diff --git a/TODO b/TODO
index c4f46e3126..097c310a79 100644
--- a/TODO
+++ b/TODO
@@ -46,6 +46,7 @@
 ✔ Replace SERIALIZE_AS_POD with BOOST_IS_BITWISE_SERIALIZABLE @started(20-01-03 13:47) @done(20-01-03 13:58) @lasted(11m22s)
 ☐ Review constructor/initialization code
 ☐ Review core timing events
+☐ Review base class serialization everywhere
 ✔ Fix CI @done(19-12-31 21:32)
 ✔ HW @done(19-08-13 15:41)
     ✔ GPU regs @done(19-08-13 15:41)
@@ -87,7 +88,7 @@
         ✔ Shared page @done(20-01-04 21:09)
         ✔ SVC @done(19-12-22 21:32)
             Nothing to do - all data is constant
-        ☐ Thread @started(19-08-13 16:45)
+        ✔ Thread @started(19-08-13 16:45) @done(20-01-06 20:01) @lasted(20w6d4h16m22s)
             This requires refactoring wakeup_callback to be an object ref
         ✔ Timer @done(19-08-13 16:45)
         ✔ VM Manager @started(19-08-13 16:46) @done(20-01-04 21:09) @lasted(20w4d5h23m42s)
diff --git a/src/common/CMakeLists.txt b/src/common/CMakeLists.txt
index ea117d4409..7ab54242d5 100644
--- a/src/common/CMakeLists.txt
+++ b/src/common/CMakeLists.txt
@@ -96,6 +96,7 @@ add_library(common STATIC
     serialization/atomic.h
     serialization/boost_discrete_interval.hpp
     serialization/boost_flat_set.h
+    serialization/boost_small_vector.hpp
     serialization/boost_vector.hpp
     string_util.cpp
     string_util.h
diff --git a/src/common/serialization/boost_small_vector.hpp b/src/common/serialization/boost_small_vector.hpp
new file mode 100644
index 0000000000..b4e07a8962
--- /dev/null
+++ b/src/common/serialization/boost_small_vector.hpp
@@ -0,0 +1,144 @@
+#ifndef BOOST_SERIALIZATION_BOOST_SMALL_VECTOR_HPP
+#define BOOST_SERIALIZATION_BOOST_SMALL_VECTOR_HPP
+
+// MS compatible compilers support #pragma once
+#if defined(_MSC_VER)
+#pragma once
+#endif
+
+/////////1/////////2/////////3/////////4/////////5/////////6/////////7/////////8
+// boost_vector.hpp: serialization for boost vector templates
+
+// (C) Copyright 2002 Robert Ramey - http://www.rrsd.com .
+// fast array serialization (C) Copyright 2005 Matthias Troyer
+// Use, modification and distribution is subject to the Boost Software
+// License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt)
+
+//  See http://www.boost.org for updates, documentation, and revision history.
+
+#include <boost/container/small_vector.hpp>
+
+#include <boost/config.hpp>
+#include <boost/detail/workaround.hpp>
+
+#include <boost/archive/detail/basic_iarchive.hpp>
+#include <boost/serialization/access.hpp>
+#include <boost/serialization/collection_size_type.hpp>
+#include <boost/serialization/item_version_type.hpp>
+#include <boost/serialization/nvp.hpp>
+
+#include <boost/mpl/bool_fwd.hpp>
+#include <boost/mpl/if.hpp>
+#include <boost/serialization/array_wrapper.hpp>
+#include <boost/serialization/collections_load_imp.hpp>
+#include <boost/serialization/collections_save_imp.hpp>
+#include <boost/serialization/split_free.hpp>
+
+// default is being compatible with version 1.34.1 files, not 1.35 files
+#ifndef BOOST_SERIALIZATION_VECTOR_VERSIONED
+#define BOOST_SERIALIZATION_VECTOR_VERSIONED(V) (V == 4 || V == 5)
+#endif
+
+namespace boost {
+namespace serialization {
+
+/////////1/////////2/////////3/////////4/////////5/////////6/////////7/////////8
+// vector< T >
+
+// the default versions
+
+template <class Archive, class U, std::size_t N>
+inline void save(Archive& ar, const boost::container::small_vector<U, N>& t,
+                 const unsigned int /* file_version */, mpl::false_) {
+    boost::serialization::stl::save_collection<Archive, boost::container::small_vector<U, N>>(ar,
+                                                                                              t);
+}
+
+template <class Archive, class U, std::size_t N>
+inline void load(Archive& ar, boost::container::small_vector<U, N>& t,
+                 const unsigned int /* file_version */, mpl::false_) {
+    const boost::archive::library_version_type library_version(ar.get_library_version());
+    // retrieve number of elements
+    item_version_type item_version(0);
+    collection_size_type count;
+    ar >> BOOST_SERIALIZATION_NVP(count);
+    if (boost::archive::library_version_type(3) < library_version) {
+        ar >> BOOST_SERIALIZATION_NVP(item_version);
+    }
+    t.reserve(count);
+    stl::collection_load_impl(ar, t, count, item_version);
+}
+
+// the optimized versions
+
+template <class Archive, class U, std::size_t N>
+inline void save(Archive& ar, const boost::container::small_vector<U, N>& t,
+                 const unsigned int /* file_version */, mpl::true_) {
+    const collection_size_type count(t.size());
+    ar << BOOST_SERIALIZATION_NVP(count);
+    if (!t.empty())
+        // explict template arguments to pass intel C++ compiler
+        ar << serialization::make_array<const U, collection_size_type>(static_cast<const U*>(&t[0]),
+                                                                       count);
+}
+
+template <class Archive, class U, std::size_t N>
+inline void load(Archive& ar, boost::container::small_vector<U, N>& t,
+                 const unsigned int /* file_version */, mpl::true_) {
+    collection_size_type count(t.size());
+    ar >> BOOST_SERIALIZATION_NVP(count);
+    t.resize(count);
+    unsigned int item_version = 0;
+    if (BOOST_SERIALIZATION_VECTOR_VERSIONED(ar.get_library_version())) {
+        ar >> BOOST_SERIALIZATION_NVP(item_version);
+    }
+    if (!t.empty())
+        // explict template arguments to pass intel C++ compiler
+        ar >> serialization::make_array<U, collection_size_type>(static_cast<U*>(&t[0]), count);
+}
+
+// dispatch to either default or optimized versions
+
+template <class Archive, class U, std::size_t N>
+inline void save(Archive& ar, const boost::container::small_vector<U, N>& t,
+                 const unsigned int file_version) {
+    typedef typename boost::serialization::use_array_optimization<Archive>::template apply<
+        typename remove_const<U>::type>::type use_optimized;
+    save(ar, t, file_version, use_optimized());
+}
+
+template <class Archive, class U, std::size_t N>
+inline void load(Archive& ar, boost::container::small_vector<U, N>& t,
+                 const unsigned int file_version) {
+#ifdef BOOST_SERIALIZATION_VECTOR_135_HPP
+    if (ar.get_library_version() == boost::archive::library_version_type(5)) {
+        load(ar, t, file_version, boost::is_arithmetic<U>());
+        return;
+    }
+#endif
+    typedef typename boost::serialization::use_array_optimization<Archive>::template apply<
+        typename remove_const<U>::type>::type use_optimized;
+    load(ar, t, file_version, use_optimized());
+}
+
+// split non-intrusive serialization function member into separate
+// non intrusive save/load member functions
+template <class Archive, class U, std::size_t N>
+inline void serialize(Archive& ar, boost::container::small_vector<U, N>& t,
+                      const unsigned int file_version) {
+    boost::serialization::split_free(ar, t, file_version);
+}
+
+// split non-intrusive serialization function member into separate
+// non intrusive save/load member functions
+template <class Archive, std::size_t N>
+inline void serialize(Archive& ar, boost::container::small_vector<bool, N>& t,
+                      const unsigned int file_version) {
+    boost::serialization::split_free(ar, t, file_version);
+}
+
+} // namespace serialization
+} // namespace boost
+
+#endif // BOOST_SERIALIZATION_BOOST_SMALL_VECTOR_HPP
diff --git a/src/common/serialization/boost_vector.hpp b/src/common/serialization/boost_vector.hpp
index d97ebd2085..55a5b9eaee 100644
--- a/src/common/serialization/boost_vector.hpp
+++ b/src/common/serialization/boost_vector.hpp
@@ -1,9 +1,9 @@
-#ifndef  BOOST_SERIALIZATION_BOOST_VECTOR_HPP
+#ifndef BOOST_SERIALIZATION_BOOST_VECTOR_HPP
 #define BOOST_SERIALIZATION_BOOST_VECTOR_HPP
 
 // MS compatible compilers support #pragma once
 #if defined(_MSC_VER)
-# pragma once
+#pragma once
 #endif
 
 /////////1/////////2/////////3/////////4/////////5/////////6/////////7/////////8
@@ -24,20 +24,20 @@
 
 #include <boost/archive/detail/basic_iarchive.hpp>
 #include <boost/serialization/access.hpp>
-#include <boost/serialization/nvp.hpp>
 #include <boost/serialization/collection_size_type.hpp>
 #include <boost/serialization/item_version_type.hpp>
+#include <boost/serialization/nvp.hpp>
 
