Skip to contentSkip to conversation

arm64: dts: rockchip: Enable TYPE-C PD for ROC-RK3399-PC

PATCH In review Jagan Teki 5 participants 10 messages Jul 19, 2023 – Sep 22, 2023

Subsystems (3)
lore
Share & tools
lore

Jagan Teki

author
#0 lore raw

PATCH 1/1 arm64: dts: rockchip: Enable TYPE-C PD for ROC-RK3399-PC

The power supply circuit in ROC-RK3399-PC is

Power Supply (or PPS) => FUSB => MP8859

VUBS pin of FUSB and IN pin of MP8859 is supplied via TYPEC_IN.

The MP8859 operated with 2.8V to 22V supply voltage and typical applications this supply can be 12V.

This patch is trying to support the PD by changing the FUSB VBUS supply to 12V and tune the I2C7 timings from downstream kernel.

Tested with PD3.0 PPS with supply voltages of 12V/3A and 20V/5A.

Cc Corentin Labbe <[email protected]>
---
 .../boot/dts/rockchip/rk3399-roc-pc.dtsi      | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)
arch/arm64/boot/dts/rockchip/rk3399-roc-pc.dtsi +16 -3
index c32913df93c3..8963b3858eae 100644
@@ -6,6 +6,7 @@
 /dts-v1/;
 #include <dt-bindings/input/linux-event-codes.h>
 #include <dt-bindings/pwm/pwm.h>
+#include "dt-bindings/usb/pd.h"
 #include "rk3399.dtsi"
 #include "rk3399-opp.dtsi"
 
@@ -524,8 +525,9 @@ &i2c3 {
 };
 
 &i2c4 {
-	i2c-scl-rising-time-ns = <600>;
-	i2c-scl-falling-time-ns = <20>;
+	clock-frequency = <400000>;
+	i2c-scl-rising-time-ns = <345>;
+	i2c-scl-falling-time-ns = <11>;
 	status = "okay";
 
 	fusb1: usb-typec@22 {
@@ -552,8 +554,19 @@ fusb0: usb-typec@22 {
 		interrupts = <2 IRQ_TYPE_LEVEL_LOW>;
 		pinctrl-names = "default";
 		pinctrl-0 = <&fusb0_int>;
-		vbus-supply = <&vcc_vbus_typec0>;
+		vbus-supply = <&dc_12v>;
 		status = "okay";
+
+		connector {
+			compatible = "usb-c-connector";
+			data-role = "dual";
+			label = "USB-C";
+			power-role = "sink";
+			try-power-role = "sink";
+			op-sink-microwatt = <10000000>;
+			sink-pdos = <PDO_FIXED(5000, 3000, PDO_FIXED_USB_COMM)
+				     PDO_VAR(5000, 20000, 5000)>;
+		};
 	};
 
 	mp8859: regulator@66 {
Signature · 7 lines
2.25.1


_______________________________________________
Linux-rockchip mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-rockchip
#1 lore raw
Quoted 6 lines · Jagan Teki

The power supply circuit in ROC-RK3399-PC is

Power Supply (or PPS) => FUSB => MP8859

VUBS pin of FUSB and IN pin of MP8859 is supplied via TYPEC_IN.

Quoted 7 lines

The MP8859 operated with 2.8V to 22V supply voltage and typical applications this supply can be 12V.

This patch is trying to support the PD by changing the FUSB VBUS supply to 12V and tune the I2C7 timings from downstream kernel.

Tested with PD3.0 PPS with supply voltages of 12V/3A and 20V/5A.

Hi Jagan,

This series works fine with a "dumb" (no PD negotiation) 5.1V Raspberry Pi PSU.

It also works fine with a Dell 45W USB-C Laptop Power Supply (model AA45NM170) which provides 5V@3A,9V@3A,15V@3A,[email protected], where Linux master fails and just tells the USB-PD PSU to power-off.

It doesn't work with a Lenovo 65W PSU (model ADLX65YLC3D) which provides 5V@2A,9V@2A,15V@3A,[email protected], after negotiation the driver turns the PD PSU off and on again, resetting the board. So it'd be great to get this fixed, but it seems like the sink-pdos should already support this PSU ?

As a side note for full transparency, another issue is that with Linux master or with this patch applied, applying power to the DC_12V header without a USB-C PSU connected (e.g. powering from the POE expansion shield), boot hangs and dumps to an initramfs shell with:

  [    7.411798] random: crng init done
  [   12.568138] platform fe3a0000.usb: deferred probe pending
  [   12.568673] platform sdio-pwrseq: deferred probe pending
  [   12.569162] platform fe3e0000.usb: deferred probe pending
  [   12.569658] platform adc-keys: deferred probe pending
  [   12.570123] i2c 7-0022: deferred probe pending
  [   12.570533] i2c 4-0022: deferred probe pending
  [   12.570944] platform ff940000.hdmi: deferred probe pending
  [   12.571448] platform vcc3v0-sd: deferred probe pending
  [   12.572000] platform vcc1v8-s3: deferred probe pending
  [   12.572475] platform sys-12v: deferred probe pending
  [   12.572933] platform vcc3v3-sys: deferred probe pending
  [   12.573412] platform fe320000.mmc: deferred probe pending
  [   12.573907] platform vcca-0v9: deferred probe pending
  [   12.574371] platform vcc5v0-host-regulator: deferred probe pending
  [   12.574935] platform ff770000.syscon:usb2phy@e450: deferred probe pending
  [   12.575552] platform vcc-vbus-typec1: deferred probe pending
  [   12.576090] platform fe300000.ethernet: deferred probe pending
  [   12.576623] platform vcc-sys: deferred probe pending
  [   12.577080] platform ff770000.syscon:usb2phy@e460: deferred probe pending
  [   12.577697] platform ff320000.syscon:io-domains: deferred probe pending
  [   12.578298] platform ff770000.syscon:io-domains: deferred probe pending
  [   12.578901] platform fe800000.usb: deferred probe pending
  [   12.579395] platform fe900000.usb: deferred probe pending
  [   12.579904] platform vdd-log: deferred probe pending
  [   12.580362] i2c 0-001b: deferred probe pending
  [   12.580772] i2c 0-0040: deferred probe pending
  [   12.581182] platform cpufreq-dt: deferred probe pending
  [   12.581663] i2c 0-0041: deferred probe pending

Thanks,

Chris

Quoted 52 lines
Cc Corentin Labbe <[email protected]>

---
 .../boot/dts/rockchip/rk3399-roc-pc.dtsi      | 19 ++++++++++++++++---

 1 file changed, 16 insertions(+), 3 deletions(-)
arch/arm64/boot/dts/rockchip/rk3399-roc-pc.dtsi +1 -0
index c32913df93c3..8963b3858eae 100644
@@ -6,6 +6,7 @@
  /dts-v1/;
  #include <dt-bindings/input/linux-event-codes.h>
  #include <dt-bindings/pwm/pwm.h>
+#include "dt-bindings/usb/pd.h"
  #include "rk3399.dtsi"
  #include "rk3399-opp.dtsi"

arch/arm64/boot/dts/rockchip/rk3399-roc-pc.dtsi +0 -0
@@ -524,8 +525,9 @@ &i2c3 {
  };


