diff mbox

[1/3] dt-bindings: add document of mediatek efuse driver

Message ID 1444984751-4572-2-git-send-email-andrew-ct.chen@mediatek.com (mailing list archive)
State New, archived
Headers show

Commit Message

andrew-ct chen Oct. 16, 2015, 8:39 a.m. UTC
From: Andrew-CT Chen <andrew-ct.chen@mediatek.com>

Add Mediatek MT8173 EFUSE Devicetree binding file

Signed-off-by: Andrew-CT Chen <andrew-ct.chen@mediatek.com>
---
 .../devicetree/bindings/nvmem/mtk-efuse.txt        | 36 ++++++++++++++++++++++
 1 file changed, 36 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/nvmem/mtk-efuse.txt

Comments

Srinivas Kandagatla Oct. 26, 2015, 9:56 a.m. UTC | #1
On 16/10/15 09:39, andrew-ct.chen@mediatek.com wrote:
> From: Andrew-CT Chen <andrew-ct.chen@mediatek.com>
>
> Add Mediatek MT8173 EFUSE Devicetree binding file
>
> Signed-off-by: Andrew-CT Chen <andrew-ct.chen@mediatek.com>

Looks good to me, minor comment below.

Will queue this for v4.5 once v4.4-rc1 is released.


> ---
>   .../devicetree/bindings/nvmem/mtk-efuse.txt        | 36 ++++++++++++++++++++++
>   1 file changed, 36 insertions(+)
>   create mode 100644 Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
>
> diff --git a/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
> new file mode 100644
> index 0000000..92edc3d
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
> @@ -0,0 +1,36 @@
> += Mediatek MTK-EFUSE device tree bindings =
> +
> +This binding is intended to represent MTK-EFUSE which is found in most Mediatek SOCs.
> +
> +Required properties:
> +- compatible: should be "mediatek,mt8173-efuse" or "mediatek,mt8135-efuse"

Can we make the compatible more generic for mediatek socs? Like 
"mediatek,efuse"  Or are they different in each SOC?

--srini
> +- reg: Should contain registers location and length
> +
> += Data cells =
> +Are child nodes of MTK-EFUSE, bindings of which as described in
> +bindings/nvmem/nvmem.txt
> +
> +Example:
> +
> +	efuse: efuse@10206000 {
> +		compatible = "mediatek,mt8173-efuse";
> +		reg	   = <0 0x10206000 0 0x1000>;
> +		#address-cells = <1>;
> +		#size-cells = <1>;
> +
> +		/* Data cells */
> +		thermal_calibration: calib@528 {
> +			reg = <0x528 0xc>;
> +		};
> +	};
> +
> += Data consumers =
> +Are device nodes which consume nvmem data cells.
> +
> +For example:
> +
> +	thermal {
> +		...
> +		nvmem-cells = <&thermal_calibration>;
> +		nvmem-cell-names = "calibration";
> +	};
>
Sascha Hauer Oct. 26, 2015, 10:23 a.m. UTC | #2
On Mon, Oct 26, 2015 at 09:56:14AM +0000, Srinivas Kandagatla wrote:
> 
> 
> On 16/10/15 09:39, andrew-ct.chen@mediatek.com wrote:
> >From: Andrew-CT Chen <andrew-ct.chen@mediatek.com>
> >
> >Add Mediatek MT8173 EFUSE Devicetree binding file
> >
> >Signed-off-by: Andrew-CT Chen <andrew-ct.chen@mediatek.com>
> 
> Looks good to me, minor comment below.
> 
> Will queue this for v4.5 once v4.4-rc1 is released.
> 
> 
> >---
> >  .../devicetree/bindings/nvmem/mtk-efuse.txt        | 36 ++++++++++++++++++++++
> >  1 file changed, 36 insertions(+)
> >  create mode 100644 Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
> >
> >diff --git a/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
> >new file mode 100644
> >index 0000000..92edc3d
> >--- /dev/null
> >+++ b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
> >@@ -0,0 +1,36 @@
> >+= Mediatek MTK-EFUSE device tree bindings =
> >+
> >+This binding is intended to represent MTK-EFUSE which is found in most Mediatek SOCs.
> >+
> >+Required properties:
> >+- compatible: should be "mediatek,mt8173-efuse" or "mediatek,mt8135-efuse"
> 
> Can we make the compatible more generic for mediatek socs? Like
> "mediatek,efuse"  Or are they different in each SOC?

