From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from bhuna.collabora.co.uk ([46.235.227.227]:47210 "EHLO bhuna.collabora.co.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753500AbdHXUOE (ORCPT ); Thu, 24 Aug 2017 16:14:04 -0400 Subject: Re: rtc: ds1307: call the platform's logic for handle IRQs. To: Heiner Kallweit , Alexandre Belloni Cc: linux-rtc@vger.kernel.org References: <832da9eb-abed-67c9-1d53-34f19d27c0d9@gmail.com> <019f5769-b3ec-2afd-9469-e722b7b0a2e5@gmail.com> From: Enric Balletbo i Serra Message-ID: Date: Thu, 24 Aug 2017 22:13:52 +0200 MIME-Version: 1.0 In-Reply-To: <019f5769-b3ec-2afd-9469-e722b7b0a2e5@gmail.com> Content-Type: text/plain; charset=utf-8 Sender: linux-rtc-owner@vger.kernel.org List-ID: Hi Heiner On 24/08/17 21:16, Heiner Kallweit wrote: > Am 24.08.2017 um 20:46 schrieb Heiner Kallweit: >> Hi Enric, >> >> I just saw your submitted patch on patchwork. As I haven't been subscribed >> to linux-rtc list yet, I can't reply to the original mail. >> >> Few remarks: >> I think the same can be achieved easier (apart from the fact that member >> irq was just removed from struct ds1307). >> The curent call to device_set_wakeup_capable has to be replaced with >> device_init_wakeup, in addition we have to call dev_pm_set_wake_irq to >> register the interrupt with the Linux wakeup core. Then the core takes >> care of everything. >> Oh I just saw your patches, thanks to advice me about them. > After further checking not even this may be necessary. > If flag I2C_CLIENT_WAKE is set for the i2c client then i2c_device_probe() > enables the device as wakeup device and registers the interrupt with the > wakeup core. > If the i2c client is defined via DT, then of_i2c_register_device() sets > this flag based on the "wakeup-source" property. > Is your device configured via DT? If yes, did you check whether wakeup > works w/o your patch with just setting property wakeup-source ? > Yes, device is configured via DT and without my patch, just setting property wakeup-source didn't work, the reason is that code: http://elixir.free-electrons.com/linux/v4.13-rc6/source/drivers/rtc/rtc-ds1307.c#L1382 I didn't test on top of your patches so let me rebase/rethink my patch on top of yours. > >> See also rtc-ds1343, although I think the calls to enable/disable_irq_wake >> are not needed there because the core takes care of this already >> (enable_irq_wake is called from dev_pm_arm_wake_irq). >> >> Rgds, Heiner >> > Thanks, Enric