Skip to contentSkip to conversation

arm: dts: rockchip: rk3399: usb: ehci: Fix EHCI probe in rk3399 to access peripherals by USB 2.

PATCH Superseded Xavier Drudis Ferran 2 participants 4 messages Jul 1, 2022 – Aug 31, 2022 Arm (32-bit)

v1 of 7 Superseded Latest is v7 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 arm: dts: rockchip: rk3399: usb: ehci: Fix EHCI probe in rk3399 to access peripherals by USB 2.

arch/arm/dts/rk3399.dtsi has a node

  usb_host0_ehci: usb@fe380000 {
       compatible = "generic-ehci";

with clocks:

       clocks = <&cru HCLK_HOST0>, <&cru HCLK_HOST0_ARB>,
                <&u2phy0>;

The first 2 refer to nodes with class UCLASS_CLK, but &u2phy0 has class UCLASS_PHY.

  u2phy0: usb2phy@e450 {
       compatible = "rockchip,rk3399-usb2phy";

Since clk_get_bulk() only looks for devices with UCLASS_CLK, it fails with -ENODEV and then ehci_usb_probe() aborts.

The consequence is peripherals connected to a USB 2 port (e.g. in a Rock Pi 4 the white port, nearer the edge) not being detected. They're detected if CONFIG_USB_OHCI_GENERIC is selected in Kconfig, because ohci_usb_probe() does not abort when one clk_get_by_index() fails, but then they work in USB 1 mode,.

rk3399.dtsi comes from linux and the u2phy0 was added[1] to the clock list in:

commit b5d1c57299734f5b54035ef2e61706b83041f20c
Author: William wu <[email protected]>
Date: Wed Dec 21 18:41:05 2016 +0800

arm64: dts: rockchip: add u2phy clock for ehci and ohci of rk3399

We found that the suspend process was blocked when it run into ehci/ohci module due to clk-480m of usb2-phy was disabled. [...]

Suspend concerns don't apply to U-Boot, and the problem with U-Boot failing to probe EHCI doesn't apply to linux, because in linux rockchip_usb2phy_clk480m_register makes u2phy0 a proper clock provider when called by rockchip_usb2phy_probe().

So I can think of a few alternative solutions:

1- Change ehci_usb_probe() to make it more similar to
ohci_usb_probe(), and survive failure to get one clock. Looks a little harder, and I don't know whether it could break something if it ignored a clock that was important for something else than suspend.

2- Change rk3399.dtsi effecttively reverting the linux commit
b5d1c57299734f5b54035ef2e61706b83041f20c. This dealigns the .dtsi from linux and seems fragile at the next synchronisation.

3- Change the clock list in rk3399-u-boot.dtsi or somewhere else.
This survives .dts* sync but may survive "too much" and miss some change from linux that we might want.

4- Enable CONFIG_USB_OHCI_GENERIC and use the ports in USB 1 mode.
This would need to be made for all boards using rk3399. In a simple test reading one file from USB storage it gave 769.5 KiB/s instead of 20.5 MiB/s with solution 2.

5- Trying to replicate linux and have usb2phy somehow provide a clk,
or have a separate clock device for usb2phy in addition to the phy device. I just can't seem to imagine how to achieve this with the U-Boot driver model, maybe because of my limited familiarity with it.

This patch is option 3. Looked like the simplest and most performant. But it might not be what's wanted, so comments wellcome.

Cc Simon Glass <[email protected]>
Cc Philipp Tomsich <[email protected]>
Cc Kever Yang <[email protected]>
Cc Lukasz Majewski <[email protected]>
Cc Sean Anderson <[email protected]>
Cc Marek Vasut <[email protected]>
---
 arch/arm/dts/rk3399-u-boot.dtsi | 8 ++++++++
 1 file changed, 8 insertions(+)
arch/arm/dts/rk3399-u-boot.dtsi +8 -0
index 716b9a433a..63c7b10405 100644
@@ -147,3 +147,11 @@
 &vopl {
 	u-boot,dm-pre-reloc;
 };
+
+&usb_host0_ehci {
+	clocks = <&cru HCLK_HOST0>, <&cru HCLK_HOST0_ARB>;
+};
+
+&usb_host1_ehci {
+	clocks = <&cru HCLK_HOST1>, <&cru HCLK_HOST1_ARB>;
+};
Signature · 1 line
2.20.1
#1 lore raw

I just checked this still works in next. No v2 needed (until feedback, because I've heard about nodes providing two devices and maybe you prefer that, so far I'd still see it too complex).

El Fri, Jul 01, 2022 at 08:59:59PM +0200, Xavier Drudis Ferran deia:

Quoted 107 lines

arch/arm/dts/rk3399.dtsi has a node

  usb_host0_ehci: usb@fe380000 {
       compatible = "generic-ehci";

with clocks:

       clocks = <&cru HCLK_HOST0>, <&cru HCLK_HOST0_ARB>,
                <&u2phy0>;

The first 2 refer to nodes with class UCLASS_CLK, but &u2phy0 has class UCLASS_PHY.

  u2phy0: usb2phy@e450 {
       compatible = "rockchip,rk3399-usb2phy";

Since clk_get_bulk() only looks for devices with UCLASS_CLK, it fails with -ENODEV and then ehci_usb_probe() aborts.

The consequence is peripherals connected to a USB 2 port (e.g. in a Rock Pi 4 the white port, nearer the edge) not being detected. They're detected if CONFIG_USB_OHCI_GENERIC is selected in Kconfig, because ohci_usb_probe() does not abort when one clk_get_by_index() fails, but then they work in USB 1 mode,.

rk3399.dtsi comes from linux and the u2phy0 was added[1] to the clock list in:

commit b5d1c57299734f5b54035ef2e61706b83041f20c
Author: William wu <[email protected]>
Date: Wed Dec 21 18:41:05 2016 +0800

arm64: dts: rockchip: add u2phy clock for ehci and ohci of rk3399

We found that the suspend process was blocked when it run into ehci/ohci module due to clk-480m of usb2-phy was disabled. [...]

Suspend concerns don't apply to U-Boot, and the problem with U-Boot failing to probe EHCI doesn't apply to linux, because in linux rockchip_usb2phy_clk480m_register makes u2phy0 a proper clock provider when called by rockchip_usb2phy_probe().

So I can think of a few alternative solutions:

1- Change ehci_usb_probe() to make it more similar to
ohci_usb_probe(), and survive failure to get one clock. Looks a little harder, and I don't know whether it could break something if it ignored a clock that was important for something else than suspend.

2- Change rk3399.dtsi effecttively reverting the linux commit
b5d1c57299734f5b54035ef2e61706b83041f20c. This dealigns the .dtsi from linux and seems fragile at the next synchronisation.

3- Change the clock list in rk3399-u-boot.dtsi or somewhere else.
This survives .dts* sync but may survive "too much" and miss some change from linux that we might want.

4- Enable CONFIG_USB_OHCI_GENERIC and use the ports in USB 1 mode.
This would need to be made for all boards using rk3399. In a simple test reading one file from USB storage it gave 769.5 KiB/s instead of 20.5 MiB/s with solution 2.

5- Trying to replicate linux and have usb2phy somehow provide a clk,
or have a separate clock device for usb2phy in addition to the phy device. I just can't seem to imagine how to achieve this with the U-Boot driver model, maybe because of my limited familiarity with it.

This patch is option 3. Looked like the simplest and most performant. But it might not be what's wanted, so comments wellcome.

Cc Simon Glass <[email protected]>
Cc Philipp Tomsich <[email protected]>
Cc Kever Yang <[email protected]>
Cc Lukasz Majewski <[email protected]>
Cc Sean Anderson <[email protected]>
Cc Marek Vasut <[email protected]>
---
 arch/arm/dts/rk3399-u-boot.dtsi | 8 ++++++++
 1 file changed, 8 insertions(+)
arch/arm/dts/rk3399-u-boot.dtsi +8 -0
index 716b9a433a..63c7b10405 100644
@@ -147,3 +147,11 @@
 &vopl {
 	u-boot,dm-pre-reloc;
 };
+
+&usb_host0_ehci {
+	clocks = <&cru HCLK_HOST0>, <&cru HCLK_HOST0_ARB>;
+};
+
+&usb_host1_ehci {
+	clocks = <&cru HCLK_HOST1>, <&cru HCLK_HOST1_ARB>;
+};
Signature · 1 line
2.20.1
#2 lore raw
In reply to Xavier Drudis Ferran

Hi Xavier,

Quoted 64 lines · Xavier Drudis Ferran

arch/arm/dts/rk3399.dtsi has a node

   usb_host0_ehci: usb@fe380000 {
        compatible = "generic-ehci";

with clocks:

        clocks = <&cru HCLK_HOST0>, <&cru HCLK_HOST0_ARB>,
                 <&u2phy0>;

The first 2 refer to nodes with class UCLASS_CLK, but &u2phy0 has class UCLASS_PHY.

   u2phy0: usb2phy@e450 {
        compatible = "rockchip,rk3399-usb2phy";

Since clk_get_bulk() only looks for devices with UCLASS_CLK, it fails with -ENODEV and then ehci_usb_probe() aborts.

The consequence is peripherals connected to a USB 2 port (e.g. in a Rock Pi 4 the white port, nearer the edge) not being detected. They're detected if CONFIG_USB_OHCI_GENERIC is selected in Kconfig, because ohci_usb_probe() does not abort when one clk_get_by_index() fails, but then they work in USB 1 mode,.

rk3399.dtsi comes from linux and the u2phy0 was added[1] to the clock list in:

commit b5d1c57299734f5b54035ef2e61706b83041f20c
Author: William wu <[email protected]>
Date: Wed Dec 21 18:41:05 2016 +0800

arm64: dts: rockchip: add u2phy clock for ehci and ohci of rk3399

We found that the suspend process was blocked when it run into ehci/ohci module due to clk-480m of usb2-phy was disabled. [...]

Suspend concerns don't apply to U-Boot, and the problem with U-Boot failing to probe EHCI doesn't apply to linux, because in linux rockchip_usb2phy_clk480m_register makes u2phy0 a proper clock provider when called by rockchip_usb2phy_probe().

So I can think of a few alternative solutions:

1- Change ehci_usb_probe() to make it more similar to
ohci_usb_probe(), and survive failure to get one clock. Looks a little harder, and I don't know whether it could break something if it ignored a clock that was important for something else than suspend.

2- Change rk3399.dtsi effecttively reverting the linux commit
b5d1c57299734f5b54035ef2e61706b83041f20c. This dealigns the .dtsi from linux and seems fragile at the next synchronisation.

3- Change the clock list in rk3399-u-boot.dtsi or somewhere else.
This survives .dts* sync but may survive "too much" and miss some change from linux that we might want.

4- Enable CONFIG_USB_OHCI_GENERIC and use the ports in USB 1 mode.
This would need to be made for all boards using rk3399. In a simple test reading one file from USB storage it gave 769.5 KiB/s instead of 20.5 MiB/s with solution 2.

Quoted 8 lines

5- Trying to replicate linux and have usb2phy somehow provide a clk,
or have a separate clock device for usb2phy in addition to the phy device. I just can't seem to imagine how to achieve this with the U-Boot driver model, maybe because of my limited familiarity with it.

This patch is option 3. Looked like the simplest and most performant. But it might not be what's wanted, so comments wellcome.

The idea for dtsi now is to sync the kernel and U-Boot dts, and Linux distribution will use the dts from U-Boot. If we change the dts property in -u-boot.dts, it will also pass to kernel, which make kernel function not available. So we would like to sync the driver for kernel and u-boot to use the same dts nodes.

So we can try option 1, output warning instead of error return for clock get fail, since the clock driver for kernel and u-boot are totally different.

Thanks,
- Kever

Quoted 31 lines
Cc Simon Glass <[email protected]>
Cc Philipp Tomsich <[email protected]>
Cc Kever Yang <[email protected]>
Cc Lukasz Majewski <[email protected]>
Cc Sean Anderson <[email protected]>
Cc Marek Vasut <[email protected]>
---
  arch/arm/dts/rk3399-u-boot.dtsi | 8 ++++++++
  1 file changed, 8 insertions(+)
arch/arm/dts/rk3399-u-boot.dtsi +8 -0
index 716b9a433a..63c7b10405 100644
@@ -147,3 +147,11 @@
  &vopl {
  	u-boot,dm-pre-reloc;
  };
+
+&usb_host0_ehci {
+	clocks = <&cru HCLK_HOST0>, <&cru HCLK_HOST0_ARB>;
+};
+
+&usb_host1_ehci {
+	clocks = <&cru HCLK_HOST1>, <&cru HCLK_HOST1_ARB>;
+};
#3 lore raw

El Sat, Aug 27, 2022 at 11:20:17AM +0800, Kever Yang deia:

Quoted 8 lines

The idea for dtsi now is to sync the kernel and U-Boot dts, and Linux distribution will
use the dts from U-Boot. If we change the dts property in -u-boot.dts, it will also pass
to kernel, which make kernel function not available. So we would like to sync the driver
for kernel and u-boot to use the same dts nodes.

That's news to me. I was thinking all the time that the reason for having some dtsi? files with -u-boot suffix was that these were exclusive for u-boot and the others were shared with linux. If now the content of .*-u-boot.dtsi? gets passed to linux too, I no longer know why we separate the files. Is this for Rockchip or ARM64 or a general rule for all of U-Boot? Should I (or someone) try to change documentation or is that already done? I thought I read something different few months ago ? Maybe I misunderstood...

Quoted 4 lines

So we can try option 1, output warning instead of error return for clock get fail, since the clock
driver for kernel and u-boot are totally different.

Ok. I hope I find time to try that some day.

Thanks for looking at the patch.