-#include <boost/serialization/collections_save_imp.hpp>
-#include <boost/serialization/collections_load_imp.hpp>
-#include <boost/serialization/split_free.hpp>
-#include <boost/serialization/array_wrapper.hpp>
 #include <boost/mpl/bool_fwd.hpp>
 #include <boost/mpl/if.hpp>
+#include <boost/serialization/array_wrapper.hpp>
+#include <boost/serialization/collections_load_imp.hpp>
+#include <boost/serialization/collections_save_imp.hpp>
+#include <boost/serialization/split_free.hpp>
 
 // default is being compatible with version 1.34.1 files, not 1.35 files
 #ifndef BOOST_SERIALIZATION_VECTOR_VERSIONED
-#define BOOST_SERIALIZATION_VECTOR_VERSIONED(V) (V==4 || V==5)
+#define BOOST_SERIALIZATION_VECTOR_VERSIONED(V) (V == 4 || V == 5)
 #endif
 
 namespace boost {
@@ -48,33 +48,23 @@ namespace serialization {
 
 // the default versions
 
-template<class Archive, class U, class Allocator>
-inline void save(
-    Archive & ar,
-    const boost::container::vector<U, Allocator> &t,
-    const unsigned int /* file_version */,
-    mpl::false_
-){
-    boost::serialization::stl::save_collection<Archive, boost::container::vector<U, Allocator> >(
-        ar, t
-    );
+template <class Archive, class U, class Allocator, class Options>
+inline void save(Archive& ar, const boost::container::vector<U, Allocator, Options>& t,
+                 const unsigned int /* file_version */, mpl::false_) {
+    boost::serialization::stl::save_collection<Archive,
+                                               boost::container::vector<U, Allocator, Options>>(ar,
+                                                                                                t);
 }
 
-template<class Archive, class U, class Allocator>
-inline void load(
-    Archive & ar,
-    boost::container::vector<U, Allocator> &t,
-    const unsigned int /* file_version */,
-    mpl::false_
-){
-    const boost::archive::library_version_type library_version(
-        ar.get_library_version()
-    );
+template <class Archive, class U, class Allocator, class Options>
+inline void load(Archive& ar, boost::container::vector<U, Allocator, Options>& t,
+                 const unsigned int /* file_version */, mpl::false_) {
+    const boost::archive::library_version_type library_version(ar.get_library_version());
     // retrieve number of elements
     item_version_type item_version(0);
     collection_size_type count;
     ar >> BOOST_SERIALIZATION_NVP(count);
-    if(boost::archive::library_version_type(3) < library_version){
+    if (boost::archive::library_version_type(3) < library_version) {
         ar >> BOOST_SERIALIZATION_NVP(item_version);
     }
     t.reserve(count);
@@ -83,107 +73,77 @@ inline void load(
 
 // the optimized versions
 
-template<class Archive, class U, class Allocator>
-inline void save(
-    Archive & ar,
-    const boost::container::vector<U, Allocator> &t,
-    const unsigned int /* file_version */,
-    mpl::true_
-){
+template <class Archive, class U, class Allocator, class Options>
+inline void save(Archive& ar, const boost::container::vector<U, Allocator, Options>& t,
+                 const unsigned int /* file_version */, mpl::true_) {
     const collection_size_type count(t.size());
     ar << BOOST_SERIALIZATION_NVP(count);
     if (!t.empty())
         // explict template arguments to pass intel C++ compiler
-        ar << serialization::make_array<const U, collection_size_type>(
-            static_cast<const U *>(&t[0]),
-            count
-        );
+        ar << serialization::make_array<const U, collection_size_type>(static_cast<const U*>(&t[0]),
+                                                                       count);
 }
 
-template<class Archive, class U, class Allocator>
-inline void load(
-    Archive & ar,
-    boost::container::vector<U, Allocator> &t,
-    const unsigned int /* file_version */,
-    mpl::true_
-){
+template <class Archive, class U, class Allocator, class Options>
+inline void load(Archive& ar, boost::container::vector<U, Allocator, Options>& t,
+                 const unsigned int /* file_version */, mpl::true_) {
     collection_size_type count(t.size());
     ar >> BOOST_SERIALIZATION_NVP(count);
     t.resize(count);
-    unsigned int item_version=0;
-    if(BOOST_SERIALIZATION_VECTOR_VERSIONED(ar.get_library_version())) {
+    unsigned int item_version = 0;
+    if (BOOST_SERIALIZATION_VECTOR_VERSIONED(ar.get_library_version())) {
         ar >> BOOST_SERIALIZATION_NVP(item_version);
     }
     if (!t.empty())
         // explict template arguments to pass intel C++ compiler
-        ar >> serialization::make_array<U, collection_size_type>(
-            static_cast<U *>(&t[0]),
-            count
-        );
-  }
+        ar >> serialization::make_array<U, collection_size_type>(static_cast<U*>(&t[0]), count);
+}
 
 // dispatch to either default or optimized versions
 
-template<class Archive, class U, class Allocator>
-inline void save(
-    Archive & ar,
-    const boost::container::vector<U, Allocator> &t,
-    const unsigned int file_version
-){
-    typedef typename
-    boost::serialization::use_array_optimization<Archive>::template apply<
-        typename remove_const<U>::type
-    >::type use_optimized;
-    save(ar,t,file_version, use_optimized());
+template <class Archive, class U, class Allocator, class Options>
+inline void save(Archive& ar, const boost::container::vector<U, Allocator, Options>& t,
+                 const unsigned int file_version) {
+    typedef typename boost::serialization::use_array_optimization<Archive>::template apply<
+        typename remove_const<U>::type>::type use_optimized;
+    save(ar, t, file_version, use_optimized());
 }
 
-template<class Archive, class U, class Allocator>
-inline void load(
-    Archive & ar,
-    boost::container::vector<U, Allocator> &t,
-    const unsigned int file_version
-){
+template <class Archive, class U, class Allocator, class Options>
+inline void load(Archive& ar, boost::container::vector<U, Allocator, Options>& t,
+                 const unsigned int file_version) {
 #ifdef BOOST_SERIALIZATION_VECTOR_135_HPP
-    if (ar.get_library_version()==boost::archive::library_version_type(5))
-    {
-      load(ar,t,file_version, boost::is_arithmetic<U>());
-      return;
+    if (ar.get_library_version() == boost::archive::library_version_type(5)) {
+        load(ar, t, file_version, boost::is_arithmetic<U>());
+        return;
     }
 #endif
-    typedef typename
-    boost::serialization::use_array_optimization<Archive>::template apply<
-        typename remove_const<U>::type
-    >::type use_optimized;
-    load(ar,t,file_version, use_optimized());
+    typedef typename boost::serialization::use_array_optimization<Archive>::template apply<
+        typename remove_const<U>::type>::type use_optimized;
+    load(ar, t, file_version, use_optimized());
 }
 
 // split non-intrusive serialization function member into separate
 // non intrusive save/load member functions
-template<class Archive, class U, class Allocator>
-inline void serialize(
-    Archive & ar,
-    boost::container::vector<U, Allocator> & t,
-    const unsigned int file_version
-){
+template <class Archive, class U, class Allocator, class Options>
+inline void serialize(Archive& ar, boost::container::vector<U, Allocator, Options>& t,
+                      const unsigned int file_version) {
     boost::serialization::split_free(ar, t, file_version);
 }
 
 // split non-intrusive serialization function member into separate
 // non intrusive save/load member functions
-template<class Archive, class Allocator>
-inline void serialize(
-    Archive & ar,
-    boost::container::vector<bool, Allocator> & t,
-    const unsigned int file_version
-){
+template <class Archive, class Allocator, class Options>
+inline void serialize(Archive& ar, boost::container::vector<bool, Allocator, Options>& t,
+                      const unsigned int file_version) {
     boost::serialization::split_free(ar, t, file_version);
 }
 
-} // serialization
+} // namespace serialization
 } // namespace boost
 
 #include <boost/serialization/collection_traits.hpp>
 
 BOOST_SERIALIZATION_COLLECTION_TRAITS(boost::container::vector)
 
-#endif // BOOST_SERIALIZATION_VECTOR_HPP
+#endif // BOOST_SERIALIZATION_BOOST_VECTOR_HPP
diff --git a/src/core/hle/kernel/address_arbiter.cpp b/src/core/hle/kernel/address_arbiter.cpp
index 7b8329c4d4..5074e352fc 100644
--- a/src/core/hle/kernel/address_arbiter.cpp
+++ b/src/core/hle/kernel/address_arbiter.cpp
@@ -80,16 +80,18 @@ std::shared_ptr<AddressArbiter> KernelSystem::CreateAddressArbiter(std::string n
     return address_arbiter;
 }
 
+void AddressArbiter::WakeUp(ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
+                            std::shared_ptr<WaitObject> object) {
+    ASSERT(reason == ThreadWakeupReason::Timeout);
+    // Remove the newly-awakened thread from the Arbiter's waiting list.
+    waiting_threads.erase(std::remove(waiting_threads.begin(), waiting_threads.end(), thread),
+                          waiting_threads.end());
+};
+
 ResultCode AddressArbiter::ArbitrateAddress(std::shared_ptr<Thread> thread, ArbitrationType type,
                                             VAddr address, s32 value, u64 nanoseconds) {
 