 &i2c4 {
-       i2c-scl-rising-time-ns = <600>;
-       i2c-scl-falling-time-ns = <20>;
+       clock-frequency = <400000>;
+       i2c-scl-rising-time-ns = <345>;
+       i2c-scl-falling-time-ns = <11>;
        status = "okay";

        fusb1: usb-typec@22 {

arch/arm64/boot/dts/rockchip/rk3399-roc-pc.dtsi +12 -1
@@ -552,8 +554,19 @@ fusb0: usb-typec@22 {
                 interrupts = <2 IRQ_TYPE_LEVEL_LOW>;
                 pinctrl-names = "default";
                 pinctrl-0 = <&fusb0_int>;
-               vbus-supply = <&vcc_vbus_typec0>;
+               vbus-supply = <&dc_12v>;
                 status = "okay";
+
+               connector {
+                       compatible = "usb-c-connector";
+                       data-role = "dual";
+                       label = "USB-C";
+                       power-role = "sink";
+                       try-power-role = "sink";
+                       op-sink-microwatt = <10000000>;
+                       sink-pdos = <PDO_FIXED(5000, 3000, PDO_FIXED_USB_COMM)
+                                    PDO_VAR(5000, 20000, 5000)>;
+               };
         };


        mp8859: regulator@66 {

Signature · 19 lines
Christopher Obbard BEng (Hons) MIET
Senior Engineer

Collabora Ltd
Platinum Building, St John's Innovation Park, Cambridge CB4 0DS, UK
Registered in England & Wales no 5513718.
This message is intended for the use of only the person(s) ("intended
recipient") to whom it is addressed. 
It may contain information that is privileged and confidential.
Accordingly, any dissemination, distribution, copying or other use of
this message or any of its content by any person other than the
intended recipient may constitute a breach of civil or criminal law and
is strictly prohibited.


_______________________________________________
Linux-rockchip mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-rockchip
#2 lore raw
Quoted 14 lines · Christopher Obbard
Quoted 12 lines · Jagan Teki

The power supply circuit in ROC-RK3399-PC is

Power Supply (or PPS) => FUSB => MP8859

VUBS pin of FUSB and IN pin of MP8859 is supplied via TYPEC_IN.

The MP8859 operated with 2.8V to 22V supply voltage and typical applications this supply can be 12V.

This patch is trying to support the PD by changing the FUSB VBUS supply to 12V and tune the I2C7 timings from downstream kernel.

Quoted 7 lines
Quoted 1 line

Tested with PD3.0 PPS with supply voltages of 12V/3A and 20V/5A.

Hi Jagan,

This series works fine with a "dumb" (no PD negotiation) 5.1V Raspberry Pi PSU.

It also works fine with a Dell 45W USB-C Laptop Power Supply (model AA45NM170) which provides 5V@3A,9V@3A,15V@3A,[email protected], where Linux master fails and just tells the USB-PD PSU to power-off.

I think this depends on the recent Rockchip TCPM changes. FUSB302 has been a pain in the USB or else this platform would have launched a lot longer ago.

Quoted 3 lines

It doesn't work with a Lenovo 65W PSU (model ADLX65YLC3D) which provides 5V@2A,9V@2A,15V@3A,[email protected], after negotiation the driver turns the PD PSU off and on again, resetting the board. So it'd be great to get this fixed, but it seems like the sink-pdos should already support this PSU ?

Something to note about USB PD that I read in the TI PD 2.0 paper and re-spewing based on memory: during transitions, the sink should self-limit to 500mA. If that is the case, the transition behavior is up to the source.

Quoted 110 lines

As a side note for full transparency, another issue is that with Linux master or with this patch applied, applying power to the DC_12V header without a USB-C PSU connected (e.g. powering from the POE expansion shield), boot hangs and dumps to an initramfs shell with:

  [    7.411798] random: crng init done
  [   12.568138] platform fe3a0000.usb: deferred probe pending
  [   12.568673] platform sdio-pwrseq: deferred probe pending
  [   12.569162] platform fe3e0000.usb: deferred probe pending
  [   12.569658] platform adc-keys: deferred probe pending
  [   12.570123] i2c 7-0022: deferred probe pending
  [   12.570533] i2c 4-0022: deferred probe pending
  [   12.570944] platform ff940000.hdmi: deferred probe pending
  [   12.571448] platform vcc3v0-sd: deferred probe pending
  [   12.572000] platform vcc1v8-s3: deferred probe pending
  [   12.572475] platform sys-12v: deferred probe pending
  [   12.572933] platform vcc3v3-sys: deferred probe pending
  [   12.573412] platform fe320000.mmc: deferred probe pending
  [   12.573907] platform vcca-0v9: deferred probe pending
  [   12.574371] platform vcc5v0-host-regulator: deferred probe pending
  [   12.574935] platform ff770000.syscon:usb2phy@e450: deferred probe pending
  [   12.575552] platform vcc-vbus-typec1: deferred probe pending
  [   12.576090] platform fe300000.ethernet: deferred probe pending
  [   12.576623] platform vcc-sys: deferred probe pending
  [   12.577080] platform ff770000.syscon:usb2phy@e460: deferred probe pending
  [   12.577697] platform ff320000.syscon:io-domains: deferred probe pending
  [   12.578298] platform ff770000.syscon:io-domains: deferred probe pending
  [   12.578901] platform fe800000.usb: deferred probe pending
  [   12.579395] platform fe900000.usb: deferred probe pending
  [   12.579904] platform vdd-log: deferred probe pending
  [   12.580362] i2c 0-001b: deferred probe pending
  [   12.580772] i2c 0-0040: deferred probe pending
  [   12.581182] platform cpufreq-dt: deferred probe pending
  [   12.581663] i2c 0-0041: deferred probe pending

Thanks,

Chris

Quoted 42 lines
Cc Corentin Labbe <[email protected]>
---
 .../boot/dts/rockchip/rk3399-roc-pc.dtsi      | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)
arch/arm64/boot/dts/rockchip/rk3399-roc-pc.dtsi +9 -3
index c32913df93c3..8963b3858eae 100644
@@ -6,6 +6,7 @@
 /dts-v1/;
 #include <dt-bindings/input/linux-event-codes.h>
 #include <dt-bindings/pwm/pwm.h>
+#include "dt-bindings/usb/pd.h"
 #include "rk3399.dtsi"
 #include "rk3399-opp.dtsi"
 
@@ -524,8 +525,9 @@ &i2c3 {
 };
 
 &i2c4 {
-       i2c-scl-rising-time-ns = <600>;
-       i2c-scl-falling-time-ns = <20>;
+       clock-frequency = <400000>;
+       i2c-scl-rising-time-ns = <345>;
+       i2c-scl-falling-time-ns = <11>;
        status = "okay";
 
        fusb1: usb-typec@22 {
@@ -552,8 +554,19 @@ fusb0: usb-typec@22 {
                interrupts = <2 IRQ_TYPE_LEVEL_LOW>;
                pinctrl-names = "default";
                pinctrl-0 = <&fusb0_int>;
-               vbus-supply = <&vcc_vbus_typec0>;
+               vbus-supply = <&dc_12v>;
                status = "okay";
+
+               connector {
+                       compatible = "usb-c-connector";
+                       data-role = "dual";

arch/arm64/boot/dts/rockchip/rk3399-roc-pc.dtsi fusb0: usb-typec@22 { · line 572

Quoted 10 lines
arch/arm64/boot/dts/rockchip/rk3399-roc-pc.dtsi +7 -0
+                       label = "USB-C";
+                       power-role = "sink";
+                       try-power-role = "sink";
+                       op-sink-microwatt = <10000000>;
+                       sink-pdos = <PDO_FIXED(5000, 3000, PDO_FIXED_USB_COMM)
+                                    PDO_VAR(5000, 20000, 5000)>;
+               };
        };
 
        mp8859: regulator@66 {

--
Christopher Obbard BEng (Hons) MIET
Senior Engineer

Collabora Ltd
Platinum Building, St John's Innovation Park, Cambridge CB4 0DS, UK Registered in England & Wales no 5513718.
This message is intended for the use of only the person(s) ("intended recipient") to whom it is addressed.
It may contain information that is privileged and confidential. Accordingly, any dissemination, distribution, copying or other use of this message or any of its content by any person other than the intended recipient may constitute a breach of civil or criminal law and is strictly prohibited.

