From: Krzysztof Kozlowski <krzk@kernel.org>
To: Martin Juecker <martin.juecker@gmail.com>
Cc: linux-samsung-soc@vger.kernel.org, kgene@kernel.org,
devicetree@vger.kernel.org
Subject: Re: [PATCH 2/5] ARM: dts: add Samsung's exynos4412-based p4note boards
Date: Tue, 27 Oct 2020 09:17:51 +0100 [thread overview]
Message-ID: <20201027081751.GA4244@kozik-lap> (raw)
In-Reply-To: <20201026232709.GA2723@adroid>
On Tue, Oct 27, 2020 at 12:27:09AM +0100, Martin Juecker wrote:
> Hello Krzysztof,
>
> thank you for the feedback.
>
> On Mon, Oct 26, 2020 at 03:48:30PM +0100, Krzysztof Kozlowski wrote:
> > On Sat, Oct 24, 2020 at 04:44:18PM +0200, Martin Jücker wrote:
>
> (...)
>
> > > +/dts-v1/;
> > > +#include "exynos4412-p4note.dtsi"
> > > +
> > > +/ {
> > > + model = "Samsung Galaxy Note 10.1 (GT-N8010/N8013) based on Exynos4412";
> > > + compatible = "samsung,n801x", "samsung,p4note", "samsung,exynos4412", "samsung,exynos4";
> >
> > Compatible without the wildcard, so samsung,n8010.
>
> I went with the n710x style compatible but I'm fine with this, too.
>
> >
> > > +
> > > + /* this is the base variant without any kind of modem */
> > > +};
> > > diff --git a/arch/arm/boot/dts/exynos4412-p4note.dtsi b/arch/arm/boot/dts/exynos4412-p4note.dtsi
> > > new file mode 100644
> > > index 000000000000..6d7e84f40e92
> > > --- /dev/null
> > > +++ b/arch/arm/boot/dts/exynos4412-p4note.dtsi
> > > @@ -0,0 +1,1057 @@
> > > +// SPDX-License-Identifier: GPL-2.0
> > > +/*
> > > + * Samsung's Exynos4412 based p4note device family base DT
> > > + *
> > > + * Copyright (c) 2020 Martin Jücker <martin.juecker@gmail.com>
> > > + */
> > > +
> > > +/dts-v1/;
> > > +#include "exynos4412.dtsi"
> > > +#include "exynos4412-ppmu-common.dtsi"
> > > +
> > > +#include <dt-bindings/input/input.h>
> > > +#include <dt-bindings/gpio/gpio.h>
> > > +#include <dt-bindings/input/gpio-keys.h>
> >
> > Looks unused.
>
> Indeed, a leftover while trying things.
>
> (...)
>
> >
> > > + tsp_reg_1: voltage-regulator-1 {
> >
> > All labels are not used.
> >
>
> Removed the labels.
>
> > > + compatible = "regulator-fixed";
> > > + regulator-name = "TSP_FIXED_VOLTAGES_1";
> >
> > Would be better to have a real name matching either schematics (which I
> > guess you do not have) or vendor sources.
> >
>
> You are right, there is no schematics, the old vendor sources only have
> gpio calls with the names TSP_LDO_ON, TSP_LDO_ON1 and TSP_LDO_ON2. Those
> regulators are for the Atmel 1664S touchscreen which has a public data
> sheet. My best guess would be that those are VDD, AVDD and XVDD but I
> have no idea how to match them or whether this is actually an accurate
> observation. My electronics knowledge is rather limited.
Then how about TSP_LDO1, TSP_LDO2, TSP_LDO3?
> > > + pinctrl-names = "default";
> > > + pinctrl-0 = <&tsp_reg_gpio_1>;
> > > + gpios = <&gpm4 5 GPIO_ACTIVE_HIGH>;
> > > + enable-active-high;
> > > + regulator-always-on;
> > > + };
> > > +
> > > + tsp_reg_2: voltage-regulator-2 {
> > > + compatible = "regulator-fixed";
> > > + regulator-name = "TSP_FIXED_VOLTAGES_2";
> > > + pinctrl-names = "default";
> > > + pinctrl-0 = <&tsp_reg_gpio_2>;
> > > + gpios = <&gpb 5 GPIO_ACTIVE_HIGH>;
> > > + enable-active-high;
> > > + regulator-always-on;
> > > + };
> > > +
> > > + tsp_reg_3: voltage-regulator-3 {
> > > + compatible = "regulator-fixed";
> > > + regulator-name = "TSP_FIXED_VOLTAGES_3";
> > > + pinctrl-names = "default";
> > > + pinctrl-0 = <&tsp_reg_gpio_3>;
> > > + gpios = <&gpb 7 GPIO_ACTIVE_HIGH>;
> > > + startup-delay-us = <20000>;
> > > + enable-active-high;
> > > + regulator-always-on;
> > > + };
> > > +
> > > + wlan_pwrseq: sdhci3-pwrseq {
> > > + compatible = "mmc-pwrseq-simple";
> > > + reset-gpios = <&gpm3 5 GPIO_ACTIVE_LOW>;
> > > + pinctrl-0 = <&wifi_reset>;
> > > + pinctrl-names = "default";
> > > + clocks = <&max77686 MAX77686_CLK_PMIC>;
> > > + clock-names = "ext_clock";
> > > + };
> > > +
> > > + i2c-gpio-10 {
> >
> > Numbering should start from 1 unless there is a reason for 10/14/19?
