On Thu, Aug 31, 2017 at 10:34:53AM +0200, Peter Zijlstra wrote: > On Thu, Aug 31, 2017 at 05:15:01PM +0900, Byungchul Park wrote: > > It's not important. Ok, check the following, instead: > > > > context X context Y > > --------- --------- > > wait_for_completion(C) > > acquire(A) > > release(A) > > process_one_work() > > acquire(B) > > release(B) > > work->fn() > > complete(C) > > > > We don't need to lose C->A and C->B dependencies unnecessarily. > > I really can't be arsed about them. Its really only the first few works > that will retain that dependency anyway, even if you were to retain > them.
Wrong. Every 'work' doing complete() for different classes of completion variable suffers from losing valuable dependencies, every time, not first few ones. Remind we are talking about dependencies wrt cross-lock, not between _holding_ locks. If you invalidate xhlock whenever work->fn(), we cannot build dependencies like C->A and C->B every time. Right? > All of that is contained in kernel/kthread and kernel/workqueue and can > be audited if needed. Its a very limited amount of code. I mean, doing it automatically w/o additional overhead is better than considering the limited amount of code manually every time changing kernel code. Do as you please.
