Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756702Ab3H2T6V (ORCPT ); Thu, 29 Aug 2013 15:58:21 -0400 Received: from hydra.sisk.pl ([212.160.235.94]:42870 "EHLO hydra.sisk.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754398Ab3H2T6T (ORCPT ); Thu, 29 Aug 2013 15:58:19 -0400 From: "Rafael J. Wysocki" To: Colin Cross Cc: Prashant Gaikwad , "linux-pm@vger.kernel.org" , "linux-kernel@vger.kernel.org" , Neil Zhang , Joseph Lo , "linux-tegra@vger.kernel.org" , "stable@vger.kernel.org" , Daniel Lezcano Subject: Re: [PATCH 1/3] cpuidle: coupled: disable interrupts after entering safe state Date: Thu, 29 Aug 2013 22:09:01 +0200 Message-ID: <3891367.RzYTp6nikM@vostro.rjw.lan> User-Agent: KMail/4.10.5 (Linux/3.11.0-rc6+; KDE/4.10.5; x86_64; ; ) In-Reply-To: References: <1377287112-12018-1-git-send-email-ccross@android.com> <521EEF26.80803@nvidia.com> MIME-Version: 1.0 Content-Transfer-Encoding: 7Bit Content-Type: text/plain; charset="utf-8" Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2468 Lines: 57 On Thursday, August 29, 2013 12:12:17 AM Colin Cross wrote: > On Wed, Aug 28, 2013 at 11:50 PM, Prashant Gaikwad wrote: > > On Saturday 24 August 2013 01:15 AM, Colin Cross wrote: > >> > >> Calling cpuidle_enter_state is expected to return with interrupts > >> enabled, but interrupts must be disabled before starting the > >> ready loop synchronization stage. Call local_irq_disable after > >> each call to cpuidle_enter_state for the safe state. > >> > >> CC: stable@vger.kernel.org > >> Signed-off-by: Colin Cross > >> --- > >> drivers/cpuidle/coupled.c | 2 ++ > >> 1 file changed, 2 insertions(+) > >> > >> diff --git a/drivers/cpuidle/coupled.c b/drivers/cpuidle/coupled.c > >> index 2a297f8..db92bcb 100644 > >> --- a/drivers/cpuidle/coupled.c > >> +++ b/drivers/cpuidle/coupled.c > >> @@ -460,6 +460,7 @@ int cpuidle_enter_state_coupled(struct cpuidle_device > >> *dev, > >> } > >> entered_state = cpuidle_enter_state(dev, drv, > >> dev->safe_state_index); > >> + local_irq_disable(); > > > > > > Colin, > > > > There is still some window where irq remains enabled after exiting safe > > state. It may introduce some corner case. > > Instead of this can we pass a parameter to cpuidle_enter_state indicating > > that irq has to be enabled or not after exit from idle state, which would be > > false when entering safe state from coupled idle. > > It's fine for irqs to be enabled when exiting the safe state, in fact > on further inspection this patch isn't strictly necessary at all - we > always enable interrupts inside cpuidle_coupled_clear_pokes soon after > cpuidle_enter_state returns, and then disable them again. It's > probably better to disable interrupts right after cpuidle_enter_state, > it makes sure interrupts are consistently disabled when calling > cpuidle_coupled_set_waiting, cpuidle_coupled_cpus_waiting and > cpuidle_coupled_clear_pokes, although that doesn't matter in the > current implementation. > > Rafael, feel free to drop the stable annotation from this patch. I will, thanks! -- I speak only for myself. Rafael J. Wysocki, Intel Open Source Technology Center. -- 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/