diff --git a/kernel/arch/i686/Thread.S b/kernel/arch/i686/Thread.S
index d7ce92f1..47ea765d 100644
--- a/kernel/arch/i686/Thread.S
+++ b/kernel/arch/i686/Thread.S
@@ -36,6 +36,8 @@ start_kernel_thread:
 
 .global start_userspace_thread
 start_userspace_thread:
+	call load_thread_sse
+
 	call get_thread_start_sp
 	movl %eax, %esp
 
diff --git a/kernel/arch/x86_64/Thread.S b/kernel/arch/x86_64/Thread.S
index 079b525c..c5b449cd 100644
--- a/kernel/arch/x86_64/Thread.S
+++ b/kernel/arch/x86_64/Thread.S
@@ -27,6 +27,8 @@ start_kernel_thread:
 
 .global start_userspace_thread
 start_userspace_thread:
+	call load_thread_sse
+
 	call get_thread_start_sp
 	movq %rax, %rsp
 
diff --git a/kernel/kernel/IDT.cpp b/kernel/kernel/IDT.cpp
index f6226750..197ef5be 100644
--- a/kernel/kernel/IDT.cpp
+++ b/kernel/kernel/IDT.cpp
@@ -354,6 +354,8 @@ namespace Kernel
 			asm volatile("cli; 1: hlt; jmp 1b");
 		}
 
+		Thread::current().save_sse();
+
 		ASSERT(InterruptController::get().is_in_service(IRQ_TIMER - IRQ_VECTOR_BASE));
 		InterruptController::get().eoi(IRQ_TIMER - IRQ_VECTOR_BASE);
 
@@ -365,6 +367,8 @@ namespace Kernel
 		auto& current_thread = Thread::current();
 		if (current_thread.can_add_signal_to_execute())
 			current_thread.handle_signal();
+
+		Thread::current().load_sse();
 	}
 
 	extern "C" void cpp_irq_handler(uint32_t irq)
diff --git a/kernel/kernel/Thread.cpp b/kernel/kernel/Thread.cpp
index 7a517af0..1f8b8e70 100644
--- a/kernel/kernel/Thread.cpp
+++ b/kernel/kernel/Thread.cpp
@@ -34,6 +34,11 @@ namespace Kernel
 		return Thread::current().userspace_stack_top() - 4 * sizeof(uintptr_t);
 	}
 
+	extern "C" void load_thread_sse()
+	{
+		Thread::current().load_sse();
+	}
+
 	static pid_t s_next_tid = 1;
 
 	alignas(16) static uint8_t s_default_sse_storage[512];
@@ -168,6 +173,7 @@ namespace Kernel
 	{
 		auto* thread = TRY(create_userspace(m_process, m_process->page_table()));
 
+		save_sse();
 		memcpy(thread->m_sse_storage, m_sse_storage, sizeof(m_sse_storage));
 
 		thread->setup_exec_impl(
@@ -202,6 +208,7 @@ namespace Kernel
 		thread->m_interrupt_stack.sp = sp;
 		thread->m_interrupt_stack.ss = 0x10;
 
+		save_sse();
 		memcpy(thread->m_sse_storage, m_sse_storage, sizeof(m_sse_storage));
 
 #if ARCH(x86_64)