[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CALvZod5ijDz=coEE8G8v_haPaKuUa5jHYzEwKvLVxHGphixsFA@mail.gmail.com>
Date: Fri, 17 Jun 2022 07:57:53 -0700
From: Shakeel Butt <shakeelb@...gle.com>
To: Waiman Long <longman@...hat.com>
Cc: Peter Zijlstra <peterz@...radead.org>,
Eric Dumazet <eric.dumazet@...il.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
linux-kernel <linux-kernel@...r.kernel.org>,
Eric Dumazet <edumazet@...gle.com>,
Ingo Molnar <mingo@...hat.com>,
Boqun Feng <boqun.feng@...il.com>,
Will Deacon <will@...nel.org>, Roman Penyaev <rpenyaev@...e.de>
Subject: Re: [PATCH] locking/rwlocks: do not starve writers
On Fri, Jun 17, 2022 at 7:43 AM Waiman Long <longman@...hat.com> wrote:
>
> On 6/17/22 08:07, Peter Zijlstra wrote:
> > On Fri, Jun 17, 2022 at 02:10:39AM -0700, Eric Dumazet wrote:
> >> --- a/kernel/locking/qrwlock.c
> >> +++ b/kernel/locking/qrwlock.c
> >> @@ -23,16 +23,6 @@ void queued_read_lock_slowpath(struct qrwlock *lock)
> >> /*
> >> * Readers come here when they cannot get the lock without waiting
> >> */
> >> - if (unlikely(in_interrupt())) {
> >> - /*
> >> - * Readers in interrupt context will get the lock immediately
> >> - * if the writer is just waiting (not holding the lock yet),
> >> - * so spin with ACQUIRE semantics until the lock is available
> >> - * without waiting in the queue.
> >> - */
> >> - atomic_cond_read_acquire(&lock->cnts, !(VAL & _QW_LOCKED));
> >> - return;
> >> - }
> >> atomic_sub(_QR_BIAS, &lock->cnts);
> >>
> >> trace_contention_begin(lock, LCB_F_SPIN | LCB_F_READ);
> > This is known to break tasklist_lock.
> >
> We certainly can't break the current usage of tasklist_lock.
>
> I am aware of this problem with networking code and is thinking about
> either relaxing the check to exclude softirq or provide a
> read_lock_unfair() variant for networking use.
read_lock_unfair() for networking use or tasklist_lock use?
> I think tasklist_lock
> isn't taken from softirq context, but I may be wrong. Providing a
> read_lock_unfair() will require quite a bit of work in the supporting
> infrastructure as well.
>
> Cheers,
> Longman
>
Powered by blists - more mailing lists