From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrzej Hajda Subject: Re: [PATCH v2 2/3] Input: add support for the STMicroelectronics FingerTip touchscreen Date: Thu, 16 Mar 2017 09:22:16 +0100 Message-ID: References: <20170210021721.12218-1-andi.shyti@samsung.com> <20170210021721.12218-3-andi.shyti@samsung.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Return-path: In-reply-to: <20170210021721.12218-3-andi.shyti@samsung.com> Sender: linux-input-owner@vger.kernel.org To: Andi Shyti , Dmitry Torokhov , Krzysztof Kozlowski , Rob Herring , Chanwoo Choi , Javier Martinez Canillas Cc: linux-input@vger.kernel.org, devicetree@vger.kernel.org, linux-kernel@vger.kernel.org, linux-samsung-soc@vger.kernel.org, Andi Shyti List-Id: linux-samsung-soc@vger.kernel.org Hi Andi, On 10.02.2017 03:17, Andi Shyti wrote: (...) > + > + err = devm_request_threaded_irq(&client->dev, client->irq, > + NULL, stmfts_irq_handler, > + IRQF_ONESHOT | IRQF_TRIGGER_LOW, > + "stmfts_irq", sdata); > + if (err) > + return err; > + > + /* > + * Disable irq they, they are not needed at this stage. > + * One possible case when an IRQ can be already rased is e.g. if the > + * regulator is set as always on and the stmfts device sends an IRQ as > + * soon as it gets powered, de-synchronizing the power on sequence. > + * During power on, the device will be reset and all the initialization > + * IRQ will be resent. > + */ > + disable_irq(sdata->client->irq); It is safer to set: irq_set_status_flags(client->irq, IRQ_NOAUTOEN); before devm_request_threaded_irq Regards Andrzej