> >
>
> I used the numbers from the vendor sources but it shouldn't really
> matter, changed it.
>
> > > + compatible = "i2c-gpio";
> > > + sda-gpios = <&gpy2 4 (GPIO_ACTIVE_HIGH | GPIO_OPEN_DRAIN)>;
> > > + scl-gpios = <&gpy2 5 (GPIO_ACTIVE_HIGH | GPIO_OPEN_DRAIN)>;
> > > + i2c-gpio,delay-us = <2>;
> > > + #address-cells = <1>;
> > > + #size-cells = <0>;
> > > +
> > > + ak8975@c {
> >
> > Node name: magnetometer
> >
>
> Done.
>
> > > + compatible = "asahi-kasei,ak8975";
> > > + reg = <0x0c>;
> > > + interrupt-parent = <&gpm4>;
> > > + interrupts = <7 IRQ_TYPE_EDGE_RISING>;
> > > + pinctrl-names = "default";
> > > + pinctrl-0 = <&ak8975_irq>;
> > > + };
> > > + };
> > > +
> > > + i2c-gpio-14 {
> > > + compatible = "i2c-gpio";
> > > + sda-gpios = <&gpy0 2 (GPIO_ACTIVE_HIGH | GPIO_OPEN_DRAIN)>;
> > > + scl-gpios = <&gpy0 3 (GPIO_ACTIVE_HIGH | GPIO_OPEN_DRAIN)>;
> > > + i2c-gpio,delay-us = <2>;
> > > + #address-cells = <1>;
> > > + #size-cells = <0>;
> > > +
> > > + max17042@36 {
> >
> > Node name: fuel-gauge
> >
>
> Done.
>
> > > + compatible = "maxim,max17042";
> > > + reg = <0x36>;
> > > + interrupt-parent = <&gpx2>;
> > > + interrupts = <3 IRQ_TYPE_EDGE_FALLING>;
> >
> > You might need to disable the pull-down on the pin, whcih is usually the
> > default.
> >
>
> Found the configuration for this in the vendor sources, it was in a
> different place than the rest.
>
> > > + maxim,rsns-microohm = <10000>;
> > > + maxim,over-heat-temp = <600>;
> > > + maxim,over-volt = <4300>;
> > > + };
> > > + };
> > > +
> > > + i2c-gpio-19 {
> > > + compatible = "i2c-gpio";
> > > + sda-gpios = <&gpm4 1 (GPIO_ACTIVE_HIGH | GPIO_OPEN_DRAIN)>;
> > > + scl-gpios = <&gpm4 0 (GPIO_ACTIVE_HIGH | GPIO_OPEN_DRAIN)>;
> > > + i2c-gpio,delay-us = <5>;
> > > + #address-cells = <1>;
> > > + #size-cells = <0>;
> > > +
> > > + stmpe_mfd@41 {
> >
> > Node name: touchscreen?
> > Although you already have a touchscreen later, so what is it?
> >
>
> The stmpe 811 is a mfd but the p4note is actually only using the adc
> part of it for checking the accessory type of the usb cable and what was
> connected to the headphone jack. I renamed it to adc@41
>
> > > + compatible = "st,stmpe811";
> > > + reg = <0x41>;
> > > + interrupt-parent = <&gpx0>;
> > > + interrupts = <1 IRQ_TYPE_LEVEL_LOW>;
> >
> > Disable the pull-down of interrupt pin?
> >
>
> Same as above.
>
> > > + interrupt-controller;
> > > + irq-trigger = <0x1>;
> > > + st,adc-freq = <3>;
> > > + st,mod-12b = <1>;
> > > + st,ref-sel = <0>;
> > > + st,sample-time = <3>;
> > > +
> > > + stmpe_adc: stmpe_adc {
> >
> > Node name: adc
> > Label seems unused, so remove it.
> >
>
> I removed the label but I cannot rename the node as the name is used by
> the driver to find and load the driver for the subcomponent.
Ok.