Quoted 4 lines

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

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

#3 lore raw

Hi Da, Jagan,

Quoted 20 lines · Da Xue
Quoted 18 lines · Christopher Obbard
Quoted 6 lines · Jagan Teki

The power supply circuit in ROC-RK3399-PC is

Power Supply (or PPS) => FUSB => MP8859

VUBS pin of FUSB and IN pin of MP8859 is supplied via TYPEC_IN.

Quoted 7 lines

The MP8859 operated with 2.8V to 22V supply voltage and typical applications this supply can be 12V.

This patch is trying to support the PD by changing the FUSB VBUS supply to 12V and tune the I2C7 timings from downstream kernel.

Tested with PD3.0 PPS with supply voltages of 12V/3A and 20V/5A.

Hi Jagan,

Quoted 7 lines
Quoted 3 lines

This series works fine with a "dumb" (no PD negotiation) 5.1V Raspberry Pi PSU.

It also works fine with a Dell 45W USB-C Laptop Power Supply (model AA45NM170) which provides 5V@3A,9V@3A,15V@3A,[email protected], where Linux master fails and just tells the USB-PD PSU to power-off.

I think this depends on the recent Rockchip TCPM changes. FUSB302 has been a pain in the USB or else this platform would have launched a lot longer ago.

Sorry, I was testing this patch on top of next-20230724 which includes https://patchwork.kernel.org/project/linux-rockchip/list/?series=757752:

8be558dcffe69b078b34b1fa93b82acaf4ce4957 ("usb: typec: tcpm: add get max power support") 1e35f074399dece73d5df11847d4a0d7a6f49434 ("usb: typec: tcpm: fix cc role at port reset")

Can you check if I am missing any other patches ?

Thank you

Quoted 126 lines
Quoted 4 lines

It doesn't work with a Lenovo 65W PSU (model ADLX65YLC3D) which provides 5V@2A,9V@2A,15V@3A,[email protected], after negotiation the driver turns the PD PSU off and on again, resetting the board. So it'd be great to get this fixed, but it seems like the sink-pdos should already support this PSU ?

Something to note about USB PD that I read in the TI PD 2.0 paper and re-spewing based on memory: during transitions, the sink should self-limit to 500mA. If that is the case, the transition behavior is up to the source.

Quoted 115 lines

As a side note for full transparency, another issue is that with Linux master or with this patch applied, applying power to the DC_12V header without a USB-C PSU connected (e.g. powering from the POE
expansion shield), boot hangs and dumps to an initramfs shell with:

  [    7.411798] random: crng init done
  [   12.568138] platform fe3a0000.usb: deferred probe pending
  [   12.568673] platform sdio-pwrseq: deferred probe pending
  [   12.569162] platform fe3e0000.usb: deferred probe pending
  [   12.569658] platform adc-keys: deferred probe pending
  [   12.570123] i2c 7-0022: deferred probe pending
  [   12.570533] i2c 4-0022: deferred probe pending
  [   12.570944] platform ff940000.hdmi: deferred probe pending
  [   12.571448] platform vcc3v0-sd: deferred probe pending
  [   12.572000] platform vcc1v8-s3: deferred probe pending
  [   12.572475] platform sys-12v: deferred probe pending
  [   12.572933] platform vcc3v3-sys: deferred probe pending
  [   12.573412] platform fe320000.mmc: deferred probe pending
  [   12.573907] platform vcca-0v9: deferred probe pending
  [   12.574371] platform vcc5v0-host-regulator: deferred probe pending [   12.574935] platform ff770000.syscon:usb2phy@e450: deferred probe pending [   12.575552] platform vcc-vbus-typec1: deferred probe pending [   12.576090] platform fe300000.ethernet: deferred probe pending [   12.576623] platform vcc-sys: deferred probe pending
  [   12.577080] platform ff770000.syscon:usb2phy@e460: deferred probe pending [   12.577697] platform ff320000.syscon:io-domains: deferred probe pending [   12.578298] platform ff770000.syscon:io-domains: deferred probe pending [   12.578901] platform fe800000.usb: deferred probe pending
  [   12.579395] platform fe900000.usb: deferred probe pending
  [   12.579904] platform vdd-log: deferred probe pending
  [   12.580362] i2c 0-001b: deferred probe pending
  [   12.580772] i2c 0-0040: deferred probe pending
  [   12.581182] platform cpufreq-dt: deferred probe pending
  [   12.581663] i2c 0-0041: deferred probe pending

Thanks,

Chris

Quoted 42 lines
Cc Corentin Labbe <[email protected]>

---
 .../boot/dts/rockchip/rk3399-roc-pc.dtsi      | 19 ++++++++++++++++---

 1 file changed, 16 insertions(+), 3 deletions(-)
arch/arm64/boot/dts/rockchip/rk3399-roc-pc.dtsi +9 -3
index c32913df93c3..8963b3858eae 100644
@@ -6,6 +6,7 @@
  /dts-v1/;
  #include <dt-bindings/input/linux-event-codes.h>
  #include <dt-bindings/pwm/pwm.h>
+#include "dt-bindings/usb/pd.h"
  #include "rk3399.dtsi"
  #include "rk3399-opp.dtsi"
 
@@ -524,8 +525,9 @@ &i2c3 {
  };
 
  &i2c4 {
-       i2c-scl-rising-time-ns = <600>;
-       i2c-scl-falling-time-ns = <20>;
+       clock-frequency = <400000>;
+       i2c-scl-rising-time-ns = <345>;
+       i2c-scl-falling-time-ns = <11>;
         status = "okay";
 
         fusb1: usb-typec@22 {
@@ -552,8 +554,19 @@ fusb0: usb-typec@22 {
                 interrupts = <2 IRQ_TYPE_LEVEL_LOW>;
                 pinctrl-names = "default";
                 pinctrl-0 = <&fusb0_int>;
-               vbus-supply = <&vcc_vbus_typec0>;
+               vbus-supply = <&dc_12v>;
                 status = "okay";
+
+               connector {
+                       compatible = "usb-c-connector";
+                       data-role = "dual";

arch/arm64/boot/dts/rockchip/rk3399-roc-pc.dtsi fusb0: usb-typec@22 { · line 572

Quoted 10 lines
arch/arm64/boot/dts/rockchip/rk3399-roc-pc.dtsi +7 -0
+                       label = "USB-C";
+                       power-role = "sink";
+                       try-power-role = "sink";
+                       op-sink-microwatt = <10000000>;
+                       sink-pdos = <PDO_FIXED(5000, 3000, PDO_FIXED_USB_COMM)
+                                    PDO_VAR(5000, 20000, 5000)>;
+               };
         };
 
         mp8859: regulator@66 {

--
Christopher Obbard BEng (Hons) MIET
Senior Engineer

Collabora Ltd
Platinum Building, St John's Innovation Park, Cambridge CB4 0DS, UK Registered in England & Wales no 5513718.
This message is intended for the use of only the person(s) ("intended recipient") to whom it is addressed.
It may contain information that is privileged and confidential. Accordingly, any dissemination, distribution, copying or other use of this message or any of its content by any person other than the intended recipient may constitute a breach of civil or criminal law and is strictly prohibited.

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

maintainer
#4 lore raw

Am Dienstag, 25. Juli 2023, 10:02:21 CEST schrieb Christopher Obbard:

Quoted 31 lines

Hi Da, Jagan,

Quoted 20 lines · Da Xue
Quoted 18 lines · Christopher Obbard
Quoted 6 lines · Jagan Teki

The power supply circuit in ROC-RK3399-PC is

Power Supply (or PPS) => FUSB => MP8859

VUBS pin of FUSB and IN pin of MP8859 is supplied via TYPEC_IN.

Quoted 7 lines

The MP8859 operated with 2.8V to 22V supply voltage and typical applications this supply can be 12V.

This patch is trying to support the PD by changing the FUSB VBUS supply to 12V and tune the I2C7 timings from downstream kernel.

Tested with PD3.0 PPS with supply voltages of 12V/3A and 20V/5A.

Hi Jagan,

Quoted 7 lines
Quoted 3 lines

This series works fine with a "dumb" (no PD negotiation) 5.1V Raspberry Pi PSU.

It also works fine with a Dell 45W USB-C Laptop Power Supply (model AA45NM170) which provides 5V@3A,9V@3A,15V@3A,[email protected], where Linux master fails and just tells the USB-PD PSU to power-off.

I think this depends on the recent Rockchip TCPM changes. FUSB302 has been a pain in the USB or else this platform would have launched a lot longer ago.

Quoted 6 lines

Sorry, I was testing this patch on top of next-20230724 which includes https://patchwork.kernel.org/project/linux-rockchip/list/?series=757752:

8be558dcffe69b078b34b1fa93b82acaf4ce4957 ("usb: typec: tcpm: add get max power support") 1e35f074399dece73d5df11847d4a0d7a6f49434 ("usb: typec: tcpm: fix cc role at port reset")

Can you check if I am missing any other patches ?

so I guess the question is, are there any new clues and/or does this patch make the situaton better or worse for the roc-rk3399-pc board as it stands now.

Thanks
Heiko

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

author
#5 lore raw

+ Frank
+ Guenter
+ linux-usb
(for any suggestions)

Quoted 35 lines · Heiko Stuebner

Am Dienstag, 25. Juli 2023, 10:02:21 CEST schrieb Christopher Obbard:

Quoted 33 lines

Hi Da, Jagan,

Quoted 20 lines · Da Xue
Quoted 18 lines · Christopher Obbard
Quoted 6 lines · Jagan Teki

The power supply circuit in ROC-RK3399-PC is

Power Supply (or PPS) => FUSB => MP8859

VUBS pin of FUSB and IN pin of MP8859 is supplied via TYPEC_IN.

Quoted 7 lines

The MP8859 operated with 2.8V to 22V supply voltage and typical applications this supply can be 12V.

This patch is trying to support the PD by changing the FUSB VBUS supply to 12V and tune the I2C7 timings from downstream kernel.

Tested with PD3.0 PPS with supply voltages of 12V/3A and 20V/5A.

Hi Jagan,

Quoted 7 lines
Quoted 3 lines

This series works fine with a "dumb" (no PD negotiation) 5.1V Raspberry Pi PSU.

It also works fine with a Dell 45W USB-C Laptop Power Supply (model AA45NM170) which provides 5V@3A,9V@3A,15V@3A,[email protected], where Linux master fails and just tells the USB-PD PSU to power-off.

I think this depends on the recent Rockchip TCPM changes. FUSB302 has been a pain in the USB or else this platform would have launched a lot longer ago.

Sorry, I was testing this patch on top of next-20230724 which includes https://patchwork.kernel.org/project/linux-rockchip/list/?series=757752:

Quoted 8 lines
Quoted 4 lines

8be558dcffe69b078b34b1fa93b82acaf4ce4957 ("usb: typec: tcpm: add get max power support") 1e35f074399dece73d5df11847d4a0d7a6f49434 ("usb: typec: tcpm: fix cc role at port reset")

Can you check if I am missing any other patches ?

so I guess the question is, are there any new clues and/or does this patch make the situaton better or worse for the roc-rk3399-pc board as it stands now.

It is working with PD3.0 (EHO 100W GaN) but seems to be inconsistent. For the non-working cases, I did check in RK3399 and RK3588 designs the PD run machine setup in tcpm with respect to the fusb302 chip receiving hard reset due to BC_LVL, handler pending.

Here, is the sample log (please don't mind, the log combined with other functions as I've enabled the static tcpm logs)

[    0.342297] FUSB: sw reset
[    0.346084] FUSB: fusb302 device ID: 0x91
[    0.346331] TCPM: Setting usb_comm capable false
[    0.351365] FUSB: pd := off
[    0.351772] FUSB: vbus is already Off
[    0.352019] FUSB: charge is already Off
[    0.352342] FUSB: vconn is already Off
[    0.352680] TCPM: Setting voltage/current limit 0 mV 0 mA
[    0.353023] TCPM: polarity 0
[    0.353497] TCPM: Requesting mux state 0, usb-role 0, orientation 0
[    0.354544] FUSB: pd header := Sink, Device
[    0.355097] TCPM: state change INVALID_STATE -> SNK_UNATTACHED
[rev1 NONE_AMS]
[    0.355469] FUSB: cc1=Open, cc2=Open
[    0.356107] TCPM: state change SNK_UNATTACHED -> PORT_RESET [rev1 NONE_AMS]
[    0.356424] TCPM: 1-0022: registered
[    0.356428] TCPM: Setting usb_comm capable false
[    0.357964] i2c 3-0011: Fixed dependency cycle(s) with
/i2s@fe470000/port/endpoint
[    0.362173] FUSB: pd := off
[    0.362182] FUSB: vbus is already Off
[    0.362445] FUSB: charge is already Off
[    0.362778] FUSB: vconn is already Off
[    0.363126] TCPM: Setting voltage/current limit 0 mV 0 mA
[    0.363479] TCPM: polarity 0
[    0.363970] TCPM: Requesting mux state 0, usb-role 0, orientation 0
[    0.364339] sdhci: Secure Digital Host Controller Interface driver
[    0.365298] FUSB: pd header := Sink, Device
[    0.365440] sdhci: Copyright(c) Pierre Ossman
[    0.366110] Synopsys Designware Multimedia Card Interface Driver
[    0.366222] TCPM: cc:=2
[    0.366758] FUSB: cc := Rd
[    0.367564] sdhci-pltfm: SDHCI platform and OF driver helper
[    0.370341] ledtrig-cpu: registered to indicate activity on CPUs
[    0.371295] scmi_protocol scmi_dev.1: Enabled polling mode TX
channel - prot_id:16
[    0.372067] arm-scmi firmware:scmi: SCMI Notifications - Core Enabled.
[    0.372670] arm-scmi firmware:scmi: SCMI Protocol v2.0 'rockchip:'
Firmware version 0x0
[    0.374620] TCPM: pending state change PORT_RESET ->
PORT_RESET_WAIT_OFF @ 100 ms [rev1 NONE_AMS]
[    0.374638] TCPM: state change PORT_RESET -> PORT_RESET_WAIT_OFF
[delayed 100 ms]
[    0.375435] TCPM: pending state change PORT_RESET_WAIT_OFF ->
SNK_UNATTACHED @ 920 ms [rev1 NONE_AMS]
[    0.379490] SMCCC: SOC_ID: ARCH_SOC_ID not implemented, skipping ....
[    0.381481] usbcore: registered new interface driver usbhid
[    0.381972] usbhid: USB HID core driver
[    0.385775] hw perfevents: enabled with armv8_cortex_a55 PMU
driver, 7 counters available
[    0.387875] hw perfevents: enabled with armv8_cortex_a76 PMU
driver, 7 counters available
[    0.390529] optee: probing for conduit method.
[    0.390926] optee: api uid mismatch
[    0.391234] optee: probe of firmware:optee failed with error -22
[    0.394071] NET: Registered PF_PACKET protocol family
[    0.394571] 9pnet: Installing 9P2000 support
[    0.394986] Key type dns_resolver registered
[    0.399904] registered taskstats version 1
[    0.400365] Loading compiled-in X.509 certificates
[    0.400933] mmc0: SDHCI controller on fe2e0000.mmc [fe2e0000.mmc] using ADMA
[    0.407446] mmc0: Failed to initialize a non-removable card
[    0.442034] rk808-regulator rk808-regulator.1.auto: there is no dvs0 gpio
[    0.443181] rk808-regulator rk808-regulator.1.auto: there is no dvs1 gpio
[    0.448771] vdd_2v0_pldo_s3: Bringing 1975000uV into 2000000-2000000uV
[    0.450335] vcc_3v3_s3: Bringing 3275000uV into 3300000-3300000uV
[    0.452301] vcc_1v8_s3: Bringing 1775000uV into 1800000-1800000uV
[    0.475134] ehci-platform fc800000.usb: EHCI Host Controller
[    0.475656] ehci-platform fc800000.usb: new USB bus registered,
assigned bus number 1
[    0.476412] ehci-platform fc800000.usb: irq 67, io mem 0xfc800000
[    0.476870] ohci-platform fc840000.usb: Generic Platform OHCI controller
[    0.477563] ohci-platform fc840000.usb: new USB bus registered,
assigned bus number 2
[    0.478298] ohci-platform fc840000.usb: irq 68, io mem 0xfc840000
[    0.480664] fan53555-regulator 0-0042: FAN53555 Option[10] Rev[1] Detected!
[    0.481424] fan53555-regulator 0-0043: FAN53555 Option[10] Rev[1] Detected!
[    0.484586] dwmmc_rockchip fe2c0000.mmc: IDMAC supports 32-bit address mode.
[    0.485274] dwmmc_rockchip fe2c0000.mmc: Using internal DMA controller.
[    0.485875] dwmmc_rockchip fe2c0000.mmc: Version ID is 270a
[    0.486417] dwmmc_rockchip fe2c0000.mmc: DW MMC controller at irq
69,32 bit host data width,256 deep fifo
[    0.488188] dwmmc_rockchip fe2c0000.mmc: Got CD GPIO
[    0.491377] cfg80211: Loading compiled-in X.509 certificates for
regulatory database
[    0.492171] ehci-platform fc880000.usb: EHCI Host Controller
[    0.492179] ohci-platform fc8c0000.usb: Generic Platform OHCI controller
[    0.492680] ehci-platform fc880000.usb: new USB bus registered,
assigned bus number 3
[    0.493271] ohci-platform fc8c0000.usb: new USB bus registered,
assigned bus number 4
[    0.493350] ehci-platform fc800000.usb: USB 2.0 started, EHCI 1.00
[    0.494008] ehci-platform fc880000.usb: irq 70, io mem 0xfc880000
[    0.494520] hub 1-0:1.0: USB hub found
[    0.494542] hub 1-0:1.0: 1 port detected
[    0.494745] ohci-platform fc8c0000.usb: irq 71, io mem 0xfc8c0000
[    0.501157] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req
400000Hz, actual 400000HZ div = 0)
[    0.508986] ehci-platform fc880000.usb: USB 2.0 started, EHCI 1.00
[    0.510401] hub 3-0:1.0: USB hub found
[    0.510779] hub 3-0:1.0: 1 port detected
[    0.542084] hub 2-0:1.0: USB hub found
[    0.542438] hub 2-0:1.0: 1 port detected
[    0.562045] hub 4-0:1.0: USB hub found
[    0.562506] hub 4-0:1.0: 1 port detected
[    0.595247] mmc_host mmc1: Bus speed (slot 0) = 198000000Hz (slot
req 200000000Hz, actual 198000000HZ div = 0)
[    0.748985] usb 1-1: new high-speed USB device number 2 using ehci-platform
[    0.906736] hub 1-1:1.0: USB hub found
[    0.907309] hub 1-1:1.0: 4 ports detected
[    0.923345] dwmmc_rockchip fe2c0000.mmc: Successfully tuned phase to 51
[    0.923962] mmc1: new ultra high speed SDR104 SDHC card at address aaaa
[    0.925357] mmcblk1: mmc1:aaaa SD32G 29.7 GiB
[    0.932234]  mmcblk1: p1 p2 p3
[    0.986929] Freeing initrd memory: 24648K
[    1.000498] Loaded X.509 cert 'sforshee: 00b28ddf47aef9cea7'
[    1.001061] clk: Disabling unused clocks
[    1.001925] ALSA device list:
[    1.002190]   No soundcards found.
[    1.002595] dw-apb-uart feb50000.serial: forbid DMA for kernel console
[    1.004969] platform regulatory.0: Direct firmware load for
regulatory.db failed with error -2
[    1.005732] cfg80211: failed to load regulatory.db
[    1.007117] Freeing unused kernel memory: 9216K
[    1.029098] Run /init as init process
Loading, please wait...
Starting version 247.3-7+deb11u1
[    1.296214] TCPM: state change PORT_RESET_WAIT_OFF ->
SNK_UNATTACHED [delayed 920 ms]
[    1.301978] FUSB: start drp toggling
[    1.302699] TCPM: state change SNK_UNATTACHED -> TOGGLING [rev1 NONE_AMS]
[    1.304531] FUSB: IRQ: VBUS_OK, vbus=On
[    1.307751] FUSB: IRQ: TOGDONE
[    1.314498] FUSB: detected cc1=Rp-3.0, cc2=Open
[    1.314880] FUSB: cc1=Rp-3.0, cc2=Open
[    1.315309] TCPM: state change TOGGLING -> SNK_ATTACH_WAIT [rev1 NONE_AMS]

Begin: Loading essential drivers ... done.
Begin: Running /scripts/init-premount ... done.
Begin: Mounting root file system ... Begin: Running /scripts/local-top ... done.
Begin: Running /scripts/local-premount ... done.
Begin: Will now check root file system ... fsck from util-linux 2.36.1 [/sbin/fsck.ext4 (1) -- /dev/mmcblk1p3] fsck.ext4 -a -C0 /dev/mmcblk1p3

[    1.315655] TCPM: pending state change SNK_ATTACH_WAIT ->
SNK_DEBOUNCED @ 200 ms [rev1 NONE_AMS]
[    1.516302] TCPM: state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED
[delayed 200 ms]
[    1.517159] TCPM: state change SNK_DEBOUNCED -> SNK_ATTACHED [rev1 NONE_AMS]
[    1.517824] TCPM: polarity 0
[    1.518454] TCPM: Requesting mux state 1, usb-role 2, orientation 1
[    1.519744] FUSB: pd header := Sink, Device
[    1.520320] TCPM: state change SNK_ATTACHED -> SNK_STARTUP [rev1 NONE_AMS]
[    1.520745] TCPM: state change SNK_STARTUP -> SNK_DISCOVERY [rev3 NONE_AMS]
[    1.521388] TCPM: Setting voltage/current limit 5000 mV 3000 mA
[    1.522016] TCPM: vbus=0 charge:=1
[    1.522547] FUSB: vbus is already Off
ram: recovering journal
[    1.522857] TCPM: state change SNK_DISCOVERY ->
SNK_WAIT_CAPABILITIES [rev3 NONE_AMS]
[    1.528181] FUSB: pd := on
ram: clean, 1521/594512 files, 74956/2374144 blocks
done.
[    1.528977] TCPM: pending state change SNK_WAIT_CAPABILITIES ->
SNK_SOFT_RESET @ 310 ms [rev3 NONE_AMS]
[    1.568848] EXT4-fs (mmcblk1p3): mounted filesystem

b550eef1-48e4-4bac-9312-2ea23c77a372 r/w with ordered data mode. Quota
mode: none.
done.
Begin: Running /scripts/local-bottom ... done.
Begin: Running /scripts/init-bottom ... done.

