Kernel: Start using boost::intrusive_ptr for lifetime management

This commit is contained in:
Yuri Kunde Schlesner 2014-12-29 10:55:30 -02:00
parent 934343368f
commit 6480d24a8c
12 changed files with 70 additions and 63 deletions

2
externals/boost vendored

@ -1 +1 @@
Subproject commit 97052c28acb141dbf3c5e14114af99045344b695
Subproject commit a1afc91d3aaa3da06bdbc13c78613e1466653405

View file

@ -30,7 +30,7 @@ public:
/// Arbitrate an address
ResultCode ArbitrateAddress(Handle handle, ArbitrationType type, u32 address, s32 value) {
Object* object = Kernel::g_handle_table.GetGeneric(handle);
Object* object = Kernel::g_handle_table.GetGeneric(handle).get();
if (object == nullptr)
return InvalidHandle(ErrorModule::Kernel);

View file

@ -53,7 +53,7 @@ public:
* @return Result of operation, 0 on success, otherwise error code
*/
ResultCode SetPermanentLock(Handle handle, const bool permanent_locked) {
Event* evt = g_handle_table.Get<Event>(handle);
Event* evt = g_handle_table.Get<Event>(handle).get();
if (evt == nullptr) return InvalidHandle(ErrorModule::Kernel);
evt->permanent_locked = permanent_locked;
@ -67,7 +67,7 @@ ResultCode SetPermanentLock(Handle handle, const bool permanent_locked) {
* @return Result of operation, 0 on success, otherwise error code
*/
ResultCode SetEventLocked(const Handle handle, const bool locked) {
Event* evt = g_handle_table.Get<Event>(handle);
Event* evt = g_handle_table.Get<Event>(handle).get();
if (evt == nullptr) return InvalidHandle(ErrorModule::Kernel);
if (!evt->permanent_locked) {
@ -82,13 +82,13 @@ ResultCode SetEventLocked(const Handle handle, const bool locked) {
* @return Result of operation, 0 on success, otherwise error code
*/
ResultCode SignalEvent(const Handle handle) {
Event* evt = g_handle_table.Get<Event>(handle);
Event* evt = g_handle_table.Get<Event>(handle).get();
if (evt == nullptr) return InvalidHandle(ErrorModule::Kernel);
// Resume threads waiting for event to signal
bool event_caught = false;
for (size_t i = 0; i < evt->waiting_threads.size(); ++i) {
Thread* thread = Kernel::g_handle_table.Get<Thread>(evt->waiting_threads[i]);
Thread* thread = Kernel::g_handle_table.Get<Thread>(evt->waiting_threads[i]).get();
if (thread != nullptr)
thread->ResumeFromWait();
@ -112,7 +112,7 @@ ResultCode SignalEvent(const Handle handle) {
* @return Result of operation, 0 on success, otherwise error code
*/
ResultCode ClearEvent(Handle handle) {
Event* evt = g_handle_table.Get<Event>(handle);
Event* evt = g_handle_table.Get<Event>(handle).get();
if (evt == nullptr) return InvalidHandle(ErrorModule::Kernel);
if (!evt->permanent_locked) {