-    auto timeout_callback = [this](ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
-                                   std::shared_ptr<WaitObject> object) {
-        ASSERT(reason == ThreadWakeupReason::Timeout);
-        // Remove the newly-awakened thread from the Arbiter's waiting list.
-        waiting_threads.erase(std::remove(waiting_threads.begin(), waiting_threads.end(), thread),
-                              waiting_threads.end());
-    };
+    auto timeout_callback = std::dynamic_pointer_cast<WakeupCallback>(shared_from_this());
 
     switch (type) {
 
diff --git a/src/core/hle/kernel/address_arbiter.h b/src/core/hle/kernel/address_arbiter.h
index 19b80315f8..85a2a065a5 100644
--- a/src/core/hle/kernel/address_arbiter.h
+++ b/src/core/hle/kernel/address_arbiter.h
@@ -12,6 +12,7 @@
 #include <boost/serialization/vector.hpp>
 #include "common/common_types.h"
 #include "core/hle/kernel/object.h"
+#include "core/hle/kernel/thread.h"
 #include "core/hle/result.h"
 
 // Address arbiters are an underlying kernel synchronization object that can be created/used via
@@ -34,7 +35,7 @@ enum class ArbitrationType : u32 {
     DecrementAndWaitIfLessThanWithTimeout,
 };
 
-class AddressArbiter final : public Object {
+class AddressArbiter final : public Object, public WakeupCallback {
 public:
     explicit AddressArbiter(KernelSystem& kernel);
     ~AddressArbiter() override;
@@ -56,6 +57,9 @@ public:
     ResultCode ArbitrateAddress(std::shared_ptr<Thread> thread, ArbitrationType type, VAddr address,
                                 s32 value, u64 nanoseconds);
 
+    void WakeUp(ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
+                std::shared_ptr<WaitObject> object);
+
 private:
     KernelSystem& kernel;
 
diff --git a/src/core/hle/kernel/hle_ipc.cpp b/src/core/hle/kernel/hle_ipc.cpp
index ab4ecfd059..b1e2f7d8b0 100644
--- a/src/core/hle/kernel/hle_ipc.cpp
+++ b/src/core/hle/kernel/hle_ipc.cpp
@@ -16,6 +16,46 @@
 
 namespace Kernel {
 
+class HLERequestContext::ThreadCallback : public Kernel::WakeupCallback {
+
+public:
+    ThreadCallback(std::shared_ptr<HLERequestContext> context_,
+                   std::shared_ptr<HLERequestContext::WakeupCallback> callback_)
+        : context(context_), callback(callback_) {}
+    void WakeUp(ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
+                std::shared_ptr<WaitObject> object) {
+        ASSERT(thread->status == ThreadStatus::WaitHleEvent);
+        if (callback) {
+            callback->WakeUp(thread, *context, reason);
+        }
+
+        auto& process = thread->owner_process;
+        // We must copy the entire command buffer *plus* the entire static buffers area, since
+        // the translation might need to read from it in order to retrieve the StaticBuffer
+        // target addresses.
+        std::array<u32_le, IPC::COMMAND_BUFFER_LENGTH + 2 * IPC::MAX_STATIC_BUFFERS> cmd_buff;
+        Memory::MemorySystem& memory = context->kernel.memory;
+        memory.ReadBlock(*process, thread->GetCommandBufferAddress(), cmd_buff.data(),
+                         cmd_buff.size() * sizeof(u32));
+        context->WriteToOutgoingCommandBuffer(cmd_buff.data(), *process);
+        // Copy the translated command buffer back into the thread's command buffer area.
+        memory.WriteBlock(*process, thread->GetCommandBufferAddress(), cmd_buff.data(),
+                          cmd_buff.size() * sizeof(u32));
+    }
+
+private:
+    ThreadCallback() = default;
+    std::shared_ptr<HLERequestContext::WakeupCallback> callback{};
+    std::shared_ptr<HLERequestContext> context{};
+
+    template <class Archive>
+    void serialize(Archive& ar, const unsigned int) {
+        ar& callback;
+        ar& context;
+    }
+    friend class boost::serialization::access;
+};
+
 SessionRequestHandler::SessionInfo::SessionInfo(std::shared_ptr<ServerSession> session,
                                                 std::unique_ptr<SessionDataBase> data)
     : session(std::move(session)), data(std::move(data)) {}
@@ -33,34 +73,16 @@ void SessionRequestHandler::ClientDisconnected(std::shared_ptr<ServerSession> se
         connected_sessions.end());
 }
 
-std::shared_ptr<Event> HLERequestContext::SleepClientThread(const std::string& reason,
-                                                            std::chrono::nanoseconds timeout,
-                                                            WakeupCallback&& callback) {
+std::shared_ptr<Event> HLERequestContext::SleepClientThread(
+    const std::string& reason, std::chrono::nanoseconds timeout,
+    std::shared_ptr<WakeupCallback> callback) {
     // Put the client thread to sleep until the wait event is signaled or the timeout expires.
-    thread->wakeup_callback = [context = *this,
-                               callback](ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
-                                         std::shared_ptr<WaitObject> object) mutable {
-        ASSERT(thread->status == ThreadStatus::WaitHleEvent);
-        callback(thread, context, reason);
-
-        auto& process = thread->owner_process;
-        // We must copy the entire command buffer *plus* the entire static buffers area, since
-        // the translation might need to read from it in order to retrieve the StaticBuffer
-        // target addresses.
-        std::array<u32_le, IPC::COMMAND_BUFFER_LENGTH + 2 * IPC::MAX_STATIC_BUFFERS> cmd_buff;
-        Memory::MemorySystem& memory = context.kernel.memory;
-        memory.ReadBlock(*process, thread->GetCommandBufferAddress(), cmd_buff.data(),
-                         cmd_buff.size() * sizeof(u32));
-        context.WriteToOutgoingCommandBuffer(cmd_buff.data(), *process);
-        // Copy the translated command buffer back into the thread's command buffer area.
-        memory.WriteBlock(*process, thread->GetCommandBufferAddress(), cmd_buff.data(),
-                          cmd_buff.size() * sizeof(u32));
-    };
+    thread->wakeup_callback = std::make_shared<ThreadCallback>(shared_from_this(), callback);
 
     auto event = kernel.CreateEvent(Kernel::ResetType::OneShot, "HLE Pause Event: " + reason);
     thread->status = ThreadStatus::WaitHleEvent;
     thread->wait_objects = {event};
-    event->AddWaitingThread(SharedFrom(thread));
+    event->AddWaitingThread(thread);
 
     if (timeout.count() > 0)
         thread->WakeAfterDelay(timeout.count());
@@ -68,8 +90,10 @@ std::shared_ptr<Event> HLERequestContext::SleepClientThread(const std::string& r
     return event;
 }
 
+HLERequestContext::HLERequestContext() : kernel(Core::Global<KernelSystem>()) {}
+
 HLERequestContext::HLERequestContext(KernelSystem& kernel, std::shared_ptr<ServerSession> session,
-                                     Thread* thread)
+                                     std::shared_ptr<Thread> thread)
     : kernel(kernel), session(std::move(session)), thread(thread) {
     cmd_buf[0] = 0;
 }
@@ -98,8 +122,9 @@ void HLERequestContext::AddStaticBuffer(u8 buffer_id, std::vector<u8> data) {
     static_buffers[buffer_id] = std::move(data);
 }
 
-ResultCode HLERequestContext::PopulateFromIncomingCommandBuffer(const u32_le* src_cmdbuf,
-                                                                Process& src_process) {
+ResultCode HLERequestContext::PopulateFromIncomingCommandBuffer(
+    const u32_le* src_cmdbuf, std::shared_ptr<Process> src_process_) {
+    auto& src_process = *src_process_;
     IPC::Header header{src_cmdbuf[0]};
 
     std::size_t untranslated_size = 1u + header.normal_params_size;
@@ -158,7 +183,7 @@ ResultCode HLERequestContext::PopulateFromIncomingCommandBuffer(const u32_le* sr
         }
         case IPC::DescriptorType::MappedBuffer: {
             u32 next_id = static_cast<u32>(request_mapped_buffers.size());
-            request_mapped_buffers.emplace_back(kernel.memory, src_process, descriptor,
+            request_mapped_buffers.emplace_back(kernel.memory, src_process_, descriptor,
                                                 src_cmdbuf[i], next_id);
             cmd_buf[i++] = next_id;
             break;
@@ -170,7 +195,7 @@ ResultCode HLERequestContext::PopulateFromIncomingCommandBuffer(const u32_le* sr
 
     if (should_record) {
         std::vector<u32> translated_cmdbuf{cmd_buf.begin(), cmd_buf.begin() + command_size};
-        kernel.GetIPCRecorder().SetRequestInfo(SharedFrom(thread), std::move(untranslated_cmdbuf),
+        kernel.GetIPCRecorder().SetRequestInfo(thread, std::move(untranslated_cmdbuf),
                                                std::move(translated_cmdbuf));
     }
 
@@ -248,7 +273,7 @@ ResultCode HLERequestContext::WriteToOutgoingCommandBuffer(u32_le* dst_cmdbuf,
 
     if (should_record) {
         std::vector<u32> translated_cmdbuf{dst_cmdbuf, dst_cmdbuf + command_size};
-        kernel.GetIPCRecorder().SetReplyInfo(SharedFrom(thread), std::move(untranslated_cmdbuf),
+        kernel.GetIPCRecorder().SetReplyInfo(thread, std::move(untranslated_cmdbuf),
                                              std::move(translated_cmdbuf));
     }
 
