The peripheral type specified in the dma phandle for each SAI node
is incorrect. Change it to specify the SAI peripheral.
Signed-off-by: Matt Porter <[email protected]>
---
arch/arm64/boot/dts/freescale/imx8mm.dtsi | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/arch/arm64/boot/dts/freescale/imx8mm.dtsi b/arch/arm64/boot/dts/freescale/imx8mm.dtsi
index 1e5e11592f7b..ddc93fc4817a 100644
--- a/arch/arm64/boot/dts/freescale/imx8mm.dtsi
+++ b/arch/arm64/boot/dts/freescale/imx8mm.dtsi
@@ -240,7 +240,7 @@
<&clk IMX8MM_CLK_SAI1_ROOT>,
<&clk IMX8MM_CLK_DUMMY>, <&clk IMX8MM_CLK_DUMMY>;
clock-names = "bus", "mclk1", "mclk2", "mclk3";
- dmas = <&sdma2 0 2 0>, <&sdma2 1 2 0>;
+ dmas = <&sdma2 0 24 0>, <&sdma2 1 24 0>;
dma-names = "rx", "tx";
status = "disabled";
};
@@ -253,7 +253,7 @@
<&clk IMX8MM_CLK_SAI2_ROOT>,
<&clk IMX8MM_CLK_DUMMY>, <&clk IMX8MM_CLK_DUMMY>;
clock-names = "bus", "mclk1", "mclk2", "mclk3";
- dmas = <&sdma2 2 2 0>, <&sdma2 3 2 0>;
+ dmas = <&sdma2 2 24 0>, <&sdma2 3 24 0>;
dma-names = "rx", "tx";
status = "disabled";
};
@@ -267,7 +267,7 @@
<&clk IMX8MM_CLK_SAI3_ROOT>,
<&clk IMX8MM_CLK_DUMMY>, <&clk IMX8MM_CLK_DUMMY>;
clock-names = "bus", "mclk1", "mclk2", "mclk3";
- dmas = <&sdma2 4 2 0>, <&sdma2 5 2 0>;
+ dmas = <&sdma2 4 24 0>, <&sdma2 5 24 0>;
dma-names = "rx", "tx";
status = "disabled";
};
@@ -280,7 +280,7 @@
<&clk IMX8MM_CLK_SAI5_ROOT>,
<&clk IMX8MM_CLK_DUMMY>, <&clk IMX8MM_CLK_DUMMY>;
clock-names = "bus", "mclk1", "mclk2", "mclk3";
- dmas = <&sdma2 8 2 0>, <&sdma2 9 2 0>;
+ dmas = <&sdma2 8 24 0>, <&sdma2 9 24 0>;
dma-names = "rx", "tx";
status = "disabled";
};
@@ -293,7 +293,7 @@
<&clk IMX8MM_CLK_SAI6_ROOT>,
<&clk IMX8MM_CLK_DUMMY>, <&clk IMX8MM_CLK_DUMMY>;
clock-names = "bus", "mclk1", "mclk2", "mclk3";
- dmas = <&sdma2 10 2 0>, <&sdma2 11 2 0>;
+ dmas = <&sdma2 10 24 0>, <&sdma2 11 24 0>;
dma-names = "rx", "tx";
status = "disabled";
};
--
2.20.1
On Tue, Mar 31, 2020 at 11:36 PM Matt Porter <[email protected]> wrote:
>
> The peripheral type specified in the dma phandle for each SAI node
> is incorrect. Change it to specify the SAI peripheral.
>
> Signed-off-by: Matt Porter <[email protected]>
Hi Matt,
Why do you think this is incorrect? AFAIK script number 2 works fine
for SAI. Can you add
more details on what bug are you encountering?
Adding Robin the owner of SDMA.
On Wed, Apr 15, 2020 at 09:04:43PM +0300, Daniel Baluta wrote:
> On Tue, Mar 31, 2020 at 11:36 PM Matt Porter <[email protected]> wrote:
> >
> > The peripheral type specified in the dma phandle for each SAI node
> > is incorrect. Change it to specify the SAI peripheral.
> >
> > Signed-off-by: Matt Porter <[email protected]>
>
> Hi Matt,
>
> Why do you think this is incorrect? AFAIK script number 2 works fine
> for SAI. Can you add
> more details on what bug are you encountering?
>
> Adding Robin the owner of SDMA.
Hi Daniel,
Thanks for the response. I was experiencing timeouts that were traced
back to dma incompletions. Changing the script at the time fixed that
issue. Now, given your response I went back and checked this again and
verified that script 2 does work for me now. I did change firmware to
the latest v4.4 from something quite old so now I'm suspecting I had
bad firmware loaded up. Does that sound plausible? In any case, I can
confirm this is not needed.
-Matt
On 2020/04/16 Matt Porter <[email protected]> wrote:d
> On Wed, Apr 15, 2020 at 09:04:43PM +0300, Daniel Baluta wrote:
> > On Tue, Mar 31, 2020 at 11:36 PM Matt Porter <[email protected]>
> wrote:
> > >
> > > The peripheral type specified in the dma phandle for each SAI node
> > > is incorrect. Change it to specify the SAI peripheral.
> > >
> > > Signed-off-by: Matt Porter <[email protected]>
> >
> > Hi Matt,
> >
> > Why do you think this is incorrect? AFAIK script number 2 works fine
> > for SAI. Can you add more details on what bug are you encountering?
> >
> > Adding Robin the owner of SDMA.
>
> Hi Daniel,
>
> Thanks for the response. I was experiencing timeouts that were traced back to
> dma incompletions. Changing the script at the time fixed that issue. Now, given
> your response I went back and checked this again and verified that script 2
> does work for me now. I did change firmware to the latest v4.4 from something
> quite old so now I'm suspecting I had bad firmware loaded up. Does that sound
> plausible? In any case, I can confirm this is not needed.
No, that's not related with firmware version since both scripts are all ROM scripts instead of RAM scripts which's in sdma firmware.