diff options
author | Peter Zijlstra <peterz@infradead.org> | 2023-09-08 18:22:51 +0200 |
---|---|---|
committer | Peter Zijlstra <peterz@infradead.org> | 2023-09-20 09:31:12 +0200 |
commit | 6b596e62ed9f90c4a97e68ae1f7b1af5beeb3c05 (patch) | |
tree | 5fdde551fcf4a48f0a1046c34b6d994e0be0ffd6 /include/linux/sched.h | |
parent | de1474b46d889ee0367f6e71d9adfeb0711e4a8d (diff) | |
download | lwn-6b596e62ed9f90c4a97e68ae1f7b1af5beeb3c05.tar.gz lwn-6b596e62ed9f90c4a97e68ae1f7b1af5beeb3c05.zip |
sched: Provide rt_mutex specific scheduler helpers
With PREEMPT_RT there is a rt_mutex recursion problem where
sched_submit_work() can use an rtlock (aka spinlock_t). More
specifically what happens is:
mutex_lock() /* really rt_mutex */
...
__rt_mutex_slowlock_locked()
task_blocks_on_rt_mutex()
// enqueue current task as waiter
// do PI chain walk
rt_mutex_slowlock_block()
schedule()
sched_submit_work()
...
spin_lock() /* really rtlock */
...
__rt_mutex_slowlock_locked()
task_blocks_on_rt_mutex()
// enqueue current task as waiter *AGAIN*
// *CONFUSION*
Fix this by making rt_mutex do the sched_submit_work() early, before
it enqueues itself as a waiter -- before it even knows *if* it will
wait.
[[ basically Thomas' patch but with different naming and a few asserts
added ]]
Originally-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20230908162254.999499-5-bigeasy@linutronix.de
Diffstat (limited to 'include/linux/sched.h')
-rw-r--r-- | include/linux/sched.h | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/include/linux/sched.h b/include/linux/sched.h index 77f01ac385f7..67623ffd4a8e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -911,6 +911,9 @@ struct task_struct { * ->sched_remote_wakeup gets used, so it can be in this word. */ unsigned sched_remote_wakeup:1; +#ifdef CONFIG_RT_MUTEXES + unsigned sched_rt_mutex:1; +#endif /* Bit to tell LSMs we're in execve(): */ unsigned in_execve:1; |