| Message ID | ce704da79c6263a4a5048af8f8e90f22e0849188.1471637309.git.ccc94453@vip.cybercity.dk (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Hi, On Sat, Aug 13, 2016 at 12:03:57AM +0200, Rask Ingemann Lambertsen wrote: > The Suncip CX-A99 board is found in at least four brands of TV boxes. > It features an Allwinner A80 SOC, with either 2 GiB DDR3 DRAM and > 16 GB eMMC or 4 GiB DDR3 DRAM and 32 GB eMMC, as well as several support > chips for Ethernet, wireless networking, video output, SATA and power > management. For details, see the linux-sunxi page about the board > <URL:https://linux-sunxi.org/Sunchip_CX-A99>. > > This patch only adds support for the SD and eMMC storage, real-time clock, > USB 2.0 ports (and by extension the SATA port), the UART port and the LEDs. > All of this relies on the boot loader to leave those parts powered up, as > I'm still working on a driver for the AXP808 PMIC. > > Signed-off-by: Rask Ingemann Lambertsen <ccc94453@vip.cybercity.dk> It looks mostly good, but I have a couple of comments though, see below. > --- > > Although the vendor U-Boot lets you boot the kernel on one of the > Cortex-A15 cores, the kernel gpio-regulator driver currently glitches > the GPIO lines to the OZ80120 regulator such that the system crashes > during startup. This part needs further work to be useful. So it doesn't power the CPU through one of the AXP regulators? Interesting design. > The driver for the AC100 real-time clock is found in Lee Jones' MFD branch > git://git.kernel.org/pub/scm/linux/kernel/git/lee/mfd.git ib-mfd-rtc-v4.9 > and is not essential for using the board. > > Changes in v4: > Node names had underscores changed to hyphens. > Changed formatting of the ac100/rtc node's clock output name list to match > that of the same node in the cubieboard4 and a80-optimus device trees. > > Changes in v3: > None. > > Changes in v2: > Fixed formatting and style issues found by scripts/checkpatch.pl. > > arch/arm/boot/dts/Makefile | 3 +- > arch/arm/boot/dts/sun9i-a80-cx-a99.dts | 261 +++++++++++++++++++++++++++++++++ > 2 files changed, 263 insertions(+), 1 deletion(-) > create mode 100644 arch/arm/boot/dts/sun9i-a80-cx-a99.dts > > diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile > index 6b3bdb6..8b77a53 100644 > --- a/arch/arm/boot/dts/Makefile > +++ b/arch/arm/boot/dts/Makefile > @@ -796,7 +796,8 @@ dtb-$(CONFIG_MACH_SUN8I) += \ > sun8i-r16-parrot.dtb > dtb-$(CONFIG_MACH_SUN9I) += \ > sun9i-a80-optimus.dtb \ > - sun9i-a80-cubieboard4.dtb > + sun9i-a80-cubieboard4.dtb \ > + sun9i-a80-cx-a99.dtb > dtb-$(CONFIG_ARCH_TANGO) += \ > tango4-vantage-1172.dtb > dtb-$(CONFIG_ARCH_TEGRA_2x_SOC) += \ > diff --git a/arch/arm/boot/dts/sun9i-a80-cx-a99.dts b/arch/arm/boot/dts/sun9i-a80-cx-a99.dts > new file mode 100644 > index 0000000..d08a35e > --- /dev/null > +++ b/arch/arm/boot/dts/sun9i-a80-cx-a99.dts > @@ -0,0 +1,261 @@ > +/* > + * sun9i-a80-cx-a99.dts - Device Tree file for the Sunchip CX-A99 board. > + * > + * Copyright (C) 2016 Rask Ingemann Lamberten <ccc94453@vip.cybercity.dk> > + * > + * This file is dual-licensed: you can use it either under the terms > + * of the GPL or the X11 license, at your option. Note that this dual > + * licensing only applies to this file, and not this project as a > + * whole. > + * > + * a) This file is free software; you can redistribute it and/or > + * modify it under the terms of the GNU General Public License as > + * published by the Free Software Foundation; either version 2 of the > + * License, or (at your option) any later version. > + * > + * This file is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + * > + * Or, alternatively, > + * > + * b) Permission is hereby granted, free of charge, to any person > + * obtaining a copy of this software and associated documentation > + * files (the "Software"), to deal in the Software without > + * restriction, including without limitation the rights to use, > + * copy, modify, merge, publish, distribute, sublicense, and/or > + * sell copies of the Software, and to permit persons to whom the > + * Software is