Skip to contentSkip to conversation

Documentation: bindings: clk: Add bindings for i.MX8MM BLK_CTL

PATCH In review Marek Vasut 5 participants 19 messages Oct 3, 2020 – Feb 4, 2021

Subsystems (9)
lore
Share & tools
lore
#0 lore raw

PATCH 1/5 Documentation: bindings: clk: Add bindings for i.MX8MM BLK_CTL

Add the i.MX8MM BLK_CTL compatible string to the list.

Cc Abel Vesa <[email protected]>
Cc Dong Aisheng <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc Guido Günther <[email protected]>
Cc Lucas Stach <[email protected]>
Cc Rob Herring <[email protected]>
Cc Shawn Guo <[email protected]>
Cc NXP Linux Team <[email protected]>
---
 Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
 1 file changed, 1 insertion(+)
Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml +1 -0
index 5e9eb402b9b6..346429f49093 100644
@@ -20,6 +20,7 @@ properties:
   compatible:
     items:
       - enum:
+         - fsl,imx8mm-dispmix-blk-ctl
          - fsl,imx8mp-audio-blk-ctl
          - fsl,imx8mp-hdmi-blk-ctl
          - fsl,imx8mp-media-blk-ctl
Signature · 1 line
2.28.0
#1 lore raw

PATCH 2/5 dt-bindings: clock: imx8mm: Add media blk_ctl clock IDs

These will be used by the imx8mm for blk_ctl driver.

Cc Abel Vesa <[email protected]>
Cc Dong Aisheng <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc Guido Günther <[email protected]>
Cc Lucas Stach <[email protected]>
Cc Rob Herring <[email protected]>
Cc Shawn Guo <[email protected]>
Cc NXP Linux Team <[email protected]>
---
 include/dt-bindings/clock/imx8mm-clock.h | 7 +++++++
 1 file changed, 7 insertions(+)
include/dt-bindings/clock/imx8mm-clock.h +7 -0
index e63a5530aed7..576b229a72f2 100644
@@ -276,4 +276,11 @@
 
 #define IMX8MM_CLK_END				252
 
+#define IMX8MM_CLK_MEDIA_BLK_CTL_LCDIF_APB		0
+#define IMX8MM_CLK_MEDIA_BLK_CTL_LCDIF_PIXEL		1
+#define IMX8MM_CLK_MEDIA_BLK_CTL_MIPI_DSI_PCLK		2
+#define IMX8MM_CLK_MEDIA_BLK_CTL_MIPI_DSI_CLKREF	3
+
+#define IMX8MM_CLK_MEDIA_BLK_CTL_END			4
+
 #endif
Signature · 1 line
2.28.0
maintainer Acked-by on v1 2/5
#2 lore raw

Re: [PATCH 2/5] dt-bindings: clock: imx8mm: Add media blk_ctl clock IDs

Quoted 12 lines · Marek Vasut

These will be used by the imx8mm for blk_ctl driver.

Cc Abel Vesa <[email protected]>
Cc Dong Aisheng <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc Guido Günther <[email protected]>
Cc Lucas Stach <[email protected]>
Cc Rob Herring <[email protected]>
Cc Shawn Guo <[email protected]>
Cc NXP Linux Team <[email protected]>
Quoted 4 lines
---
 include/dt-bindings/clock/imx8mm-clock.h | 7 +++++++
 1 file changed, 7 insertions(+)
#3 lore raw

PATCH 3/5 dt-bindings: reset: imx8mm: Add media blk_ctl reset IDs

In reply to Marek Vasut

These will be used by the imx8mm for blk_ctl driver.

Cc Abel Vesa <[email protected]>
Cc Dong Aisheng <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc Guido Günther <[email protected]>
Cc Lucas Stach <[email protected]>
Cc Rob Herring <[email protected]>
Cc Shawn Guo <[email protected]>
Cc NXP Linux Team <[email protected]>
---
 include/dt-bindings/reset/imx8mm-reset.h | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
 create mode 100644 include/dt-bindings/reset/imx8mm-reset.h
include/dt-bindings/reset/imx8mm-reset.h new file +14 -0
new file mode 100644
index 000000000000..1dda59d30c06
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2020 Marek Vasut <[email protected]>
+ */
+
+#ifndef DT_BINDING_RESET_IMX8MM_H
+#define DT_BINDING_RESET_IMX8MM_H
+
+#define IMX8MM_MEDIA_BLK_CTL_RESET_MIPI_DSI_I_PRESET	0
+#define IMX8MM_MEDIA_BLK_CTL_RESET_MIPI_M_RESET		1
+
+#define IMX8MM_MEDIA_BLK_CTL_RESET_NUM			2
+
+#endif
Signature · 1 line
2.28.0
maintainer Acked-by on v1 3/5
#4 lore raw

Re: [PATCH 3/5] dt-bindings: reset: imx8mm: Add media blk_ctl reset IDs

Quoted 12 lines · Marek Vasut

These will be used by the imx8mm for blk_ctl driver.

Cc Abel Vesa <[email protected]>
Cc Dong Aisheng <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc Guido Günther <[email protected]>
Cc Lucas Stach <[email protected]>
Cc Rob Herring <[email protected]>
Cc Shawn Guo <[email protected]>
Cc NXP Linux Team <[email protected]>
Quoted 5 lines
---
 include/dt-bindings/reset/imx8mm-reset.h | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
 create mode 100644 include/dt-bindings/reset/imx8mm-reset.h
#5 lore raw

PATCH 4/5 clk: imx: Fix rewriting of hws by resets in generic blk-ctl driver

In reply to Marek Vasut

Do not call imx_blk_ctl_register_one_clock() if the hw type is a reset, only call it for clock, otherwise hws[hw->id] content might be corrupted or NULL if the above is also called for hw type reset.

