[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170222113832.GQ6515@twins.programming.kicks-ass.net>
Date: Wed, 22 Feb 2017 12:38:32 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Stafford Horne <shorne@...il.com>
Cc: Jonas Bonn <jonas@...thpole.se>,
Stefan Kristiansson <stefan.kristiansson@...nalahti.fi>,
linux@...ck-us.net, openrisc@...ts.librecores.org,
linux-kernel@...r.kernel.org, Ingo Molnar <mingo@...hat.com>
Subject: Re: [PATCH v3 10/25] openrisc: add spinlock implementation
On Wed, Feb 22, 2017 at 04:11:39AM +0900, Stafford Horne wrote:
> + lockval.tickets.owner = ACCESS_ONCE(lock->tickets.owner);
> + return !arch_spin_value_unlocked(ACCESS_ONCE(*lock));
> + struct __raw_tickets tickets = ACCESS_ONCE(lock->tickets);
> +#define arch_write_can_lock(x) (ACCESS_ONCE((x)->lock) == 0)
> +#define arch_read_can_lock(x) (ACCESS_ONCE((x)->lock) < 0x80000000)
Please, no ACCESS_ONCE() in new code, use READ_ONCE() or WRITE_ONCE().
Powered by blists - more mailing lists