furnished to do so, subject to the following > + * conditions: > + * > + * The above copyright notice and this permission notice shall be > + * included in all copies or substantial portions of the Software. > + * > + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, > + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES > + * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND > + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT > + * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, > + * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING > + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR > + * OTHER DEALINGS IN THE SOFTWARE. > + */ > + > +/* The Sunchip CX-A99 board is found in several similar Android TV box > + * products, such as: > + * > + * Instabox Fantasy A8 (no external antenna) > + * Jesurun CS-Q8 (ships with larger remote control) > + * Jesurun Maxone > + * Rikomagic (RKM) MK80/MK80LE > + * Tronsmart Draco AW80 Meta/Telos > + * > + * See <URL:https://linux-sunxi.org/Sunchip_cx_a99> for more information. > + */ > + > +/dts-v1/; > +#include "sun9i-a80.dtsi" > +#include "sunxi-common-regulators.dtsi" > + > +#include <dt-bindings/gpio/gpio.h> > +#include <dt-bindings/pinctrl/sun4i-a10.h> > + > +/ { > + model = "Sunchip CX-A99"; > + compatible = "sunchip,cx-a99", "allwinner,sun9i-a80"; > + > + aliases { > + serial0 = &uart0; > + }; > + > + chosen { > + stdout-path = "serial0:115200n8"; > + }; > + > + leds { > + compatible = "gpio-leds"; > + pinctrl-names = "default"; > + pinctrl-0 = <&led_pins_cx_a99>; > + > + blue { > + gpios = <&pio 6 10 GPIO_ACTIVE_HIGH>; /* PG10 */ > + label = "cx-a99:blue:normal"; > + default-state = "on"; > + }; > + > + red { > + gpios = <&pio 6 11 GPIO_ACTIVE_HIGH>; /* PG11 */ > + label = "cx-a99:red:standby"; The last part of the label should be the function. > + }; > + }; > + > + /* Power supply for the four Cortex-A15 cores. */ > + reg_vdd_cpub: oz80120 { > + compatible = "regulator-gpio"; > + regulator-name = "vdd-cpub"; > + > + regulator-boot-on; > + regulator-always-on; > + regulator-type = "voltage"; > + regulator-min-microvolt = < 800000>; > + regulator-max-microvolt = <1100000>; > + > + pinctrl-names = "default"; > + pinctrl-0 = <&vdd_cpub_r_pins_cx_a99>; > + enable-gpio = <&r_pio 0 2 GPIO_ACTIVE_HIGH>; /* PL2 */ > + enable-active-high; > + gpios = <&r_pio 0 3 GPIO_ACTIVE_HIGH>, /* PL3 */ > + <&r_pio 0 4 GPIO_ACTIVE_HIGH>, /* PL4 */ > + <&r_pio 0 5 GPIO_ACTIVE_HIGH>; /* PL5 */ > + /* There appears to be an external pull-up on PL3. */ > + gpios-states = <1 0 0>; > + states = < 750000 0x7 > + 800000 0x3 > + 850000 0x5 > + 900000 0x1 > + 950000 0x6 > + 1000000 0x2 > + 1100000 0x4 > + 1200000 0x0>; > + }; > +}; > + > +&ehci0 { > + status = "okay"; > +}; > + > +&ehci2 { > + status = "okay"; > +}; > + > +&mmc0 { > + pinctrl-names = "default"; > + pinctrl-0 = <&mmc0_pins>, <&mmc0_cd_pin_cx_a99>; > + vmmc-supply = <®_vcc3v3>; > + bus-width = <4>; > + cd-gpios = <&pio 7 17 GPIO_ACTIVE_LOW>; /* PH17 */ > + status = "okay"; > +}; > + > +&mmc2 { > + pinctrl-names = "default"; > + pinctrl-0 = <&mmc2_8bit_pins>; > + vmmc-supply = <®_vcc3v3>; > + bus-width = <8>; > + non-removable; > + status = "okay"; > +}; > + > +&osc32k { > + clocks = <&ac100_rtc 0>; > +}; > + > +&pio { > + led_pins_cx_a99: led-pins { > + allwinner,pins = "PG10", "PG11"; > + allwinner,function = "gpio_out"; > + allwinner,drive = <SUN4I_PINCTRL_10_MA>; > + allwinner,pull = <SUN4I_PINCTRL_NO_PULL>; > + }; > + > + mmc0_cd_pin_cx_a99: mmc0-cd { > + allwinner,pins = "PH17"; > + allwinner,function = "gpio_in"; > + allwinner,pull = <SUN4I_PINCTRL_PULL_UP>; > + }; > +}; > + > +/* Each GPIO controls VBUS for a port on the GL850G hub connected to ehci0; > + * PL7 for port 1, the USB connector closest to the 12 V power connector, and > + * PL8 for port 2, the USB connector next to the (micro)SD card slot. > + * Note: The regulators are not chained like this in reality, but > + * regulator-fixed doesn't support a gpio list, and allwinner,sun9i-a80-usb-phy > + * doesn't support more than one supply, so this