[2/2] ARM: dts: qcom: ipq4019-ap.dk01.1: align flash node with bindings

Message ID 20230811210142.403160-2-robimarko@gmail.com
State New
Headers
Series [1/2] ARM: dts: qcom: ipq4019-ap.dk01.1: use existing labels for nodes |

Commit Message

Robert Marko Aug. 11, 2023, 9:01 p.m. UTC
  Rename the SPI-NOR node to flash@0, remove #address-cells and #size-cells
as they should be under the partitions subnode and use the generic
jedec,spi-nor compatible.

Signed-off-by: Robert Marko <robimarko@gmail.com>
---
 arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
  

Comments

Konrad Dybcio Aug. 11, 2023, 9:28 p.m. UTC | #1
On 11.08.2023 23:01, Robert Marko wrote:
> Rename the SPI-NOR node to flash@0, remove #address-cells and #size-cells
> as they should be under the partitions subnode and use the generic
> jedec,spi-nor compatible.
> 
> Signed-off-by: Robert Marko <robimarko@gmail.com>
> ---
You can also do "nandmanufacturer,mx25l25635e", "jedec,spi-nor"

Konrad

>  arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi | 6 ++----
>  1 file changed, 2 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi b/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi
> index 0714616c9e45..f7ac8f9d0b6f 100644
> --- a/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi
> +++ b/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi
> @@ -75,11 +75,9 @@ &blsp1_spi1 {
>  	status = "okay";
>  	cs-gpios = <&tlmm 54 GPIO_ACTIVE_HIGH>;
>  
> -	mx25l25635e@0 {
> -		#address-cells = <1>;
> -		#size-cells = <1>;
> +	flash@0 {
>  		reg = <0>;
> -		compatible = "mx25l25635e";
> +		compatible = "jedec,spi-nor";
>  		spi-max-frequency = <24000000>;
>  	};
>  };
  
Robert Marko Aug. 11, 2023, 9:35 p.m. UTC | #2
On Fri, 11 Aug 2023 at 23:28, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>
> On 11.08.2023 23:01, Robert Marko wrote:
> > Rename the SPI-NOR node to flash@0, remove #address-cells and #size-cells
> > as they should be under the partitions subnode and use the generic
> > jedec,spi-nor compatible.
> >
> > Signed-off-by: Robert Marko <robimarko@gmail.com>
> > ---
> You can also do "nandmanufacturer,mx25l25635e", "jedec,spi-nor"

Hi,
I grepped the vendor U-Boot to make sure it's not being triggered off
the mx25l25635e
compatible but the only hit is the IC support itself.
MX25L25635 was just the original NOR IC Qualcomm used on the board so
to me it made
most sense to just use the JEDEC compatible as NOR itself is JEDEC NOR
compatible.

Regards,
Robert
>
> Konrad
>
> >  arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi | 6 ++----
> >  1 file changed, 2 insertions(+), 4 deletions(-)
> >
> > diff --git a/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi b/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi
> > index 0714616c9e45..f7ac8f9d0b6f 100644
> > --- a/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi
> > +++ b/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi
> > @@ -75,11 +75,9 @@ &blsp1_spi1 {
> >       status = "okay";
> >       cs-gpios = <&tlmm 54 GPIO_ACTIVE_HIGH>;
> >
> > -     mx25l25635e@0 {
> > -             #address-cells = <1>;
> > -             #size-cells = <1>;
> > +     flash@0 {
> >               reg = <0>;
> > -             compatible = "mx25l25635e";
> > +             compatible = "jedec,spi-nor";
> >               spi-max-frequency = <24000000>;
> >       };
> >  };
  
Konrad Dybcio Aug. 11, 2023, 10:56 p.m. UTC | #3
On 11.08.2023 23:35, Robert Marko wrote:
> On Fri, 11 Aug 2023 at 23:28, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>>
>> On 11.08.2023 23:01, Robert Marko wrote:
>>> Rename the SPI-NOR node to flash@0, remove #address-cells and #size-cells
>>> as they should be under the partitions subnode and use the generic
>>> jedec,spi-nor compatible.
>>>
>>> Signed-off-by: Robert Marko <robimarko@gmail.com>
>>> ---
>> You can also do "nandmanufacturer,mx25l25635e", "jedec,spi-nor"
> 
> Hi,
> I grepped the vendor U-Boot to make sure it's not being triggered off
> the mx25l25635e
> compatible but the only hit is the IC support itself.
> MX25L25635 was just the original NOR IC Qualcomm used on the board so
> to me it made
> most sense to just use the JEDEC compatible as NOR itself is JEDEC NOR
> compatible.
OK if dynamic identification works fine

Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>

Konrad
  
Robert Marko Aug. 12, 2023, 9:55 a.m. UTC | #4
On Sat, 12 Aug 2023 at 00:56, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>
> On 11.08.2023 23:35, Robert Marko wrote:
> > On Fri, 11 Aug 2023 at 23:28, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
> >>
> >> On 11.08.2023 23:01, Robert Marko wrote:
> >>> Rename the SPI-NOR node to flash@0, remove #address-cells and #size-cells
> >>> as they should be under the partitions subnode and use the generic
> >>> jedec,spi-nor compatible.
> >>>
> >>> Signed-off-by: Robert Marko <robimarko@gmail.com>
> >>> ---
> >> You can also do "nandmanufacturer,mx25l25635e", "jedec,spi-nor"
> >
> > Hi,
> > I grepped the vendor U-Boot to make sure it's not being triggered off
> > the mx25l25635e
> > compatible but the only hit is the IC support itself.
> > MX25L25635 was just the original NOR IC Qualcomm used on the board so
> > to me it made
> > most sense to just use the JEDEC compatible as NOR itself is JEDEC NOR
> > compatible.
> OK if dynamic identification works fine

It should work fine, datasheet is clear that its JEDEC compatible.
That being said, I dont actually have the board, just figured it was
time for a cleanup as
OpenWrt has been patching DK01 and DK04 for ages.

Regards,
Robert
>
> Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
>
> Konrad
  
Konrad Dybcio Aug. 12, 2023, 10:47 a.m. UTC | #5
On 12.08.2023 11:55, Robert Marko wrote:
> On Sat, 12 Aug 2023 at 00:56, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>>
>> On 11.08.2023 23:35, Robert Marko wrote:
>>> On Fri, 11 Aug 2023 at 23:28, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>>>>
>>>> On 11.08.2023 23:01, Robert Marko wrote:
>>>>> Rename the SPI-NOR node to flash@0, remove #address-cells and #size-cells
>>>>> as they should be under the partitions subnode and use the generic
>>>>> jedec,spi-nor compatible.
>>>>>
>>>>> Signed-off-by: Robert Marko <robimarko@gmail.com>
>>>>> ---
>>>> You can also do "nandmanufacturer,mx25l25635e", "jedec,spi-nor"
>>>
>>> Hi,
>>> I grepped the vendor U-Boot to make sure it's not being triggered off
>>> the mx25l25635e
>>> compatible but the only hit is the IC support itself.
>>> MX25L25635 was just the original NOR IC Qualcomm used on the board so
>>> to me it made
>>> most sense to just use the JEDEC compatible as NOR itself is JEDEC NOR
>>> compatible.
>> OK if dynamic identification works fine
> 
> It should work fine, datasheet is clear that its JEDEC compatible.
> That being said, I dont actually have the board, just figured it was
> time for a cleanup as
> OpenWrt has been patching DK01 and DK04 for ages.
Hm. Do we know whether there are still users of this boards?

Konrad
  
Robert Marko Aug. 12, 2023, 2:07 p.m. UTC | #6
On Sat, 12 Aug 2023 at 12:47, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>
> On 12.08.2023 11:55, Robert Marko wrote:
> > On Sat, 12 Aug 2023 at 00:56, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
> >>
> >> On 11.08.2023 23:35, Robert Marko wrote:
> >>> On Fri, 11 Aug 2023 at 23:28, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
> >>>>
> >>>> On 11.08.2023 23:01, Robert Marko wrote:
> >>>>> Rename the SPI-NOR node to flash@0, remove #address-cells and #size-cells
> >>>>> as they should be under the partitions subnode and use the generic
> >>>>> jedec,spi-nor compatible.
> >>>>>
> >>>>> Signed-off-by: Robert Marko <robimarko@gmail.com>
> >>>>> ---
> >>>> You can also do "nandmanufacturer,mx25l25635e", "jedec,spi-nor"
> >>>
> >>> Hi,
> >>> I grepped the vendor U-Boot to make sure it's not being triggered off
> >>> the mx25l25635e
> >>> compatible but the only hit is the IC support itself.
> >>> MX25L25635 was just the original NOR IC Qualcomm used on the board so
> >>> to me it made
> >>> most sense to just use the JEDEC compatible as NOR itself is JEDEC NOR
> >>> compatible.
> >> OK if dynamic identification works fine
> >
> > It should work fine, datasheet is clear that its JEDEC compatible.
> > That being said, I dont actually have the board, just figured it was
> > time for a cleanup as
> > OpenWrt has been patching DK01 and DK04 for ages.
> Hm. Do we know whether there are still users of this boards?

