Partition currently called "uboot" does not only contain U-boot, but
rather it contains TF-A, U-boot and U-boot environment.
So, to avoid accidentally deleting the U-boot environment which is
located at 0x180000 split the partition.
"uboot" is not the correct name as you can't boot these boards with U-boot
only, TF-A must be present as well, so rename the "uboot" partition to
"firmware".
While we are here, describe the NOR node as "spi-flash@0" instead of
"m25p80@0" which is the old SPI-NOR driver name.
This won't break booting for existing devices as the SoC-s BootROM is not
partition aware at all, it will simply try booting from 0x0 of the
boot device that is set by bootstrap pins.
This will however prevent accidental or automated flashing of just U-boot
to the partition.
Signed-off-by: Robert Marko <[email protected]>
---
Changes in v2:
* Update the commit description by adressing ABI breaking concerns
---
arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts b/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
index 95d46e8d081c..ac64949bb53e 100644
--- a/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
+++ b/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
@@ -99,7 +99,7 @@ &spi0 {
pinctrl-names = "default";
pinctrl-0 = <&spi_quad_pins>;
- m25p80@0 {
+ spi-flash@0 {
compatible = "jedec,spi-nor";
reg = <0>;
spi-max-frequency = <54000000>;
@@ -108,10 +108,15 @@ partitions {
compatible = "fixed-partitions";
#address-cells = <1>;
#size-cells = <1>;
- /* only bootloader is located on the SPI */
+
partition@0 {
- label = "uboot";
- reg = <0 0x400000>;
+ label = "firmware";
+ reg = <0x0 0x180000>;
+ };
+
+ partition@180000 {
+ label = "u-boot-env";
+ reg = <0x180000 0x10000>;
};
};
};
--
2.35.1
uDPU has a pair of NCT375 temperature sensors, which are TMP75C compatible
as far as the driver is concerned.
The current LM75 compatible worked as all of the LM75 compatible sensors
are backwards compatible with the original part, but it meant that lower
resolution and incorrect sample rate was being used.
The "lm75" compatible has been deprecated anyway and is meant as fallback
in order to keep older DTS-es working.
Signed-off-by: Robert Marko <[email protected]>
---
Changes in v2:
* None
---
arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts b/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
index ac64949bb53e..1f534c0c65f7 100644
--- a/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
+++ b/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
@@ -153,15 +153,15 @@ &i2c1 {
scl-gpios = <&gpionb 2 (GPIO_ACTIVE_HIGH | GPIO_OPEN_DRAIN)>;
sda-gpios = <&gpionb 3 (GPIO_ACTIVE_HIGH | GPIO_OPEN_DRAIN)>;
- lm75@48 {
+ nct375@48 {
status = "okay";
- compatible = "lm75";
+ compatible = "ti,tmp75c";
reg = <0x48>;
};
- lm75@49 {
+ nct375@49 {
status = "okay";
- compatible = "lm75";
+ compatible = "ti,tmp75c";
reg = <0x49>;
};
};
--
2.35.1
On Tue, Mar 22, 2022 at 11:58:57AM +0100, Robert Marko wrote:
> uDPU has a pair of NCT375 temperature sensors, which are TMP75C compatible
> as far as the driver is concerned.
>
> The current LM75 compatible worked as all of the LM75 compatible sensors
> are backwards compatible with the original part, but it meant that lower
> resolution and incorrect sample rate was being used.
>
> The "lm75" compatible has been deprecated anyway and is meant as fallback
> in order to keep older DTS-es working.
>
> Signed-off-by: Robert Marko <[email protected]>
Reviewed-by: Andrew Lunn <[email protected]>
Andrew
On Tue, Mar 22, 2022 at 11:58:56AM +0100, Robert Marko wrote:
> Partition currently called "uboot" does not only contain U-boot, but
> rather it contains TF-A, U-boot and U-boot environment.
>
> So, to avoid accidentally deleting the U-boot environment which is
> located at 0x180000 split the partition.
>
> "uboot" is not the correct name as you can't boot these boards with U-boot
> only, TF-A must be present as well, so rename the "uboot" partition to
> "firmware".
>
> While we are here, describe the NOR node as "spi-flash@0" instead of
> "m25p80@0" which is the old SPI-NOR driver name.
