Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753706AbaFRQqb (ORCPT ); Wed, 18 Jun 2014 12:46:31 -0400 Received: from mail-ig0-f177.google.com ([209.85.213.177]:40629 "EHLO mail-ig0-f177.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751853AbaFRQq3 (ORCPT ); Wed, 18 Jun 2014 12:46:29 -0400 Date: Wed, 18 Jun 2014 17:46:21 +0100 From: Lee Jones To: Doug Anderson Cc: Andrew Bresticker , Stephen Warren , Olof Johansson , Sonny Rao , linux-samsung-soc , Javier Martinez Canillas , Bill Richardson , Simon Glass , Wolfram Sang , "broonie@kernel.org" , Samuel Ortiz , "linux-kernel@vger.kernel.org" Subject: Re: [PATCH 02/10] mfd: cros_ec: IRQs for cros_ec should be optional Message-ID: <20140618164621.GB25353@lee--X1> References: <1402954800-28215-1-git-send-email-dianders@chromium.org> <1402954800-28215-3-git-send-email-dianders@chromium.org> <20140618075551.GJ21030@lee--X1> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 18 Jun 2014, Doug Anderson wrote: > Lee, > > On Wed, Jun 18, 2014 at 12:55 AM, Lee Jones wrote: > > On Mon, 16 Jun 2014, Doug Anderson wrote: > > > >> From: Bill Richardson > >> > >> Preparing the way for the LPC device, which is just a plaform_device without > >> interrupts. > >> > >> Signed-off-by: Bill Richardson > >> Signed-off-by: Doug Anderson > >> --- > >> drivers/mfd/cros_ec.c | 26 +++++++++++++------------- > >> 1 file changed, 13 insertions(+), 13 deletions(-) > >> > >> diff --git a/drivers/mfd/cros_ec.c b/drivers/mfd/cros_ec.c > >> index 38fe9bf..bd6f936 100644 > >> --- a/drivers/mfd/cros_ec.c > >> +++ b/drivers/mfd/cros_ec.c > >> @@ -119,17 +119,15 @@ int cros_ec_register(struct cros_ec_device *ec_dev) > >> return -ENOMEM; > >> } > >> > >> - if (!ec_dev->irq) { > >> - dev_dbg(dev, "no valid IRQ: %d\n", ec_dev->irq); > >> - return err; > >> - } > >> - > >> - err = request_threaded_irq(ec_dev->irq, NULL, ec_irq_thread, > >> - IRQF_TRIGGER_LOW | IRQF_ONESHOT, > >> - "chromeos-ec", ec_dev); > >> - if (err) { > >> - dev_err(dev, "request irq %d: error %d\n", ec_dev->irq, err); > >> - return err; > >> + if (ec_dev->irq) { > >> + err = request_threaded_irq(ec_dev->irq, NULL, ec_irq_thread, > >> + IRQF_TRIGGER_LOW | IRQF_ONESHOT, > >> + "chromeos-ec", ec_dev); > > > > Is there anything stopping you using the devm_* version? > > I'm generally quite hesitant about using the devm_ IRQ functions. > Requesting an IRQ enables the IRQ at the time of request and freeing > it disables it, right? Leaving it up to the the devm subsystem to > disable your IRQ tends to be a race waiting to happen if an IRQ > happens after you've freed all your memory / cleaned up all your > state. > > Looking at cros_ec in particular though: > > * Right now the last thing done in cros_ec_remove() (which is the last > thing in both cros_ec_i2c_remove and cros_ec_spi_remove) is to free > the IRQ. That means that you're right: we could switch to devm_ in > this case and it wouldn't introduce any new bugs. > > * ...but I'm not convinced that the location of the free_irq() today > is quite right. I couldn't actually get it to crash or hang, but > there is a time period where we've called mfd_remove_devices() and the > IRQ is still active. That doesn't seem like a super great situation > to be in. I'll add a move of the irq_free to the patch series. I guess if you're concerned about ordering you could always use devm_free_irq() in the places where you think it should be freed earlier than release. I'm pretty sure that discludes the failure patch in probe() though, so we'd still be able to rid a few lines. -- Lee Jones Linaro STMicroelectronics Landing Team Lead Linaro.org │ Open source software for ARM SoCs Follow Linaro: Facebook | Twitter | Blog -- 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/