Date: Thu, 9 Apr 2015 20:28:38 +0200
From: Peter Zijlstra <peterz@infradead.org>
To: Kirill Tkhai <ktkhai@parallels.com>
Cc: linux-kernel@vger.kernel.org, Oleg Nesterov <oleg@redhat.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Ingo Molnar <mingo@redhat.com>, Michal Hocko <mhocko@suse.cz>,
        Rik van Riel <riel@redhat.com>, Ionut Alexa <ionut.m.alexa@gmail.com>,
        Peter Hurley <peter@hurleysoftware.com>,
        Kirill Tkhai <tkhai@yandex.ru>
Subject: Re: [PATCH] exit: Use read lock for do_notify_parent() instead of
 write lock
Message-ID: <20150409182838.GZ5029@twins.programming.kicks-ass.net>
References: <1428602348.12166.29.camel@parallels.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1428602348.12166.29.camel@parallels.com>
User-Agent: Mutt/1.5.21 (2012-12-30)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 829
Lines: 21

On Thu, Apr 09, 2015 at 08:59:08PM +0300, Kirill Tkhai wrote:

I've not really read your email yet, however:

> Also, in the future we may think about new rwlock primitive, which atomically
> drops write lock and acquires read lock. Something like this for example:
> 
> include/asm-generic/qrwlock.h:
> static inline void queue_reduce_locked_write_to_read(struct qrwlock *lock)
> {
> 	smp_mb__before_atomic();
> 	atomic_add(_QR_BIAS - _QW_LOCKED, &lock->cnts);
> }

we actually have that for the rwsems: downgrade_write(). So the
consistent naming would be: queue_downgrade_write().
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/