[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20160413174302.GC21722@arm.com>
Date: Wed, 13 Apr 2016 18:43:04 +0100
From: Will Deacon <will.deacon@....com>
To: Jason Low <jason.low2@...com>
Cc: Peter Zijlstra <peterz@...radead.org>,
Linus Torvalds <torvalds@...ux-foundation.org>,
linux-kernel@...r.kernel.org, mingo@...hat.com,
paulmck@...ux.vnet.ibm.com, terry.rudd@....com,
waiman.long@....com, boqun.feng@...il.com, dave@...olabs.net,
kbuild test robot <lkp@...el.com>, kbuild-all@...org
Subject: Re: [PATCH v2] MCS spinlock: Use smp_cond_load_acquire()
On Tue, Apr 12, 2016 at 08:02:17PM -0700, Jason Low wrote:
> For qspinlocks on ARM64, we would like to use WFE instead
> of purely spinning. Qspinlocks internally have lock
> contenders spin on an MCS lock.
>
> Update arch_mcs_spin_lock_contended() such that it uses
> the new smp_cond_load_acquire() so that ARM64 can also
> override this spin loop with its own implementation using WFE.
>
> On x86, it can also cheaper to use this than spinning on
> smp_load_acquire().
>
> Signed-off-by: Jason Low <jason.low2@...com>
FWIW, we just override arch_mcs_spin_lock_contended entirely for arch/arm/
and use wfe there so we could do the same for arm64 in mainline already.
Will
Powered by blists - more mailing lists