diff mbox

[V2,3/3] ARM: davinci: da850: add NAND driver entries

Message ID 50ED66F2.8050006@ti.com (mailing list archive)
State New, archived
Headers show

Commit Message

Sekhar Nori Jan. 9, 2013, 12:47 p.m. UTC
On 1/8/2013 1:50 PM, Kumar, Anil wrote:
> Add NAND driver entries to export NAND functionality on da850 EVM and
> NAND pinctrl node to do pin mux according to pinctrl-single driver.

Subject line should have DT and da850 evm somewhere.
Otherwise NAND has been supported on DA850 for long.

> 
> Signed-off-by: Kumar, Anil <anilkumar.v@ti.com>
> ---
> :100644 100644 c7609d0... 382a7da... M	arch/arm/boot/dts/da850-evm.dts
> :100644 100644 e9c6e82... 16e2ac2... M	arch/arm/boot/dts/da850.dtsi
>  arch/arm/boot/dts/da850-evm.dts |   19 +++++++++++++++++++
>  arch/arm/boot/dts/da850.dtsi    |   13 +++++++++++++
>  2 files changed, 32 insertions(+), 0 deletions(-)
> 
> diff --git a/arch/arm/boot/dts/da850-evm.dts b/arch/arm/boot/dts/da850-evm.dts
> index c7609d0..382a7da 100644
> --- a/arch/arm/boot/dts/da850-evm.dts
> +++ b/arch/arm/boot/dts/da850-evm.dts
> @@ -28,4 +28,23 @@
>  			status = "okay";
>  		};
>  	};
> +	nand_cs3@62000000 {
> +		status = "okay";
> +	};
> +};
> +&pmx_core{
> +	pinctrl-names = "default";
> +	pinctrl-0 = <
> +		&nand_cs3_pins
> +	>;

This means that the NAND pins are configured even if NAND is not
probed. Right? This can be moved into the nand_cs3 node to avoid that. 
And then when used with Linus Walleij's patch "drivers/pinctrl: grab 
default handles from device core" which should be accepted soon, the 
pins will be automatically setup when the NAND gets probed.

> +
> +	nand_cs3_pins: pinmux_nand_pins {
> +		pinctrl-single,bits = <
> +			0x1c 0x00110000  0x00ff0000     /* EMA_OE, EMA_WE */
> +			0x1c 0x00000110  0x00000ff0     /* EMA_CS[4],EMA_CS[3]*/
> +			0x24 0x11111111  0xffffffff     /* EMA_D[0], EMA_D[1], EMA_D[2], EMA_D[3],
> +							   EMA_D[4], EMA_D[5], EMA_D[6], EMA_D[7] */
> +			0x30 0x01100000  0x0ff00000     /* EMA_A[1], EMA_A[2] */
> +		>;
> +	};

I suspect a number of other boards using NAND on CS3 would have to 
repeat these configurations. It will be better to move them into the 
da850.dtsi file so it can be reused.

Since I am new to the DT and pinctrl stuff, I had to actually try these to 
make sure what I am suggesting isn't nonsense. I came up with this 
patch on top of your series. Please test it at your end and also see if you 
are happy with the changes.

Thanks,
Sekhar

---8<----

Comments

Kumar, Anil Jan. 10, 2013, 7:37 a.m. UTC | #1
On Wed, Jan 09, 2013 at 18:17:46, Nori, Sekhar wrote:
> On 1/8/2013 1:50 PM, Kumar, Anil wrote:
> > Add NAND driver entries to export NAND functionality on da850 EVM and
> > NAND pinctrl node to do pin mux according to pinctrl-single driver.
> 
> Subject line should have DT and da850 evm somewhere.
> Otherwise NAND has been supported on DA850 for long.

I will update it in V3 of this patch series.

> 
> > 
> > Signed-off-by: Kumar, Anil <anilkumar.v@ti.com>
> > ---
> > :100644 100644 c7609d0... 382a7da... M	arch/arm/boot/dts/da850-evm.dts
> > :100644 100644 e9c6e82... 16e2ac2... M	arch/arm/boot/dts/da850.dtsi
> >  arch/arm/boot/dts/da850-evm.dts |   19 +++++++++++++++++++
> >  arch/arm/boot/dts/da850.dtsi    |   13 +++++++++++++
> >  2 files changed, 32 insertions(+), 0 deletions(-)
> > 
> > diff --git a/arch/arm/boot/dts/da850-evm.dts b/arch/arm/boot/dts/da850-evm.dts
> > index c7609d0..382a7da 100644
> > --- a/arch/arm/boot/dts/da850-evm.dts
> > +++ b/arch/arm/boot/dts/da850-evm.dts
> > @@ -28,4 +28,23 @@
> >  			status = "okay";
> >  		};
> >  	};
> > +	nand_cs3@62000000 {
> > +		status = "okay";
> > +	};
> > +};
> > +&pmx_core{
> > +	pinctrl-names = "default";
> > +	pinctrl-0 = <
> > +		&nand_cs3_pins
> > +	>;
> 
> This means that the NAND pins are configured even if NAND is not
> probed. Right?

Right 

> This can be moved into the nand_cs3 node to avoid that. 
> And then when used with Linus Walleij's patch "drivers/pinctrl: grab 
> default handles from device core" which should be accepted soon, the 
> pins will be automatically setup when the NAND gets probed.

It is better. I will do it in V3 of this patch series.

> 
> > +
> > +	nand_cs3_pins: pinmux_nand_pins {
> > +		pinctrl-single,bits = <
> > +			0x1c 0x00110000  0x00ff0000     /* EMA_OE, EMA_WE */
> > +			0x1c 0x00000110  0x00000ff0     /* EMA_CS[4],EMA_CS[3]*/
> > +			0x24 0x11111111  0xffffffff     /* EMA_D[0], EMA_D[1], EMA_D[2], EMA_D[3],
> > +							   EMA_D[4], EMA_D[5], EMA_D[6], EMA_D[7] */
> > +			0x30 0x01100000  0x0ff00000     /* EMA_A[1], EMA_A[2] */
> > +		>;
> > +	};
> 
> I suspect a number of other boards using NAND on CS3 would have to 
> repeat these configurations. It will be better to move them into the 
> da850.dtsi file so it can be reused.

> 
> Since I am new to the DT and pinctrl stuff, I had to actually try these to 
> make sure what I am suggesting isn't nonsense. I came up with this 
> patch on top of your series. Please test it at your end

Tested. It is working. 

> and also see if you 
> are happy with the changes.

I do not think that it is good idea to move NAND pin mux information
into da850.dtsi because this information is evm specific.
if we will use this approach then we must use the same approach for
other modules also as ASoC etc.   

Thanks,
Anil

> ---8<----
> diff --git a/arch/arm/boot/dts/da850-evm.dts b/arch/arm/boot/dts/da850-evm.dts
> index 382a7da..52abb9d 100644
> --- a/arch/arm/boot/dts/da850-evm.dts
> +++ b/arch/arm/boot/dts/da850-evm.dts
> @@ -30,21 +30,8 @@
>  	};
>  	nand_cs3@62000000 {
>  		status = "okay";
> +		pinctrl-names = "default";
> +		pinctrl-0 = <&nand_cs3_pins>;
>  	};
>  };
> -&pmx_core{
> -	pinctrl-names = "default";
> -	pinctrl-0 = <
> -		&nand_cs3_pins
> -	>;
>  
> -	nand_cs3_pins: pinmux_nand_pins {
> -		pinctrl-single,bits = <
> -			0x1c 0x00110000  0x00ff0000     /* EMA_OE, EMA_WE */
> -			0x1c 0x00000110  0x00000ff0     /* EMA_CS[4],EMA_CS[3]*/
> -			0x24 0x11111111  0xffffffff     /* EMA_D[0], EMA_D[1], EMA_D[2], EMA_D[3],
> -							   EMA_D[4], EMA_D[5], EMA_D[6], EMA_D[7] */
> -			0x30 0x01100000  0x0ff00000     /* EMA_A[1], EMA_A[2] */
> -		>;
> -	};
> -};
> diff --git a/arch/arm/boot/dts/da850.dtsi b/arch/arm/boot/dts/da850.dtsi
> index 6ab9730..c4b21a8 100644
> --- a/arch/arm/boot/dts/da850.dtsi
> +++ b/arch/arm/boot/dts/da850.dtsi
> @@ -38,7 +38,25 @@
>  			pinctrl-single,register-width = <32>;
>  			pinctrl-single,function-mask = <0xffffffff>;
>  			status = "disabled";
> +
> +			nand_cs3_pins: pinmux_nand_pins {
> +				pinctrl-single,bits = <
> +					/* EMA_OE, EMA_WE */
> +					0x1c 0x00110000  0x00ff0000
> +					/* EMA_CS[4],EMA_CS[3]*/
> +					0x1c 0x00000110  0x00000ff0
> +					/*
> +					 * EMA_D[0], EMA_D[1], EMA_D[2],
> +					 * EMA_D[3], EMA_D[4], EMA_D[5],
> +					 * EMA_D[6], EMA_D[7]
> +					 */
> +					0x24 0x11111111  0xffffffff
> +					/* EMA_A[1], EMA_A[2] */
> +					0x30 0x01100000  0x0ff00000
> +				>;
> +			};
>  		};
> +
>  		serial0: serial@1c42000 {
>  			compatible = "ns16550a";
>  			reg = <0x42000 0x100>;
>
Sekhar Nori Jan. 10, 2013, 12:19 p.m. UTC | #2
On 1/10/2013 1:07 PM, Kumar, Anil wrote:
> On Wed, Jan 09, 2013 at 18:17:46, Nori, Sekhar wrote:

> I do not think that it is good idea to move NAND pin mux information
> into da850.dtsi because this information is evm specific.
> if we will use this approach then we must use the same approach for
> other modules also as ASoC etc.   

Why do you consider this EVM specific. IOW, which pins do you see
changing on another board? BTW, if there are additional pins needed than
what are listed, we can always add more pinux entries in the .dts files.
The pins present in dtsi file should the base case.

Thanks,
Sekhar
Kumar, Anil Jan. 15, 2013, 10:36 a.m. UTC | #3
On Thu, Jan 10, 2013 at 17:49:13, Nori, Sekhar wrote:
> On 1/10/2013 1:07 PM, Kumar, Anil wrote:
> > On Wed, Jan 09, 2013 at 18:17:46, Nori, Sekhar wrote:
> 
> > I do not think that it is good idea to move NAND pin mux information
> > into da850.dtsi because this information is evm specific.
> > if we will use this approach then we must use the same approach for
> > other modules also as ASoC etc.   
> 
> Why do you consider this EVM specific. IOW, which pins do you see
> changing on another board? BTW, if there are additional pins needed than
> what are listed, we can always add more pinux entries in the .dts files.
> The pins present in dtsi file should the base case.
 
Ok, we can use this approach for DaVinci as its SoC modules do not
have multiple pin configuration option. I will do the changes in 
next patch series.

Thanks,
Anil
Sekhar Nori Jan. 15, 2013, 11:11 a.m. UTC | #4
On 1/15/2013 4:06 PM, Kumar, Anil wrote:
> On Thu, Jan 10, 2013 at 17:49:13, Nori, Sekhar wrote:
>> On 1/10/2013 1:07 PM, Kumar, Anil wrote:
>>> On Wed, Jan 09, 2013 at 18:17:46, Nori, Sekhar wrote:
>>
>>> I do not think that it is good idea to move NAND pin mux information
>>> into da850.dtsi because this information is evm specific.
>>> if we will use this approach then we must use the same approach for
>>> other modules also as ASoC etc.   
>>
>> Why do you consider this EVM specific. IOW, which pins do you see
>> changing on another board? BTW, if there are additional pins needed than
>> what are listed, we can always add more pinux entries in the .dts files.
>> The pins present in dtsi file should the base case.
>  
> Ok, we can use this approach for DaVinci as its SoC modules do not
> have multiple pin configuration option. I will do the changes in 
> next patch series.

