Extend Freescale eDMA driver bindings to support eDMA3 IP blocks in
i.MX8QM and i.MX8QXP SoCs. In i.MX93, both eDMA3 and eDMA4 are now.
Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
.../devicetree/bindings/dma/fsl,edma.yaml | 118 +++++++++++++++---
1 file changed, 100 insertions(+), 18 deletions(-)
diff --git a/Documentation/devicetree/bindings/dma/fsl,edma.yaml b/Documentation/devicetree/bindings/dma/fsl,edma.yaml
index 5fd8fc604261..2f79492fb332 100644
--- a/Documentation/devicetree/bindings/dma/fsl,edma.yaml
+++ b/Documentation/devicetree/bindings/dma/fsl,edma.yaml
@@ -21,32 +21,20 @@ properties:
- enum:
- fsl,vf610-edma
- fsl,imx7ulp-edma
+ - fsl,imx8qm-adma
+ - fsl,imx8qm-edma
+ - fsl,imx93-edma3
+ - fsl,imx93-edma4
- items:
- const: fsl,ls1028a-edma
- const: fsl,vf610-edma
- reg:
- minItems: 2
- maxItems: 3
-
- interrupts:
- minItems: 2
- maxItems: 17
-
- interrupt-names:
- minItems: 2
- maxItems: 17
-
- "#dma-cells":
- const: 2
-
- dma-channels:
- const: 32
-
clocks:
+ minItems: 1
maxItems: 2
clock-names:
+ minItems: 1
maxItems: 2
big-endian:
@@ -55,6 +43,43 @@ properties:
eDMA are implemented in big endian mode, otherwise in little mode.
type: boolean
+if:
+ properties:
+ compatible:
+ contains:
+ enum:
+ - fsl,imx8qm-adma
+ - fsl,imx8qm-edma
+ - fsl,imx93-edma3
+ - fsl,imx93-edma4
+then:
+ properties:
+ reg:
+ maxItems: 1
+ interrupts:
+ minItems: 1
+ maxItems: 64
+ "#dma-cells":
+ const: 3
+ dma-channels:
+ minItems: 1
+ maxItems: 64
+else:
+ properties:
+ reg:
+ minItems: 2
+ maxItems: 3
+ interrupts:
+ minItems: 2
+ maxItems: 17
+ interrupt-names:
+ minItems: 2
+ maxItems: 17
+ "#dma-cells":
+ const: 2
+ dma-channels:
+ const: 32
+
required:
- "#dma-cells"
- compatible
@@ -101,6 +126,19 @@ allOf:
reg:
maxItems: 2
+ - if:
+ properties:
+ compatible:
+ contains:
+ enum:
+ - fsl,imx93-edma3
+ - fsl,imx93-edma4
+ then:
+ properties:
+ clock-names:
+ const: dma
+
+
unevaluatedProperties: false
examples:
@@ -153,3 +191,47 @@ examples:
clock-names = "dma", "dmamux0";
clocks = <&pcc2 IMX7ULP_CLK_DMA1>, <&pcc2 IMX7ULP_CLK_DMA_MUX1>;
};
+
+ - |
+ #include <dt-bindings/interrupt-controller/arm-gic.h>
+ #include <dt-bindings/clock/imx93-clock.h>
+
+ dma-controller@44000000 {
+ compatible = "fsl,imx93-edma3";
+ reg = <0x44000000 0x200000>;
+ #dma-cells = <3>;
+ dma-channels = <31>;
+ interrupts = <GIC_SPI 95 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 96 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 97 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 98 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 99 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 100 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 101 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 102 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 103 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 104 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 105 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 106 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 107 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 108 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 109 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 110 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 111 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 112 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 113 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 114 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 115 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 116 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 117 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 118 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 119 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 120 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 121 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 122 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 123 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 124 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 125 IRQ_TYPE_LEVEL_HIGH>;
+ clocks = <&clk IMX93_CLK_EDMA1_GATE>;
+ clock-names = "dma";
+ };
--
2.34.1On 13/06/2023 23:31, Frank Li wrote: > Extend Freescale eDMA driver bindings to support eDMA3 IP blocks in > i.MX8QM and i.MX8QXP SoCs. In i.MX93, both eDMA3 and eDMA4 are now. > > Signed-off-by: Frank Li <Frank.Li@nxp.com> > --- > .../devicetree/bindings/dma/fsl,edma.yaml | 118 +++++++++++++++--- > 1 file changed, 100 insertions(+), 18 deletions(-) > > diff --git a/Documentation/devicetree/bindings/dma/fsl,edma.yaml b/Documentation/devicetree/bindings/dma/fsl,edma.yaml > index 5fd8fc604261..2f79492fb332 100644 > --- a/Documentation/devicetree/bindings/dma/fsl,edma.yaml > +++ b/Documentation/devicetree/bindings/dma/fsl,edma.yaml > @@ -21,32 +21,20 @@ properties: > - enum: > - fsl,vf610-edma > - fsl,imx7ulp-edma > + - fsl,imx8qm-adma > + - fsl,imx8qm-edma > + - fsl,imx93-edma3 > + - fsl,imx93-edma4 > - items: > - const: fsl,ls1028a-edma > - const: fsl,vf610-edma > > - reg: > - minItems: 2 > - maxItems: 3 > - > - interrupts: > - minItems: 2 > - maxItems: 17 What is happening here? > - > - interrupt-names: > - minItems: 2 > - maxItems: 17 > - > - "#dma-cells": > - const: 2 > - > - dma-channels: > - const: 32 No, why all these are being removed? > - > clocks: > + minItems: 1 > maxItems: 2 > > clock-names: > + minItems: 1 > maxItems: 2 > > big-endian: > @@ -55,6 +43,43 @@ properties: > eDMA are implemented in big endian mode, otherwise in little mode. > type: boolean > > +if: This should not be outside of your allOf. This patch looks entirely different than your v4 and I don't really understand why. > + properties: > + compatible: > + contains: > + enum: > + - fsl,imx8qm-adma > + - fsl,imx8qm-edma > + - fsl,imx93-edma3 > + - fsl,imx93-edma4 > +then: > + properties: > + reg: > + maxItems: 1 > + interrupts: > + minItems: 1 > + maxItems: 64 What's more, you don't have these properties defined in top-level. Sorry, they should not be moved. I did not ask for this. Best regards, Krzysztof
On Tue, Jun 13, 2023 at 11:43:31PM +0200, Krzysztof Kozlowski wrote:
> On 13/06/2023 23:31, Frank Li wrote:
> > Extend Freescale eDMA driver bindings to support eDMA3 IP blocks in
> > i.MX8QM and i.MX8QXP SoCs. In i.MX93, both eDMA3 and eDMA4 are now.
> >
> > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > ---
> > .../devicetree/bindings/dma/fsl,edma.yaml | 118 +++++++++++++++---
> > 1 file changed, 100 insertions(+), 18 deletions(-)
> >
> > diff --git a/Documentation/devicetree/bindings/dma/fsl,edma.yaml b/Documentation/devicetree/bindings/dma/fsl,edma.yaml
> > index 5fd8fc604261..2f79492fb332 100644
> > --- a/Documentation/devicetree/bindings/dma/fsl,edma.yaml
> > +++ b/Documentation/devicetree/bindings/dma/fsl,edma.yaml
> > @@ -21,32 +21,20 @@ properties:
> > - enum:
> > - fsl,vf610-edma
> > - fsl,imx7ulp-edma
> > + - fsl,imx8qm-adma
> > + - fsl,imx8qm-edma
> > + - fsl,imx93-edma3
> > + - fsl,imx93-edma4
> > - items:
> > - const: fsl,ls1028a-edma
> > - const: fsl,vf610-edma
> >
> > - reg:
> > - minItems: 2
> > - maxItems: 3
> > -
> > - interrupts:
> > - minItems: 2
> > - maxItems: 17
>
> What is happening here?
I found dt_check always check this part firstly, then check allOf.
>
> > -
> > - interrupt-names:
> > - minItems: 2
> > - maxItems: 17
> > -
> > - "#dma-cells":
> > - const: 2
> > -
> > - dma-channels:
> > - const: 32
>
> No, why all these are being removed?
I move common part ahead of if-then-else branch to read early.
>
> > -
> > clocks:
> > + minItems: 1
> > maxItems: 2
> >
> > clock-names:
> > + minItems: 1
> > maxItems: 2
> >
> > big-endian:
> > @@ -55,6 +43,43 @@ properties:
> > eDMA are implemented in big endian mode, otherwise in little mode.
> > type: boolean
> >
> > +if:
>
> This should not be outside of your allOf. This patch looks entirely
> different than your v4 and I don't really understand why.
>
allOf looks like addtional constraints addition to previous define.
for example:
if previous interrupts is 17, I can't overwrite to bigger value 64
in this sesson.
previous version: dts check report two error,
first: interrupt is too long. (look like check top one)
then: interrupt is too short. (look like check allOf part)
>
> > + properties:
> > + compatible:
> > + contains:
> > + enum:
> > + - fsl,imx8qm-adma
> > + - fsl,imx8qm-edma
> > + - fsl,imx93-edma3
> > + - fsl,imx93-edma4
> > +then:
> > + properties:
> > + reg:
> > + maxItems: 1
> > + interrupts:
> > + minItems: 1
> > + maxItems: 64
>
> What's more, you don't have these properties defined in top-level.
> Sorry, they should not be moved. I did not ask for this.
It is there.
if-then-else before "required"
>
> Best regards,
> Krzysztof
>
On 14/06/2023 00:01, Frank Li wrote: > On Tue, Jun 13, 2023 at 11:43:31PM +0200, Krzysztof Kozlowski wrote: >> On 13/06/2023 23:31, Frank Li wrote: >>> Extend Freescale eDMA driver bindings to support eDMA3 IP blocks in >>> i.MX8QM and i.MX8QXP SoCs. In i.MX93, both eDMA3 and eDMA4 are now. >>> >>> Signed-off-by: Frank Li <Frank.Li@nxp.com> >>> --- >>> .../devicetree/bindings/dma/fsl,edma.yaml | 118 +++++++++++++++--- >>> 1 file changed, 100 insertions(+), 18 deletions(-) >>> >>> diff --git a/Documentation/devicetree/bindings/dma/fsl,edma.yaml b/Documentation/devicetree/bindings/dma/fsl,edma.yaml >>> index 5fd8fc604261..2f79492fb332 100644 >>> --- a/Documentation/devicetree/bindings/dma/fsl,edma.yaml >>> +++ b/Documentation/devicetree/bindings/dma/fsl,edma.yaml >>> @@ -21,32 +21,20 @@ properties: >>> - enum: >>> - fsl,vf610-edma >>> - fsl,imx7ulp-edma >>> + - fsl,imx8qm-adma >>> + - fsl,imx8qm-edma >>> + - fsl,imx93-edma3 >>> + - fsl,imx93-edma4 >>> - items: >>> - const: fsl,ls1028a-edma >>> - const: fsl,vf610-edma >>> >>> - reg: >>> - minItems: 2 >>> - maxItems: 3 >>> - >>> - interrupts: >>> - minItems: 2 >>> - maxItems: 17 >> >> What is happening here? > > I found dt_check always check this part firstly, then check allOf. > >> >>> - >>> - interrupt-names: >>> - minItems: 2 >>> - maxItems: 17 >>> - >>> - "#dma-cells": >>> - const: 2 >>> - >>> - dma-channels: >>> - const: 32 >> >> No, why all these are being removed? > > I move common part ahead of if-then-else branch to read early. > >> >>> - >>> clocks: >>> + minItems: 1 >>> maxItems: 2 >>> >>> clock-names: >>> + minItems: 1 >>> maxItems: 2 >>> >>> big-endian: >>> @@ -55,6 +43,43 @@ properties: >>> eDMA are implemented in big endian mode, otherwise in little mode. >>> type: boolean >>> >>> +if: >> >> This should not be outside of your allOf. This patch looks entirely >> different than your v4 and I don't really understand why. >> > > allOf looks like addtional constraints addition to previous define. > for example: > if previous interrupts is 17, I can't overwrite to bigger value 64 > in this sesson. Yes, because the top-level had wrong constraint. Fix top-level, don't remove it. > > previous version: dts check report two error, > first: interrupt is too long. (look like check top one) > then: interrupt is too short. (look like check allOf part) > >> >>> + properties: >>> + compatible: >>> + contains: >>> + enum: >>> + - fsl,imx8qm-adma >>> + - fsl,imx8qm-edma >>> + - fsl,imx93-edma3 >>> + - fsl,imx93-edma4 >>> +then: >>> + properties: >>> + reg: >>> + maxItems: 1 >>> + interrupts: >>> + minItems: 1 >>> + maxItems: 64 >> >> What's more, you don't have these properties defined in top-level. >> Sorry, they should not be moved. I did not ask for this. > > It is there. > if-then-else before "required" It's in if, not in top-level properties. Best regards, Krzysztof
© 2016 - 2026 Red Hat, Inc.