[NPTL docs] pthread robustness and pthread_mutex_consistent

[Florian Weimer]

On 08/07/2017 11:35 AM, Yubin Ruan wrote:
> A programmer can call pthread_kill(), or call pthread_exit() wrongly (at the
> wrong place). I don't know. I am writing some library so I have to
> cover all the possibility.

But in this case, merely recovering the lock does not roll back any
partial updates that might have happened since the acquisition of the lock.

Florian