will have to do (for now). > + */ > +®_usb1_vbus { > + pinctrl-0 = <&usb1_vbus_r_pin_cx_a99>; > + gpio = <&r_pio 0 7 GPIO_ACTIVE_HIGH>; /* PL7 */ > + status = "okay"; > +}; > + > +®_usb2_vbus { > + pinctrl-0 = <&usb2_vbus_r_pin_cx_a99>; > + gpio = <&r_pio 0 8 GPIO_ACTIVE_HIGH>; /* PL8 */ I'd really prefer it to be modelled properly, and not attached to the wrong device. I have some work pending to allow multiple regulators in the same property, but that won't be ready soon. For the time being, I would suggest having two usb1 regulators defined, each with their respective GPIOs, and one set to always on (and keep that great comment). Thanks, Maxime
On Mon, Aug 22, 2016 at 08:57:45PM +0200, Maxime Ripard wrote: > Hi, > > On Sat, Aug 13, 2016 at 12:03:57AM +0200, Rask Ingemann Lambertsen wrote: > > The Suncip CX-A99 board is found in at least four brands of TV boxes. > > It features an Allwinner A80 SOC, with either 2 GiB DDR3 DRAM and > > 16 GB eMMC or 4 GiB DDR3 DRAM and 32 GB eMMC, as well as several support > > chips for Ethernet, wireless networking, video output, SATA and power > > management. For details, see the linux-sunxi page about the board > > <URL:https://linux-sunxi.org/Sunchip_CX-A99>. > > > > This patch only adds support for the SD and eMMC storage, real-time clock, > > USB 2.0 ports (and by extension the SATA port), the UART port and the LEDs. > > All of this relies on the boot loader to leave those parts powered up, as > > I'm still working on a driver for the AXP808 PMIC. > > > > Signed-off-by: Rask Ingemann Lambertsen <ccc94453@vip.cybercity.dk> > > It looks mostly good, but I have a couple of comments though, see below. > > > --- > > > > Although the vendor U-Boot lets you boot the kernel on one of the > > Cortex-A15 cores, the kernel gpio-regulator driver currently glitches > > the GPIO lines to the OZ80120 regulator such that the system crashes > > during startup. This part needs further work to be useful. > > So it doesn't power the CPU through one of the AXP regulators? > Interesting design. The Cortex-A7 cores are powered by the dcdca regulator on the AXP 808 PMIC. Right now, I'm using the AXP 806 driver that Chen-Yu Tsai posted to drive the AXP 808 and so far, it looks good. [snip] > > + leds { > > + compatible = "gpio-leds"; > > + pinctrl-names = "default"; > > + pinctrl-0 = <&led_pins_cx_a99>; > > + > > + blue { > > + gpios = <&pio 6 10 GPIO_ACTIVE_HIGH>; /* PG10 */ > > + label = "cx-a99:blue:normal"; > > + default-state = "on"; > > + }; > > + > > + red { > > + gpios = <&pio 6 11 GPIO_ACTIVE_HIGH>; /* PG11 */ > > + label = "cx-a99:red:standby"; > > The last part of the label should be the function. I'm not sure what else to label them. They form a bi-colour LED emitting through the front of the device. The stock OS lights the blue LED when the device is powered on and lights the red LED when the device is suspended. There is no label on the front of the device telling what the colours mean. Documentation/devicetree/bindings/leds/common.txt and Documentation/devicetree/bindings/leds/leds-gpio.txt don't provide much in the way of examples. Suggestions are welcome. [snip] > > + > > +/* Each GPIO controls VBUS for a port on the GL850G hub connected to ehci0; > > + * PL7 for port 1, the USB connector closest to the 12 V power connector, and > > + * PL8 for port 2, the USB connector next to the (micro)SD card slot. > > + * Note: The regulators are not chained like this in reality, but > > + * regulator-fixed doesn't support a gpio list, and allwinner,sun9i-a80-usb-phy > > + * doesn't support more than one supply, so this will have to do (for now). > > + */ > > +®_usb1_vbus { > > + pinctrl-0 = <&usb1_vbus_r_pin_cx_a99>; > > + gpio = <&r_pio 0 7 GPIO_ACTIVE_HIGH>; /* PL7 */ > > + status = "okay"; > > +}; > > + > > +®_usb2_vbus { > > + pinctrl-0 = <&usb2_vbus_r_pin_cx_a99>; > > + gpio = <&r_pio 0 8 GPIO_ACTIVE_HIGH>; /* PL8 */ > > I'd really prefer it to be modelled properly, and not attached to the > wrong device. > > I have some work pending to allow multiple regulators in the same > property, but that won't be ready soon. > > For the time being, I would suggest having two usb1 regulators > defined, each with their respective GPIOs, and one set to always on > (and keep that great comment). Will do if I don't come up with something better. I gave it a shot to describe the hub as a child of ehci0 with a child node for each of the two ports, but it didn't work. Also, using the phy-supply property for the vbus-supply is an ugly hack, in the first place, isn't it? Shouldn't it be more like this? &usbphy1 { phy-supply = <®_vcc33_usbh>; }; &ehci0 { vcc-supply = <®_vdd09_usbh>; phy = <&usbphy1>; hub@1 { port@1 { vbus-supply = <®_usb1_vbus>; } port@2 { vbus-supply = <®_usb2_vbus>; } }; }; It would generally be great to be able to describe regulators that should be kept in sync, because the Wifi+BT module's Vbat pin is supplied by two regulators in parallel. IIRC, Hans de Goede ran into that as well on one of his boards.
On Mon, Aug 22, 2016 at 11:29:59PM +0200, Rask Ingemann Lambertsen wrote: > On Mon, Aug 22, 2016 at 08:57:45PM +0200, Maxime Ripard wrote: > > Hi, > > > > On Sat, Aug 13, 2016 at 12:03:57AM +0200, Rask Ingemann Lambertsen wrote: > > > The Suncip CX-A99 board is found in at least four brands of TV boxes. > > > It features an Allwinner A80 SOC, with either 2 GiB DDR3 DRAM and > > > 16 GB eMMC or 4 GiB DDR3 DRAM and 32 GB eMMC, as well as several support > > > chips for Ethernet, wireless networking, video output, SATA and power > > > management. For details, see the linux-sunxi page about the board > > > <URL:https://linux-sunxi.org/Sunchip_CX-A99>. > > > > > > This patch only adds support for the SD and eMMC storage, real-time clock, > > > USB 2.0 ports (and by extension the SATA port), the UART port and the LEDs. > > > All of this relies on the boot loader to leave those parts powered up, as > > > I'm still working on a driver for the AXP808 PMIC. > > > > > > Signed-off-by: Rask Ingemann Lambertsen <ccc94453@vip.cybercity.dk> > > > > It looks mostly good, but I have a couple of comments though, see below. > > > > > --- > > > > > > Although the vendor U-Boot lets you boot the kernel on one of the > > > Cortex-A15 cores, the kernel gpio-regulator driver currently glitches > > > the GPIO lines to the OZ80120 regulator such that the system crashes > > > during startup. This part needs further work to be useful. > > > > So it doesn't power the CPU through one of the AXP regulators? > > Interesting design. > > The Cortex-A7 cores are powered by the dcdca regulator on the AXP 808 PMIC. > Right now, I'm using the AXP 806 driver that Chen-Yu Tsai posted to drive > the AXP 808 and so far, it looks good. Ok. > > > + leds { > > > + compatible = "gpio-leds"; > > > + pinctrl-names = "default"; > > > + pinctrl-0 = <&led_pins_cx_a99>; > > > + > > > + blue { > > > + gpios = <&pio 6 10 GPIO_ACTIVE_HIGH>; /* PG10 */ > > > + label = "cx-a99:blue:normal"; > > > + default-state = "on"; > > > + }; > > > + > > > + red { > > > + gpios = <&pio 6 11 GPIO_ACTIVE_HIGH>; /* PG11 */ > > > + label = "cx-a99:red:standby"; > > > > The last part of the label should be the function. > > I'm not sure what else to label them. They form a bi-colour LED emitting > through the front of the device. The stock OS lights the blue LED when > the device is powered on and lights the red LED when the device is > suspended. There is no label on the front of the device telling what the > colours mean. Documentation/devicetree/bindings/leds/common.txt and > Documentation/devicetree/bindings/leds/leds-gpio.txt don't provide much in > the way of examples. Suggestions are welcome. Hmmmm. status for both then? > [snip] > > > + > > > +/* Each GPIO controls VBUS for a port on the GL850G hub connected to ehci0; > > > + * PL7 for port 1, the USB connector closest to the 12 V power connector, and > > > + * PL8 for port 2, the USB connector next to the (micro)SD card slot. > > > + * Note: The regulators are not chained like this in reality, but > > > + * regulator-fixed doesn't support a gpio list, and allwinner,sun9i-a80-usb-phy > > > + * doesn't support more than one supply, so this will have to do (for now). > > > + */ > > > +®_usb1_vbus { > > > + pinctrl-0 = <&usb1_vbus_r_pin_cx_a99>; > > > + gpio = <&r_pio 0 7 GPIO_ACTIVE_HIGH>; /* PL7 */ > > > + status = "okay"; > > > +}; > > > + > > > +®_usb2_vbus { > > > + pinctrl-0 = <&usb2_vbus_r_pin_cx_a99>; > > > + gpio = <&r_pio 0 8 GPIO_ACTIVE_HIGH>; /* PL8 */ > > > > I'd really prefer it to be modelled properly, and not attached to the > > wrong device. > > > > I have some work pending to allow multiple regulators in the same > > property, but that won't be ready soon. > > > > For the time being, I would suggest having two usb1 regulators > > defined, each with their respective GPIOs, and one set to always on > > (and keep that great comment). > > Will do if I don't come up with something better. I gave it a shot to > describe the hub as a child of ehci0 with a child node for each of the > two ports, but it didn't work. > > Also, using the phy-supply property for the vbus-supply is an ugly hack, > in the first place, isn't it? Shouldn't it be more like this? > > &usbphy1 { > phy-supply = <®_vcc33_usbh>; > }; > > &ehci0 { > vcc-supply = <®_vdd09_usbh>; > phy = <&usbphy1>; > > hub@1 { > port@1 { > vbus-supply = <®_usb1_vbus>; > } > > port@2 { > vbus-supply = <®_usb2_vbus>; > } > }; > }; It looks great to me. I'm not really sure how happy the DT maintainers are going to be about it, and how easy it would be to support without breaking the existing users. > It would generally be great to be able to describe regulators that should > be kept in sync, because the Wifi+BT module's Vbat pin is supplied by two > regulators in parallel. IIRC, Hans de Goede ran into that as well on one > of his boards. Yes, we got the same issue on the CHIP. I'm the one with the hot potato, but that would require a significant rework of the regulator framework, that I haven't had the time to do yet. Thanks! Maxime
On Thu, Aug 25, 2016 at 3:44 AM, Maxime Ripard <maxime.ripard@free-electrons.com> wrote: > On Mon, Aug 22, 2016 at 11:29:59PM +0200, Rask Ingemann Lambertsen wrote: >> On Mon, Aug 22, 2016 at 08:57:45PM +0200, Maxime Ripard wrote: >> > Hi, >> > >> > On Sat, Aug 13, 2016 at 12:03:57AM +0200, Rask Ingemann Lambertsen wrote: >> > > The Suncip CX-A99 board is found in at least four brands of TV boxes. >> > > It features an Allwinner A80 SOC, with either 2 GiB DDR3 DRAM and >> > > 16 GB eMMC or 4 GiB DDR3 DRAM and 32 GB eMMC, as well as several support >> > > chips for Ethernet, wireless networking, video output, SATA and power >> > > management. For details, see the linux-sunxi page about the board >> > > <URL:https://linux-sunxi.org/Sunchip_CX-A99>. >> > > >> > > This patch only adds support for the SD and eMMC storage, real-time clock, >> > > USB 2.0 ports (and by extension the SATA port), the UART port and the LEDs. >> > > All of this relies on the boot loader to leave those parts powered up, as >> > > I'm still working on a driver for the AXP808 PMIC. >> > > >> > > Signed-off-by: Rask Ingemann Lambertsen <ccc94453@vip.cybercity.dk> >> > >> > It looks mostly good, but I have a couple of comments though, see below. >> > >> > > --- >> > > >> > > Although the vendor U-Boot lets you boot the kernel on one of the >> > > Cortex-A15 cores, the kernel gpio-regulator driver currently glitches >> > > the GPIO lines to the OZ80120 regulator such that the system crashes >> > > during startup. This part needs further work to be useful. >> > >> > So it doesn't power the CPU through one of the AXP regulators? >> > Interesting design. >> >> The Cortex-A7 cores are powered by the dcdca regulator on the AXP 808 PMIC. >> Right now, I'm using the AXP 806 driver that Chen-Yu Tsai posted to drive >> the AXP 808 and so far, it looks good. > > Ok. Could you look around for AXP808 datasheets? > >> > > + leds { >> > > + compatible = "gpio-leds"; >> > > + pinctrl-names = "default"; >> > > + pinctrl-0 = <&led_pins_cx_a99>; >> > > + >> > > + blue { >> > > + gpios = <&pio 6 10 GPIO_ACTIVE_HIGH>; /* PG10 */ >> > > + label = "cx-a99:blue:normal"; >> > > + default-state = "on"; >> > > + }; >> > > + >> > > + red { >> > > + gpios = <&pio 6 11 GPIO_ACTIVE_HIGH>; /* PG11 */ >> > > + label = "cx-a99:red:standby"; >> > >> > The last part of the label should be the function. >> >> I'm not sure what else to label them. They form a bi-colour LED emitting >> through the front of the device. The stock OS lights the blue LED when >> the device is powered on and lights the red LED when the device is >> suspended. There is no label on the front of the device telling what the >> colours mean. Documentation/devicetree/bindings/leds/common.txt and >> Documentation/devicetree/bindings/leds/leds-gpio.txt don't provide much in >> the way of examples. Suggestions are welcome. > > Hmmmm. status for both then? > >> [snip] >> > > + >> > > +/* Each GPIO controls VBUS for a port on the GL850G hub connected to ehci0; >> > > + * PL7 for port 1, the USB connector closest to the 12 V power connector, and >> > > + * PL8 for port 2, the USB connector next to the (micro)SD card slot. >> > > + * Note: The regulators are not chained like this in reality, but >> > > + * regulator-fixed doesn't support a gpio list, and allwinner,sun9i-a80-usb-phy >> > > + * doesn't support more than one supply, so this will have to do (for now). >> > > + */ >> > > +®_usb1_vbus { >> > > + pinctrl-0 = <&usb1_vbus_r_pin_cx_a99>; >> > > + gpio = <&r_pio 0 7 GPIO_ACTIVE_HIGH>; /* PL7 */ >> > > + status = "okay"; >> > > +}; >> > > + >> > > +®_usb2_vbus { >> > > + pinctrl-0 = <&usb2_vbus_r_pin_cx_a99>; >> > > + gpio = <&r_pio 0 8 GPIO_ACTIVE_HIGH>; /* PL8 */ >> > >> > I'd really prefer it to be modelled properly, and not attached to the >> > wrong device. >> > >> > I have some work pending to allow multiple regulators in the same >> > property, but that won't be ready soon. >> > >> > For the time being, I would suggest having two usb1 regulators >> > defined, each with their respective GPIOs, and one set to always on >> > (and keep that great comment). >> >> Will do if I don't come up with something better. I gave it a shot to >> describe the hub as a child of ehci0 with a child node for each of the >> two ports, but it didn't work. >> >> Also, using the phy-supply property for the vbus-supply is an ugly hack, >> in the first place, isn't it? Shouldn't it be more like this? Is it. Back then when I added USB support I didn't notice the PHYs had separate power supplies, and 2 even. This will be reworked in the future. >> >> &usbphy1 { >> phy-supply = <®_vcc33_usbh>; >> }; >> >> &ehci0 { >> vcc-supply = <®_vdd09_usbh>; >> phy = <&usbphy1>; >> >> hub@1 { >> port@1 { >> vbus-supply = <®_usb1_vbus>; >> } >> >> port@2 { >> vbus-supply = <®_usb2_vbus>; >> } >> }; >> }; > > It looks great to me. I'm not really sure how happy the DT maintainers > are going to be about it, and how easy it would be to support without > breaking the existing users. The hub and device power sequencing stuff is being worked on, though I've lost track of the progress. I don't think regulators were included in the initial proposal though. ChenYu >> It would generally be great to be able to describe regulators that should >> be kept in sync, because the Wifi+BT module's Vbat pin is supplied by two >> regulators in parallel. IIRC, Hans de Goede ran into that as well on one >> of his boards. > > Yes, we got the same issue on the CHIP. I'm the one with the hot > potato, but that would require a significant rework of the regulator > framework, that I haven't had the time to do yet. > > Thanks! > Maxime > > -- > Maxime Ripard, Free Electrons > Embedded Linux and Kernel engineering > http://free-electrons.com
diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile index 6b3bdb6..8b77a53 100644 --- a/arch/arm/boot/dts/Makefile +++ b/arch/arm/boot/dts/Makefile @@ -796,7 +796,8 @@ dtb-$(CONFIG_MACH_SUN8I) += \ sun8i-r16-parrot.dtb dtb-$(CONFIG_MACH_SUN9I) += \ sun9i-a80-optimus.dtb \ - sun9i-a80-cubieboard4.dtb + sun9i-a80-cubieboard4.dtb \ + sun9i-a80-cx-a99.dtb dtb-$(CONFIG_ARCH_TANGO) += \ tango4-vantage-1172.dtb dtb-$(CONFIG_ARCH_TEGRA_2x_SOC) += \ diff --git a/arch/arm/boot/dts/sun9i-a80-cx-a99.dts b/arch/arm/boot/dts/sun9i-a80-cx-a99.dts new file mode 100644 index 0000000..d08a35e --- /dev/null +++ b/arch/arm/boot/dts/sun9i-a80-cx-a99.dts @@ -0,0 +1,261 @@ +/* + * sun9i-a80-cx-a99.dts - Device Tree file for the Sunchip CX-A99 board. + * + * Copyright (C) 2016 Rask Ingemann Lamberten <ccc94453@vip.cybercity.dk> + * + * This file is dual-licensed: you can use it either under the terms + * of the GPL or the X11 license, at your option. Note that this dual + * licensing only applies to this file, and not this project as a + * whole. + * + * a) This file is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; either version 2 of the + * License, or (at your option) any later version. + * + * This file is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * Or, alternatively, + * + * b) Permission is hereby granted, free of charge, to any person + * obtaining a copy of this software and associated documentation + * files (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, + * copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following + * conditions: + * + * The above copyright notice and this permission notice shall be + * included in all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES + * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT + * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, + * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ + +/* The Sunchip CX-A99 board is found in several similar Android TV box + * products, such as: + * + * Instabox Fantasy A8 (no external antenna) + * Jesurun CS-Q8 (ships with larger remote control) + * Jesurun Maxone + * Rikomagic (RKM) MK80/MK80LE + * Tronsmart Draco AW80 Meta/Telos + * + * See <URL:https://linux-sunxi.org/Sunchip_cx_a99> for more information. + */ + +/dts-v1/; +#include "sun9i-a80.dtsi" +#include "sunxi-common-regulators.dtsi" + +#include <dt-bindings/gpio/gpio.h> +#include <dt-bindings/pinctrl/sun4i-a10.h> + +/ { + model = "Sunchip CX-A99"; + compatible = "sunchip,cx-a99", "allwinner,sun9i-a80"; + + aliases { + serial0 = &uart0; + }; + + chosen { + stdout-path = "serial0:115200n8"; + }; + + leds { + compatible = "gpio-leds"; + pinctrl-names = "default"; + pinctrl-0 = <&led_pins_cx_a99>; + + blue { + gpios = <&pio 6 10 GPIO_ACTIVE_HIGH>; /* PG10 */ + label = "cx-a99:blue:normal"; + default-state = "on"; + }; + + red { + gpios = <&pio 6 11 GPIO_ACTIVE_HIGH>; /* PG11 */ + label = "cx-a99:red:standby"; + }; + }; + + /* Power supply for the four Cortex-A15 cores. */ + reg_vdd_cpub: oz80120 { + compatible = "regulator-gpio"; + regulator-name = "vdd-cpub"; + + regulator-boot-on; + regulator-always-on; + regulator-type = "voltage"; + regulator-min-microvolt = < 800000>; + regulator-max-microvolt = <1100000>; + + pinctrl-names = "default"; + pinctrl-0 = <&vdd_cpub_r_pins_cx_a99>; + enable-gpio = <&r_pio 0 2 GPIO_ACTIVE_HIGH>; /* PL2 */ + enable-active-high; + gpios = <&r_pio 0 3 GPIO_ACTIVE_HIGH>, /* PL3 */ + <&r_pio 0 4 GPIO_ACTIVE_HIGH>, /* PL4 */ + <&r_pio 0 5 GPIO_ACTIVE_HIGH>; /* PL5 */ + /* There appears to be an external pull-up on PL3. */ + gpios-states = <1 0 0>; + states = < 750000 0x7 + 800000 0x3 + 850000 0x5 + 900000 0x1 + 950000 0x6 + 1000000 0x2 + 1100000 0x4 + 1200000 0x0>; + }; +}; + +&ehci0 { + status = "okay"; +}; + +&ehci2 { + status = "okay"; +}; + +&mmc0 { + pinctrl-names = "default"; + pinctrl-0 = <&mmc0_pins>, <&mmc0_cd_pin_cx_a99>; + vmmc-supply = <®_vcc3v3>; + bus-width = <4>; + cd-gpios = <&pio 7 17 GPIO_ACTIVE_LOW>; /* PH17 */ + status = "okay"; +}; + +&mmc2 { + pinctrl-names = "default"; + pinctrl-0 = <&mmc2_8bit_pins>; + vmmc-supply = <®_vcc3v3>; + bus-width = <8>; + non-removable; + status = "okay"; +}; + +&osc32k { + clocks = <&ac100_rtc 0>; +}; + +&pio { + led_pins_cx_a99: led-pins { + allwinner,pins = "PG10", "PG11"; + allwinner,function = "gpio_out"; + allwinner,drive = <SUN4I_PINCTRL_10_MA>; + allwinner,pull = <SUN4I_PINCTRL_NO_PULL>; + }; + + mmc0_cd_pin_cx_a99: mmc0-cd { + allwinner,pins = "PH17"; + allwinner,function = "gpio_in"; + allwinner,pull = <SUN4I_PINCTRL_PULL_UP>; + }; +}; + +/* Each GPIO