You could do this even if this was the case by defining multiple pin groups.

Thanks,
Sekhar
Linus Walleij Jan. 17, 2013, 10:30 a.m. UTC | #5
On Wed, Jan 9, 2013 at 1:47 PM, Sekhar Nori <nsekhar@ti.com> wrote:
> On 1/8/2013 1:50 PM, Kumar, Anil wrote:

>> +&pmx_core{
>> +     pinctrl-names = "default";
>> +     pinctrl-0 = <
>> +             &nand_cs3_pins
>> +     >;
>
> This means that the NAND pins are configured even if NAND is not
> probed. Right? This can be moved into the nand_cs3 node to avoid that.
> And then when used with Linus Walleij's patch "drivers/pinctrl: grab
> default handles from device core" which should be accepted soon, the
> pins will be automatically setup when the NAND gets probed.

That is correct, I am waiting for Greg's ACK on the core grab patch
(maybe it's higher up in my mailbox) and some indication from Stephen
Warren that it doesn't break the Tegra, and I'll put it into linux-next.

Yours,
Linus Walleij
diff mbox

Patch

diff --git a/arch/arm/boot/dts/da850-evm.dts b/arch/arm/boot/dts/da850-evm.dts
index 382a7da..52abb9d 100644
--- a/arch/arm/boot/dts/da850-evm.dts
+++ b/arch/arm/boot/dts/da850-evm.dts
@@ -30,21 +30,8 @@ 
 	};
 	nand_cs3@62000000 {
 		status = "okay";
+		pinctrl-names = "default";
+		pinctrl-0 = <&nand_cs3_pins>;
 	};
 };
-&pmx_core{
-	pinctrl-names = "default";
-	pinctrl-0 = <
-		&nand_cs3_pins
-	>;
 
-	nand_cs3_pins: pinmux_nand_pins {
-		pinctrl-single,bits = <
-			0x1c 0x00110000  0x00ff0000     /* EMA_OE, EMA_WE */
-			0x1c 0x00000110  0x00000ff0     /* EMA_CS[4],EMA_CS[3]*/
-			0x24 0x11111111  0xffffffff     /* EMA_D[0], EMA_D[1], EMA_D[2], EMA_D[3],
-							   EMA_D[4], EMA_D[5], EMA_D[6], EMA_D[7] */
-			0x30 0x01100000  0x0ff00000     /* EMA_A[1], EMA_A[2] */
-		>;
-	};
-};
diff --git a/arch/arm/boot/dts/da850.dtsi b/arch/arm/boot/dts/da850.dtsi
index 6ab9730..c4b21a8 100644
--- a/arch/arm/boot/dts/da850.dtsi
+++ b/arch/arm/boot/dts/da850.dtsi
@@ -38,7 +38,25 @@ 
 			pinctrl-single,register-width = <32>;
 			pinctrl-single,function-mask = <0xffffffff>;
 			status = "disabled";
+
+			nand_cs3_pins: pinmux_nand_pins {
+				pinctrl-single,bits = <
+					/* EMA_OE, EMA_WE */
+					0x1c 0x00110000  0x00ff0000
+					/* EMA_CS[4],EMA_CS[3]*/
+					0x1c 0x00000110  0x00000ff0
+					/*
+					 * EMA_D[0], EMA_D[1], EMA_D[2],
+					 * EMA_D[3], EMA_D[4], EMA_D[5],
+					 * EMA_D[6], EMA_D[7]
+					 */
+					0x24 0x11111111  0xffffffff
+					/* EMA_A[1], EMA_A[2] */
+					0x30 0x01100000  0x0ff00000
+				>;
+			};
 		};
+
 		serial0: serial@1c42000 {
 			compatible = "ns16550a";
 			reg = <0x42000 0x100>;