From mboxrd@z Thu Jan 1 00:00:00 1970 From: Laxman Dewangan Subject: Re: [PATCH 3/6] soc/tegra: pmc: Add support for IO pads power state and voltage Date: Tue, 3 May 2016 18:01:39 +0530 Message-ID: <57289A2B.7040501@nvidia.com> References: <1462191434-28933-1-git-send-email-ldewangan@nvidia.com> <1462191434-28933-4-git-send-email-ldewangan@nvidia.com> <57289AC0.4090604@nvidia.com> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <57289AC0.4090604-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org> Sender: linux-tegra-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Jon Hunter , swarren-3lzwWm7+Weoh9ZMKESR00Q@public.gmane.org, linus.walleij-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org, gnurou-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org, robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org, mark.rutland-5wv7dgnIgG8@public.gmane.org, thierry.reding-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org Cc: linux-tegra-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-gpio-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: devicetree@vger.kernel.org On Tuesday 03 May 2016 06:04 PM, Jon Hunter wrote: > On 02/05/16 13:17, Laxman Dewangan wrote: >> + >> + return tegra_io_rail_power_on(dpd_bit); > From a readability standpoint the above seems weird because > tegra_io_pads_power_enable() takes an ID as the argument, translates it > to a bit value and passes it to tegra_io_rail_power_on() which also > takes an ID for the argument. Yaah, I did not want to duplicate the implementation and hence it is there. We will get rid of the older APIs once this is IN and new mechanism there. >> + >> + bval = (io_volt_uv == 3300000) ? BIT(pwr_bit) : 0; >> + >> + mutex_lock(&pmc->powergates_lock); >> + tegra_pmc_read_modify_write(PMC_PWR_DET, BIT(pwr_bit), BIT(pwr_bit)); >> + tegra_pmc_read_modify_write(PMC_PWR_DET_VAL, BIT(pwr_bit), bval); >> + mutex_unlock(&pmc->powergates_lock); > There are 4 instances of BIT(pwr_bit). May be we should do this once or > have tegra_io_pads_to_power_val() return the bit? OK, will re-write this part. > >> +int tegra_io_pads_power_disable(int io_pad_id); >> +int tegra_io_pads_power_is_enabled(int io_pad_id); > What I don't like here, is now we have two public APIs to do the same > job because tegra_io_pads_power_enable/disable() calls > tegra_io_rail_power_off/on() internally. Furthermore, the two APIs use > different ID definitions to accomplish the same job. This shouldn't be > necessary. Currently SOR driver is using the tegra_io_rail_power_off/on() APIs. Once the proper interface available then I will move sor driver to use new method and then we can full get rid of older APIs and macros. Till that, we need to have this. >