Cc Abel Vesa <[email protected]>
Cc Dong Aisheng <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc Guido Günther <[email protected]>
Cc Lucas Stach <[email protected]>
Cc NXP Linux Team <[email protected]>
Cc Shawn Guo <[email protected]>
---
 drivers/clk/imx/clk-blk-ctl.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/clk/imx/clk-blk-ctl.c +3 -0
index 1a6f1eb49f69..8dc59868984f 100644
@@ -221,6 +221,9 @@ static int imx_blk_ctl_register_clock_controller(struct device *dev)
 	for (i = 0; i < dev_data->hws_num; i++) {
 		struct imx_blk_ctl_hw *hw = &dev_data->hws[i];
 
+		if (hw->type == BLK_CTL_RESET)
+			continue;
+
 		hws[hw->id] = imx_blk_ctl_register_one_clock(dev, hw);
 		WARN(IS_ERR(hws[hw->id]), "failed to register clock %d", hw->id);
 	}
Signature · 7 lines
2.28.0


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
#6 lore raw

PATCH 5/5 clk: imx: Add blk-ctl driver for i.MX8MM

In reply to Marek Vasut

The i.MX8MM platform also has a BLK_CTL, however it is not documented in the documentation at all. Enable the generic blk_ctl driver so the IP can be controlled the same way as on MX8MP, and add the clock and reset entries for MX8MM.

Cc Abel Vesa <[email protected]>
Cc Dong Aisheng <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc Guido Günther <[email protected]>
Cc Lucas Stach <[email protected]>
Cc NXP Linux Team <[email protected]>
Cc Shawn Guo <[email protected]>
---
 drivers/clk/imx/Makefile             |  2 +-
 drivers/clk/imx/clk-blk-ctl-imx8mm.c | 66 ++++++++++++++++++++++++++++
 2 files changed, 67 insertions(+), 1 deletion(-)
 create mode 100644 drivers/clk/imx/clk-blk-ctl-imx8mm.c
drivers/clk/imx/Makefile +1 -1
index 6e70e6821727..5e7cc7ec29b6 100644
@@ -25,7 +25,7 @@ obj-$(CONFIG_MXC_CLK_SCU) += \
 	clk-scu.o \
 	clk-lpcg-scu.o
 
-obj-$(CONFIG_CLK_IMX8MM) += clk-imx8mm.o
+obj-$(CONFIG_CLK_IMX8MM) += clk-imx8mm.o clk-blk-ctl.o clk-blk-ctl-imx8mm.o
 obj-$(CONFIG_CLK_IMX8MN) += clk-imx8mn.o
 obj-$(CONFIG_CLK_IMX8MP) += clk-imx8mp.o clk-blk-ctl.o clk-blk-ctl-imx8mp.o
 obj-$(CONFIG_CLK_IMX8MQ) += clk-imx8mq.o
drivers/clk/imx/clk-blk-ctl-imx8mm.c new file +66 -0

66 lines not on this page · raw message

Signature · 7 lines
2.28.0


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
maintainer Acked-by on v1 1/5
#7 lore raw
In reply to Marek Vasut
Quoted 12 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

Cc Abel Vesa <[email protected]>
Cc Dong Aisheng <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc Guido Günther <[email protected]>
Cc Lucas Stach <[email protected]>
Cc Rob Herring <[email protected]>
Cc Shawn Guo <[email protected]>
Cc NXP Linux Team <[email protected]>
Quoted 4 lines
---
 Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
 1 file changed, 1 insertion(+)
#8 lore raw
In reply to Marek Vasut
Quoted 13 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

Cc Abel Vesa <[email protected]>
Cc Dong Aisheng <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc Guido Günther <[email protected]>
Cc Lucas Stach <[email protected]>
Cc Rob Herring <[email protected]>
Cc Shawn Guo <[email protected]>
Cc NXP Linux Team <[email protected]>
Quoted 4 lines
---
 Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
 1 file changed, 1 insertion(+)

Is there a DTSI change part of this patch? I was going to try to test it, but I am not seeing a change to the imx8mm.dtsi, and I am not sure where to put the node.

adam

Quoted 20 lines
Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml +1 -1
index 5e9eb402b9b6..346429f49093 100644
@@ -20,6 +20,7 @@ properties:
   compatible:
     items:
       - enum:
+         - fsl,imx8mm-dispmix-blk-ctl
          - fsl,imx8mp-audio-blk-ctl
          - fsl,imx8mp-hdmi-blk-ctl
          - fsl,imx8mp-media-blk-ctl
--

2.28.0

_______________________________________________
linux-arm-kernel mailing list [email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

#9 lore raw
Quoted 3 lines · Adam Ford
Quoted 2 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

[...]

Quoted 8 lines
Quoted 4 lines
---
 Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
 1 file changed, 1 insertion(+)

Is there a DTSI change part of this patch? I was going to try to test it, but I am not seeing a change to the imx8mm.dtsi, and I am not sure where to put the node.

There are in fact quite a few other pieces you need to have in place, this patchset in itself is not particularly useful, it is just infra for the LCDIF and MIPI DSIM block control. You might want to wait until they all land in next and test that result.

#10 lore raw
Quoted 11 lines · Marek Vasut
Quoted 3 lines · Adam Ford
Quoted 2 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

[...]

Quoted 5 lines
Quoted 4 lines
---
 Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
 1 file changed, 1 insertion(+)
Quoted 8 lines
Quoted 3 lines

Is there a DTSI change part of this patch? I was going to try to test it, but I am not seeing a change to the imx8mm.dtsi, and I am not sure where to put the node.