@@ -262,13 +287,15 @@ MappedBuffer& HLERequestContext::GetMappedBuffer(u32 id_from_cmdbuf) {
 
 void HLERequestContext::ReportUnimplemented() const {
     if (kernel.GetIPCRecorder().IsEnabled()) {
-        kernel.GetIPCRecorder().SetHLEUnimplemented(SharedFrom(thread));
+        kernel.GetIPCRecorder().SetHLEUnimplemented(thread);
     }
 }
 
-MappedBuffer::MappedBuffer(Memory::MemorySystem& memory, const Process& process, u32 descriptor,
-                           VAddr address, u32 id)
-    : memory(&memory), id(id), address(address), process(&process) {
+MappedBuffer::MappedBuffer() : memory(&Core::Global<Core::System>().Memory()) {}
+
+MappedBuffer::MappedBuffer(Memory::MemorySystem& memory, std::shared_ptr<Process> process,
+                           u32 descriptor, VAddr address, u32 id)
+    : memory(&memory), id(id), address(address), process(process) {
     IPC::MappedBufferDescInfo desc{descriptor};
     size = desc.size;
     perms = desc.perms;
@@ -287,3 +314,5 @@ void MappedBuffer::Write(const void* src_buffer, std::size_t offset, std::size_t
 }
 
 } // namespace Kernel
+
+SERIALIZE_EXPORT_IMPL(Kernel::HLERequestContext::ThreadCallback)
diff --git a/src/core/hle/kernel/hle_ipc.h b/src/core/hle/kernel/hle_ipc.h
index 2177b733ea..56c6d8ce1c 100644
--- a/src/core/hle/kernel/hle_ipc.h
+++ b/src/core/hle/kernel/hle_ipc.h
@@ -16,6 +16,7 @@
 #include <boost/serialization/unique_ptr.hpp>
 #include <boost/serialization/vector.hpp>
 #include "common/common_types.h"
+#include "common/serialization/boost_small_vector.hpp"
 #include "common/swap.h"
 #include "core/hle/ipc.h"
 #include "core/hle/kernel/object.h"
@@ -127,7 +128,7 @@ private:
 
 class MappedBuffer {
 public:
-    MappedBuffer(Memory::MemorySystem& memory, const Process& process, u32 descriptor,
+    MappedBuffer(Memory::MemorySystem& memory, std::shared_ptr<Process> process, u32 descriptor,
                  VAddr address, u32 id);
 
     // interface for service
@@ -151,9 +152,21 @@ private:
     Memory::MemorySystem* memory;
     u32 id;
     VAddr address;
-    const Process* process;
-    std::size_t size;
+    std::shared_ptr<Process> process;
+    u32 size;
     IPC::MappedBufferPermissions perms;
+
+    MappedBuffer();
+
+    template <class Archive>
+    void serialize(Archive& ar, const unsigned int) {
+        ar& id;
+        ar& address;
+        ar& process;
+        ar& size;
+        ar& perms;
+    }
+    friend class boost::serialization::access;
 };
 
 /**
@@ -185,9 +198,10 @@ private:
  * id of the memory interface and let kernel convert it back to client vaddr. No real unmapping is
  * needed in this case, though.
  */
-class HLERequestContext {
+class HLERequestContext : std::enable_shared_from_this<HLERequestContext> {
 public:
-    HLERequestContext(KernelSystem& kernel, std::shared_ptr<ServerSession> session, Thread* thread);
+    HLERequestContext(KernelSystem& kernel, std::shared_ptr<ServerSession> session,
+                      std::shared_ptr<Thread> thread);
     ~HLERequestContext();
 
     /// Returns a pointer to the IPC command buffer for this request.
@@ -203,8 +217,12 @@ public:
         return session;
     }
 
-    using WakeupCallback = std::function<void(
-        std::shared_ptr<Thread> thread, HLERequestContext& context, ThreadWakeupReason reason)>;
+    class WakeupCallback {
+    public:
+        virtual ~WakeupCallback() = default;
+        virtual void WakeUp(std::shared_ptr<Thread> thread, HLERequestContext& context,
+                            ThreadWakeupReason reason) = 0;
+    };
 
     /**
      * Puts the specified guest thread to sleep until the returned event is signaled or until the
@@ -219,7 +237,7 @@ public:
      */
     std::shared_ptr<Event> SleepClientThread(const std::string& reason,
                                              std::chrono::nanoseconds timeout,
-                                             WakeupCallback&& callback);
+                                             std::shared_ptr<WakeupCallback> callback);
 
     /**
      * Resolves a object id from the request command buffer into a pointer to an object. See the
@@ -259,26 +277,43 @@ public:
     MappedBuffer& GetMappedBuffer(u32 id_from_cmdbuf);
 
     /// Populates this context with data from the requesting process/thread.
-    ResultCode PopulateFromIncomingCommandBuffer(const u32_le* src_cmdbuf, Process& src_process);
+    ResultCode PopulateFromIncomingCommandBuffer(const u32_le* src_cmdbuf,
+                                                 std::shared_ptr<Process> src_process);
     /// Writes data from this context back to the requesting process/thread.
     ResultCode WriteToOutgoingCommandBuffer(u32_le* dst_cmdbuf, Process& dst_process) const;
 
     /// Reports an unimplemented function.
     void ReportUnimplemented() const;
 
+    class ThreadCallback;
+    friend class ThreadCallback;
+
 private:
     KernelSystem& kernel;
     std::array<u32, IPC::COMMAND_BUFFER_LENGTH> cmd_buf;
     std::shared_ptr<ServerSession> session;
-    Thread* thread;
+    std::shared_ptr<Thread> thread;
     // TODO(yuriks): Check common usage of this and optimize size accordingly
     boost::container::small_vector<std::shared_ptr<Object>, 8> request_handles;
     // The static buffers will be created when the IPC request is translated.
     std::array<std::vector<u8>, IPC::MAX_STATIC_BUFFERS> static_buffers;
     // The mapped buffers will be created when the IPC request is translated
     boost::container::small_vector<MappedBuffer, 8> request_mapped_buffers;
+
+    HLERequestContext();
+    template <class Archive>
+    void serialize(Archive& ar, const unsigned int) {
+        ar& cmd_buf;
+        ar& session;
+        ar& thread;
+        ar& request_handles;
+        ar& static_buffers;
+        ar& request_mapped_buffers;
+    }
+    friend class boost::serialization::access;
 };
 
 } // namespace Kernel
 
 BOOST_CLASS_EXPORT_KEY(Kernel::SessionRequestHandler::SessionDataBase)
+BOOST_CLASS_EXPORT_KEY(Kernel::HLERequestContext::ThreadCallback)
diff --git a/src/core/hle/kernel/ipc.cpp b/src/core/hle/kernel/ipc.cpp
index e39732c149..eb14888404 100644
--- a/src/core/hle/kernel/ipc.cpp
+++ b/src/core/hle/kernel/ipc.cpp
@@ -72,7 +72,7 @@ ResultCode TranslateCommandBuffer(Kernel::KernelSystem& kernel, Memory::MemorySy
                 if (handle == CurrentThread) {
                     object = src_thread;
                 } else if (handle == CurrentProcess) {
-                    object = SharedFrom(src_process);
+                    object = src_process;
                 } else if (handle != 0) {
                     object = src_process->handle_table.GetGeneric(handle);
                     if (descriptor == IPC::DescriptorType::MoveHandle) {
diff --git a/src/core/hle/kernel/ipc_debugger/recorder.cpp b/src/core/hle/kernel/ipc_debugger/recorder.cpp
index 968815c5bd..f1e4a09f16 100644
--- a/src/core/hle/kernel/ipc_debugger/recorder.cpp
+++ b/src/core/hle/kernel/ipc_debugger/recorder.cpp
@@ -52,7 +52,7 @@ void Recorder::RegisterRequest(const std::shared_ptr<Kernel::ClientSession>& cli
 
     RequestRecord record = {/* id */ ++record_count,
                             /* status */ RequestStatus::Sent,
-                            /* client_process */ GetObjectInfo(client_thread->owner_process),
+                            /* client_process */ GetObjectInfo(client_thread->owner_process.get()),
                             /* client_thread */ GetObjectInfo(client_thread.get()),
                             /* client_session */ GetObjectInfo(client_session.get()),
                             /* client_port */ GetObjectInfo(client_session->parent->port.get()),
@@ -82,7 +82,7 @@ void Recorder::SetRequestInfo(const std::shared_ptr<Kernel::Thread>& client_thre
     record.translated_request_cmdbuf = std::move(translated_cmdbuf);
 
     if (server_thread) {
-        record.server_process = GetObjectInfo(server_thread->owner_process);
+        record.server_process = GetObjectInfo(server_thread->owner_process.get());
         record.server_thread = GetObjectInfo(server_thread.get());
     } else {
         record.is_hle = true;
diff --git a/src/core/hle/kernel/kernel.h b/src/core/hle/kernel/kernel.h
index c07bfdbcaf..e7b7314d62 100644
--- a/src/core/hle/kernel/kernel.h
+++ b/src/core/hle/kernel/kernel.h
@@ -134,7 +134,7 @@ public:
      */
     ResultVal<std::shared_ptr<Thread>> CreateThread(std::string name, VAddr entry_point,
                                                     u32 priority, u32 arg, s32 processor_id,
-                                                    VAddr stack_top, Process& owner_process);
+                                                    VAddr stack_top, std::shared_ptr<Process> owner_process);
 
