MXS getting AUART to work with DT.

MXS getting AUART to work with DT.

[Subodh Nijsure]

Hi,

I am testing my mx28 hardware with arm-soc/for-next branch. I have made  
additions shown below to my dt file.

What I see is only ttyAPP0 device gets created but not the ttyAPP1, do 
folks who know DT much better than I, see any issues with this DT 
definition below?

auart0_pins_a: auart0 at 0 {
         reg = <0>;
         fsl,pinmux-ids = <0x3000 0x3010 0x3020 0x3030>;
         fsl,drive-strength = <0>;
         fsl,voltage = <1>;
         fsl,pull-up = <0>;
};
auart1_pins_a: auart1 at 0 {
         reg = <0>;
         fsl,pinmux-ids = <0x3040 0x3050>;
         fsl,drive-strength = <0>;
         fsl,voltage = <1>;
         fsl,pull-up = <0>;
};
auart0: serial at 8006a000 {
          pinctrl-names = "default";
          pinctrl-0 = <&auart0_pins_a>;
          status = "okay";
};
auart1: serial at 8006c000 {
           pinctrl-names = "default";
           pinctrl-0 = <&auart1_pins_a>;
           status = "okay";
  };

-Subodh

MXS getting AUART to work with DT.

[Fabio Estevam]

Hi Subodh,

On Tue, Jun 12, 2012 at 10:42 PM, Subodh Nijsure <snijsure@grid-net.com> wrote:
> Hi,
>
> I am testing my mx28 hardware with arm-soc/for-next branch. I have made
> ?additions shown below to my dt file.
>
> What I see is only ttyAPP0 device gets created but not the ttyAPP1, do folks
> who know DT much better than I, see any issues with this DT definition
> below?

I just enabled CONFIG_SERIAL_MXS_AUART and booted my mx28evk (non-DT
case) that registers
auart0 and auart2, but I don't see any ttyAPPx node.

Is there anything I am missing?

I would like to get ttyAPP0/2 working first on non-DT case.

Thanks,

Fabio Estevam

MXS getting AUART to work with DT.

[Marek Vasut]

Dear Fabio Estevam,

> Hi Subodh,
> 
> On Tue, Jun 12, 2012 at 10:42 PM, Subodh Nijsure <snijsure@grid-net.com> 
wrote:
> > Hi,
> > 
> > I am testing my mx28 hardware with arm-soc/for-next branch. I have made
> >  additions shown below to my dt file.
> > 
> > What I see is only ttyAPP0 device gets created but not the ttyAPP1, do
> > folks who know DT much better than I, see any issues with this DT
> > definition below?
> 
> I just enabled CONFIG_SERIAL_MXS_AUART and booted my mx28evk (non-DT
> case) that registers
> auart0 and auart2, but I don't see any ttyAPPx node.
> 
> Is there anything I am missing?
> 
> I would like to get ttyAPP0/2 working first on non-DT case.

I think the kernel for non-DT case is already FUBAR ... just do a simple DT 
conversion :-(

> Thanks,
> 
> Fabio Estevam

Best regards,
your utmost DT supporter and friend
Marek Vasut

MXS getting AUART to work with DT.

[Fabio Estevam]

On Wed, Jun 13, 2012 at 10:32 PM, Marek Vasut <marex@denx.de> wrote:

> I think the kernel for non-DT case is already FUBAR ... just do a simple DT
> conversion :-(

So does it mean mxs-auart has never been used?

MXS getting AUART to work with DT.

[Fabio Estevam]

On Wed, Jun 13, 2012 at 11:33 PM, Fabio Estevam <festevam@gmail.com> wrote:
> On Wed, Jun 13, 2012 at 10:32 PM, Marek Vasut <marex@denx.de> wrote:
>
>> I think the kernel for non-DT case is already FUBAR ... just do a simple DT
>> conversion :-(
>
> So does it mean mxs-auart has never been used?

Ok, just tested a 3.4.2 (non-DT kernel) and I see:

[    0.570000] mxs-auart.0: ttyAPP0 at MMIO 0x8006a000 (irq = 112) is
a mxs-auart.0
[    0.580000] mxs-auart mxs-auart.0: Found APPUART 3.1.0
[    0.590000] mxs-auart.3: ttyAPP3 at MMIO 0x80070000 (irq = 115) is
a mxs-auart.3
[    0.600000] mxs-auart mxs-auart.3: Found APPUART 3.1.0

,so it means that something broke mxs-auart on 3.5-rc.

