RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-31 11:28:12 +00:00
Code Issues Activity Actions

Kernel: Fix a few deadlocks with Thread::m_lock and g_scheduler_lock

Browse source 
g_scheduler_lock cannot safely be acquired after Thread::m_lock
because another processor may already hold g_scheduler_lock and wait
for the same Thread::m_lock.
This commit is contained in:
Tom 2020-10-25 20:22:59 -06:00 • committed by Andreas Kling
parent 8c764319ad
commit 1e2e3eed62
6 changed files with 55 additions and 37 deletions
Download patch file Download diff file Expand all files Collapse all files

4
Kernel/Syscalls/thread.cpp
View file

@ -70,7 +70,6 @@ int Process::sys$create_thread(void* (*entry)(void*), Userspace<const Syscall::S
builder.appendf("[%d]", thread->tid().value());
thread->set_name(builder.to_string());
thread->set_priority(requested_thread_priority);
if (!is_thread_joinable)
thread->detach();
@ -83,6 +82,9 @@ int Process::sys$create_thread(void* (*entry)(void*), Userspace<const Syscall::S
auto tsr_result = thread->make_thread_specific_region({});
if (tsr_result.is_error())
return tsr_result.error();
ScopedSpinLock lock(g_scheduler_lock);
thread->set_priority(requested_thread_priority);
thread->set_state(Thread::State::Runnable);
return thread->tid().value();
}