     /**
      * Creates a semaphore.
diff --git a/src/core/hle/kernel/server_session.cpp b/src/core/hle/kernel/server_session.cpp
index 4b393b63d9..8bb82fc839 100644
--- a/src/core/hle/kernel/server_session.cpp
+++ b/src/core/hle/kernel/server_session.cpp
@@ -71,12 +71,12 @@ ResultCode ServerSession::HandleSyncRequest(std::shared_ptr<Thread> thread) {
     // If this ServerSession has an associated HLE handler, forward the request to it.
     if (hle_handler != nullptr) {
         std::array<u32_le, IPC::COMMAND_BUFFER_LENGTH + 2 * IPC::MAX_STATIC_BUFFERS> cmd_buf;
-        Kernel::Process* current_process = thread->owner_process;
+        auto current_process = thread->owner_process;
         kernel.memory.ReadBlock(*current_process, thread->GetCommandBufferAddress(), cmd_buf.data(),
                                 cmd_buf.size() * sizeof(u32));
 
-        Kernel::HLERequestContext context(kernel, SharedFrom(this), thread.get());
-        context.PopulateFromIncomingCommandBuffer(cmd_buf.data(), *current_process);
+        Kernel::HLERequestContext context(kernel, SharedFrom(this), thread);
+        context.PopulateFromIncomingCommandBuffer(cmd_buf.data(), current_process);
 
         hle_handler->HandleSyncRequest(context);
 
diff --git a/src/core/hle/kernel/svc.cpp b/src/core/hle/kernel/svc.cpp
index b5ebaf936e..7f7cc12725 100644
--- a/src/core/hle/kernel/svc.cpp
+++ b/src/core/hle/kernel/svc.cpp
@@ -282,7 +282,7 @@ void SVC::ExitProcess() {
     // Stop all the process threads that are currently waiting for objects.
     auto& thread_list = kernel.GetThreadManager().GetThreadList();
     for (auto& thread : thread_list) {
-        if (thread->owner_process != current_process.get())
+        if (thread->owner_process != current_process)
             continue;
 
         if (thread.get() == kernel.GetThreadManager().GetCurrentThread())
@@ -403,6 +403,73 @@ ResultCode SVC::CloseHandle(Handle handle) {
     return kernel.GetCurrentProcess()->handle_table.Close(handle);
 }
 
+static ResultCode ReceiveIPCRequest(Kernel::KernelSystem& kernel, Memory::MemorySystem& memory,
+                                    std::shared_ptr<ServerSession> server_session,
+                                    std::shared_ptr<Thread> thread);
+
+class SVC_SyncCallback : public Kernel::WakeupCallback {
+public:
+    SVC_SyncCallback(bool do_output_) : do_output(do_output_) {}
+    void WakeUp(ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
+                std::shared_ptr<WaitObject> object) {
+
+        if (reason == ThreadWakeupReason::Timeout) {
+            thread->SetWaitSynchronizationResult(RESULT_TIMEOUT);
+            return;
+        }
+
+        ASSERT(reason == ThreadWakeupReason::Signal);
+
+        thread->SetWaitSynchronizationResult(RESULT_SUCCESS);
+
+        // The wait_all case does not update the output index.
+        if (do_output) {
+            thread->SetWaitSynchronizationOutput(thread->GetWaitObjectIndex(object.get()));
+        }
+    }
+
+private:
+    bool do_output;
+
+    SVC_SyncCallback() = default;
+    template <class Archive>
+    void serialize(Archive& ar, const unsigned int) {
+        ar& do_output;
+    }
+    friend class boost::serialization::access;
+};
+
+class SVC_IPCCallback : public Kernel::WakeupCallback {
+public:
+    SVC_IPCCallback(Core::System& system_) : system(system_) {}
+
+    void WakeUp(ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
+                std::shared_ptr<WaitObject> object) {
+
+        ASSERT(thread->status == ThreadStatus::WaitSynchAny);
+        ASSERT(reason == ThreadWakeupReason::Signal);
+
+        ResultCode result = RESULT_SUCCESS;
+
+        if (object->GetHandleType() == HandleType::ServerSession) {
+            auto server_session = DynamicObjectCast<ServerSession>(object);
+            result = ReceiveIPCRequest(system.Kernel(), system.Memory(), server_session, thread);
+        }
+
+        thread->SetWaitSynchronizationResult(result);
+        thread->SetWaitSynchronizationOutput(thread->GetWaitObjectIndex(object.get()));
+    }
+
+private:
+    Core::System& system;
+
+    SVC_IPCCallback() : system(Core::Global<Core::System>()) {}
+
+    template <class Archive>
+    void serialize(Archive& ar, const unsigned int) {}
+    friend class boost::serialization::access;
+};
+
 /// Wait for a handle to synchronize, timeout after the specified nanoseconds
 ResultCode SVC::WaitSynchronization1(Handle handle, s64 nano_seconds) {
     auto object = kernel.GetCurrentProcess()->handle_table.Get<WaitObject>(handle);
@@ -426,21 +493,7 @@ ResultCode SVC::WaitSynchronization1(Handle handle, s64 nano_seconds) {
         // Create an event to wake the thread up after the specified nanosecond delay has passed
         thread->WakeAfterDelay(nano_seconds);
 
-        thread->wakeup_callback = [](ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
-                                     std::shared_ptr<WaitObject> object) {
-            ASSERT(thread->status == ThreadStatus::WaitSynchAny);
-
-            if (reason == ThreadWakeupReason::Timeout) {
-                thread->SetWaitSynchronizationResult(RESULT_TIMEOUT);
-                return;
-            }
-
-            ASSERT(reason == ThreadWakeupReason::Signal);
-            thread->SetWaitSynchronizationResult(RESULT_SUCCESS);
-
-            // WaitSynchronization1 doesn't have an output index like WaitSynchronizationN, so we
-            // don't have to do anything else here.
-        };
+        thread->wakeup_callback = std::make_shared<SVC_SyncCallback>(false);
 
         system.PrepareReschedule();
 
@@ -515,20 +568,7 @@ ResultCode SVC::WaitSynchronizationN(s32* out, VAddr handles_address, s32 handle
         // Create an event to wake the thread up after the specified nanosecond delay has passed
         thread->WakeAfterDelay(nano_seconds);
 
-        thread->wakeup_callback = [](ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
-                                     std::shared_ptr<WaitObject> object) {
-            ASSERT(thread->status == ThreadStatus::WaitSynchAll);
-
-            if (reason == ThreadWakeupReason::Timeout) {
-                thread->SetWaitSynchronizationResult(RESULT_TIMEOUT);
-                return;
-            }
-
-            ASSERT(reason == ThreadWakeupReason::Signal);
-
-            thread->SetWaitSynchronizationResult(RESULT_SUCCESS);
-            // The wait_all case does not update the output index.
-        };
+        thread->wakeup_callback = std::make_shared<SVC_SyncCallback>(false);
 
         system.PrepareReschedule();
 
@@ -575,20 +615,7 @@ ResultCode SVC::WaitSynchronizationN(s32* out, VAddr handles_address, s32 handle
         // Create an event to wake the thread up after the specified nanosecond delay has passed
         thread->WakeAfterDelay(nano_seconds);
 
-        thread->wakeup_callback = [](ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
-                                     std::shared_ptr<WaitObject> object) {
-            ASSERT(thread->status == ThreadStatus::WaitSynchAny);
-
-            if (reason == ThreadWakeupReason::Timeout) {
-                thread->SetWaitSynchronizationResult(RESULT_TIMEOUT);
-                return;
-            }
-
-            ASSERT(reason == ThreadWakeupReason::Signal);
-
-            thread->SetWaitSynchronizationResult(RESULT_SUCCESS);
-            thread->SetWaitSynchronizationOutput(thread->GetWaitObjectIndex(object.get()));
-        };
+        thread->wakeup_callback = std::make_shared<SVC_SyncCallback>(true);
 
         system.PrepareReschedule();
 
@@ -730,22 +757,7 @@ ResultCode SVC::ReplyAndReceive(s32* index, VAddr handles_address, s32 handle_co
 
     thread->wait_objects = std::move(objects);
 
-    thread->wakeup_callback = [& kernel = this->kernel, &memory = this->memory](
-                                  ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
-                                  std::shared_ptr<WaitObject> object) {
-        ASSERT(thread->status == ThreadStatus::WaitSynchAny);
-        ASSERT(reason == ThreadWakeupReason::Signal);
-
-        ResultCode result = RESULT_SUCCESS;
-
-        if (object->GetHandleType() == HandleType::ServerSession) {
-            auto server_session = DynamicObjectCast<ServerSession>(object);
-            result = ReceiveIPCRequest(kernel, memory, server_session, thread);
-        }
-
-        thread->SetWaitSynchronizationResult(result);
-        thread->SetWaitSynchronizationOutput(thread->GetWaitObjectIndex(object.get()));
-    };
+    thread->wakeup_callback = std::make_shared<SVC_IPCCallback>(system);
 
     system.PrepareReschedule();
 
@@ -911,7 +923,7 @@ ResultCode SVC::CreateThread(Handle* out_handle, u32 entry_point, u32 arg, VAddr
 