controls VBUS for a port on the GL850G hub connected to ehci0; + * PL7 for port 1, the USB connector closest to the 12 V power connector, and + * PL8 for port 2, the USB connector next to the (micro)SD card slot. + * Note: The regulators are not chained like this in reality, but + * regulator-fixed doesn't support a gpio list, and allwinner,sun9i-a80-usb-phy + * doesn't support more than one supply, so this will have to do (for now). + */ +®_usb1_vbus { + pinctrl-0 = <&usb1_vbus_r_pin_cx_a99>; + gpio = <&r_pio 0 7 GPIO_ACTIVE_HIGH>; /* PL7 */ + status = "okay"; +}; + +®_usb2_vbus { + pinctrl-0 = <&usb2_vbus_r_pin_cx_a99>; + gpio = <&r_pio 0 8 GPIO_ACTIVE_HIGH>; /* PL8 */ + vin-supply = <®_usb1_vbus>; + status = "okay"; +}; + +&r_ir { + status = "okay"; +}; + +&r_pio { + usb1_vbus_r_pin_cx_a99: usb0-vbus-pin { + allwinner,pins = "PL7"; + allwinner,function = "gpio_out"; + allwinner,drive = <SUN4I_PINCTRL_10_MA>; + allwinner,pull = <SUN4I_PINCTRL_NO_PULL>; + }; + + usb2_vbus_r_pin_cx_a99: usb1-vbus-pin { + allwinner,pins = "PL8"; + allwinner,function = "gpio_out"; + allwinner,drive = <SUN4I_PINCTRL_10_MA>; + allwinner,pull = <SUN4I_PINCTRL_NO_PULL>; + }; + + vdd_cpub_r_pins_cx_a99: regulator-r-pins { + allwinner,pins = "PL2", "PL3", "PL4", "PL5"; + allwinner,function = "gpio_out"; + allwinner,drive = <SUN4I_PINCTRL_10_MA>; + allwinner,pull = <SUN4I_PINCTRL_NO_PULL>; + }; +}; + +&r_rsb { + status = "okay"; + + ac100: codec@e89 { + compatible = "x-powers,ac100"; + reg = <0xe89>; + + ac100_codec: codec { + compatible = "x-powers,ac100-codec"; + interrupt-parent = <&r_pio>; + interrupts = <0 9 IRQ_TYPE_LEVEL_LOW>; /* PL9 */ + #clock-cells = <0>; + clock-output-names = "4M_adda"; + }; + + ac100_rtc: rtc { + compatible = "x-powers,ac100-rtc"; + interrupt-parent = <&nmi_intc>; + interrupts = <0 IRQ_TYPE_LEVEL_LOW>; + clocks = <&ac100_codec>; + #clock-cells = <1>; + clock-output-names = "cko1_rtc", + "cko2_rtc", + "cko3_rtc"; + }; + }; +}; + +/* 5-pin connector opposite of the SD card slot: + * 1 = GND (pointed to by small triangle), 2 = GND, 3 = 3.3 V, 4 = RX, 5 = TX. + */ +&uart0 { + pinctrl-names = "default"; + pinctrl-0 = <&uart0_pins_a>; + status = "okay"; +}; + +/* The port has two V_bus supplies. See workaround at reg_usb1_vbus. */ +&usbphy1 { + phy-supply = <®_usb2_vbus>; + status = "okay"; +}; + +&usbphy3 { + status = "okay"; +};
The Suncip CX-A99 board is found in at least four brands of TV boxes. It features an Allwinner A80 SOC, with either 2 GiB DDR3 DRAM and 16 GB eMMC or 4 GiB DDR3 DRAM and 32 GB eMMC, as well as several support chips for Ethernet, wireless networking, video output, SATA and power management. For details, see the linux-sunxi page about the board <URL:https://linux-sunxi.org/Sunchip_CX-A99>. This patch only adds support for the SD and eMMC storage, real-time clock, USB 2.0 ports (and by extension the SATA port), the UART port and the LEDs. All of this relies on the boot loader to leave those parts powered up, as I'm still working on a driver for the AXP808 PMIC. Signed-off-by: Rask Ingemann Lambertsen <ccc94453@vip.cybercity.dk> --- Although the vendor U-Boot lets you boot the kernel on one of the Cortex-A15 cores, the kernel gpio-regulator driver currently glitches the GPIO lines to the OZ80120 regulator such that the system crashes during startup. This part needs further work to be useful. The driver for the AC100 real-time clock is found in Lee Jones' MFD branch git://git.kernel.org/pub/scm/linux/kernel/git/lee/mfd.git ib-mfd-rtc-v4.9 and is not essential for using the board. Changes in v4: Node names had underscores changed to hyphens. Changed formatting of the ac100/rtc node's clock output name list to match that of the same node in the cubieboard4 and a80-optimus device trees. Changes in v3: None. Changes in v2: Fixed formatting and style issues found by scripts/checkpatch.pl. arch/arm/boot/dts/Makefile | 3 +- arch/arm/boot/dts/sun9i-a80-cx-a99.dts | 261 +++++++++++++++++++++++++++++++++ 2 files changed, 263 insertions(+), 1 deletion(-) create mode 100644 arch/arm/boot/dts/sun9i-a80-cx-a99.dts