[LEDE-DEV] [PATCH 4/4] ramips/RT5350F-OLINUXINO(-EVB) dts: enable ttyS1

Martin Blumenstingl martin.blumenstingl at googlemail.com
Sat Sep 16 07:19:47 PDT 2017


Hi Zoltan,

On Wed, Sep 6, 2017 at 2:25 AM, Zoltan Gyarmati
<mr.zoltan.gyarmati at gmail.com> wrote:
> Dear John&All,
>
> thanks for the review, see my response inline
>
> On 08/29/2017 09:37 AM, John Crispin wrote:
>> Hi,
>>
>> comment inline
>>
>>
>> On 26/08/17 21:54, Zoltan Gyarmati wrote:
>>>   The RT5350F's second UART pins are available on the base module and on
>>> the EVB as well, so enable it in the device tree.
>>>   Additionaly, the uartlite at c00 and uart at 500 nodes swapped in
>>> rt5350.dtsi
>>> to keep the serial console as ttyS0.
>>>
>>> Signed-off-by: Zoltan Gyarmati <mr.zoltan.gyarmati at gmail.com>
>>> ---
>>>   target/linux/ramips/dts/RT5350F-OLINUXINO.dtsi | 11 +++++++++-
>>>   target/linux/ramips/dts/rt5350.dtsi            | 30
>>> +++++++++++++-------------
>>>   2 files changed, 25 insertions(+), 16 deletions(-)
>>>
>>> diff --git a/target/linux/ramips/dts/RT5350F-OLINUXINO.dtsi
>>> b/target/linux/ramips/dts/RT5350F-OLINUXINO.dtsi
>>> index 955a13cddd..1632f3c085 100644
>>> --- a/target/linux/ramips/dts/RT5350F-OLINUXINO.dtsi
>>> +++ b/target/linux/ramips/dts/RT5350F-OLINUXINO.dtsi
>>> @@ -46,9 +46,13 @@
>>>   &pinctrl {
>>>       state_default: pinctrl0 {
>>>           gpio {
>>> -            ralink,group = "jtag", "rgmii", "mdio", "uartf";
>>> +            ralink,group = "jtag", "rgmii", "mdio";
>>>               ralink,function = "gpio";
>>>           };
>>> +        uartf_gpio {
>>> +            ralink,group = "uartf";
>>> +            ralink,function = "gpio uartf";
>>> +        };
>>>       };
>>>   };
>>>   @@ -77,3 +81,8 @@
>>>   &i2c {
>>>       status = "okay";
>>>   };
>>> +
>>> +&uart {
>>> +    status = "okay";
>>> +};
>>> +
>>> diff --git a/target/linux/ramips/dts/rt5350.dtsi
>>> b/target/linux/ramips/dts/rt5350.dtsi
>>> index a92c113043..f027e17d9d 100644
>>> --- a/target/linux/ramips/dts/rt5350.dtsi
>>> +++ b/target/linux/ramips/dts/rt5350.dtsi
>>> @@ -83,21 +83,6 @@
>>>               interrupts = <3>;
>>>           };
>>>   -        uart: uart at 500 {
>>> -            compatible = "ralink,rt5350-uart", "ralink,rt2880-uart",
>>> "ns16550a";
>>> -            reg = <0x500 0x100>;
>>> -
>>> -            resets = <&rstctrl 12>;
>>> -            reset-names = "uart";
>>> -
>>> -            interrupt-parent = <&intc>;
>>> -            interrupts = <5>;
>>> -
>>> -            reg-shift = <2>;
>>> -
>>> -            status = "disabled";
>>> -        };
>>> -
>>>           gpio0: gpio at 600 {
>>>               compatible = "ralink,rt5350-gpio", "ralink,rt2880-gpio";
>>>               reg = <0x600 0x34>;
>>> @@ -221,6 +206,21 @@
>>>               reg-shift = <2>;
>>>           };
>>>   +        uart: uart at 500 {
>>> +            compatible = "ralink,rt5350-uart", "ralink,rt2880-uart",
>>> "ns16550a";
>>> +            reg = <0x500 0x100>;
>>> +
>>> +            resets = <&rstctrl 12>;
>>> +            reset-names = "uart";
>>> +
>>> +            interrupt-parent = <&intc>;
>>> +            interrupts = <5>;
>>> +
>>> +            reg-shift = <2>;
>>> +
>>> +            status = "disabled";
>>> +        };
>>
>> moving the node makes no sense. the change is a no-op. please remove
>> it and update the patch description
>
>
> I've just double checked it now, and the order of these nodes does
> change the order of the serial port numbering, therefore it's not no-op.
> Please see the relevant sections from both dmesg outputs:
have you tried using the devicetree aliases node (see [0] for example)
to configure the order?

> #### With original rt5350.dtsi, uart at 500 enabled
> [ 0.565407] gpio-export gpio_export: 3 gpio(s) exported
> [ 0.576247] Serial: 8250/16550 driver, 2 ports, IRQ sharing disabled
> [ 0.590942] console [ttyS0] disabled
> [    0.598130] 10000500.uart: ttyS0 at MMIO 0x10000500 (irq = 13,
> base_baud = 2500000) is a Palmchip BK-3103
> [    0.617160] console [ttyS0] enabled
> [    0.630939] bootconsole [early0] disabled
> [    0.647968] 10000c00.uartlite: ttyS1 at MMIO 0x10000c00 (irq = 20,
> base_baud = 2500000) is a Palmchip BK-3103
> [    0.680207] spi spi0.0: force spi mode3
>
> #### With swapped items in rt5350.dtsi
> [    0.564356] gpio-export gpio_export: 3 gpio(s) exported
> [    0.575201] Serial: 8250/16550 driver, 2 ports, IRQ sharing disabled
> [    0.590006] console [ttyS0] disabled
> [    0.597183] 10000c00.uartlite: ttyS0 at MMIO 0x10000c00 (irq = 20,
> base_baud = 2500000) is a Palmchip BK-3103
> [    0.616906] console [ttyS0] enabled
> [    0.630692] bootconsole [early0] disabled
> [    0.647677] 10000500.uart: ttyS1 at MMIO 0x10000500 (irq = 13,
> base_baud = 2500000) is a Palmchip BK-3103
> [    0.678972] spi spi0.0: force spi mode3
>
> Consequently (given that ttyS0 is configured as console in the kernel
> command line),
> the serial console moves to the pins of uart500 in the second (swapped)
> case.
>  Do you have any suggestion how to solve this on the level of
> RT5350F-OLINUXINO.dtsi, without touching rt5350.dtsi?
>
>
>>
>>     John
>>
>>> +
>>>           systick: systick at d00 {
>>>               compatible = "ralink,rt5350-systick",
>>> "ralink,cevt-systick";
>>>               reg = <0xd00 0x10>;
>>
>
> Thanks, regards,
>
> Zoltan Gyarmati
> https://zgyarmati.de
>
>
>
> _______________________________________________
> Lede-dev mailing list
> Lede-dev at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/lede-dev
>


[0] http://elinux.org/Device_Tree_Usage#aliases_Node



More information about the Lede-dev mailing list