGroup>
     <ClInclude Include="Crypto\aes.h">
@@ -1267,9 +1267,6 @@
     <ClInclude Include="Emu\Event.h">
       <Filter>Emu\SysCalls</Filter>
     </ClInclude>
-    <ClInclude Include="..\Utilities\SSemaphore.h">
-      <Filter>Utilities</Filter>
-    </ClInclude>
     <ClInclude Include="Emu\SysCalls\lv2\sys_process.h">
       <Filter>Emu\SysCalls\lv2</Filter>
     </ClInclude>
@@ -1738,5 +1735,8 @@
     <ClInclude Include="Emu\SysCalls\Modules\sceNpUtil.h">
       <Filter>Emu\SysCalls\Modules</Filter>
     </ClInclude>
+    <ClInclude Include="..\Utilities\Semaphore.h">
+      <Filter>Utilities</Filter>
+    </ClInclude>
   </ItemGroup>
 </Project>
\ No newline at end of file