While we can add an additional "mediatek,efuse" compatible we should
always add the most specific compatible first. Otherwise we cannot
differentiate between SoCs later without changing the device tree.

Sascha
Srinivas Kandagatla Oct. 26, 2015, 10:55 a.m. UTC | #3
On 26/10/15 10:23, Sascha Hauer wrote:
> On Mon, Oct 26, 2015 at 09:56:14AM +0000, Srinivas Kandagatla wrote:
>>
>>
>> On 16/10/15 09:39, andrew-ct.chen@mediatek.com wrote:
>>> From: Andrew-CT Chen <andrew-ct.chen@mediatek.com>
>>>
>>> Add Mediatek MT8173 EFUSE Devicetree binding file
>>>
>>> Signed-off-by: Andrew-CT Chen <andrew-ct.chen@mediatek.com>
>>
>> Looks good to me, minor comment below.
>>
>> Will queue this for v4.5 once v4.4-rc1 is released.
>>
>>
>>> ---
>>>   .../devicetree/bindings/nvmem/mtk-efuse.txt        | 36 ++++++++++++++++++++++
>>>   1 file changed, 36 insertions(+)
>>>   create mode 100644 Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
>>>
>>> diff --git a/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
>>> new file mode 100644
>>> index 0000000..92edc3d
>>> --- /dev/null
>>> +++ b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
>>> @@ -0,0 +1,36 @@
>>> += Mediatek MTK-EFUSE device tree bindings =
>>> +
>>> +This binding is intended to represent MTK-EFUSE which is found in most Mediatek SOCs.
>>> +
>>> +Required properties:
>>> +- compatible: should be "mediatek,mt8173-efuse" or "mediatek,mt8135-efuse"
>>
>> Can we make the compatible more generic for mediatek socs? Like
>> "mediatek,efuse"  Or are they different in each SOC?
>
> While we can add an additional "mediatek,efuse" compatible we should
> always add the most specific compatible first. Otherwise we cannot
> differentiate between SoCs later without changing the device tree.
Yes, that’s correct.

AFAIK, Ideally the driver should have more generic compatible string 
unless there is any SOC specific thing, and DT can still use the order 
of specific, generic compatible strings. If not we would end up adding 
compatible strings into the driver for every new SOC.

--srini

>
> Sascha
>
andrew-ct chen Oct. 27, 2015, 5:32 a.m. UTC | #4
On Mon, 2015-10-26 at 10:55 +0000, Srinivas Kandagatla wrote:
> 
> On 26/10/15 10:23, Sascha Hauer wrote:
> > On Mon, Oct 26, 2015 at 09:56:14AM +0000, Srinivas Kandagatla wrote:
> >>
> >>
> >> On 16/10/15 09:39, andrew-ct.chen@mediatek.com wrote:
> >>> From: Andrew-CT Chen <andrew-ct.chen@mediatek.com>
> >>>
> >>> Add Mediatek MT8173 EFUSE Devicetree binding file
> >>>
> >>> Signed-off-by: Andrew-CT Chen <andrew-ct.chen@mediatek.com>
> >>
> >> Looks good to me, minor comment below.
> >>
> >> Will queue this for v4.5 once v4.4-rc1 is released.
> >>
> >>
> >>> ---
> >>>   .../devicetree/bindings/nvmem/mtk-efuse.txt        | 36 ++++++++++++++++++++++
> >>>   1 file changed, 36 insertions(+)
> >>>   create mode 100644 Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
> >>>
> >>> diff --git a/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
> >>> new file mode 100644
> >>> index 0000000..92edc3d
> >>> --- /dev/null
> >>> +++ b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
> >>> @@ -0,0 +1,36 @@
> >>> += Mediatek MTK-EFUSE device tree bindings =
> >>> +
> >>> +This binding is intended to represent MTK-EFUSE which is found in most Mediatek SOCs.
> >>> +
> >>> +Required properties:
> >>> +- compatible: should be "mediatek,mt8173-efuse" or "mediatek,mt8135-efuse"
> >>
> >> Can we make the compatible more generic for mediatek socs? Like
> >> "mediatek,efuse"  Or are they different in each SOC?
> >
> > While we can add an additional "mediatek,efuse" compatible we should
> > always add the most specific compatible first. Otherwise we cannot
> > differentiate between SoCs later without changing the device tree.
> Yes, that’s correct.
> 
> AFAIK, Ideally the driver should have more generic compatible string 
> unless there is any SOC specific thing, and DT can still use the order 
> of specific, generic compatible strings. If not we would end up adding 
> compatible strings into the driver for every new SOC.