[    1.650560] EXT4-fs (mmcblk1p3): re-mounted

b550eef1-48e4-4bac-9312-2ea23c77a372 r/w. Quota mode: none. Starting syslogd: OK
Starting klogd: OK
Running sysctl: OK
Initializing random number generator: OK
Saving random seed: [ 1.839341] TCPM: state change
SNK_WAIT_CAPABILITIES -> SNK_SOFT_RESET [delayed 310 ms]

[    1.839362] TCPM: AMS SOFT_RESET_AMS start
[    1.840087] TCPM: state change SNK_SOFT_RESET -> AMS_START [rev3
SOFT_RESET_AMS]
[    1.840461] TCPM: state change AMS_START -> SOFT_RESET_SEND [rev3
SOFT_RESET_AMS]
[    1.841161] TCPM: PD TX, header: 0x8d
[    1.844492] FUSB: sending PD message header: 8d
[    1.844836] FUSB: sending PD message len: 0
[    1.847576] FUSB: IRQ: BC_LVL, handler pending
[    1.847964] FUSB: IRQ: PD tx success
[    1.850111] FUSB: PD message header: 161
[    1.850445] FUSB: PD message len: 0
[    1.850802] TCPM: PD TX complete, status: 0
[    1.851211] TCPM: pending state change SOFT_RESET_SEND ->
HARD_RESET_SEND @ 60 ms [rev3 SOFT_RESET_AMS]
[    1.853070] FUSB: IRQ: BC_LVL, handler pending
[    1.853914] FUSB: IRQ: PD sent good CRC
[    1.856233] FUSB: PD message header: 1a3
[    1.856590] FUSB: PD message len: 0
[    1.857065] TCPM: PD RX, header: 0x1a3 [1]
[    1.857402] TCPM: AMS SOFT_RESET_AMS finished
[    1.857774] TCPM: state change SOFT_RESET_SEND ->
SNK_WAIT_CAPABILITIES [rev3 NONE_AMS]
[    1.864223] FUSB: pd := on
[    1.864967] TCPM: pending state change SNK_WAIT_CAPABILITIES ->
HARD_RESET_SEND @ 310 ms [rev3 NONE_AMS]
[    1.877754] FUSB: IRQ: BC_LVL, handler pending
[    1.878606] FUSB: IRQ: PD received hardreset
[    1.879902] TCPM: state change SNK_WAIT_CAPABILITIES ->
HARD_RESET_START [rev3 HARD_RESET]
[    1.887331] FUSB: pd := off
[    1.888078] TCPM: state change HARD_RESET_START ->
SNK_HARD_RESET_SINK_OFF [rev3 HARD_RESET]
[    1.888339] TCPM: vconn:=0
[    1.889118] FUSB: vconn is already Off
[    1.889370] TCPM: Requesting mux state 1, usb-role 2, orientation 1
[    1.890621] FUSB: pd header := Sink, Device

Look like Frank sent some patch related(not 100% sure) to this in linux-usb [1], but the result is the same even with this change. I'm still debugging tcpm, any pointers on this would be useful.

[1] https://patchwork.kernel.org/project/linux-usb/patch/[email protected]/

Thanks,
Jagan.

#6 lore raw

Hi Jagan,

In my view, it is the problem that the fusb302 missed the Source caps of PD charger (I assume the charger is normal), maybe the RX interrupt was missed or its FIFO had been touched.

The patch you mentioned ([1] in your mail) may be able to fix your issue if the FIFO of the fusb302 is flushed by set_pd_rx() invoked in TCPM, but it has no effort for RX interrupt missed.

Anyway, It is the defect of the fusb302 driver and should be fixed in it .  Hope my analysis is helpful to you.

BR,
Frank

Quoted 265 lines · Jagan Teki

+ Frank
+ Guenter
+ linux-usb
(for any suggestions)

Quoted 30 lines · Heiko Stuebner

Am Dienstag, 25. Juli 2023, 10:02:21 CEST schrieb Christopher Obbard:

Quoted 29 lines

Hi Da, Jagan,

Quoted 16 lines · Da Xue
Quoted 14 lines · Christopher Obbard
Quoted 13 lines · Jagan Teki

The power supply circuit in ROC-RK3399-PC is

Power Supply (or PPS) => FUSB => MP8859

VUBS pin of FUSB and IN pin of MP8859 is supplied via TYPEC_IN.

The MP8859 operated with 2.8V to 22V supply voltage and typical applications this supply can be 12V.

This patch is trying to support the PD by changing the FUSB VBUS supply to 12V and tune the I2C7 timings from downstream kernel.

Tested with PD3.0 PPS with supply voltages of 12V/3A and 20V/5A.

Quoted 8 lines
Quoted 5 lines

Hi Jagan,

This series works fine with a "dumb" (no PD negotiation) 5.1V Raspberry Pi PSU.

It also works fine with a Dell 45W USB-C Laptop Power Supply (model AA45NM170) which provides 5V@3A,9V@3A,15V@3A,[email protected], where Linux master fails and just tells the USB-PD PSU to power-off.

I think this depends on the recent Rockchip TCPM changes. FUSB302 has been a pain in the USB or else this platform would have launched a lot longer ago.

Sorry, I was testing this patch on top of next-20230724 which includes https://patchwork.kernel.org/project/linux-rockchip/list/?series=757752:

Quoted 7 lines
Quoted 4 lines

8be558dcffe69b078b34b1fa93b82acaf4ce4957 ("usb: typec: tcpm: add get max power support") 1e35f074399dece73d5df11847d4a0d7a6f49434 ("usb: typec: tcpm: fix cc role at port reset")

Can you check if I am missing any other patches ?

so I guess the question is, are there any new clues and/or does this patch make the situaton better or worse for the roc-rk3399-pc board as it stands now.

It is working with PD3.0 (EHO 100W GaN) but seems to be inconsistent. For the non-working cases, I did check in RK3399 and RK3588 designs the PD run machine setup in tcpm with respect to the fusb302 chip receiving hard reset due to BC_LVL, handler pending.

Here, is the sample log (please don't mind, the log combined with other functions as I've enabled the static tcpm logs)

[    0.342297] FUSB: sw reset
[    0.346084] FUSB: fusb302 device ID: 0x91
[    0.346331] TCPM: Setting usb_comm capable false
[    0.351365] FUSB: pd := off
[    0.351772] FUSB: vbus is already Off
[    0.352019] FUSB: charge is already Off
[    0.352342] FUSB: vconn is already Off
[    0.352680] TCPM: Setting voltage/current limit 0 mV 0 mA
[    0.353023] TCPM: polarity 0
[    0.353497] TCPM: Requesting mux state 0, usb-role 0, orientation 0
[    0.354544] FUSB: pd header := Sink, Device
[    0.355097] TCPM: state change INVALID_STATE -> SNK_UNATTACHED
[rev1 NONE_AMS]
[    0.355469] FUSB: cc1=Open, cc2=Open
[    0.356107] TCPM: state change SNK_UNATTACHED -> PORT_RESET [rev1 NONE_AMS]
[    0.356424] TCPM: 1-0022: registered
[    0.356428] TCPM: Setting usb_comm capable false
[    0.357964] i2c 3-0011: Fixed dependency cycle(s) with
/i2s@fe470000/port/endpoint
[    0.362173] FUSB: pd := off
[    0.362182] FUSB: vbus is already Off
[    0.362445] FUSB: charge is already Off
[    0.362778] FUSB: vconn is already Off
[    0.363126] TCPM: Setting voltage/current limit 0 mV 0 mA
[    0.363479] TCPM: polarity 0
[    0.363970] TCPM: Requesting mux state 0, usb-role 0, orientation 0
[    0.364339] sdhci: Secure Digital Host Controller Interface driver
[    0.365298] FUSB: pd header := Sink, Device
[    0.365440] sdhci: Copyright(c) Pierre Ossman
[    0.366110] Synopsys Designware Multimedia Card Interface Driver
[    0.366222] TCPM: cc:=2
[    0.366758] FUSB: cc := Rd
[    0.367564] sdhci-pltfm: SDHCI platform and OF driver helper
[    0.370341] ledtrig-cpu: registered to indicate activity on CPUs
[    0.371295] scmi_protocol scmi_dev.1: Enabled polling mode TX
channel - prot_id:16
[    0.372067] arm-scmi firmware:scmi: SCMI Notifications - Core Enabled.
[    0.372670] arm-scmi firmware:scmi: SCMI Protocol v2.0 'rockchip:'
Firmware version 0x0
[    0.374620] TCPM: pending state change PORT_RESET ->
PORT_RESET_WAIT_OFF @ 100 ms [rev1 NONE_AMS]
[    0.374638] TCPM: state change PORT_RESET -> PORT_RESET_WAIT_OFF
[delayed 100 ms]
[    0.375435] TCPM: pending state change PORT_RESET_WAIT_OFF ->
SNK_UNATTACHED @ 920 ms [rev1 NONE_AMS]
[    0.379490] SMCCC: SOC_ID: ARCH_SOC_ID not implemented, skipping ....
[    0.381481] usbcore: registered new interface driver usbhid
[    0.381972] usbhid: USB HID core driver
[    0.385775] hw perfevents: enabled with armv8_cortex_a55 PMU
driver, 7 counters available
[    0.387875] hw perfevents: enabled with armv8_cortex_a76 PMU
driver, 7 counters available
[    0.390529] optee: probing for conduit method.
[    0.390926] optee: api uid mismatch
[    0.391234] optee: probe of firmware:optee failed with error -22
[    0.394071] NET: Registered PF_PACKET protocol family
[    0.394571] 9pnet: Installing 9P2000 support
[    0.394986] Key type dns_resolver registered
[    0.399904] registered taskstats version 1
[    0.400365] Loading compiled-in X.509 certificates
[    0.400933] mmc0: SDHCI controller on fe2e0000.mmc [fe2e0000.mmc] using ADMA
[    0.407446] mmc0: Failed to initialize a non-removable card
[    0.442034] rk808-regulator rk808-regulator.1.auto: there is no dvs0 gpio
[    0.443181] rk808-regulator rk808-regulator.1.auto: there is no dvs1 gpio
[    0.448771] vdd_2v0_pldo_s3: Bringing 1975000uV into 2000000-2000000uV
[    0.450335] vcc_3v3_s3: Bringing 3275000uV into 3300000-3300000uV
[    0.452301] vcc_1v8_s3: Bringing 1775000uV into 1800000-1800000uV
[    0.475134] ehci-platform fc800000.usb: EHCI Host Controller
[    0.475656] ehci-platform fc800000.usb: new USB bus registered,
assigned bus number 1
[    0.476412] ehci-platform fc800000.usb: irq 67, io mem 0xfc800000
[    0.476870] ohci-platform fc840000.usb: Generic Platform OHCI controller
[    0.477563] ohci-platform fc840000.usb: new USB bus registered,
assigned bus number 2
[    0.478298] ohci-platform fc840000.usb: irq 68, io mem 0xfc840000
[    0.480664] fan53555-regulator 0-0042: FAN53555 Option[10] Rev[1] Detected!
[    0.481424] fan53555-regulator 0-0043: FAN53555 Option[10] Rev[1] Detected!
[    0.484586] dwmmc_rockchip fe2c0000.mmc: IDMAC supports 32-bit address mode.
[    0.485274] dwmmc_rockchip fe2c0000.mmc: Using internal DMA controller.
[    0.485875] dwmmc_rockchip fe2c0000.mmc: Version ID is 270a
[    0.486417] dwmmc_rockchip fe2c0000.mmc: DW MMC controller at irq
69,32 bit host data width,256 deep fifo
[    0.488188] dwmmc_rockchip fe2c0000.mmc: Got CD GPIO
[    0.491377] cfg80211: Loading compiled-in X.509 certificates for
regulatory database
[    0.492171] ehci-platform fc880000.usb: EHCI Host Controller
[    0.492179] ohci-platform fc8c0000.usb: Generic Platform OHCI controller
[    0.492680] ehci-platform fc880000.usb: new USB bus registered,
assigned bus number 3
[    0.493271] ohci-platform fc8c0000.usb: new USB bus registered,
assigned bus number 4
[    0.493350] ehci-platform fc800000.usb: USB 2.0 started, EHCI 1.00
[    0.494008] ehci-platform fc880000.usb: irq 70, io mem 0xfc880000
[    0.494520] hub 1-0:1.0: USB hub found
[    0.494542] hub 1-0:1.0: 1 port detected
[    0.494745] ohci-platform fc8c0000.usb: irq 71, io mem 0xfc8c0000
[    0.501157] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req
400000Hz, actual 400000HZ div = 0)
[    0.508986] ehci-platform fc880000.usb: USB 2.0 started, EHCI 1.00
[    0.510401] hub 3-0:1.0: USB hub found
[    0.510779] hub 3-0:1.0: 1 port detected
[    0.542084] hub 2-0:1.0: USB hub found
[    0.542438] hub 2-0:1.0: 1 port detected
[    0.562045] hub 4-0:1.0: USB hub found
[    0.562506] hub 4-0:1.0: 1 port detected
[    0.595247] mmc_host mmc1: Bus speed (slot 0) = 198000000Hz (slot
req 200000000Hz, actual 198000000HZ div = 0)
[    0.748985] usb 1-1: new high-speed USB device number 2 using ehci-platform
[    0.906736] hub 1-1:1.0: USB hub found
[    0.907309] hub 1-1:1.0: 4 ports detected
[    0.923345] dwmmc_rockchip fe2c0000.mmc: Successfully tuned phase to 51
[    0.923962] mmc1: new ultra high speed SDR104 SDHC card at address aaaa
[    0.925357] mmcblk1: mmc1:aaaa SD32G 29.7 GiB
[    0.932234]  mmcblk1: p1 p2 p3
[    0.986929] Freeing initrd memory: 24648K
[    1.000498] Loaded X.509 cert 'sforshee: 00b28ddf47aef9cea7'
[    1.001061] clk: Disabling unused clocks
[    1.001925] ALSA device list:
[    1.002190]   No soundcards found.
[    1.002595] dw-apb-uart feb50000.serial: forbid DMA for kernel console
[    1.004969] platform regulatory.0: Direct firmware load for
regulatory.db failed with error -2
[    1.005732] cfg80211: failed to load regulatory.db
[    1.007117] Freeing unused kernel memory: 9216K
[    1.029098] Run /init as init process
Loading, please wait...
Starting version 247.3-7+deb11u1
[    1.296214] TCPM: state change PORT_RESET_WAIT_OFF ->
SNK_UNATTACHED [delayed 920 ms]
[    1.301978] FUSB: start drp toggling
[    1.302699] TCPM: state change SNK_UNATTACHED -> TOGGLING [rev1 NONE_AMS]
[    1.304531] FUSB: IRQ: VBUS_OK, vbus=On
[    1.307751] FUSB: IRQ: TOGDONE
[    1.314498] FUSB: detected cc1=Rp-3.0, cc2=Open
[    1.314880] FUSB: cc1=Rp-3.0, cc2=Open
[    1.315309] TCPM: state change TOGGLING -> SNK_ATTACH_WAIT [rev1 NONE_AMS]