     CASCADE_RESULT(std::shared_ptr<Thread> thread,
                    kernel.CreateThread(name, entry_point, priority, arg, processor_id, stack_top,
-                                       *current_process));
+                                       current_process));
 
     thread->context->SetFpscr(FPSCR_DEFAULT_NAN | FPSCR_FLUSH_TO_ZERO |
                               FPSCR_ROUND_TOZERO); // 0x03C00000
@@ -1020,7 +1032,7 @@ ResultCode SVC::GetProcessIdOfThread(u32* process_id, Handle thread_handle) {
     if (thread == nullptr)
         return ERR_INVALID_HANDLE;
 
-    const std::shared_ptr<Process> process = SharedFrom(thread->owner_process);
+    const std::shared_ptr<Process> process = thread->owner_process;
 
     ASSERT_MSG(process != nullptr, "Invalid parent process for thread={:#010X}", thread_handle);
 
@@ -1611,3 +1623,6 @@ void SVCContext::CallSVC(u32 immediate) {
 }
 
 } // namespace Kernel
+
+SERIALIZE_EXPORT_IMPL(Kernel::SVC_SyncCallback)
+SERIALIZE_EXPORT_IMPL(Kernel::SVC_IPCCallback)
diff --git a/src/core/hle/kernel/svc.h b/src/core/hle/kernel/svc.h
index efddff9e88..4cf2654001 100644
--- a/src/core/hle/kernel/svc.h
+++ b/src/core/hle/kernel/svc.h
@@ -5,6 +5,7 @@
 #pragma once
 
 #include <memory>
+#include <boost/serialization/export.hpp>
 #include "common/common_types.h"
 
 namespace Core {
@@ -25,4 +26,10 @@ private:
     std::unique_ptr<SVC> impl;
 };
 
+class SVC_SyncCallback;
+class SVC_IPCCallback;
+
 } // namespace Kernel
+
+BOOST_CLASS_EXPORT_KEY(Kernel::SVC_SyncCallback)
+BOOST_CLASS_EXPORT_KEY(Kernel::SVC_IPCCallback)
diff --git a/src/core/hle/kernel/thread.cpp b/src/core/hle/kernel/thread.cpp
index 1af2d4f4d6..13a3017a53 100644
--- a/src/core/hle/kernel/thread.cpp
+++ b/src/core/hle/kernel/thread.cpp
@@ -48,7 +48,7 @@ void Thread::serialize(Archive& ar, const unsigned int file_version) {
     ar& wait_objects;
     ar& wait_address;
     ar& name;
-    // TODO: How the hell to do wakeup_callback
+    ar& wakeup_callback;
 }
 
 SERIALIZE_IMPL(Thread)
@@ -138,8 +138,8 @@ void ThreadManager::SwitchContext(Thread* new_thread) {
         ready_queue.remove(new_thread->current_priority, new_thread);
         new_thread->status = ThreadStatus::Running;
 
-        if (previous_process.get() != current_thread->owner_process) {
-            kernel.SetCurrentProcess(SharedFrom(current_thread->owner_process));
+        if (previous_process != current_thread->owner_process) {
+            kernel.SetCurrentProcess(current_thread->owner_process);
         }
 
         cpu->LoadContext(new_thread->context);
@@ -196,7 +196,7 @@ void ThreadManager::ThreadWakeupCallback(u64 thread_id, s64 cycles_late) {
 
         // Invoke the wakeup callback before clearing the wait objects
         if (thread->wakeup_callback)
-            thread->wakeup_callback(ThreadWakeupReason::Timeout, thread, nullptr);
+            thread->wakeup_callback->WakeUp(ThreadWakeupReason::Timeout, thread, nullptr);
 
         // Remove the thread from each of its waiting objects' waitlists
         for (auto& object : thread->wait_objects)
@@ -313,10 +313,9 @@ static void ResetThreadContext(const std::unique_ptr<ARM_Interface::ThreadContex
     context->SetCpsr(USER32MODE | ((entry_point & 1) << 5)); // Usermode and THUMB mode
 }
 
-ResultVal<std::shared_ptr<Thread>> KernelSystem::CreateThread(std::string name, VAddr entry_point,
-                                                              u32 priority, u32 arg,
-                                                              s32 processor_id, VAddr stack_top,
-                                                              Process& owner_process) {
+ResultVal<std::shared_ptr<Thread>> KernelSystem::CreateThread(
+    std::string name, VAddr entry_point, u32 priority, u32 arg, s32 processor_id, VAddr stack_top,
+    std::shared_ptr<Process> owner_process) {
     // Check if priority is in ranged. Lowest priority -> highest priority id.
     if (priority > ThreadPrioLowest) {
         LOG_ERROR(Kernel_SVC, "Invalid thread priority: {}", priority);
@@ -330,7 +329,7 @@ ResultVal<std::shared_ptr<Thread>> KernelSystem::CreateThread(std::string name,
 
     // TODO(yuriks): Other checks, returning 0xD9001BEA
 
-    if (!Memory::IsValidVirtualAddress(owner_process, entry_point)) {
+    if (!Memory::IsValidVirtualAddress(*owner_process, entry_point)) {
         LOG_ERROR(Kernel_SVC, "(name={}): invalid entry {:08x}", name, entry_point);
         // TODO: Verify error
         return ResultCode(ErrorDescription::InvalidAddress, ErrorModule::Kernel,
@@ -353,10 +352,10 @@ ResultVal<std::shared_ptr<Thread>> KernelSystem::CreateThread(std::string name,
     thread->wait_address = 0;
     thread->name = std::move(name);
     thread_manager->wakeup_callback_table[thread->thread_id] = thread.get();
-    thread->owner_process = &owner_process;
+    thread->owner_process = owner_process;
 
     // Find the next available TLS index, and mark it as used
-    auto& tls_slots = owner_process.tls_slots;
+    auto& tls_slots = owner_process->tls_slots;
 
     auto [available_page, available_slot, needs_allocation] = GetFreeThreadLocalSlot(tls_slots);
 
@@ -372,13 +371,13 @@ ResultVal<std::shared_ptr<Thread>> KernelSystem::CreateThread(std::string name,
                       "Not enough space in region to allocate a new TLS page for thread");
             return ERR_OUT_OF_MEMORY;
         }
-        owner_process.memory_used += Memory::PAGE_SIZE;
+        owner_process->memory_used += Memory::PAGE_SIZE;
 
         tls_slots.emplace_back(0); // The page is completely available at the start
         available_page = tls_slots.size() - 1;
         available_slot = 0; // Use the first slot in the new page
 
-        auto& vm_manager = owner_process.vm_manager;
+        auto& vm_manager = owner_process->vm_manager;
 
         // Map the page to the current process' address space.
         vm_manager.MapBackingMemory(Memory::TLS_AREA_VADDR + available_page * Memory::PAGE_SIZE,
@@ -391,7 +390,7 @@ ResultVal<std::shared_ptr<Thread>> KernelSystem::CreateThread(std::string name,
     thread->tls_address = Memory::TLS_AREA_VADDR + available_page * Memory::PAGE_SIZE +
                           available_slot * Memory::TLS_ENTRY_SIZE;
 
-    memory.ZeroBlock(owner_process, thread->tls_address, Memory::TLS_ENTRY_SIZE);
+    memory.ZeroBlock(*owner_process, thread->tls_address, Memory::TLS_ENTRY_SIZE);
 
     // TODO(peachum): move to ScheduleThread() when scheduler is added so selected core is used
     // to initialize the context
@@ -438,7 +437,7 @@ std::shared_ptr<Thread> SetupMainThread(KernelSystem& kernel, u32 entry_point, u
     // Initialize new "main" thread
     auto thread_res =
         kernel.CreateThread("main", entry_point, priority, 0, owner_process->ideal_processor,
-                            Memory::HEAP_VADDR_END, *owner_process);
+                            Memory::HEAP_VADDR_END, owner_process);
 
     std::shared_ptr<Thread> thread = std::move(thread_res).Unwrap();
 
diff --git a/src/core/hle/kernel/thread.h b/src/core/hle/kernel/thread.h
index 219db9f2d9..9941c76db5 100644
--- a/src/core/hle/kernel/thread.h
+++ b/src/core/hle/kernel/thread.h
@@ -59,6 +59,15 @@ enum class ThreadWakeupReason {
     Timeout // The thread was woken up due to a wait timeout.
 };
 
+class Thread;
+
+class WakeupCallback {
+public:
+    virtual ~WakeupCallback() = default;
+    virtual void WakeUp(ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
+                        std::shared_ptr<WaitObject> object) = 0;
+};
+
 class ThreadManager {
 public:
     explicit ThreadManager(Kernel::KernelSystem& kernel);
@@ -300,7 +309,7 @@ public:
     /// Mutexes that this thread is currently waiting for.
     boost::container::flat_set<std::shared_ptr<Mutex>> pending_mutexes;
 
-    Process* owner_process; ///< Process that owns this thread
+    std::shared_ptr<Process> owner_process; ///< Process that owns this thread
 
     /// Objects that the thread is waiting on, in the same order as they were
     // passed to WaitSynchronization1/N.
@@ -310,12 +319,10 @@ public:
 
     std::string name;
 