>
> > > + compatible = "st,stmpe-adc";
> > > + #io-channel-cells = <1>;
> > > + st,norequest-mask = <0x2F>;
> > > + };
> > > + };
> > > + };
> > > +};
> > > +
> > > +&adc {
> > > + status = "okay";
> >
> > Missing vdd-supply
> >
>
> I'm not sure about the supply here, it's not actually in the vendor
> sources. It seems the bootloader is setting this up. The regulator
> naming suggests, that it would be LDO23. What should I do here?
> Currently no driver is using it but it's needed for the backlight
> dimming which I currently still have some trouble with.
The LDO23 is most likely for the touch panel ADC, not for the SoC one.
Here it could be LDO3. You have to put something otherwise dtschema will
complain.
Go with the LDO3 and maybe a comment like: "TBD, not confirmed".
>
> > > +};
>
> (...)
>
> > > +
> > > +&i2c_7 {
> > > + samsung,i2c-sda-delay = <100>;
> > > + samsung,i2c-slave-addr = <0x10>;
> > > + samsung,i2c-max-bus-freq = <400000>;
> > > + pinctrl-0 = <&i2c7_bus>;
> > > + pinctrl-names = "default";
> > > + status = "okay";
> > > +
> > > + max77686: max77686_pmic@9 {
> > > + compatible = "maxim,max77686";
> > > + interrupt-parent = <&gpx0>;
> > > + interrupts = <7 IRQ_TYPE_NONE>;
> > > + pinctrl-0 = <&max77686_irq>;
> > > + pinctrl-names = "default";
> > > + reg = <0x09>;
> > > + #clock-cells = <1>;
> > > +
> > > + voltage-regulators {
> >
> > Just "regulators" and no empty line after this.
> >
> > You should define all regulators here. If some are unknown, keep the
> > min-max the same as in driver.
> >
>
> I used the driver to read all the voltages, it also reads values for
> the LDOs that are not defined in the DT, so I have values for all of
> them setup by the bootloader. What I don't know is their name and what
> they are used for. Interestingly apart from two LDOs, the voltages did
> match exactly with those from the midas DT, so it could be very
> similar. The two non-matching were only off by 0.1V. How should I go
> about this?
Actually, let's skip the constraints (min-max) for unknown regulators.
Just add the entries with generic names, like:
arch/arm/boot/dts/exynos5410-odroidxu.dts
I would not put here min-max constraints from bootloader because pretty
often it has only an initial setup for specific frequency - just to make
it booting.
If you have the constraints (min-max, always on) from vendor sources,
you could add them.
Best regards,
Krzysztof
next prev parent reply other threads:[~2020-10-27 8:17 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-10-24 14:44 [PATCH 0/5] add exynos4412-based p4note device family support Martin Jücker
2020-10-24 14:44 ` [PATCH 1/5] dt-bindings: samsung: document bindings for p4note family boards Martin Jücker
2020-10-30 18:28 ` Rob Herring
2020-10-24 14:44 ` [PATCH 2/5] ARM: dts: add Samsung's exynos4412-based p4note boards Martin Jücker
2020-10-26 11:55 ` Marek Szyprowski
2020-10-26 14:48 ` Krzysztof Kozlowski
2020-10-26 16:00 ` Krzysztof Kozlowski
2020-10-26 23:27 ` Martin Juecker
2020-10-27 8:17 ` Krzysztof Kozlowski [this message]
2020-10-29 21:40 ` Martin Juecker
2020-10-30 7:41 ` Marek Szyprowski
2020-10-24 14:44 ` [PATCH 3/5] ARM: defconfig: add stmpe adc driver for p4note Martin Jücker
2020-10-24 14:44 ` [PATCH 4/5] ARM: defconfig: compile Atmel MXT touchscreeen as module Martin Jücker
2020-10-24 14:44 ` [PATCH 5/5] ARM: exynos: extend cpuidle support to p4note boards Martin Jücker
-- strict thread matches above, loose matches on Subject: below --
2020-10-31 17:58 [PATCH v2 0/5] add exynos4412-based p4note device family support Martin Jücker
2020-10-31 17:58 ` [PATCH 2/5] ARM: dts: add Samsung's exynos4412-based p4note boards Martin Jücker
2020-11-02 19:18 ` Krzysztof Kozlowski
2020-11-02 21:21 ` Martin Juecker
2020-11-03 7:53 ` Krzysztof Kozlowski
2020-11-03 9:51 ` Marek Szyprowski
2020-11-03 10:38 ` Martin Juecker
2020-11-05 11:50 ` Marek Szyprowski
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20201027081751.GA4244@kozik-lap \
--to=krzk@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=kgene@kernel.org \
--cc=linux-samsung-soc@vger.kernel.org \
--cc=martin.juecker@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).