On 4/28/19 11:57 AM, Waiman Long wrote:
During my rwsem testing, it was found that after a down_read(), the reader count may occasionally become 0 or even negative. Consequently, a writer may steal the lock at that time and execute with the reader in parallel thus breaking the mutual exclusion guarantee of the write lock. In other words, both readers and writer can become rwsem owners simultaneously.
I would like to have this particular patch merged in the next merge window, if possible, so that I can backport it downstream.
Thanks, Longman