-    using WakeupCallback = void(ThreadWakeupReason reason, std::shared_ptr<Thread> thread,
-                                std::shared_ptr<WaitObject> object);
     // Callback that will be invoked when the thread is resumed from a waiting state. If the thread
     // was waiting via WaitSynchronizationN then the object will be the last object that became
     // available. In case of a timeout, the object will be nullptr.
-    std::function<WakeupCallback> wakeup_callback;
+    std::shared_ptr<WakeupCallback> wakeup_callback;
 
 private:
     ThreadManager& thread_manager;
diff --git a/src/core/hle/kernel/wait_object.cpp b/src/core/hle/kernel/wait_object.cpp
index 94ae632cde..6fe75b6d82 100644
--- a/src/core/hle/kernel/wait_object.cpp
+++ b/src/core/hle/kernel/wait_object.cpp
@@ -80,7 +80,7 @@ void WaitObject::WakeupAllWaitingThreads() {
 
         // Invoke the wakeup callback before clearing the wait objects
         if (thread->wakeup_callback)
-            thread->wakeup_callback(ThreadWakeupReason::Signal, thread, SharedFrom(this));
+            thread->wakeup_callback->WakeUp(ThreadWakeupReason::Signal, thread, SharedFrom(this));
 
         for (auto& object : thread->wait_objects)
             object->RemoveWaitingThread(thread.get());
diff --git a/src/core/hle/service/fs/file.cpp b/src/core/hle/service/fs/file.cpp
index 946e27211c..c8727314b8 100644
--- a/src/core/hle/service/fs/file.cpp
+++ b/src/core/hle/service/fs/file.cpp
@@ -71,12 +71,7 @@ void File::Read(Kernel::HLERequestContext& ctx) {
     rb.PushMappedBuffer(buffer);
 
     std::chrono::nanoseconds read_timeout_ns{backend->GetReadDelayNs(length)};
-    ctx.SleepClientThread("file::read", read_timeout_ns,
-                          [](std::shared_ptr<Kernel::Thread> /*thread*/,
-                             Kernel::HLERequestContext& /*ctx*/,
-                             Kernel::ThreadWakeupReason /*reason*/) {
-                              // Nothing to do here
-                          });
+    ctx.SleepClientThread("file::read", read_timeout_ns, nullptr);
 }
 
 void File::Write(Kernel::HLERequestContext& ctx) {
diff --git a/src/core/hle/service/fs/fs_user.cpp b/src/core/hle/service/fs/fs_user.cpp
index e027e837fe..010600ee2d 100644
--- a/src/core/hle/service/fs/fs_user.cpp
+++ b/src/core/hle/service/fs/fs_user.cpp
@@ -76,12 +76,7 @@ void FS_USER::OpenFile(Kernel::HLERequestContext& ctx) {
         LOG_ERROR(Service_FS, "failed to get a handle for file {}", file_path.DebugStr());
     }
 
-    ctx.SleepClientThread("fs_user::open", open_timeout_ns,
-                          [](std::shared_ptr<Kernel::Thread> /*thread*/,
-                             Kernel::HLERequestContext& /*ctx*/,
-                             Kernel::ThreadWakeupReason /*reason*/) {
-                              // Nothing to do here
-                          });
+    ctx.SleepClientThread("fs_user::open", open_timeout_ns, nullptr);
 }
 
 void FS_USER::OpenFileDirectly(Kernel::HLERequestContext& ctx) {
@@ -134,12 +129,7 @@ void FS_USER::OpenFileDirectly(Kernel::HLERequestContext& ctx) {
                   file_path.DebugStr(), mode.hex, attributes);
     }
 
-    ctx.SleepClientThread("fs_user::open_directly", open_timeout_ns,
-                          [](std::shared_ptr<Kernel::Thread> /*thread*/,
-                             Kernel::HLERequestContext& /*ctx*/,
-                             Kernel::ThreadWakeupReason /*reason*/) {
-                              // Nothing to do here
-                          });
+    ctx.SleepClientThread("fs_user::open_directly", open_timeout_ns, nullptr);
 }
 
 void FS_USER::DeleteFile(Kernel::HLERequestContext& ctx) {
diff --git a/src/core/hle/service/nwm/nwm_uds.cpp b/src/core/hle/service/nwm/nwm_uds.cpp
index dddf55333e..eef8cbd1d3 100644
--- a/src/core/hle/service/nwm/nwm_uds.cpp
+++ b/src/core/hle/service/nwm/nwm_uds.cpp
@@ -1170,6 +1170,29 @@ void NWM_UDS::GetChannel(Kernel::HLERequestContext& ctx) {
     LOG_DEBUG(Service_NWM, "called");
 }
 
+class NWM_UDS::ThreadCallback : public Kernel::HLERequestContext::WakeupCallback {
+public:
+    ThreadCallback(u16 command_id_) : command_id(command_id_) {}
+
+    void WakeUp(std::shared_ptr<Kernel::Thread> thread, Kernel::HLERequestContext& ctx,
+                Kernel::ThreadWakeupReason reason) {
+        // TODO(B3N30): Add error handling for host full and timeout
+        IPC::RequestBuilder rb(ctx, command_id, 1, 0);
+        rb.Push(RESULT_SUCCESS);
+        LOG_DEBUG(Service_NWM, "connection sequence finished");
+    }
+
+private:
+    ThreadCallback() = default;
+    u16 command_id;
+
+    template <class Archive>
+    void serialize(Archive& ar, const unsigned int) {
+        ar& command_id;
+    }
+    friend class boost::serialization::access;
+};
+
 void NWM_UDS::ConnectToNetwork(Kernel::HLERequestContext& ctx, u16 command_id,
                                const u8* network_info_buffer, std::size_t network_info_size,
                                u8 connection_type, std::vector<u8> passphrase) {
@@ -1183,15 +1206,8 @@ void NWM_UDS::ConnectToNetwork(Kernel::HLERequestContext& ctx, u16 command_id,
     // Since this timing is handled by core_timing it could differ from the 'real world' time
     static constexpr std::chrono::nanoseconds UDSConnectionTimeout{300000000};
 
-    connection_event = ctx.SleepClientThread(
-        "uds::ConnectToNetwork", UDSConnectionTimeout,
-        [command_id](std::shared_ptr<Kernel::Thread> thread, Kernel::HLERequestContext& ctx,
-                     Kernel::ThreadWakeupReason reason) {
-            // TODO(B3N30): Add error handling for host full and timeout
-            IPC::RequestBuilder rb(ctx, command_id, 1, 0);
-            rb.Push(RESULT_SUCCESS);
-            LOG_DEBUG(Service_NWM, "connection sequence finished");
-        });
+    connection_event = ctx.SleepClientThread("uds::ConnectToNetwork", UDSConnectionTimeout,
+                                             std::make_shared<ThreadCallback>(command_id));
 }
 
 void NWM_UDS::ConnectToNetwork(Kernel::HLERequestContext& ctx) {
@@ -1418,3 +1434,5 @@ NWM_UDS::~NWM_UDS() {
 }
 
 } // namespace Service::NWM
+
+SERIALIZE_EXPORT_IMPL(Service::NWM::NWM_UDS::ThreadCallback)
diff --git a/src/core/hle/service/nwm/nwm_uds.h b/src/core/hle/service/nwm/nwm_uds.h
index 07dd7e9ba1..b2ab1d76a6 100644
--- a/src/core/hle/service/nwm/nwm_uds.h
+++ b/src/core/hle/service/nwm/nwm_uds.h
@@ -15,6 +15,7 @@
 #include <unordered_map>
 #include <vector>
 #include <boost/optional.hpp>
+#include <boost/serialization/export.hpp>
 #include "common/common_types.h"
 #include "common/swap.h"
 #include "core/hle/service/service.h"
@@ -127,6 +128,8 @@ public:
     explicit NWM_UDS(Core::System& system);
     ~NWM_UDS();
 
+    class ThreadCallback;
+
 private:
     Core::System& system;
 
@@ -560,3 +563,4 @@ private:
 
 SERVICE_CONSTRUCT(Service::NWM::NWM_UDS)
 BOOST_CLASS_EXPORT_KEY(Service::NWM::NWM_UDS)
+BOOST_CLASS_EXPORT_KEY(Service::NWM::NWM_UDS::ThreadCallback)
diff --git a/src/core/hle/service/sm/srv.cpp b/src/core/hle/service/sm/srv.cpp
index 396bd3559e..10179f4163 100644
--- a/src/core/hle/service/sm/srv.cpp
+++ b/src/core/hle/service/sm/srv.cpp
@@ -3,6 +3,7 @@
 // Refer to the license.txt file included.
 
 #include <tuple>
+#include "common/archives.h"
 #include "common/common_types.h"
 #include "common/logging/log.h"
 #include "core/core.h"
@@ -71,6 +72,46 @@ void SRV::EnableNotification(Kernel::HLERequestContext& ctx) {
     LOG_WARNING(Service_SRV, "(STUBBED) called");
 }
 
+class SRV::ThreadCallback : public Kernel::HLERequestContext::WakeupCallback {
+
+public:
+    ThreadCallback(Core::System& system_, std::string name_) : system(system_), name(name_) {}
+
+    void WakeUp(std::shared_ptr<Kernel::Thread> thread, Kernel::HLERequestContext& ctx,
+                Kernel::ThreadWakeupReason reason) {
+        LOG_ERROR(Service_SRV, "called service={} wakeup", name);
+        auto client_port = system.ServiceManager().GetServicePort(name);
+
+        auto session = client_port.Unwrap()->Connect();
+        if (session.Succeeded()) {
+            LOG_DEBUG(Service_SRV, "called service={} -> session={}", name,
+                      (*session)->GetObjectId());
+            IPC::RequestBuilder rb(ctx, 0x5, 1, 2);
+            rb.Push(session.Code());
+            rb.PushMoveObjects(std::move(session).Unwrap());
+        } else if (session.Code() == Kernel::ERR_MAX_CONNECTIONS_REACHED) {
+            LOG_ERROR(Service_SRV, "called service={} -> ERR_MAX_CONNECTIONS_REACHED", name);
+            UNREACHABLE();
+        } else {
+            LOG_ERROR(Service_SRV, "called service={} -> error 0x{:08X}", name, session.Code().raw);
+            IPC::RequestBuilder rb(ctx, 0x5, 1, 0);
+            rb.Push(session.Code());
+        }
+    }
+
+private:
+    Core::System& system;
+    std::string name;
+
+    ThreadCallback() : system(Core::Global<Core::System>()) {}
+
+    template <class Archive>
+    void serialize(Archive& ar, const unsigned int) {
+        ar& name;
+    }
+    friend class boost::serialization::access;
+};
+
 /**
  * SRV::GetServiceHandle service function
  *  Inputs:
@@ -100,28 +141,7 @@ void SRV::GetServiceHandle(Kernel::HLERequestContext& ctx) {
 
     // TODO(yuriks): Permission checks go here
 
-    auto get_handle = [name, this](std::shared_ptr<Kernel::Thread> thread,
-                                   Kernel::HLERequestContext& ctx,
-                                   Kernel::ThreadWakeupReason reason) {
-        LOG_ERROR(Service_SRV, "called service={} wakeup", name);
-        auto client_port = system.ServiceManager().GetServicePort(name);
-
-        auto session = client_port.Unwrap()->Connect();
-        if (session.Succeeded()) {
-            LOG_DEBUG(Service_SRV, "called service={} -> session={}", name,
-                      (*session)->GetObjectId());
-            IPC::RequestBuilder rb(ctx, 0x5, 1, 2);
-            rb.Push(session.Code());
-            rb.PushMoveObjects(std::move(session).Unwrap());
-        } else if (session.Code() == Kernel::ERR_MAX_CONNECTIONS_REACHED) {
-            LOG_ERROR(Service_SRV, "called service={} -> ERR_MAX_CONNECTIONS_REACHED", name);
-            UNREACHABLE();
-        } else {
-            LOG_ERROR(Service_SRV, "called service={} -> error 0x{:08X}", name, session.Code().raw);
-            IPC::RequestBuilder rb(ctx, 0x5, 1, 0);
-            rb.Push(session.Code());
-        }
-    };
+    auto get_handle = std::make_shared<ThreadCallback>(system, name);
 
     auto client_port = system.ServiceManager().GetServicePort(name);
     if (client_port.Failed()) {
@@ -266,3 +286,5 @@ SRV::SRV(Core::System& system) : ServiceFramework("srv:", 4), system(system) {
 SRV::~SRV() = default;
 
 } // namespace Service::SM
+
+SERIALIZE_EXPORT_IMPL(Service::SM::SRV::ThreadCallback)
diff --git a/src/core/hle/service/sm/srv.h b/src/core/hle/service/sm/srv.h
index 2382f48425..7d17f87a58 100644
--- a/src/core/hle/service/sm/srv.h
+++ b/src/core/hle/service/sm/srv.h
@@ -6,6 +6,7 @@
 
 #include <memory>
 #include <unordered_map>
+#include <boost/serialization/export.hpp>
 #include "core/hle/service/service.h"
 
 namespace Core {
@@ -25,6 +26,8 @@ public:
     explicit SRV(Core::System& system);
     ~SRV();
 
+    class ThreadCallback;
+
 private:
     void RegisterClient(Kernel::HLERequestContext& ctx);
     void EnableNotification(Kernel::HLERequestContext& ctx);
@@ -40,3 +43,5 @@ private:
 };
 
 } // namespace Service::SM
+
+BOOST_CLASS_EXPORT_KEY(Service::SM::SRV::ThreadCallback)
diff --git a/src/tests/core/hle/kernel/hle_ipc.cpp b/src/tests/core/hle/kernel/hle_ipc.cpp
index 414d64021f..a4f7c8062c 100644
--- a/src/tests/core/hle/kernel/hle_ipc.cpp
+++ b/src/tests/core/hle/kernel/hle_ipc.cpp
@@ -37,7 +37,7 @@ TEST_CASE("HLERequestContext::PopulateFromIncomingCommandBuffer", "[core][kernel
             IPC::MakeHeader(0x1234, 0, 0),
         };
 
-        context.PopulateFromIncomingCommandBuffer(input, *process);
+        context.PopulateFromIncomingCommandBuffer(input, process);
 
         REQUIRE(context.CommandBuffer()[0] == 0x12340000);
     }
@@ -50,7 +50,7 @@ TEST_CASE("HLERequestContext::PopulateFromIncomingCommandBuffer", "[core][kernel
             0xAABBCCDD,
         };
 
-        context.PopulateFromIncomingCommandBuffer(input, *process);
+        context.PopulateFromIncomingCommandBuffer(input, process);
 
         auto* output = context.CommandBuffer();
         REQUIRE(output[1] == 0x12345678);
@@ -67,7 +67,7 @@ TEST_CASE("HLERequestContext::PopulateFromIncomingCommandBuffer", "[core][kernel
             a_handle,
         };
 
-        context.PopulateFromIncomingCommandBuffer(input, *process);
+        context.PopulateFromIncomingCommandBuffer(input, process);
 
         auto* output = context.CommandBuffer();
         REQUIRE(context.GetIncomingHandle(output[2]) == a);
@@ -83,7 +83,7 @@ TEST_CASE("HLERequestContext::PopulateFromIncomingCommandBuffer", "[core][kernel
             a_handle,
         };
 
-        context.PopulateFromIncomingCommandBuffer(input, *process);
+        context.PopulateFromIncomingCommandBuffer(input, process);
 
         auto* output = context.CommandBuffer();
         REQUIRE(context.GetIncomingHandle(output[2]) == a);
@@ -103,7 +103,7 @@ TEST_CASE("HLERequestContext::PopulateFromIncomingCommandBuffer", "[core][kernel
             process->handle_table.Create(c).Unwrap(),
         };
 
-        context.PopulateFromIncomingCommandBuffer(input, *process);
+        context.PopulateFromIncomingCommandBuffer(input, process);
 
         auto* output = context.CommandBuffer();
         REQUIRE(context.GetIncomingHandle(output[2]) == a);
@@ -118,7 +118,7 @@ TEST_CASE("HLERequestContext::PopulateFromIncomingCommandBuffer", "[core][kernel
             0,
         };
 
-        auto result = context.PopulateFromIncomingCommandBuffer(input, *process);
+        auto result = context.PopulateFromIncomingCommandBuffer(input, process);
 
         REQUIRE(result == RESULT_SUCCESS);
         auto* output = context.CommandBuffer();
@@ -132,7 +132,7 @@ TEST_CASE("HLERequestContext::PopulateFromIncomingCommandBuffer", "[core][kernel
             0x98989898,
         };
 
-        context.PopulateFromIncomingCommandBuffer(input, *process);
+        context.PopulateFromIncomingCommandBuffer(input, process);
 
         REQUIRE(context.CommandBuffer()[2] == process->process_id);
     }
@@ -153,7 +153,7 @@ TEST_CASE("HLERequestContext::PopulateFromIncomingCommandBuffer", "[core][kernel
             target_address,
         };
 
-        context.PopulateFromIncomingCommandBuffer(input, *process);
+        context.PopulateFromIncomingCommandBuffer(input, process);
 
         CHECK(context.GetStaticBuffer(0) == mem->Vector());
 
@@ -175,7 +175,7 @@ TEST_CASE("HLERequestContext::PopulateFromIncomingCommandBuffer", "[core][kernel
             target_address,
         };
 
-        context.PopulateFromIncomingCommandBuffer(input, *process);
+        context.PopulateFromIncomingCommandBuffer(input, process);
 
         std::vector<u8> other_buffer(buffer.GetSize());
         context.GetMappedBuffer(0).Read(other_buffer.data(), 0, buffer.GetSize());
@@ -219,7 +219,7 @@ TEST_CASE("HLERequestContext::PopulateFromIncomingCommandBuffer", "[core][kernel
             target_address_mapped,
         };
 
-        context.PopulateFromIncomingCommandBuffer(input, *process);
+        context.PopulateFromIncomingCommandBuffer(input, process);
 
         auto* output = context.CommandBuffer();
         CHECK(output[1] == 0x12345678);
@@ -365,7 +365,7 @@ TEST_CASE("HLERequestContext::WriteToOutgoingCommandBuffer", "[core][kernel]") {
             target_address,
         };
 
-        context.PopulateFromIncomingCommandBuffer(input_cmdbuff, *process);
+        context.PopulateFromIncomingCommandBuffer(input_cmdbuff, process);
 
         context.GetMappedBuffer(0).Write(input_buffer.data(), 0, input_buffer.size());