>
> This won't break booting for existing devices as the SoC-s BootROM is not
> partition aware at all, it will simply try booting from 0x0 of the
> boot device that is set by bootstrap pins.
>
> This will however prevent accidental or automated flashing of just U-boot
> to the partition.
>
> Signed-off-by: Robert Marko <[email protected]>
Reviewed-by: Andrew Lunn <[email protected]>
Andrew
On Tuesday 22 March 2022 11:58:56 Robert Marko wrote:
> Partition currently called "uboot" does not only contain U-boot, but
> rather it contains TF-A, U-boot and U-boot environment.
>
> So, to avoid accidentally deleting the U-boot environment which is
> located at 0x180000 split the partition.
>
> "uboot" is not the correct name as you can't boot these boards with U-boot
> only, TF-A must be present as well, so rename the "uboot" partition to
> "firmware".
>
> While we are here, describe the NOR node as "spi-flash@0" instead of
> "m25p80@0" which is the old SPI-NOR driver name.
>
> This won't break booting for existing devices as the SoC-s BootROM is not
> partition aware at all, it will simply try booting from 0x0 of the
> boot device that is set by bootstrap pins.
>
> This will however prevent accidental or automated flashing of just U-boot
> to the partition.
>
> Signed-off-by: Robert Marko <[email protected]>
Reviewed-by: Pali Rohár <[email protected]>
> ---
> Changes in v2:
> * Update the commit description by adressing ABI breaking concerns
> ---
> arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts | 13 +++++++++----
> 1 file changed, 9 insertions(+), 4 deletions(-)
>
> diff --git a/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts b/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
> index 95d46e8d081c..ac64949bb53e 100644
> --- a/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
> +++ b/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
> @@ -99,7 +99,7 @@ &spi0 {
> pinctrl-names = "default";
> pinctrl-0 = <&spi_quad_pins>;
>
> - m25p80@0 {
> + spi-flash@0 {
> compatible = "jedec,spi-nor";
> reg = <0>;
> spi-max-frequency = <54000000>;
> @@ -108,10 +108,15 @@ partitions {
> compatible = "fixed-partitions";
> #address-cells = <1>;
> #size-cells = <1>;
> - /* only bootloader is located on the SPI */
> +
> partition@0 {
> - label = "uboot";
> - reg = <0 0x400000>;
> + label = "firmware";
> + reg = <0x0 0x180000>;
> + };
> +
> + partition@180000 {
> + label = "u-boot-env";
> + reg = <0x180000 0x10000>;
> };
> };
> };
> --
> 2.35.1
>
On Tuesday 22 March 2022 11:58:57 Robert Marko wrote:
> uDPU has a pair of NCT375 temperature sensors, which are TMP75C compatible
> as far as the driver is concerned.
>
> The current LM75 compatible worked as all of the LM75 compatible sensors
> are backwards compatible with the original part, but it meant that lower
> resolution and incorrect sample rate was being used.
>
> The "lm75" compatible has been deprecated anyway and is meant as fallback
> in order to keep older DTS-es working.
>
> Signed-off-by: Robert Marko <[email protected]>
Reviewed-by: Pali Rohár <[email protected]>
> ---
> Changes in v2:
> * None
> ---
> arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts b/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
> index ac64949bb53e..1f534c0c65f7 100644
> --- a/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
> +++ b/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
> @@ -153,15 +153,15 @@ &i2c1 {
> scl-gpios = <&gpionb 2 (GPIO_ACTIVE_HIGH | GPIO_OPEN_DRAIN)>;
> sda-gpios = <&gpionb 3 (GPIO_ACTIVE_HIGH | GPIO_OPEN_DRAIN)>;
>
> - lm75@48 {
> + nct375@48 {
> status = "okay";
> - compatible = "lm75";
> + compatible = "ti,tmp75c";
> reg = <0x48>;
> };
>
> - lm75@49 {
> + nct375@49 {
> status = "okay";
> - compatible = "lm75";
> + compatible = "ti,tmp75c";
> reg = <0x49>;
> };
> };
> --
> 2.35.1
>
Robert Marko <[email protected]> writes:
> Partition currently called "uboot" does not only contain U-boot, but
> rather it contains TF-A, U-boot and U-boot environment.
>
> So, to avoid accidentally deleting the U-boot environment which is
> located at 0x180000 split the partition.
>
> "uboot" is not the correct name as you can't boot these boards with U-boot
> only, TF-A must be present as well, so rename the "uboot" partition to
> "firmware".
>
> While we are here, describe the NOR node as "spi-flash@0" instead of
> "m25p80@0" which is the old SPI-NOR driver name.
>
> This won't break booting for existing devices as the SoC-s BootROM is not
> partition aware at all, it will simply try booting from 0x0 of the
> boot device that is set by bootstrap pins.
>
> This will however prevent accidental or automated flashing of just U-boot
> to the partition.
>
> Signed-off-by: Robert Marko <[email protected]>
> ---
> Changes in v2:
> * Update the commit description by adressing ABI breaking concerns
Applied on mvebu/dt64
Thanks,
Gregory
> ---
> arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts | 13 +++++++++----
> 1 file changed, 9 insertions(+), 4 deletions(-)
>
> diff --git a/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts b/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
> index 95d46e8d081c..ac64949bb53e 100644
> --- a/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
> +++ b/arch/arm64/boot/dts/marvell/armada-3720-uDPU.dts
> @@ -99,7 +99,7 @@ &spi0 {
> pinctrl-names = "default";
> pinctrl-0 = <&spi_quad_pins>;
>
> - m25p80@0 {
> + spi-flash@0 {
> compatible = "jedec,spi-nor";
> reg = <0>;
> spi-max-frequency = <54000000>;
> @@ -108,10 +108,15 @@ partitions {
> compatible = "fixed-partitions";
> #address-cells = <1>;
> #size-cells = <1>;
> - /* only bootloader is located on the SPI */
> +
> partition@0 {
> - label = "uboot";
> - reg = <0 0x400000>;
> + label = "firmware";
> + reg = <0x0 0x180000>;
> + };
> +
> + partition@180000 {
> + label = "u-boot-env";
> + reg = <0x180000 0x10000>;
> };
> };
> };
> --
> 2.35.1
>
--
Gregory Clement, Bootlin
Embedded Linux and Kernel engineering
http://bootlin.com
On Tue, May 10, 2022 at 3:56 PM Krzysztof Kozlowski
<[email protected]> wrote:
>
> On 22/03/2022 11:58, Robert Marko wrote:
> > Partition currently called "uboot" does not only contain U-boot, but
> > rather it contains TF-A, U-boot and U-boot environment.
> >
> > So, to avoid accidentally deleting the U-boot environment which is
> > located at 0x180000 split the partition.
> >
> > "uboot" is not the correct name as you can't boot these boards with U-boot
> > only, TF-A must be present as well, so rename the "uboot" partition to
> > "firmware".
> >
> > While we are here, describe the NOR node as "spi-flash@0" instead of
> > "m25p80@0" which is the old SPI-NOR driver name.
>
> This patch was applied, but it was never tested with dtbs_check (nor
> rebased on my cleanups). spi-flash is still wrong.
Hi Krzysztof, sorry for that, I did run dtbs_check on it but there is
a whole forest of
warnings about schemas not being matched as most of the bindings are
still plaintext
and I missed this.
I did correct this to flash@0 in the eDPU series that was sent out today.
Regards,
Robert
>
> Best regards,
> Krzysztof
--
Robert Marko
Staff Embedded Linux Engineer
Sartura Ltd.
Lendavska ulica 16a
10000 Zagreb, Croatia
Email: [email protected]
Web: http://www.sartura.hr
On 22/03/2022 11:58, Robert Marko wrote:
> Partition currently called "uboot" does not only contain U-boot, but
> rather it contains TF-A, U-boot and U-boot environment.
>
> So, to avoid accidentally deleting the U-boot environment which is
> located at 0x180000 split the partition.
>
> "uboot" is not the correct name as you can't boot these boards with U-boot
> only, TF-A must be present as well, so rename the "uboot" partition to
> "firmware".
>
> While we are here, describe the NOR node as "spi-flash@0" instead of
> "m25p80@0" which is the old SPI-NOR driver name.
This patch was applied, but it was never tested with dtbs_check (nor
rebased on my cleanups). spi-flash is still wrong.
Best regards,
Krzysztof