I honestly doubt it as they have been broken in OpenWrt for years and
nobody complained.
So we are currently removing support for them, but I still wanted to
at least fixup the DTS state
upstream.
These boards are not obtainable anymore.

Regards,
Robert
>
> Konrad
  
Konrad Dybcio Aug. 12, 2023, 2:08 p.m. UTC | #7
On 12.08.2023 16:07, Robert Marko wrote:
> On Sat, 12 Aug 2023 at 12:47, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>>
>> On 12.08.2023 11:55, Robert Marko wrote:
>>> On Sat, 12 Aug 2023 at 00:56, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>>>>
>>>> On 11.08.2023 23:35, Robert Marko wrote:
>>>>> On Fri, 11 Aug 2023 at 23:28, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>>>>>>
>>>>>> On 11.08.2023 23:01, Robert Marko wrote:
>>>>>>> Rename the SPI-NOR node to flash@0, remove #address-cells and #size-cells
>>>>>>> as they should be under the partitions subnode and use the generic
>>>>>>> jedec,spi-nor compatible.
>>>>>>>
>>>>>>> Signed-off-by: Robert Marko <robimarko@gmail.com>
>>>>>>> ---
>>>>>> You can also do "nandmanufacturer,mx25l25635e", "jedec,spi-nor"
>>>>>
>>>>> Hi,
>>>>> I grepped the vendor U-Boot to make sure it's not being triggered off
>>>>> the mx25l25635e
>>>>> compatible but the only hit is the IC support itself.
>>>>> MX25L25635 was just the original NOR IC Qualcomm used on the board so
>>>>> to me it made
>>>>> most sense to just use the JEDEC compatible as NOR itself is JEDEC NOR
>>>>> compatible.
>>>> OK if dynamic identification works fine
>>>
>>> It should work fine, datasheet is clear that its JEDEC compatible.
>>> That being said, I dont actually have the board, just figured it was
>>> time for a cleanup as
>>> OpenWrt has been patching DK01 and DK04 for ages.
>> Hm. Do we know whether there are still users of this boards?
> 
> I honestly doubt it as they have been broken in OpenWrt for years and
> nobody complained.
> So we are currently removing support for them, but I still wanted to
> at least fixup the DTS state
> upstream.
> These boards are not obtainable anymore.
I also noticed they were detached from the other snapdragons in u-boot
for no good reason (at first glance anyway).

If we don't get any hands up from users, I think deprecation is in order.

Konrad
  
Robert Marko Aug. 12, 2023, 2:12 p.m. UTC | #8
On Sat, 12 Aug 2023 at 16:08, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>
> On 12.08.2023 16:07, Robert Marko wrote:
> > On Sat, 12 Aug 2023 at 12:47, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
> >>
> >> On 12.08.2023 11:55, Robert Marko wrote:
> >>> On Sat, 12 Aug 2023 at 00:56, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
> >>>>
> >>>> On 11.08.2023 23:35, Robert Marko wrote:
> >>>>> On Fri, 11 Aug 2023 at 23:28, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
> >>>>>>
> >>>>>> On 11.08.2023 23:01, Robert Marko wrote:
> >>>>>>> Rename the SPI-NOR node to flash@0, remove #address-cells and #size-cells
> >>>>>>> as they should be under the partitions subnode and use the generic
> >>>>>>> jedec,spi-nor compatible.
> >>>>>>>
> >>>>>>> Signed-off-by: Robert Marko <robimarko@gmail.com>
> >>>>>>> ---
> >>>>>> You can also do "nandmanufacturer,mx25l25635e", "jedec,spi-nor"
> >>>>>
> >>>>> Hi,
> >>>>> I grepped the vendor U-Boot to make sure it's not being triggered off
> >>>>> the mx25l25635e
> >>>>> compatible but the only hit is the IC support itself.
> >>>>> MX25L25635 was just the original NOR IC Qualcomm used on the board so
> >>>>> to me it made
> >>>>> most sense to just use the JEDEC compatible as NOR itself is JEDEC NOR
> >>>>> compatible.
> >>>> OK if dynamic identification works fine
> >>>
> >>> It should work fine, datasheet is clear that its JEDEC compatible.
> >>> That being said, I dont actually have the board, just figured it was
> >>> time for a cleanup as
> >>> OpenWrt has been patching DK01 and DK04 for ages.
> >> Hm. Do we know whether there are still users of this boards?
> >
> > I honestly doubt it as they have been broken in OpenWrt for years and
> > nobody complained.
> > So we are currently removing support for them, but I still wanted to
> > at least fixup the DTS state
> > upstream.
> > These boards are not obtainable anymore.
> I also noticed they were detached from the other snapdragons in u-boot
> for no good reason (at first glance anyway).