View file

@ -13,7 +13,7 @@
namespace Kernel {
Thread* g_main_thread = nullptr;
smart_ptr<Thread> g_main_thread = nullptr;
HandleTable g_handle_table;
u64 g_program_id = 0;
@ -22,7 +22,7 @@ HandleTable::HandleTable() {
Clear();
}
ResultVal<Handle> HandleTable::Create(Object* obj) {
ResultVal<Handle> HandleTable::Create(smart_ptr<Object> obj) {
_dbg_assert_(Kernel, obj != nullptr);
u16 slot = next_free_slot;
@ -38,22 +38,22 @@ ResultVal<Handle> HandleTable::Create(Object* obj) {
// CTR-OS doesn't use generation 0, so skip straight to 1.
if (next_generation >= (1 << 15)) next_generation = 1;
generations[slot] = generation;
intrusive_ptr_add_ref(obj);
objects[slot] = obj;
Handle handle = generation | (slot << 15);
obj->handle = handle;
generations[slot] = generation;
objects[slot] = std::move(obj);
return MakeResult<Handle>(handle);
}
ResultVal<Handle> HandleTable::Duplicate(Handle handle) {
Object* object = GetGeneric(handle);
smart_ptr<Object> object = GetGeneric(handle);
if (object == nullptr) {
LOG_ERROR(Kernel, "Tried to duplicate invalid handle: %08X", handle);
return ERR_INVALID_HANDLE;
}
return Create(object);
return Create(std::move(object));
}
ResultCode HandleTable::Close(Handle handle) {
@ -63,7 +63,6 @@ ResultCode HandleTable::Close(Handle handle) {
size_t slot = GetSlot(handle);
u16 generation = GetGeneration(handle);
intrusive_ptr_release(objects[slot]);
objects[slot] = nullptr;
generations[generation] = next_free_slot;
@ -78,7 +77,7 @@ bool HandleTable::IsValid(Handle handle) const {
return slot < MAX_COUNT && objects[slot] != nullptr && generations[slot] == generation;
}
Object* HandleTable::GetGeneric(Handle handle) const {
smart_ptr<Object> HandleTable::GetGeneric(Handle handle) const {
if (handle == CurrentThread) {
return GetCurrentThread();
} else if (handle == CurrentProcess) {
@ -95,8 +94,6 @@ Object* HandleTable::GetGeneric(Handle handle) const {
void HandleTable::Clear() {
for (size_t i = 0; i < MAX_COUNT; ++i) {
generations[i] = i + 1;
if (objects[i] != nullptr)
intrusive_ptr_release(objects[i]);
objects[i] = nullptr;
}
next_free_slot = 0;

View file

@ -4,6 +4,8 @@
#pragma once
#include <boost/intrusive_ptr.hpp>
#include <array>
#include <string>
#include "common/common.h"
@ -75,7 +77,7 @@ private:
unsigned int ref_count = 0;
};
// Special functions that will later be used by boost::instrusive_ptr to do automatic ref-counting
// Special functions used by boost::instrusive_ptr to do automatic ref-counting
inline void intrusive_ptr_add_ref(Object* object) {
++object->ref_count;
}
@ -86,6 +88,9 @@ inline void intrusive_ptr_release(Object* object) {
}
}
template <typename T>
using smart_ptr = boost::intrusive_ptr<T>;
/**
* This class allows the creation of Handles, which are references to objects that can be tested
* for validity and looked up. Here they are used to pass references to kernel objects to/from the
@ -118,7 +123,7 @@ public:
* @return The created Handle or one of the following errors:
* - `ERR_OUT_OF_HANDLES`: the maximum number of handles has been exceeded.
*/
ResultVal<Handle> Create(Object* obj);
ResultVal<Handle> Create(smart_ptr<Object> obj);
/**
* Returns a new handle that points to the same object as the passed in handle.
@ -142,7 +147,7 @@ public:
* Looks up a handle.
* @returns Pointer to the looked-up object, or `nullptr` if the handle is not valid.
*/
Object* GetGeneric(Handle handle) const;
smart_ptr<Object> GetGeneric(Handle handle) const;
/**
* Looks up a handle while verifying its type.
@ -150,10 +155,10 @@ public:
* type differs from the handle type `T::HANDLE_TYPE`.
*/
template <class T>
T* Get(Handle handle) const {
Object* object = GetGeneric(handle);
smart_ptr<T> Get(Handle handle) const {
smart_ptr<Object> object = GetGeneric(handle);
if (object != nullptr && object->GetHandleType() == T::HANDLE_TYPE) {
return static_cast<T*>(object);
return boost::static_pointer_cast<T>(object);
}
return nullptr;
}
@ -172,7 +177,7 @@ private:
static u16 GetGeneration(Handle handle) { return handle & 0x7FFF; }
/// Stores the Object referenced by the handle or null if the slot is empty.
std::array<Object*, MAX_COUNT> objects;
std::array<smart_ptr<Object>, MAX_COUNT> objects;
/**
* The value of `next_generation` when the handle was created, used to check for validity. For
@ -191,7 +196,7 @@ private:
};
extern HandleTable g_handle_table;
extern Thread* g_main_thread;
extern smart_ptr<Thread> g_main_thread;
/// The ID code of the currently running game
/// TODO(Subv): This variable should not be here,

View file

@ -48,7 +48,7 @@ void MutexAcquireLock(Mutex* mutex, Handle thread = GetCurrentThread()->GetHandl
bool ReleaseMutexForThread(Mutex* mutex, Handle thread_handle) {
MutexAcquireLock(mutex, thread_handle);
Thread* thread = Kernel::g_handle_table.Get<Thread>(thread_handle);
Thread* thread = Kernel::g_handle_table.Get<Thread>(thread_handle).get();
if (thread != nullptr)
thread->ResumeFromWait();
return true;
@ -90,7 +90,7 @@ void ReleaseThreadMutexes(Handle thread) {
// Release every mutex that the thread holds, and resume execution on the waiting threads
for (MutexMap::iterator iter = locked.first; iter != locked.second; ++iter) {
Mutex* mutex = g_handle_table.Get<Mutex>(iter->second);
Mutex* mutex = g_handle_table.Get<Mutex>(iter->second).get();
ResumeWaitingThread(mutex);
}
@ -118,7 +118,7 @@ bool ReleaseMutex(Mutex* mutex) {
* @param handle Handle to mutex to release
*/
ResultCode ReleaseMutex(Handle handle) {
Mutex* mutex = Kernel::g_handle_table.Get<Mutex>(handle);
Mutex* mutex = Kernel::g_handle_table.Get<Mutex>(handle).get();
if (mutex == nullptr) return InvalidHandle(ErrorModule::Kernel);
if (!ReleaseMutex(mutex)) {

View file

@ -70,7 +70,7 @@ ResultCode CreateSemaphore(Handle* handle, s32 initial_count,
}
ResultCode ReleaseSemaphore(s32* count, Handle handle, s32 release_count) {
Semaphore* semaphore = g_handle_table.Get<Semaphore>(handle);
Semaphore* semaphore = g_handle_table.Get<Semaphore>(handle).get();
if (semaphore == nullptr)
return InvalidHandle(ErrorModule::Kernel);
@ -84,7 +84,7 @@ ResultCode ReleaseSemaphore(s32* count, Handle handle, s32 release_count) {
// Notify some of the threads that the semaphore has been released
// stop once the semaphore is full again or there are no more waiting threads
while (!semaphore->waiting_threads.empty() && semaphore->IsAvailable()) {
Thread* thread = Kernel::g_handle_table.Get<Thread>(semaphore->waiting_threads.front());
Thread* thread = Kernel::g_handle_table.Get<Thread>(semaphore->waiting_threads.front()).get();
if (thread != nullptr)
thread->ResumeFromWait();
semaphore->waiting_threads.pop();

View file

@ -61,7 +61,7 @@ ResultCode MapSharedMemory(u32 handle, u32 address, MemoryPermission permissions
return ResultCode(ErrorDescription::InvalidAddress, ErrorModule::Kernel,
ErrorSummary::InvalidArgument, ErrorLevel::Permanent);
}
SharedMemory* shared_memory = Kernel::g_handle_table.Get<SharedMemory>(handle);
SharedMemory* shared_memory = Kernel::g_handle_table.Get<SharedMemory>(handle).get();
if (shared_memory == nullptr) return InvalidHandle(ErrorModule::Kernel);
shared_memory->base_address = address;
@ -72,7 +72,7 @@ ResultCode MapSharedMemory(u32 handle, u32 address, MemoryPermission permissions
}
ResultVal<u8*> GetSharedMemoryPointer(Handle handle, u32 offset) {
SharedMemory* shared_memory = Kernel::g_handle_table.Get<SharedMemory>(handle);
SharedMemory* shared_memory = Kernel::g_handle_table.Get<SharedMemory>(handle).get();
if (shared_memory == nullptr) return InvalidHandle(ErrorModule::Kernel);
if (0 != shared_memory->base_address)