There are in fact quite a few other pieces you need to have in place, this patchset in itself is not particularly useful, it is just infra for the LCDIF and MIPI DSIM block control. You might want to wait until they all land in next and test that result.

I have several patches in place, the GPCv2, this block driver, enabling GPU DT node, I'm also working on the DSIM patch you posted. I was hoping to test them all together and reply to the various threads with tested-by. I also want to get my device tree stuff ready on the beacon boards so when everything lands, I can post DTS updates to enable the LCDIF, DSI, and the HDMI bridge.

If you have a repo somewhere that has all these combined, I can just work on the final layer to enable the device tree plumbing on my board. I just need the imx8mm.dtsi changes for this, DSIM, and the LCDIF so I can finish the task.

adam

#11 lore raw
Quoted 29 lines · Adam Ford
Quoted 11 lines · Marek Vasut
Quoted 3 lines · Adam Ford
Quoted 2 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

[...]

Quoted 5 lines
Quoted 4 lines
---
 Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
 1 file changed, 1 insertion(+)
Quoted 8 lines
Quoted 3 lines

Is there a DTSI change part of this patch? I was going to try to test it, but I am not seeing a change to the imx8mm.dtsi, and I am not sure where to put the node.

There are in fact quite a few other pieces you need to have in place, this patchset in itself is not particularly useful, it is just infra for the LCDIF and MIPI DSIM block control. You might want to wait until they all land in next and test that result.

I have several patches in place, the GPCv2, this block driver, enabling GPU DT node, I'm also working on the DSIM patch you posted. I was hoping to test them all together and reply to the various threads with tested-by. I also want to get my device tree stuff ready on the beacon boards so when everything lands, I can post DTS updates to enable the LCDIF, DSI, and the HDMI bridge.

Quoted 4 lines

If you have a repo somewhere that has all these combined, I can just work on the final layer to enable the device tree plumbing on my board. I just need the imx8mm.dtsi changes for this, DSIM, and the LCDIF so I can finish the task.

On that note, I also have a i.MX8M Nano board which is similar to my 8MM. If I understood the 8MM clock block driver better, I hope to adapt your changes for the Nano too. Once the GPCv2 driver is accepted, I was also going to look at updating it to support the Nano as well which also has the same DSIM and LCDIF as the 8MM as well and a better GPU than the Mini but lacking the VPU.

adam

Quoted 2 lines

adam

#12 lore raw
Quoted 35 lines · Adam Ford
Quoted 29 lines · Adam Ford
Quoted 11 lines · Marek Vasut
Quoted 3 lines · Adam Ford
Quoted 2 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

[...]

Quoted 5 lines
Quoted 4 lines
---
 Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
 1 file changed, 1 insertion(+)
Quoted 8 lines
Quoted 3 lines

Is there a DTSI change part of this patch? I was going to try to test it, but I am not seeing a change to the imx8mm.dtsi, and I am not sure where to put the node.

There are in fact quite a few other pieces you need to have in place, this patchset in itself is not particularly useful, it is just infra for the LCDIF and MIPI DSIM block control. You might want to wait until they all land in next and test that result.

I have several patches in place, the GPCv2, this block driver, enabling GPU DT node, I'm also working on the DSIM patch you posted. I was hoping to test them all together and reply to the various threads with tested-by. I also want to get my device tree stuff ready on the beacon boards so when everything lands, I can post DTS updates to enable the LCDIF, DSI, and the HDMI bridge.

Quoted 4 lines

If you have a repo somewhere that has all these combined, I can just work on the final layer to enable the device tree plumbing on my board. I just need the imx8mm.dtsi changes for this, DSIM, and the LCDIF so I can finish the task.

Quoted 6 lines

On that note, I also have a i.MX8M Nano board which is similar to my 8MM. If I understood the 8MM clock block driver better, I hope to adapt your changes for the Nano too. Once the GPCv2 driver is accepted, I was also going to look at updating it to support the Nano as well which also has the same DSIM and LCDIF as the 8MM as well and a better GPU than the Mini but lacking the VPU.

I don't have a branch, but I sent you the collected patches off-list.

#13 lore raw

Hi,

Quoted 36 lines · Marek Vasut
Quoted 35 lines · Adam Ford
Quoted 33 lines · Adam Ford
Quoted 11 lines · Marek Vasut
Quoted 3 lines · Adam Ford
Quoted 2 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

[...]

Quoted 5 lines
Quoted 4 lines
---
  Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
  1 file changed, 1 insertion(+)
Quoted 8 lines
Quoted 3 lines

Is there a DTSI change part of this patch? I was going to try to test it, but I am not seeing a change to the imx8mm.dtsi, and I am not sure where to put the node.

There are in fact quite a few other pieces you need to have in place, this patchset in itself is not particularly useful, it is just infra for the LCDIF and MIPI DSIM block control. You might want to wait until they all land in next and test that result.

I have several patches in place, the GPCv2, this block driver, enabling GPU DT node, I'm also working on the DSIM patch you posted. I was hoping to test them all together and reply to the various threads with tested-by. I also want to get my device tree stuff ready on the beacon boards so when everything lands, I can post DTS updates to enable the LCDIF, DSI, and the HDMI bridge.

If you have a repo somewhere that has all these combined, I can just work on the final layer to enable the device tree plumbing on my board. I just need the imx8mm.dtsi changes for this, DSIM, and the LCDIF so I can finish the task.

Quoted 9 lines
Quoted 6 lines

On that note, I also have a i.MX8M Nano board which is similar to my 8MM. If I understood the 8MM clock block driver better, I hope to adapt your changes for the Nano too. Once the GPCv2 driver is accepted, I was also going to look at updating it to support the Nano as well which also has the same DSIM and LCDIF as the 8MM as well and a better GPU than the Mini but lacking the VPU.

