kernel: Refactor thread_local variable usage

On MSVC at least, there seems to be a non-trivial overhead to calling GetHostThreadId().

This slightly reworks the host_thread_id variable to reduce some of the complexity around its usage, along with some small refactors around current_thread and dummy thread
This commit is contained in:
ameerj 2023-02-10 20:43:06 -05:00
parent 7bad3a7e5e
commit 5e746da981

View file

@ -367,23 +367,21 @@ struct KernelCore::Impl {
current_process = process;
}
static inline thread_local u32 host_thread_id = UINT32_MAX;
static inline thread_local u8 host_thread_id = UINT8_MAX;
/// Gets the host thread ID for the caller, allocating a new one if this is the first time
u32 GetHostThreadId(std::size_t core_id) {
if (host_thread_id == UINT32_MAX) {
// The first four slots are reserved for CPU core threads
ASSERT(core_id < Core::Hardware::NUM_CPU_CORES);
host_thread_id = static_cast<u32>(core_id);
}
/// Sets the host thread ID for the caller.
u32 SetHostThreadId(std::size_t core_id) {
// This should only be called during core init.
ASSERT(host_thread_id == UINT8_MAX);
// The first four slots are reserved for CPU core threads
ASSERT(core_id < Core::Hardware::NUM_CPU_CORES);
host_thread_id = static_cast<u8>(core_id);
return host_thread_id;
}
/// Gets the host thread ID for the caller, allocating a new one if this is the first time
u32 GetHostThreadId() {
if (host_thread_id == UINT32_MAX) {
host_thread_id = next_host_thread_id++;
}
/// Gets the host thread ID for the caller
u32 GetHostThreadId() const {
return host_thread_id;
}
@ -391,23 +389,19 @@ struct KernelCore::Impl {
KThread* GetHostDummyThread(KThread* existing_thread) {
auto initialize = [this](KThread* thread) {
ASSERT(KThread::InitializeDummyThread(thread, nullptr).IsSuccess());
thread->SetName(fmt::format("DummyThread:{}", GetHostThreadId()));
thread->SetName(fmt::format("DummyThread:{}", next_host_thread_id++));
return thread;
};
thread_local KThread raw_thread{system.Kernel()};
thread_local KThread* thread = nullptr;
if (thread == nullptr) {
thread = (existing_thread == nullptr) ? initialize(&raw_thread) : existing_thread;
}
thread_local KThread* thread = existing_thread ? existing_thread : initialize(&raw_thread);
return thread;
}
/// Registers a CPU core thread by allocating a host thread ID for it
void RegisterCoreThread(std::size_t core_id) {
ASSERT(core_id < Core::Hardware::NUM_CPU_CORES);
const auto this_id = GetHostThreadId(core_id);
const auto this_id = SetHostThreadId(core_id);
if (!is_multicore) {
single_core_thread_id = this_id;
}
@ -415,7 +409,6 @@ struct KernelCore::Impl {
/// Registers a new host thread by allocating a host thread ID for it
void RegisterHostThread(KThread* existing_thread) {
[[maybe_unused]] const auto this_id = GetHostThreadId();
[[maybe_unused]] const auto dummy_thread = GetHostDummyThread(existing_thread);
}
@ -445,11 +438,9 @@ struct KernelCore::Impl {
static inline thread_local KThread* current_thread{nullptr};
KThread* GetCurrentEmuThread() {
const auto thread_id = GetCurrentHostThreadID();
if (thread_id >= Core::Hardware::NUM_CPU_CORES) {
return GetHostDummyThread(nullptr);
if (!current_thread) {
current_thread = GetHostDummyThread(nullptr);
}
return current_thread;
}
@ -1002,7 +993,7 @@ const Kernel::PhysicalCore& KernelCore::CurrentPhysicalCore() const {
}
Kernel::KScheduler* KernelCore::CurrentScheduler() {
u32 core_id = impl->GetCurrentHostThreadID();
const u32 core_id = impl->GetCurrentHostThreadID();
if (core_id >= Core::Hardware::NUM_CPU_CORES) {
// This is expected when called from not a guest thread
return {};