View file

@ -35,7 +35,7 @@ ResultVal<bool> Thread::WaitSynchronization() {
}
// Lists all thread ids that aren't deleted/etc.
static std::vector<Thread*> thread_queue; // TODO(yuriks): Owned
static std::vector<smart_ptr<Thread>> thread_queue;
// Lists only ready thread ids.
static Common::ThreadQueueList<Thread*> thread_ready_queue;
@ -109,8 +109,8 @@ void Thread::Stop(const char* reason) {
ChangeReadyState(this, false);
status = THREADSTATUS_DORMANT;
for (Thread* waiting_thread : waiting_threads) {
if (CheckWaitType(waiting_thread, WAITTYPE_THREADEND, this))
for (auto& waiting_thread : waiting_threads) {
if (CheckWaitType(waiting_thread.get(), WAITTYPE_THREADEND, this))
waiting_thread->ResumeFromWait();
}
waiting_threads.clear();
@ -142,15 +142,15 @@ Thread* ArbitrateHighestPriorityThread(Object* arbiter, u32 address) {
s32 priority = THREADPRIO_LOWEST;
// Iterate through threads, find highest priority thread that is waiting to be arbitrated...
for (Thread* thread : thread_queue) {
if (!CheckWaitType(thread, WAITTYPE_ARB, arbiter, address))
for (auto& thread : thread_queue) {
if (!CheckWaitType(thread.get(), WAITTYPE_ARB, arbiter, address))
continue;
if (thread == nullptr)
continue; // TODO(yuriks): Thread handle will hang around forever. Should clean up.
if(thread->current_priority <= priority) {
highest_priority_thread = thread;
highest_priority_thread = thread.get();
priority = thread->current_priority;
}
}
@ -167,8 +167,8 @@ Thread* ArbitrateHighestPriorityThread(Object* arbiter, u32 address) {
void ArbitrateAllThreads(Object* arbiter, u32 address) {
// Iterate through threads, find highest priority thread that is waiting to be arbitrated...
for (Thread* thread : thread_queue) {
if (CheckWaitType(thread, WAITTYPE_ARB, arbiter, address))
for (auto& thread : thread_queue) {
if (CheckWaitType(thread.get(), WAITTYPE_ARB, arbiter, address))
thread->ResumeFromWait();
}
}
@ -251,15 +251,15 @@ static void DebugThreadQueue() {
return;
}
LOG_DEBUG(Kernel, "0x%02X 0x%08X (current)", thread->current_priority, GetCurrentThread()->GetHandle());
for (Thread* t : thread_queue) {
s32 priority = thread_ready_queue.contains(t);
for (auto& t : thread_queue) {
s32 priority = thread_ready_queue.contains(t.get());
if (priority != -1) {
LOG_DEBUG(Kernel, "0x%02X 0x%08X", priority, t->GetHandle());
}
}
}
ResultVal<Thread*> Thread::Create(const char* name, u32 entry_point, s32 priority, u32 arg,
ResultVal<smart_ptr<Thread>> Thread::Create(const char* name, u32 entry_point, s32 priority, u32 arg,
s32 processor_id, u32 stack_top, int stack_size) {
_dbg_assert_(Kernel, name != nullptr);
@ -313,7 +313,7 @@ ResultVal<Thread*> Thread::Create(const char* name, u32 entry_point, s32 priorit
ResetThread(thread, arg, 0);
CallThread(thread);
return MakeResult<Thread*>(thread);
return MakeResult<smart_ptr<Thread>>(thread);
}
/// Set the priority of the thread specified by handle
@ -343,13 +343,13 @@ void Thread::SetPriority(s32 priority) {
}
/// Sets up the primary application thread
Thread* SetupMainThread(s32 priority, int stack_size) {
smart_ptr<Thread> SetupMainThread(s32 priority, int stack_size) {
// Initialize new "main" thread
ResultVal<Thread*> thread_res = Thread::Create("main", Core::g_app_core->GetPC(), priority, 0,
ResultVal<smart_ptr<Thread>> thread_res = Thread::Create("main", Core::g_app_core->GetPC(), priority, 0,
THREADPROCESSORID_0, Memory::SCRATCHPAD_VADDR_END, stack_size);
// TODO(yuriks): Propagate error
_dbg_assert_(Kernel, thread_res.Succeeded());
Thread* thread = *thread_res;
smart_ptr<Thread> thread = std::move(*thread_res);
// If running another thread already, set it to "ready" state
Thread* cur = GetCurrentThread();
@ -358,7 +358,7 @@ Thread* SetupMainThread(s32 priority, int stack_size) {
}
// Run new "main" thread
current_thread = thread;
current_thread = thread.get();
thread->status = THREADSTATUS_RUNNING;
Core::g_app_core->LoadContext(thread->context);
@ -378,7 +378,7 @@ void Reschedule() {
} else {
LOG_TRACE(Kernel, "cannot context switch from 0x%08X, no higher priority thread!", prev->GetHandle());
for (Thread* thread : thread_queue) {
for (auto& thread : thread_queue) {
LOG_TRACE(Kernel, "\thandle=0x%08X prio=0x%02X, status=0x%08X wait_type=0x%08X wait_handle=0x%08X",
thread->GetHandle(), thread->current_priority, thread->status, thread->wait_type, thread->wait_object->GetHandle());
}