I don't have a branch, but I sent you the collected patches off-list.

I would also be interested in the patch collection for BLK_CTL, DSIM, etc. Marek, would you mind sending me those, too?

Adam, did you already set up a branch and do some tests with the full stack?

Thanks,
Frieder

#14 lore raw
Quoted 47 lines · Frieder Schrempf

Hi,

Quoted 43 lines · Marek Vasut
Quoted 41 lines · Adam Ford
Quoted 29 lines · Adam Ford
Quoted 11 lines · Marek Vasut
Quoted 3 lines · Adam Ford
Quoted 2 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

[...]

Quoted 5 lines
Quoted 4 lines
---
  Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
  1 file changed, 1 insertion(+)
Quoted 8 lines
Quoted 3 lines

Is there a DTSI change part of this patch? I was going to try to test it, but I am not seeing a change to the imx8mm.dtsi, and I am not sure where to put the node.

There are in fact quite a few other pieces you need to have in place, this patchset in itself is not particularly useful, it is just infra for the LCDIF and MIPI DSIM block control. You might want to wait until they all land in next and test that result.

I have several patches in place, the GPCv2, this block driver, enabling GPU DT node, I'm also working on the DSIM patch you posted. I was hoping to test them all together and reply to the various threads with tested-by. I also want to get my device tree stuff ready on the beacon boards so when everything lands, I can post DTS updates to enable the LCDIF, DSI, and the HDMI bridge.

Quoted 4 lines

If you have a repo somewhere that has all these combined, I can just work on the final layer to enable the device tree plumbing on my board. I just need the imx8mm.dtsi changes for this, DSIM, and the LCDIF so I can finish the task.

On that note, I also have a i.MX8M Nano board which is similar to my 8MM. If I understood the 8MM clock block driver better, I hope to adapt your changes for the Nano too. Once the GPCv2 driver is accepted, I was also going to look at updating it to support the Nano as well which also has the same DSIM and LCDIF as the 8MM as well and a better GPU than the Mini but lacking the VPU.

Quoted 7 lines
Quoted 2 lines

I don't have a branch, but I sent you the collected patches off-list.

I would also be interested in the patch collection for BLK_CTL, DSIM, etc. Marek, would you mind sending me those, too?

Adam, did you already set up a branch and do some tests with the full stack?

Frieder,

I have been monitoring some of the activity on the BLK_CTL. It seems like there is some disagreement on how to connect the power domain controller with the BLK_CTL. Someone reported that it causes a hang on the 8MP, so until that gets resolved, I doubt we'll be able to use the display system. Some of the DSIM changes happening are being pushed back for further changes, so it seems like having the full integration might be a while.

adam

Quoted 3 lines

Thanks,
Frieder

#15 lore raw

Hi,

Quoted 59 lines · Adam Ford
Quoted 47 lines · Frieder Schrempf

Hi,

Quoted 43 lines · Marek Vasut
Quoted 41 lines · Adam Ford
Quoted 29 lines · Adam Ford
Quoted 11 lines · Marek Vasut
Quoted 3 lines · Adam Ford
Quoted 2 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

[...]

Quoted 5 lines
Quoted 4 lines
---
   Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
   1 file changed, 1 insertion(+)
Quoted 8 lines
Quoted 3 lines

Is there a DTSI change part of this patch? I was going to try to test it, but I am not seeing a change to the imx8mm.dtsi, and I am not sure where to put the node.

There are in fact quite a few other pieces you need to have in place, this patchset in itself is not particularly useful, it is just infra for the LCDIF and MIPI DSIM block control. You might want to wait until they all land in next and test that result.

I have several patches in place, the GPCv2, this block driver, enabling GPU DT node, I'm also working on the DSIM patch you posted. I was hoping to test them all together and reply to the various threads with tested-by. I also want to get my device tree stuff ready on the beacon boards so when everything lands, I can post DTS updates to enable the LCDIF, DSI, and the HDMI bridge.

Quoted 4 lines

If you have a repo somewhere that has all these combined, I can just work on the final layer to enable the device tree plumbing on my board. I just need the imx8mm.dtsi changes for this, DSIM, and the LCDIF so I can finish the task.

On that note, I also have a i.MX8M Nano board which is similar to my 8MM. If I understood the 8MM clock block driver better, I hope to adapt your changes for the Nano too. Once the GPCv2 driver is accepted, I was also going to look at updating it to support the Nano as well which also has the same DSIM and LCDIF as the 8MM as well and a better GPU than the Mini but lacking the VPU.

Quoted 7 lines
Quoted 2 lines

I don't have a branch, but I sent you the collected patches off-list.

I would also be interested in the patch collection for BLK_CTL, DSIM, etc. Marek, would you mind sending me those, too?

Adam, did you already set up a branch and do some tests with the full stack?

Frieder,

Quoted 7 lines

I have been monitoring some of the activity on the BLK_CTL. It seems like there is some disagreement on how to connect the power domain controller with the BLK_CTL. Someone reported that it causes a hang on the 8MP, so until that gets resolved, I doubt we'll be able to use the display system. Some of the DSIM changes happening are being pushed back for further changes, so it seems like having the full integration might be a while.

I have pulled all the latest patches, including Marek's off-list patches together in one branch based on v5.10-rc7 [1] if anyone is interested.

I added some fixes on top, that I needed to get my display behind another Toshiba DSI-DPI bridge working. Those are probably not upstreamable at all and need further investigation.

I'm hoping to reply to the individual threads for more feedback. I see that there are some blocking issues, but we hopefully get them resolved somehow.

Thanks
Frieder

[1] https://github.com/fschrempf/linux/commits/v5.10-mx8mm-graphics

#16 lore raw

'On Thu, Dec 10, 2020 at 7:15 AM Frieder Schrempf

Quoted 78 lines · <[email protected]> wrote:

Hi,

Quoted 59 lines · Adam Ford
Quoted 47 lines · Frieder Schrempf

Hi,

Quoted 43 lines · Marek Vasut
Quoted 41 lines · Adam Ford
Quoted 29 lines · Adam Ford
Quoted 11 lines · Marek Vasut
Quoted 3 lines · Adam Ford
Quoted 2 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

[...]

Quoted 5 lines
Quoted 4 lines
---
   Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
   1 file changed, 1 insertion(+)
Quoted 8 lines
Quoted 3 lines

Is there a DTSI change part of this patch? I was going to try to test it, but I am not seeing a change to the imx8mm.dtsi, and I am not sure where to put the node.

There are in fact quite a few other pieces you need to have in place, this patchset in itself is not particularly useful, it is just infra for the LCDIF and MIPI DSIM block control. You might want to wait until they all land in next and test that result.

I have several patches in place, the GPCv2, this block driver, enabling GPU DT node, I'm also working on the DSIM patch you posted. I was hoping to test them all together and reply to the various threads with tested-by. I also want to get my device tree stuff ready on the beacon boards so when everything lands, I can post DTS updates to enable the LCDIF, DSI, and the HDMI bridge.

Quoted 4 lines

If you have a repo somewhere that has all these combined, I can just work on the final layer to enable the device tree plumbing on my board. I just need the imx8mm.dtsi changes for this, DSIM, and the LCDIF so I can finish the task.

On that note, I also have a i.MX8M Nano board which is similar to my 8MM. If I understood the 8MM clock block driver better, I hope to adapt your changes for the Nano too. Once the GPCv2 driver is accepted, I was also going to look at updating it to support the Nano as well which also has the same DSIM and LCDIF as the 8MM as well and a better GPU than the Mini but lacking the VPU.

Quoted 7 lines
Quoted 2 lines

I don't have a branch, but I sent you the collected patches off-list.

I would also be interested in the patch collection for BLK_CTL, DSIM, etc. Marek, would you mind sending me those, too?

Adam, did you already set up a branch and do some tests with the full stack?

Frieder,

Quoted 7 lines

I have been monitoring some of the activity on the BLK_CTL. It seems like there is some disagreement on how to connect the power domain controller with the BLK_CTL. Someone reported that it causes a hang on the 8MP, so until that gets resolved, I doubt we'll be able to use the display system. Some of the DSIM changes happening are being pushed back for further changes, so it seems like having the full integration might be a while.

I have pulled all the latest patches, including Marek's off-list patches together in one branch based on v5.10-rc7 [1] if anyone is interested.

I added some fixes on top, that I needed to get my display behind another Toshiba DSI-DPI bridge working. Those are probably not upstreamable at all and need further investigation.

Quoted 9 lines

I'm hoping to reply to the individual threads for more feedback. I see that there are some blocking issues, but we hopefully get them resolved somehow.

Thanks
Frieder

[1] https://github.com/fschrempf/linux/commits/v5.10-mx8mm-graphics

Frieder,

Thanks for sharing your repo as it's getting hard to track these patchsets (gpc/blk-ctl/power-domain/exynos/dsim). I'm also working on display support for IMX8MM and in my case I'm trying to connect to a RaspberryPi 7in display which I see Marek has been doing some work on to split out drivers/gpu/drm/panel/panel-raspberrypi-touchscreen.c to separate bridge, regulator/backlight, and simple-panel driver.

Marek,

Thanks for your recent work on splitting out the rpi display driver so that it can be bound via device-tree. I have found that I need to move the tc358762_init to enable vs pre-enable when using it with the in-progress samsung-dsim driver else the driver fails writes due to not being enabled yet:

(quoted diff) +0 -0

b/drivers/gpu/drm/bridge/tc358762.c
index 1bfdfc6..0d88e61 100644

drivers/gpu/drm/bridge/tc358762.c +9 -2
@@ -153,11 +153,17 @@ static void tc358762_pre_enable(struct drm_bridge *bridge)
        if (ret < 0)
                dev_err(ctx->dev, "error enabling regulators (%d)\n", ret);
 
+       ctx->pre_enabled = true;
+}
+
+static void tc358762_enable(struct drm_bridge *bridge)
+{
+       struct tc358762 *ctx = bridge_to_tc358762(bridge);
+       int ret;
+
        ret = tc358762_init(ctx);
        if (ret < 0)
                dev_err(ctx->dev, "error initializing bridge (%d)\n", ret);
-
-       ctx->pre_enabled = true;
 }
 
 static int tc358762_attach(struct drm_bridge *bridge,
@@ -172,6 +178,7 @@ static int tc358762_attach(struct drm_bridge *bridge,
 static const struct drm_bridge_funcs tc358762_bridge_funcs = {
        .post_disable = tc358762_post_disable,
        .pre_enable = tc358762_pre_enable,
+       .enable = tc358762_enable,
        .attach = tc358762_attach,
 };
 

Frieder, I did find that your "drm/exynos: Fix PLL PMS offset for P value bitfield" patch breaks the samsung_dsim_host_transfer for me with the tc358762 bridge in the rpi panel. If I have that patch I get a timeout on the transfer with some added debugging:

[    4.386387] tc358762_write 0x0210=0x00000003 0
[    4.387031] samsung_dsim_host_transfer ret: 0
[    4.387038] tc358762_write 0x0164=0x00000005 0
[    4.387375] samsung_dsim_host_transfer ret: 0
[    4.387379] tc358762_write 0x0168=0x00000005 0
[    4.387409] samsung_dsim_host_transfer ret: 0
[    4.387413] tc358762_write 0x0144=0x00000000 0
[    4.387741] samsung_dsim_host_transfer ret: 0
[    4.387745] tc358762_write 0x0148=0x00000000 0
[    4.387773] samsung_dsim_host_transfer ret: 0
[    4.387777] tc358762_write 0x0114=0x00000003 0
[    4.387804] samsung_dsim_host_transfer ret: 0
[    4.387808] tc358762_write 0x0450=0x00000000 0
[    4.387834] samsung_dsim_host_transfer ret: 0
[    4.387838] tc358762_write 0x0420=0x00100150 0
[    4.388168] samsung_dsim_host_transfer ret: 0
[    4.388172] tc358762_write 0x0464=0x0000040f 0
[    4.388200] samsung_dsim_host_transfer ret: 0
[    4.493346] tc358762_write 0x0104=0x00000001 0
[    5.509341] imx-dsim-dsi 32e10000.mipi_dsi: xfer timed out: 29 06
00 00 04 01 01 00 00 00
[    5.509345] samsung_dsim_host_transfer ret: -110
[    5.509348] tc358762_write mipi_dsi_generic_write failed err=-110
[    5.509352] tc358762_write 0x0204=0x00000001 -110
[    5.617336] tc358762_init failed err=-110
[    5.617344] tc358762 32e10000.mipi_dsi.0: error initializing bridge (-110)

Here is your patch which causes this issue for me:

(quoted diff) +0 -0

b/drivers/gpu/drm/bridge/samsung-dsim.c
index cb1ec3c..fc7c1d0 100644

drivers/gpu/drm/bridge/samsung-dsim.c +1 -1
@@ -174,7 +174,7 @@
 /* DSIM_PLLCTRL */
 #define DSIM_FREQ_BAND(x)              ((x) << 24)
 #define DSIM_PLL_EN                    (1 << 23)
-#define DSIM_PLL_P(x)                  ((x) << 13)
+#define DSIM_PLL_P(x)                  ((x) << 14)
 #define DSIM_PLL_M(x)                  ((x) << 4)
 #define DSIM_PLL_S(x)                  ((x) << 1)
 

I'm not very knowledgeable about MIPI DSI and find it strange that several writes in tc35872_init succeed until the failing writes:

        tc358762_write(ctx, PPI_STARTPPI, PPI_START_FUNCTION);
        tc358762_write(ctx, DSI_STARTDSI, DSI_RX_START);

For what its worth I've backported Marek's rpi backlight/reglator and simple-pannel driver to the NXP imx_5.4.47_2.2.0 kernel and do not see any MIPI DSI write failure there, although I have the same behavior of the display not showing anything.

Marek, are you using the rpi panel with IMX8MM? While I now have the drivers probing without error and have a functional backlight, regulator I see nothing on the display.

here is my dt fragment for my IMX8MM:

/ {
        panel {
                compatible = "powertip,ph800480t013-idf02";
                power-supply = <&attiny>;
                backlight = <&attiny>;
                port {
                        panel_in: endpointpanelin {
                                remote-endpoint = <&bridge_out>;
                        };
                };
        };
};
&i2c3 {
        edt-ft5x06@38 {
                compatible = "edt,edt-ft5x06";
                reg = <0x38>;
                pinctrl-0 = <&pinctrl_touchscreen>;
                interrupt-parent = <&gpio1>;
                interrupts = <7 IRQ_TYPE_EDGE_FALLING>;
                vcc-supply = <&attiny>;
                invert;
                screen-x = <800>;
                screen-y = <480>;
        };

        attiny: regulator@45 {
                compatible = "raspberrypi,7inch-touchscreen-panel-regulator";
                reg = <0x45>;
        };
};
&mipi_dsi {
        #address-cells = <1>;
        #size-cells = <0>;
        status = "okay";

        bridge@0 {
                compatible = "toshiba,tc358762";
                reg = <0>;
                vddc-supply = <&attiny>;
                #address-cells = <1>;
                #size-cells = <0>;
                status = "okay";

                ports {
                        #address-cells = <1>;
                        #size-cells = <0>;

                        port@0 {
                                reg = <0>;
                                bridge_in: endpoint {
                                        remote-endpoint = <&dsi_out>;
                                };
                        };

                        port@1 {
                                reg = <1>;
                                bridge_out: endpoint {
                                        remote-endpoint = <&panel_in>;
                                };
                        };
                };
        };

        ports {
                port@1 {
                        reg = <1>;
                        dsi_out: endpoint {
                                remote-endpoint = <&bridge_in>;
                        };
                };
        };
};

And after booting I have:
/sys/class/backlight/7inch-touchscreen-panel-bl/ - backlight controller /sys/class/regulator/regulator.9/ - tc358762-power /sys/class/input/input1/ /dev/input/event1 - generic ft5x06 (79) /sys/class/drm/card0
/sys/class/drm/card0-DPI-1
/sys/class/graphics/fb0 - mxsfb_drv.c

I'm using 'video=DPI-1:800x480@65M' in my kernel cmdline although I don't think thats needed for fb display.

# fbset -i

mode "800x480"

    geometry 800 480 800 480 32
    timings 0 0 0 0 0 0 0
    accel true
    rgba 8/16,8/8,8/0,0/0

endmode

Frame buffer device information:

    Name        : mxsfb-drmdrmfb
    Address     : 0
    Size        : 1536000
    Type        : PACKED PIXELS
    Visual      : TRUECOLOR
    XPanStep    : 1
    YPanStep    : 1
    YWrapStep   : 0
    LineLength  : 3200

Accelerator : No

# gst-launch-1.0 videotestsrc ! fbdevsink  # just shows a blank (but
backlit) display

Any idea what could be going wrong here?

Also Marek do you know why the edt-ft5x06 driver never seems to assert it's interrupt? I haven't gotten that working even though I've wired the INT pin from the display to a DIO on the IMX8MM.

Best Regards,

Tim

#17 lore raw

Hi Tim,

Quoted 90 lines · Tim Harvey

'On Thu, Dec 10, 2020 at 7:15 AM Frieder Schrempf

Quoted 78 lines · <[email protected]> wrote:

Hi,

Quoted 59 lines · Adam Ford
Quoted 47 lines · Frieder Schrempf

Hi,

Quoted 43 lines · Marek Vasut
Quoted 41 lines · Adam Ford
Quoted 29 lines · Adam Ford
Quoted 11 lines · Marek Vasut
Quoted 3 lines · Adam Ford
Quoted 2 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

[...]

Quoted 5 lines
Quoted 4 lines
---
    Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
    1 file changed, 1 insertion(+)
Quoted 8 lines
Quoted 3 lines

Is there a DTSI change part of this patch? I was going to try to test it, but I am not seeing a change to the imx8mm.dtsi, and I am not sure where to put the node.

There are in fact quite a few other pieces you need to have in place, this patchset in itself is not particularly useful, it is just infra for the LCDIF and MIPI DSIM block control. You might want to wait until they all land in next and test that result.

I have several patches in place, the GPCv2, this block driver, enabling GPU DT node, I'm also working on the DSIM patch you posted. I was hoping to test them all together and reply to the various threads with tested-by. I also want to get my device tree stuff ready on the beacon boards so when everything lands, I can post DTS updates to enable the LCDIF, DSI, and the HDMI bridge.

Quoted 4 lines

If you have a repo somewhere that has all these combined, I can just work on the final layer to enable the device tree plumbing on my board. I just need the imx8mm.dtsi changes for this, DSIM, and the LCDIF so I can finish the task.

On that note, I also have a i.MX8M Nano board which is similar to my 8MM. If I understood the 8MM clock block driver better, I hope to adapt your changes for the Nano too. Once the GPCv2 driver is accepted, I was also going to look at updating it to support the Nano as well which also has the same DSIM and LCDIF as the 8MM as well and a better GPU than the Mini but lacking the VPU.

Quoted 7 lines
Quoted 2 lines

I don't have a branch, but I sent you the collected patches off-list.

I would also be interested in the patch collection for BLK_CTL, DSIM, etc. Marek, would you mind sending me those, too?

Adam, did you already set up a branch and do some tests with the full stack?

Frieder,

Quoted 7 lines

I have been monitoring some of the activity on the BLK_CTL. It seems like there is some disagreement on how to connect the power domain controller with the BLK_CTL. Someone reported that it causes a hang on the 8MP, so until that gets resolved, I doubt we'll be able to use the display system. Some of the DSIM changes happening are being pushed back for further changes, so it seems like having the full integration might be a while.

I have pulled all the latest patches, including Marek's off-list patches together in one branch based on v5.10-rc7 [1] if anyone is interested.

I added some fixes on top, that I needed to get my display behind another Toshiba DSI-DPI bridge working. Those are probably not upstreamable at all and need further investigation.

Quoted 9 lines

I'm hoping to reply to the individual threads for more feedback. I see that there are some blocking issues, but we hopefully get them resolved somehow.

Thanks
Frieder

[1] https://eur04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Ffschrempf%2Flinux%2Fcommits%2Fv5.10-mx8mm-graphics&amp;data=04%7C01%7Cfrieder.schrempf%40kontron.de%7C4b06e39a1030405300be08d8a20913f6%7C8c9d3c973fd941c8a2b1646f3942daf1%7C0%7C0%7C637437507175831939%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=n1Y4HwamMfPukFuVEurgt7KOM9SMLkercFhgFMuE6ro%3D&amp;reserved=0

Quoted 8 lines

Frieder,

Thanks for sharing your repo as it's getting hard to track these patchsets (gpc/blk-ctl/power-domain/exynos/dsim). I'm also working on display support for IMX8MM and in my case I'm trying to connect to a RaspberryPi 7in display which I see Marek has been doing some work on to split out drivers/gpu/drm/panel/panel-raspberrypi-touchscreen.c to separate bridge, regulator/backlight, and simple-panel driver.

thanks for the feedback and good to know of other people caring about upstream support.

Quoted 79 lines

Marek,

Thanks for your recent work on splitting out the rpi display driver so that it can be bound via device-tree. I have found that I need to move the tc358762_init to enable vs pre-enable when using it with the in-progress samsung-dsim driver else the driver fails writes due to not being enabled yet:

drivers/gpu/drm/bridge/tc358762.c +9 -2
index 1bfdfc6..0d88e61 100644
@@ -153,11 +153,17 @@ static void tc358762_pre_enable(struct drm_bridge *bridge)
         if (ret < 0)
                 dev_err(ctx->dev, "error enabling regulators (%d)\n", ret);
 
+       ctx->pre_enabled = true;
+}
+
+static void tc358762_enable(struct drm_bridge *bridge)
+{
+       struct tc358762 *ctx = bridge_to_tc358762(bridge);
+       int ret;
+
         ret = tc358762_init(ctx);
         if (ret < 0)
                 dev_err(ctx->dev, "error initializing bridge (%d)\n", ret);
-
-       ctx->pre_enabled = true;
  }
 
  static int tc358762_attach(struct drm_bridge *bridge,
@@ -172,6 +178,7 @@ static int tc358762_attach(struct drm_bridge *bridge,
  static const struct drm_bridge_funcs tc358762_bridge_funcs = {
         .post_disable = tc358762_post_disable,
         .pre_enable = tc358762_pre_enable,
+       .enable = tc358762_enable,
         .attach = tc358762_attach,
  };
 

Frieder, I did find that your "drm/exynos: Fix PLL PMS offset for P value bitfield" patch breaks the samsung_dsim_host_transfer for me with the tc358762 bridge in the rpi panel. If I have that patch I get a timeout on the transfer with some added debugging:

[    4.386387] tc358762_write 0x0210=0x00000003 0
[    4.387031] samsung_dsim_host_transfer ret: 0
[    4.387038] tc358762_write 0x0164=0x00000005 0
[    4.387375] samsung_dsim_host_transfer ret: 0
[    4.387379] tc358762_write 0x0168=0x00000005 0
[    4.387409] samsung_dsim_host_transfer ret: 0
[    4.387413] tc358762_write 0x0144=0x00000000 0
[    4.387741] samsung_dsim_host_transfer ret: 0
[    4.387745] tc358762_write 0x0148=0x00000000 0
[    4.387773] samsung_dsim_host_transfer ret: 0
[    4.387777] tc358762_write 0x0114=0x00000003 0
[    4.387804] samsung_dsim_host_transfer ret: 0
[    4.387808] tc358762_write 0x0450=0x00000000 0
[    4.387834] samsung_dsim_host_transfer ret: 0
[    4.387838] tc358762_write 0x0420=0x00100150 0
[    4.388168] samsung_dsim_host_transfer ret: 0
[    4.388172] tc358762_write 0x0464=0x0000040f 0
[    4.388200] samsung_dsim_host_transfer ret: 0
[    4.493346] tc358762_write 0x0104=0x00000001 0
[    5.509341] imx-dsim-dsi 32e10000.mipi_dsi: xfer timed out: 29 06
00 00 04 01 01 00 00 00
[    5.509345] samsung_dsim_host_transfer ret: -110
[    5.509348] tc358762_write mipi_dsi_generic_write failed err=-110
[    5.509352] tc358762_write 0x0204=0x00000001 -110
[    5.617336] tc358762_init failed err=-110
[    5.617344] tc358762 32e10000.mipi_dsi.0: error initializing bridge (-110)

Here is your patch which causes this issue for me:

drivers/gpu/drm/bridge/samsung-dsim.c +0 -0
index cb1ec3c..fc7c1d0 100644

drivers/gpu/drm/bridge/samsung-dsim.c line 179

Quoted 7 lines
drivers/gpu/drm/bridge/samsung-dsim.c +1 -1
@@ -174,7 +174,7 @@
  /* DSIM_PLLCTRL */
  #define DSIM_FREQ_BAND(x)              ((x) << 24)
  #define DSIM_PLL_EN                    (1 << 23)
-#define DSIM_PLL_P(x)                  ((x) << 13)
+#define DSIM_PLL_P(x)                  ((x) << 14)
  #define DSIM_PLL_M(x)                  ((x) << 4)
  #define DSIM_PLL_S(x)                  ((x) << 1)

As I already mentioned in the commit message of this change, I have no idea how the "correct" fix should look like or if there even is anything to fix here at all. It's just what I needed to get my setup working and I found it really odd that the NXP vendor implementation differs from the upstream Exynos driver in this place.

I have some other hardware setups with different bridges (LVDS/HDMI) behind the DSI and if I find some time, I will try them and see if they behave differently. Unfortunately I don't have any hardware to connect the RPi display to the i.MX8MM to test your setup.

Best regards
Frieder

Quoted 14 lines

I'm not very knowledgeable about MIPI DSI and find it strange that several writes in tc35872_init succeed until the failing writes:

         tc358762_write(ctx, PPI_STARTPPI, PPI_START_FUNCTION);
         tc358762_write(ctx, DSI_STARTDSI, DSI_RX_START);

For what its worth I've backported Marek's rpi backlight/reglator and simple-pannel driver to the NXP imx_5.4.47_2.2.0 kernel and do not see any MIPI DSI write failure there, although I have the same behavior of the display not showing anything.

Marek, are you using the rpi panel with IMX8MM? While I now have the drivers probing without error and have a functional backlight, regulator I see nothing on the display.

#18 lore raw
In reply to Marek Vasut
Quoted 2 lines · Marek Vasut

Add the i.MX8MM BLK_CTL compatible string to the list.

It seems that NXP has updated the TRM to Rev 3 for the i.MX8M. For what it's worth, section 13.2 calls this DISPLAY_BLK_CTRL

They better document the missing registers.

adam

Quoted 35 lines
Cc Abel Vesa <[email protected]>
Cc Dong Aisheng <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc Guido Günther <[email protected]>
Cc Lucas Stach <[email protected]>
Cc Rob Herring <[email protected]>
Cc Shawn Guo <[email protected]>
Cc NXP Linux Team <[email protected]>
---
 Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml | 1 +
 1 file changed, 1 insertion(+)
Documentation/devicetree/bindings/clock/fsl,imx-blk-ctl.yaml +1 -1
index 5e9eb402b9b6..346429f49093 100644
@@ -20,6 +20,7 @@ properties:
   compatible:
     items:
       - enum:
+         - fsl,imx8mm-dispmix-blk-ctl
          - fsl,imx8mp-audio-blk-ctl
          - fsl,imx8mp-hdmi-blk-ctl
          - fsl,imx8mp-media-blk-ctl
--

2.28.0

_______________________________________________
linux-arm-kernel mailing list [email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel