Re: Priority Inversion and Unlimited Spin of pthread_rwlock_t

[Florian Weimer <fweimer@redhat.com>]

* Peng Zheng:

> And the SCHED_OTHER writer was just about to enable the `__wrphase_futex` in
> `__pthread_rwlock_wrlock_full` (just one ARM instruction away) 
> but never able to do that (the two readers ate nearly all available CPUs):
>
>   while ((r & PTHREAD_RWLOCK_WRPHASE) == 0
>      && (r >> PTHREAD_RWLOCK_READER_SHIFT) == 0)
>     {
>       if (atomic_compare_exchange_weak_acquire (&rwlock->__data.__readers,
>                         &r, r | PTHREAD_RWLOCK_WRPHASE))
>     {
>       atomic_store_relaxed (&rwlock->__data.__wrphase_futex, 1);  /* writer was stuck HERE! */
>
>       goto done;
>     }
>       /* TODO Back-off.  */
>     }

Is this about filling in the TODO?

> Unlimited user space spin is too dangerous to be used, how about
> limiting the total number of spins before suspending using futex? Or
> using rseq as mentioned in the LWN artible?

> [0] https://lwn.net/Articles/944895/

The LWN article does not talk about priority inversion.  With rseq,
userspace can detect preemption, but that may never happen with certain
realtime scheduling priorities.

Thanks,
Florian