View file

@ -53,7 +53,7 @@ namespace Kernel {
class Thread : public Kernel::Object {
public:
static ResultVal<Thread*> Create(const char* name, u32 entry_point, s32 priority, u32 arg,
static ResultVal<smart_ptr<Thread>> Create(const char* name, u32 entry_point, s32 priority, u32 arg,
s32 processor_id, u32 stack_top, int stack_size = Kernel::DEFAULT_STACK_SIZE);
std::string GetName() const override { return name; }
@ -97,7 +97,7 @@ public:
Object* wait_object;
VAddr wait_address;
std::vector<Thread*> waiting_threads; // TODO(yuriks): Owned
std::vector<smart_ptr<Thread>> waiting_threads;
std::string name;
@ -106,7 +106,7 @@ private:
};
/// Sets up the primary application thread
Thread* SetupMainThread(s32 priority, int stack_size = Kernel::DEFAULT_STACK_SIZE);
smart_ptr<Thread> SetupMainThread(s32 priority, int stack_size = Kernel::DEFAULT_STACK_SIZE);
/// Reschedules to the next available thread (call after current thread is suspended)
void Reschedule();

View file

@ -71,7 +71,8 @@ void Manager::DeleteService(const std::string& port_name) {
/// Get a Service Interface from its Handle
Interface* Manager::FetchFromHandle(Handle handle) {
return Kernel::g_handle_table.Get<Interface>(handle);
// TODO(yuriks): This function is very suspicious and should probably be exterminated.
return Kernel::g_handle_table.Get<Interface>(handle).get();
}
/// Get a Service Interface from its port

View file

@ -26,6 +26,8 @@
namespace SVC {
using Kernel::smart_ptr;
enum ControlMemoryOperation {
MEMORY_OPERATION_HEAP = 0x00000003,
MEMORY_OPERATION_GSP_HEAP = 0x00010003,
@ -93,7 +95,7 @@ static Result ConnectToPort(Handle* out, const char* port_name) {
/// Synchronize to an OS service
static Result SendSyncRequest(Handle handle) {
Kernel::Session* session = Kernel::g_handle_table.Get<Kernel::Session>(handle);
smart_ptr<Kernel::Session> session = Kernel::g_handle_table.Get<Kernel::Session>(handle);
if (session == nullptr) {
return InvalidHandle(ErrorModule::Kernel).raw;
}
@ -120,7 +122,7 @@ static Result WaitSynchronization1(Handle handle, s64 nano_seconds) {
// TODO(bunnei): Do something with nano_seconds, currently ignoring this
bool wait_infinite = (nano_seconds == -1); // Used to wait until a thread has terminated
Kernel::Object* object = Kernel::g_handle_table.GetGeneric(handle);
smart_ptr<Kernel::Object> object = Kernel::g_handle_table.GetGeneric(handle);
if (object == nullptr)
return InvalidHandle(ErrorModule::Kernel).raw;
@ -149,7 +151,7 @@ static Result WaitSynchronizationN(s32* out, Handle* handles, s32 handle_count,
// Iterate through each handle, synchronize kernel object
for (s32 i = 0; i < handle_count; i++) {
Kernel::Object* object = Kernel::g_handle_table.GetGeneric(handles[i]);
smart_ptr<Kernel::Object> object = Kernel::g_handle_table.GetGeneric(handles[i]);
if (object == nullptr)
return InvalidHandle(ErrorModule::Kernel).raw;
@ -221,6 +223,8 @@ static Result GetResourceLimitCurrentValues(s64* values, Handle resource_limit,
/// Creates a new thread
static Result CreateThread(u32 priority, u32 entry_point, u32 arg, u32 stack_top, u32 processor_id) {
using Kernel::Thread;
std::string name;
if (Symbols::HasSymbol(entry_point)) {
TSymbol symbol = Symbols::GetSymbol(entry_point);
@ -229,11 +233,11 @@ static Result CreateThread(u32 priority, u32 entry_point, u32 arg, u32 stack_top
name = Common::StringFromFormat("unknown-%08x", entry_point);
}
ResultVal<Kernel::Thread*> thread_res = Kernel::Thread::Create(name.c_str(), entry_point, priority, arg,
ResultVal<smart_ptr<Thread>> thread_res = Kernel::Thread::Create(name.c_str(), entry_point, priority, arg,
processor_id, stack_top);
if (thread_res.Failed())
return thread_res.Code().raw;
Kernel::Thread* thread = *thread_res;
smart_ptr<Thread> thread = std::move(*thread_res);
Core::g_app_core->SetReg(1, thread->GetHandle());
@ -254,7 +258,7 @@ static void ExitThread() {
/// Gets the priority for the specified thread
static Result GetThreadPriority(s32* priority, Handle handle) {
const Kernel::Thread* thread = Kernel::g_handle_table.Get<Kernel::Thread>(handle);
const smart_ptr<Kernel::Thread> thread = Kernel::g_handle_table.Get<Kernel::Thread>(handle);
if (thread == nullptr)
return InvalidHandle(ErrorModule::Kernel).raw;
@ -264,7 +268,7 @@ static Result GetThreadPriority(s32* priority, Handle handle) {
/// Sets the priority for the specified thread
static Result SetThreadPriority(Handle handle, s32 priority) {
Kernel::Thread* thread = Kernel::g_handle_table.Get<Kernel::Thread>(handle);
smart_ptr<Kernel::Thread> thread = Kernel::g_handle_table.Get<Kernel::Thread>(handle);
if (thread == nullptr)
return InvalidHandle(ErrorModule::Kernel).raw;
@ -291,7 +295,7 @@ static Result ReleaseMutex(Handle handle) {
static Result GetThreadId(u32* thread_id, Handle handle) {
LOG_TRACE(Kernel_SVC, "called thread=0x%08X", handle);
const Kernel::Thread* thread = Kernel::g_handle_table.Get<Kernel::Thread>(handle);
const smart_ptr<Kernel::Thread> thread = Kernel::g_handle_table.Get<Kernel::Thread>(handle);
if (thread == nullptr)
return InvalidHandle(ErrorModule::Kernel).raw;