Will look more on this tomorrow.

MXS getting AUART to work with DT.

[Marek Vasut]

Dear Fabio Estevam,

> On Wed, Jun 13, 2012 at 11:33 PM, Fabio Estevam <festevam@gmail.com> wrote:
> > On Wed, Jun 13, 2012 at 10:32 PM, Marek Vasut <marex@denx.de> wrote:
> >> I think the kernel for non-DT case is already FUBAR ... just do a simple
> >> DT conversion :-(
> > 
> > So does it mean mxs-auart has never been used?
> 
> Ok, just tested a 3.4.2 (non-DT kernel) and I see:
> 
> [    0.570000] mxs-auart.0: ttyAPP0 at MMIO 0x8006a000 (irq = 112) is
> a mxs-auart.0
> [    0.580000] mxs-auart mxs-auart.0: Found APPUART 3.1.0
> [    0.590000] mxs-auart.3: ttyAPP3 at MMIO 0x80070000 (irq = 115) is
> a mxs-auart.3
> [    0.600000] mxs-auart mxs-auart.3: Found APPUART 3.1.0
> 
> ,so it means that something broke mxs-auart on 3.5-rc.

I tested it on 3.4-rc6, it worked back there with non-DT kernel.

> Will look more on this tomorrow.

Best regards,
Marek Vasut

MXS getting AUART to work with DT.

[Marek Vasut]

Dear Subodh Nijsure,

> Hi,
> 
> I am testing my mx28 hardware with arm-soc/for-next branch. I have made
> additions shown below to my dt file.
> 
> What I see is only ttyAPP0 device gets created but not the ttyAPP1, do
> folks who know DT much better than I, see any issues with this DT
> definition below?

Last time I checked, it wasn't that easy to get AUART flipped over to DT. Did 
someone do that already?

> 
> auart0_pins_a: auart0 at 0 {
>          reg = <0>;
>          fsl,pinmux-ids = <0x3000 0x3010 0x3020 0x3030>;
>          fsl,drive-strength = <0>;
>          fsl,voltage = <1>;
>          fsl,pull-up = <0>;
> };
> auart1_pins_a: auart1 at 0 {
>          reg = <0>;
>          fsl,pinmux-ids = <0x3040 0x3050>;
>          fsl,drive-strength = <0>;
>          fsl,voltage = <1>;
>          fsl,pull-up = <0>;
> };
> auart0: serial at 8006a000 {
>           pinctrl-names = "default";
>           pinctrl-0 = <&auart0_pins_a>;
>           status = "okay";
> };
> auart1: serial at 8006c000 {
>            pinctrl-names = "default";
>            pinctrl-0 = <&auart1_pins_a>;
>            status = "okay";
>   };
> 
> -Subodh

Best regards,
Marek Vasut

MXS getting AUART to work with DT.

[Fabio Estevam]

On Tue, Jun 12, 2012 at 10:42 PM, Subodh Nijsure <snijsure@grid-net.com> wrote:
> Hi,
>
> I am testing my mx28 hardware with arm-soc/for-next branch. I have made
> ?additions shown below to my dt file.
>
> What I see is only ttyAPP0 device gets created but not the ttyAPP1, do folks
> who know DT much better than I, see any issues with this DT definition
> below?

Would something like the patch below be needed?

---
 arch/arm/boot/dts/imx28.dtsi   |    5 +++++
 drivers/tty/serial/mxs-auart.c |    9 +++++++++
 2 files changed, 14 insertions(+), 0 deletions(-)

diff --git a/arch/arm/boot/dts/imx28.dtsi b/arch/arm/boot/dts/imx28.dtsi
index 4634cb8..d524377 100644
--- a/arch/arm/boot/dts/imx28.dtsi
+++ b/arch/arm/boot/dts/imx28.dtsi
@@ -404,30 +404,35 @@
 			};

 			auart0: serial at 8006a000 {
+				compatible = "fsl,mxs-auart";
 				reg = <0x8006a000 0x2000>;
 				interrupts = <112 70 71>;
 				status = "disabled";
 			};

 			auart1: serial at 8006c000 {
+				compatible = "fsl,mxs-auart";
 				reg = <0x8006c000 0x2000>;
 				interrupts = <113 72 73>;
 				status = "disabled";
 			};

 			auart2: serial at 8006e000 {
+				compatible = "fsl,mxs-auart";
 				reg = <0x8006e000 0x2000>;
 				interrupts = <114 74 75>;
 				status = "disabled";
 			};

 			auart3: serial at 80070000 {
+				compatible = "fsl,mxs-auart";
 				reg = <0x80070000 0x2000>;
 				interrupts = <115 76 77>;
 				status = "disabled";
 			};

 			auart4: serial at 80072000 {
+				compatible = "fsl,mxs-auart";
 				reg = <0x80072000 0x2000>;
 				interrupts = <116 78 79>;
 				status = "disabled";
diff --git a/drivers/tty/serial/mxs-auart.c b/drivers/tty/serial/mxs-auart.c
index ec56d83..6540f28 100644
--- a/drivers/tty/serial/mxs-auart.c
+++ b/drivers/tty/serial/mxs-auart.c
@@ -33,6 +33,8 @@
 #include <linux/delay.h>
 #include <linux/io.h>
 #include <linux/pinctrl/consumer.h>
+#include <linux/of.h>
+#include <linux/of_address.h>

 #include <asm/cacheflush.h>

@@ -769,12 +771,19 @@ static int __devexit mxs_auart_remove(struct
platform_device *pdev)
 	return 0;
 }

+static struct of_device_id mxs_auart_dt_ids[] = {
+	{.compatible = "fsl,mxs-auart", },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, mxs_auart_of_match);
+
 static struct platform_driver mxs_auart_driver = {
 	.probe = mxs_auart_probe,
 	.remove = __devexit_p(mxs_auart_remove),
 	.driver = {
 		.name = "mxs-auart",
 		.owner = THIS_MODULE,
+		.of_match_table = of_match_ptr(mxs_auart_dt_ids),
 	},
 };

-- 
1.7.1

MXS getting AUART to work with DT.

[Subodh Nijsure]

On 06/13/2012 07:57 PM, Fabio Estevam wrote:
> On Tue, Jun 12, 2012 at 10:42 PM, Subodh Nijsure<snijsure@grid-net.com>  wrote:
>> Hi,
>>
>> I am testing my mx28 hardware with arm-soc/for-next branch. I have made
>>   additions shown below to my dt file.
>>
>> What I see is only ttyAPP0 device gets created but not the ttyAPP1, do folks
>> who know DT much better than I, see any issues with this DT definition
>> below?
> Would something like the patch below be needed?
>
> ---
>   arch/arm/boot/dts/imx28.dtsi   |    5 +++++
>   drivers/tty/serial/mxs-auart.c |    9 +++++++++
>   2 files changed, 14 insertions(+), 0 deletions(-)
>
> diff --git a/arch/arm/boot/dts/imx28.dtsi b/arch/arm/boot/dts/imx28.dtsi
> index 4634cb8..d524377 100644
> --- a/arch/arm/boot/dts/imx28.dtsi
> +++ b/arch/arm/boot/dts/imx28.dtsi
> @@ -404,30 +404,35 @@
>   			};
>
>   			auart0: serial at 8006a000 {
> +				compatible = "fsl,mxs-auart";
>   				reg =<0x8006a000 0x2000>;
>   				interrupts =<112 70 71>;
>   				status = "disabled";
>   			};
>
>   			auart1: serial at 8006c000 {
> +				compatible = "fsl,mxs-auart";
>   				reg =<0x8006c000 0x2000>;
>   				interrupts =<113 72 73>;
>   				status = "disabled";
>   			};
>
>   			auart2: serial at 8006e000 {
> +				compatible = "fsl,mxs-auart";
>   				reg =<0x8006e000 0x2000>;
>   				interrupts =<114 74 75>;
>   				status = "disabled";
>   			};
>
>   			auart3: serial at 80070000 {
> +				compatible = "fsl,mxs-auart";
>   				reg =<0x80070000 0x2000>;
>   				interrupts =<115 76 77>;
>   				status = "disabled";
>   			};
>
>   			auart4: serial at 80072000 {
> +				compatible = "fsl,mxs-auart";
>   				reg =<0x80072000 0x2000>;
>   				interrupts =<116 78 79>;
>   				status = "disabled";
> diff --git a/drivers/tty/serial/mxs-auart.c b/drivers/tty/serial/mxs-auart.c
> index ec56d83..6540f28 100644
> --- a/drivers/tty/serial/mxs-auart.c
> +++ b/drivers/tty/serial/mxs-auart.c
> @@ -33,6 +33,8 @@
>   #include<linux/delay.h>
>   #include<linux/io.h>
>   #include<linux/pinctrl/consumer.h>
> +#include<linux/of.h>
> +#include<linux/of_address.h>
>
>   #include<asm/cacheflush.h>
>
> @@ -769,12 +771,19 @@ static int __devexit mxs_auart_remove(struct
> platform_device *pdev)
>   	return 0;
>   }
>
> +static struct of_device_id mxs_auart_dt_ids[] = {
> +	{.compatible = "fsl,mxs-auart", },
> +	{ /* sentinel */ }
> +};
> +MODULE_DEVICE_TABLE(of, mxs_auart_of_match);
> +
>   static struct platform_driver mxs_auart_driver = {
>   	.probe = mxs_auart_probe,
>   	.remove = __devexit_p(mxs_auart_remove),
>   	.driver = {
>   		.name = "mxs-auart",
>   		.owner = THIS_MODULE,
> +		.of_match_table = of_match_ptr(mxs_auart_dt_ids),
>   	},
>   };
>
With this change you will only see AUART0 being created all subsequent 
attempts fail.  I think we need to have dt specific probe, my colleague 
Matt Burtch has implemented that dt specific probe for AUART will 
request him to send his diff tomorrow AM.

I think something is broken in 3.5 as far as I can tell AUART goes it 
had worked for me in 3.4, I am novice at device-tree and was not sure if 
its because of DT or something else.

-Subodh

MXS getting AUART to work with DT.

[Fabio Estevam]

On Thu, Jun 14, 2012 at 1:40 AM, Subodh Nijsure <snijsure@grid-net.com> wrote:

> With this change you will only see AUART0 being created all subsequent
> attempts fail. ?I think we need to have dt specific probe, my colleague Matt
> Burtch has implemented that dt specific probe for AUART will request him to
> send his diff tomorrow AM.
>
> I think something is broken in 3.5 as far as I can tell AUART goes it had
> worked for me in 3.4, I am novice at device-tree and was not sure if its
> because of DT or something else.

Ok, just tested linux-next and I see that mxs-auart probes correctly
for the non-dt kernel.

We need to fix mxs-auart for dt now.

Regards,

Fabio Estevam

MXS getting AUART to work with DT.

[Matt Burtch]

Here is the patch I put together to get the AUART recognized with the dt.  I'm pretty new to dt development, so this was a quick first pass. I'm sure there are more proper ways to get the device id and pass literals between the dt and probe function...but this works for now.

Regards,
-Matt Burtch
________________________________________
From: Fabio Estevam [festevam at gmail.com]
Sent: Wednesday, June 13, 2012 9:54 PM
To: Subodh Nijsure
Cc: linux-arm-kernel at lists.infradead.org; Marek Va?ut; Shawn Guo; Matt Burtch
Subject: Re: MXS getting AUART to work with DT.

On Thu, Jun 14, 2012 at 1:40 AM, Subodh Nijsure <snijsure@grid-net.com> wrote:

> With this change you will only see AUART0 being created all subsequent
> attempts fail.  I think we need to have dt specific probe, my colleague Matt
> Burtch has implemented that dt specific probe for AUART will request him to
> send his diff tomorrow AM.
>
> I think something is broken in 3.5 as far as I can tell AUART goes it had
> worked for me in 3.4, I am novice at device-tree and was not sure if its
> because of DT or something else.

Ok, just tested linux-next and I see that mxs-auart probes correctly
for the non-dt kernel.

We need to fix mxs-auart for dt now.

Regards,

Fabio Estevam
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mxs-auart.patch
Type: text/x-patch
Size: 3155 bytes
Desc: mxs-auart.patch
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20120614/9fae73f8/attachment-0001.bin>

MXS getting AUART to work with DT.

[Fabio Estevam]

Hi Matt,

On Thu, Jun 14, 2012 at 1:49 PM, Matt Burtch <matt@grid-net.com> wrote:
> Here is the patch I put together to get the AUART recognized with the dt. ?I'm pretty new to dt development, so this was a quick first pass. I'm sure there are more proper ways to get the device id and pass literals between the dt and probe function...but this works for now.

Ok, I managed to fix it properly now.

Will submit the patches tomorrow.

Regards,

Fabio Estevam