We will modify for generic compatible string like "mediatek,efuse" and
rebase it to v4.4-rc1.

> 
> --srini
> 
> >
> > Sascha
> >
andrew-ct chen Oct. 27, 2015, 9:27 a.m. UTC | #5
On Tue, 2015-10-27 at 13:32 +0800, andrew-ct chen wrote:
> On Mon, 2015-10-26 at 10:55 +0000, Srinivas Kandagatla wrote:
> > 
> > On 26/10/15 10:23, Sascha Hauer wrote:
> > > On Mon, Oct 26, 2015 at 09:56:14AM +0000, Srinivas Kandagatla wrote:
> > >>
> > >>
> > >> On 16/10/15 09:39, andrew-ct.chen@mediatek.com wrote:
> > >>> From: Andrew-CT Chen <andrew-ct.chen@mediatek.com>
> > >>>
> > >>> Add Mediatek MT8173 EFUSE Devicetree binding file
> > >>>
> > >>> Signed-off-by: Andrew-CT Chen <andrew-ct.chen@mediatek.com>
> > >>
> > >> Looks good to me, minor comment below.
> > >>
> > >> Will queue this for v4.5 once v4.4-rc1 is released.
> > >>
> > >>
> > >>> ---
> > >>>   .../devicetree/bindings/nvmem/mtk-efuse.txt        | 36 ++++++++++++++++++++++
> > >>>   1 file changed, 36 insertions(+)
> > >>>   create mode 100644 Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
> > >>>
> > >>> diff --git a/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
> > >>> new file mode 100644
> > >>> index 0000000..92edc3d
> > >>> --- /dev/null
> > >>> +++ b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
> > >>> @@ -0,0 +1,36 @@
> > >>> += Mediatek MTK-EFUSE device tree bindings =
> > >>> +
> > >>> +This binding is intended to represent MTK-EFUSE which is found in most Mediatek SOCs.
> > >>> +
> > >>> +Required properties:
> > >>> +- compatible: should be "mediatek,mt8173-efuse" or "mediatek,mt8135-efuse"
> > >>
> > >> Can we make the compatible more generic for mediatek socs? Like
> > >> "mediatek,efuse"  Or are they different in each SOC?
> > >
> > > While we can add an additional "mediatek,efuse" compatible we should
> > > always add the most specific compatible first. Otherwise we cannot
> > > differentiate between SoCs later without changing the device tree.
> > Yes, that’s correct.
> > 
> > AFAIK, Ideally the driver should have more generic compatible string 
> > unless there is any SOC specific thing, and DT can still use the order 
> > of specific, generic compatible strings. If not we would end up adding 
> > compatible strings into the driver for every new SOC.
> 
> We will modify for generic compatible string like "mediatek,efuse" and
> rebase it to v4.4-rc1.

For more clearly, we will modify the compatible string to
"mediatek,mt8173-efuse" and "mediatek,efuse".

> 
> > 
> > --srini
> > 
> > >
> > > Sascha
> > >
>
diff mbox

Patch

diff --git a/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
new file mode 100644
index 0000000..92edc3d
--- /dev/null
+++ b/Documentation/devicetree/bindings/nvmem/mtk-efuse.txt
@@ -0,0 +1,36 @@ 
+= Mediatek MTK-EFUSE device tree bindings =
+
+This binding is intended to represent MTK-EFUSE which is found in most Mediatek SOCs.
+
+Required properties:
+- compatible: should be "mediatek,mt8173-efuse" or "mediatek,mt8135-efuse"
+- reg: Should contain registers location and length
+
+= Data cells =
+Are child nodes of MTK-EFUSE, bindings of which as described in
+bindings/nvmem/nvmem.txt
+
+Example:
+
+	efuse: efuse@10206000 {
+		compatible = "mediatek,mt8173-efuse";
+		reg	   = <0 0x10206000 0 0x1000>;
+		#address-cells = <1>;
+		#size-cells = <1>;
+
+		/* Data cells */
+		thermal_calibration: calib@528 {
+			reg = <0x528 0xc>;
+		};
+	};
+
+= Data consumers =
+Are device nodes which consume nvmem data cells.
+
+For example:
+
+	thermal {
+		...
+		nvmem-cells = <&thermal_calibration>;
+		nvmem-cell-names = "calibration";
+	};