diff --git a/DynamicHooks/registers.cpp b/DynamicHooks/registers.cpp
index c9d28d3..77fe503 100644
--- a/DynamicHooks/registers.cpp
+++ b/DynamicHooks/registers.cpp
@@ -151,14 +151,15 @@ CRegisters::CRegisters(ke::Vector<Register_t> registers)
 	// ========================================================================
 	// >> 128-bit XMM registers
 	// ========================================================================
-	m_xmm0 = CreateRegister(registers, XMM0, 16);
-	m_xmm1 = CreateRegister(registers, XMM1, 16);
-	m_xmm2 = CreateRegister(registers, XMM2, 16);
-	m_xmm3 = CreateRegister(registers, XMM3, 16);
-	m_xmm4 = CreateRegister(registers, XMM4, 16);
-	m_xmm5 = CreateRegister(registers, XMM5, 16);
-	m_xmm6 = CreateRegister(registers, XMM6, 16);
-	m_xmm7 = CreateRegister(registers, XMM7, 16);
+	// Copying data from xmm0-7 requires the memory address to be 16-byte aligned.
+	m_xmm0 = CreateRegister(registers, XMM0, 16, 16);
+	m_xmm1 = CreateRegister(registers, XMM1, 16, 16);
+	m_xmm2 = CreateRegister(registers, XMM2, 16, 16);
+	m_xmm3 = CreateRegister(registers, XMM3, 16, 16);
+	m_xmm4 = CreateRegister(registers, XMM4, 16, 16);
+	m_xmm5 = CreateRegister(registers, XMM5, 16, 16);
+	m_xmm6 = CreateRegister(registers, XMM6, 16, 16);
+	m_xmm7 = CreateRegister(registers, XMM7, 16, 16);
 
 	// 64-bit mode only
 	/*
@@ -360,13 +361,13 @@ CRegisters::~CRegisters()
 	DeleteRegister(m_st7);
 }
 
-CRegister* CRegisters::CreateRegister(ke::Vector<Register_t>& registers, Register_t reg, int iSize)
+CRegister* CRegisters::CreateRegister(ke::Vector<Register_t>& registers, Register_t reg, uint16_t iSize, uint16_t iAlignment)
 {
 	for(size_t i = 0; i < registers.length(); i++)
 	{
 		if (registers[i] == reg)
 		{
-			return new CRegister(iSize);
+			return new CRegister(iSize, iAlignment);
 		}
 	}
 	return NULL;
diff --git a/DynamicHooks/registers.h b/DynamicHooks/registers.h
index 049d66b..acb0642 100644
--- a/DynamicHooks/registers.h
+++ b/DynamicHooks/registers.h
@@ -216,15 +216,31 @@ enum Register_t
 class CRegister
 {
 public:
-	CRegister(int iSize)
+	CRegister(uint16_t iSize, uint16_t iAlignment = 0)
 	{
 		m_iSize = iSize;
-		m_pAddress = malloc(iSize);
+		m_iAlignment = iAlignment;
+		if (iAlignment > 0)
+#ifdef KE_WINDOWS
+			m_pAddress = _aligned_malloc(iSize, iAlignment);
+#else
+			m_pAddress = aligned_alloc(iAlignment, iSize);
+#endif
+		else
+			m_pAddress = malloc(iSize);
 	}
 
 	~CRegister()
 	{
+
+#ifdef KE_WINDOWS
+		if (m_iAlignment > 0)
+			_aligned_free(m_pAddress);
+		else
+			free(m_pAddress);
+#else
 		free(m_pAddress);
+#endif
 	}
 
 	template<class T>
@@ -252,7 +268,8 @@ public:
 	}
 
 public:
-	int m_iSize;
+	uint16_t m_iSize;
+	uint16_t m_iAlignment;
 	void* m_pAddress;
 };
 
@@ -269,7 +286,7 @@ public:
 	CRegister* GetRegister(Register_t reg);
 
 private:
-	CRegister* CreateRegister(ke::Vector<Register_t>& registers, Register_t reg, int iSize);
+	CRegister* CreateRegister(ke::Vector<Register_t>& registers, Register_t reg, uint16_t iSize, uint16_t iAlignment = 0);
 	void DeleteRegister(CRegister* pRegister);
 
 public: