RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-25 15:17:36 +00:00
Code Issues Activity Actions

Kernel: Don't hold scheduler lock while setting up blocker in Thread::block

Browse source 
This fixes a deadlock when one processor is trying to block while another is
trying to unblock the same.
This commit is contained in:
Tom 2021-01-23 11:08:34 -07:00 • committed by Andreas Kling
parent 250a310454
commit ccaeb47401
1 changed files with 8 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

17
Kernel/Thread.h
View file

@ -821,13 +821,14 @@ public:
ASSERT(this == Thread::current());
ScopedCritical critical;
ASSERT(!s_mm_lock.own_lock());
ScopedSpinLock scheduler_lock(g_scheduler_lock);
ScopedSpinLock block_lock(m_block_lock);
// We need to hold m_block_lock so that nobody can unblock a blocker as soon
// as it is constructed and registered elsewhere
m_in_block = true;
T t(forward<Args>(args)...);
ScopedSpinLock scheduler_lock(g_scheduler_lock);
// Relaxed semantics are fine for timeout_unblocked because we
// synchronize on the spin locks already.
Atomic<bool, AK::MemoryOrder::memory_order_relaxed> timeout_unblocked(false);
@ -881,23 +882,20 @@ public:
set_state(Thread::Blocked);
}
scheduler_lock.unlock();
block_lock.unlock();
dbgln<THREAD_DEBUG>("Thread {} blocking on {} ({}) -->", *this, &t, t.state_string());
bool did_timeout = false;
auto previous_locked = LockMode::Unlocked;
u32 lock_count_to_restore = 0;
auto previous_locked = unlock_process_if_locked(lock_count_to_restore);
for (;;) {
scheduler_lock.unlock();
// Yield to the scheduler, and wait for us to resume unblocked.
if (previous_locked == LockMode::Unlocked)
previous_locked = unlock_process_if_locked(lock_count_to_restore);
ASSERT(!g_scheduler_lock.own_lock());
ASSERT(Processor::current().in_critical());
yield_while_not_holding_big_lock();
ASSERT(Processor::current().in_critical());
scheduler_lock.lock();
ScopedSpinLock block_lock2(m_block_lock);
if (should_be_stopped() || state() == Stopped) {
dbgln("Thread should be stopped, current state: {}", state_string());
@ -918,11 +916,13 @@ public:
ASSERT(m_blocker == &t);
m_blocker = nullptr;
}
dbgln<THREAD_DEBUG>("<-- Thread {} unblocked from {} ({})", *this, &t, t.state_string());
m_in_block = false;
break;
}
if (t.was_interrupted_by_signal()) {
ScopedSpinLock scheduler_lock(g_scheduler_lock);
ScopedSpinLock lock(m_lock);
dispatch_one_pending_signal();
}
@ -931,7 +931,6 @@ public:
// to clean up now while we're still holding m_lock
auto result = t.end_blocking({}, did_timeout); // calls was_unblocked internally
scheduler_lock.unlock();
if (timer && !did_timeout) {
// Cancel the timer while not holding any locks. This allows
// the timer function to complete before we remove it