If you are talking about the mainline U-Boot then yeah, my basic port was done
years ago and I knew way less about the SoC then now.
Currently its on my TODO to merge them with Snapdragon and add some proper
GPIO and pinctrl drivers as well as using the Linux DTS.

>
> If we don't get any hands up from users, I think deprecation is in order.
>
> Konrad
  
Konrad Dybcio Aug. 12, 2023, 2:15 p.m. UTC | #9
On 12.08.2023 16:12, Robert Marko wrote:
> On Sat, 12 Aug 2023 at 16:08, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>>
>> On 12.08.2023 16:07, Robert Marko wrote:
>>> On Sat, 12 Aug 2023 at 12:47, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>>>>
>>>> On 12.08.2023 11:55, Robert Marko wrote:
>>>>> On Sat, 12 Aug 2023 at 00:56, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>>>>>>
>>>>>> On 11.08.2023 23:35, Robert Marko wrote:
>>>>>>> On Fri, 11 Aug 2023 at 23:28, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>>>>>>>>
>>>>>>>> On 11.08.2023 23:01, Robert Marko wrote:
>>>>>>>>> Rename the SPI-NOR node to flash@0, remove #address-cells and #size-cells
>>>>>>>>> as they should be under the partitions subnode and use the generic
>>>>>>>>> jedec,spi-nor compatible.
>>>>>>>>>
>>>>>>>>> Signed-off-by: Robert Marko <robimarko@gmail.com>
>>>>>>>>> ---
>>>>>>>> You can also do "nandmanufacturer,mx25l25635e", "jedec,spi-nor"
>>>>>>>
>>>>>>> Hi,
>>>>>>> I grepped the vendor U-Boot to make sure it's not being triggered off
>>>>>>> the mx25l25635e
>>>>>>> compatible but the only hit is the IC support itself.
>>>>>>> MX25L25635 was just the original NOR IC Qualcomm used on the board so
>>>>>>> to me it made
>>>>>>> most sense to just use the JEDEC compatible as NOR itself is JEDEC NOR
>>>>>>> compatible.
>>>>>> OK if dynamic identification works fine
>>>>>
>>>>> It should work fine, datasheet is clear that its JEDEC compatible.
>>>>> That being said, I dont actually have the board, just figured it was
>>>>> time for a cleanup as
>>>>> OpenWrt has been patching DK01 and DK04 for ages.
>>>> Hm. Do we know whether there are still users of this boards?
>>>
>>> I honestly doubt it as they have been broken in OpenWrt for years and
>>> nobody complained.
>>> So we are currently removing support for them, but I still wanted to
>>> at least fixup the DTS state
>>> upstream.
>>> These boards are not obtainable anymore.
>> I also noticed they were detached from the other snapdragons in u-boot
>> for no good reason (at first glance anyway).
> 
> If you are talking about the mainline U-Boot then yeah, my basic port was done
> years ago and I knew way less about the SoC then now.
> Currently its on my TODO to merge them with Snapdragon and add some proper
> GPIO and pinctrl drivers as well as using the Linux DTS.
Take a look at this branch of mine [1], I already did some of that.

If you wish to upstream that, please coordinate with Caleb (+CC) who
may be interested in the same in parallel.

