From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bjorn Andersson Subject: Re: [PATCH 2/3] pinctrl: qcom-spmi-gpio: Fix input value report Date: Thu, 9 Apr 2015 09:01:11 -0700 Message-ID: <20150409160111.GH19186@sonymobile.com> References: <1428592717-21706-1-git-send-email-ivan.ivanov@linaro.org> <1428592717-21706-2-git-send-email-ivan.ivanov@linaro.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Return-path: Content-Disposition: inline In-Reply-To: <1428592717-21706-2-git-send-email-ivan.ivanov@linaro.org> Sender: linux-kernel-owner@vger.kernel.org To: "Ivan T. Ivanov" Cc: Linus Walleij , Arnd Bergmann , Soren Brinkmann , "linux-gpio@vger.kernel.org" , "linux-kernel@vger.kernel.org" , "linux-arm-msm@vger.kernel.org" List-Id: linux-arm-msm@vger.kernel.org On Thu 09 Apr 08:18 PDT 2015, Ivan T. Ivanov wrote: > Read input buffer when input is enabled, not when it is > disabled. Also fix interpretation of the pmic_gpio_read() > return code, negative value means an error. > > Signed-off-by: Ivan T. Ivanov > --- > drivers/pinctrl/qcom/pinctrl-spmi-gpio.c | 11 ++++++----- > 1 file changed, 6 insertions(+), 5 deletions(-) > > diff --git a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c b/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c > index 4b21aac..e8b74c6 100644 > --- a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c > +++ b/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c > @@ -466,12 +466,13 @@ static void pmic_gpio_config_dbg_show(struct pinctrl_dev *pctldev, > seq_puts(s, " ---"); > } else { > > - if (!pad->input_enabled) { > + if (pad->input_enabled) { > ret = pmic_gpio_read(state, pad, PMIC_MPP_REG_RT_STS); > - if (!ret) { > - ret &= PMIC_MPP_REG_RT_STS_VAL_MASK; > - pad->out_value = ret; > - } > + if (ret < 0) > + return; > + > + ret &= PMIC_MPP_REG_RT_STS_VAL_MASK; > + pad->out_value = ret; I'm not too happy on pmic_gpio_read() returning the register value as an signed int, sorry for not picking on this back when we merged the driver. As far as I can see bit 31 is reserved in all registers in the block though, so there shouldn't be any practical issues with this _today_. But please consider revising pmic_gpio_read() to take a u32* parameter for the data instead. Reviewed-by: Bjorn Andersson Regards, Bjorn