[1] https://github.com/konradybcio/u-boot/commits/konrad/rb1_forcepushing

Konrad
  
Robert Marko Aug. 14, 2023, 10:54 a.m. UTC | #10
On Sat, 12 Aug 2023 at 16:15, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
>
> On 12.08.2023 16:12, Robert Marko wrote:
> > On Sat, 12 Aug 2023 at 16:08, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
> >>
> >> On 12.08.2023 16:07, Robert Marko wrote:
> >>> On Sat, 12 Aug 2023 at 12:47, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
> >>>>
> >>>> On 12.08.2023 11:55, Robert Marko wrote:
> >>>>> On Sat, 12 Aug 2023 at 00:56, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
> >>>>>>
> >>>>>> On 11.08.2023 23:35, Robert Marko wrote:
> >>>>>>> On Fri, 11 Aug 2023 at 23:28, Konrad Dybcio <konrad.dybcio@linaro.org> wrote:
> >>>>>>>>
> >>>>>>>> On 11.08.2023 23:01, Robert Marko wrote:
> >>>>>>>>> Rename the SPI-NOR node to flash@0, remove #address-cells and #size-cells
> >>>>>>>>> as they should be under the partitions subnode and use the generic
> >>>>>>>>> jedec,spi-nor compatible.
> >>>>>>>>>
> >>>>>>>>> Signed-off-by: Robert Marko <robimarko@gmail.com>
> >>>>>>>>> ---
> >>>>>>>> You can also do "nandmanufacturer,mx25l25635e", "jedec,spi-nor"
> >>>>>>>
> >>>>>>> Hi,
> >>>>>>> I grepped the vendor U-Boot to make sure it's not being triggered off
> >>>>>>> the mx25l25635e
> >>>>>>> compatible but the only hit is the IC support itself.
> >>>>>>> MX25L25635 was just the original NOR IC Qualcomm used on the board so
> >>>>>>> to me it made
> >>>>>>> most sense to just use the JEDEC compatible as NOR itself is JEDEC NOR
> >>>>>>> compatible.
> >>>>>> OK if dynamic identification works fine
> >>>>>
> >>>>> It should work fine, datasheet is clear that its JEDEC compatible.
> >>>>> That being said, I dont actually have the board, just figured it was
> >>>>> time for a cleanup as
> >>>>> OpenWrt has been patching DK01 and DK04 for ages.
> >>>> Hm. Do we know whether there are still users of this boards?
> >>>
> >>> I honestly doubt it as they have been broken in OpenWrt for years and
> >>> nobody complained.
> >>> So we are currently removing support for them, but I still wanted to
> >>> at least fixup the DTS state
> >>> upstream.
> >>> These boards are not obtainable anymore.
> >> I also noticed they were detached from the other snapdragons in u-boot
> >> for no good reason (at first glance anyway).
> >
> > If you are talking about the mainline U-Boot then yeah, my basic port was done
> > years ago and I knew way less about the SoC then now.
> > Currently its on my TODO to merge them with Snapdragon and add some proper
> > GPIO and pinctrl drivers as well as using the Linux DTS.
> Take a look at this branch of mine [1], I already did some of that.
>
> If you wish to upstream that, please coordinate with Caleb (+CC) who
> may be interested in the same in parallel.
>
> [1] https://github.com/konradybcio/u-boot/commits/konrad/rb1_forcepushing

That is great, though this is very low hanging fruit for me, I have
plenty more of IPQ807x
stuff that needs to make its way upstream.

Regards,
Robert
>
> Konrad
  

Patch

diff --git a/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi b/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi
index 0714616c9e45..f7ac8f9d0b6f 100644
--- a/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi
+++ b/arch/arm/boot/dts/qcom/qcom-ipq4019-ap.dk01.1.dtsi
@@ -75,11 +75,9 @@  &blsp1_spi1 {
 	status = "okay";
 	cs-gpios = <&tlmm 54 GPIO_ACTIVE_HIGH>;
 
-	mx25l25635e@0 {
-		#address-cells = <1>;
-		#size-cells = <1>;
+	flash@0 {
 		reg = <0>;
-		compatible = "mx25l25635e";
+		compatible = "jedec,spi-nor";
 		spi-max-frequency = <24000000>;
 	};
 };