Begin: Loading essential drivers ... done.
Begin: Running /scripts/init-premount ... done.
Begin: Mounting root file system ... Begin: Running /scripts/local-top ... done.
Begin: Running /scripts/local-premount ... done.
Begin: Will now check root file system ... fsck from util-linux 2.36.1 [/sbin/fsck.ext4 (1) -- /dev/mmcblk1p3] fsck.ext4 -a -C0 /dev/mmcblk1p3

[    1.315655] TCPM: pending state change SNK_ATTACH_WAIT ->
SNK_DEBOUNCED @ 200 ms [rev1 NONE_AMS]
[    1.516302] TCPM: state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED
[delayed 200 ms]
[    1.517159] TCPM: state change SNK_DEBOUNCED -> SNK_ATTACHED [rev1 NONE_AMS]
[    1.517824] TCPM: polarity 0
[    1.518454] TCPM: Requesting mux state 1, usb-role 2, orientation 1
[    1.519744] FUSB: pd header := Sink, Device
[    1.520320] TCPM: state change SNK_ATTACHED -> SNK_STARTUP [rev1 NONE_AMS]
[    1.520745] TCPM: state change SNK_STARTUP -> SNK_DISCOVERY [rev3 NONE_AMS]
[    1.521388] TCPM: Setting voltage/current limit 5000 mV 3000 mA
[    1.522016] TCPM: vbus=0 charge:=1
[    1.522547] FUSB: vbus is already Off
ram: recovering journal
[    1.522857] TCPM: state change SNK_DISCOVERY ->
SNK_WAIT_CAPABILITIES [rev3 NONE_AMS]
[    1.528181] FUSB: pd := on
ram: clean, 1521/594512 files, 74956/2374144 blocks
done.
[    1.528977] TCPM: pending state change SNK_WAIT_CAPABILITIES ->
SNK_SOFT_RESET @ 310 ms [rev3 NONE_AMS]
[    1.568848] EXT4-fs (mmcblk1p3): mounted filesystem

b550eef1-48e4-4bac-9312-2ea23c77a372 r/w with ordered data mode. Quota
mode: none.
done.
Begin: Running /scripts/local-bottom ... done.
Begin: Running /scripts/init-bottom ... done.

[    1.650560] EXT4-fs (mmcblk1p3): re-mounted

b550eef1-48e4-4bac-9312-2ea23c77a372 r/w. Quota mode: none. Starting syslogd: OK
Starting klogd: OK
Running sysctl: OK
Initializing random number generator: OK
Saving random seed: [ 1.839341] TCPM: state change
SNK_WAIT_CAPABILITIES -> SNK_SOFT_RESET [delayed 310 ms]

[    1.839362] TCPM: AMS SOFT_RESET_AMS start
[    1.840087] TCPM: state change SNK_SOFT_RESET -> AMS_START [rev3
SOFT_RESET_AMS]
[    1.840461] TCPM: state change AMS_START -> SOFT_RESET_SEND [rev3
SOFT_RESET_AMS]
[    1.841161] TCPM: PD TX, header: 0x8d
[    1.844492] FUSB: sending PD message header: 8d
[    1.844836] FUSB: sending PD message len: 0
[    1.847576] FUSB: IRQ: BC_LVL, handler pending
[    1.847964] FUSB: IRQ: PD tx success
[    1.850111] FUSB: PD message header: 161
[    1.850445] FUSB: PD message len: 0
[    1.850802] TCPM: PD TX complete, status: 0
[    1.851211] TCPM: pending state change SOFT_RESET_SEND ->
HARD_RESET_SEND @ 60 ms [rev3 SOFT_RESET_AMS]
[    1.853070] FUSB: IRQ: BC_LVL, handler pending
[    1.853914] FUSB: IRQ: PD sent good CRC
[    1.856233] FUSB: PD message header: 1a3
[    1.856590] FUSB: PD message len: 0
[    1.857065] TCPM: PD RX, header: 0x1a3 [1]
[    1.857402] TCPM: AMS SOFT_RESET_AMS finished
[    1.857774] TCPM: state change SOFT_RESET_SEND ->
SNK_WAIT_CAPABILITIES [rev3 NONE_AMS]
[    1.864223] FUSB: pd := on
[    1.864967] TCPM: pending state change SNK_WAIT_CAPABILITIES ->
HARD_RESET_SEND @ 310 ms [rev3 NONE_AMS]
[    1.877754] FUSB: IRQ: BC_LVL, handler pending
[    1.878606] FUSB: IRQ: PD received hardreset
[    1.879902] TCPM: state change SNK_WAIT_CAPABILITIES ->
HARD_RESET_START [rev3 HARD_RESET]
[    1.887331] FUSB: pd := off
[    1.888078] TCPM: state change HARD_RESET_START ->
SNK_HARD_RESET_SINK_OFF [rev3 HARD_RESET]
[    1.888339] TCPM: vconn:=0
[    1.889118] FUSB: vconn is already Off
[    1.889370] TCPM: Requesting mux state 1, usb-role 2, orientation 1
[    1.890621] FUSB: pd header := Sink, Device
Quoted 8 lines

Look like Frank sent some patch related(not 100% sure) to this in linux-usb [1], but the result is the same even with this change. I'm still debugging tcpm, any pointers on this would be useful.

[1] https://patchwork.kernel.org/project/linux-usb/patch/[email protected]/

Thanks,
Jagan.

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

#7 lore raw
Quoted 6 lines · Frank Wang

Hi Jagan,

In my view, it is the problem that the fusb302 missed the Source caps of PD charger (I assume the charger is normal), maybe the RX interrupt was missed or its FIFO had been touched.

I can confirm these findings. After logging the flushes, I see multiple tcpm states that experience flushes where relevant PDOs sent by the chargers are flushed instead of acted upon due to the timing. This causes multiple failures in the state machine resulting in HARD_RESETs.

Since the driver is not privy to the tcpm_states, do we add explicit flush functions to TCPM and split them from set_pd_rx?

Or is there any implementation suggestions instead of hacking away at TCPM before set_pd_rx to check state.

Quoted 5 lines

The patch you mentioned ([1] in your mail) may be able to fix your issue if the FIFO of the fusb302 is flushed by set_pd_rx() invoked in TCPM, but it has no effort for RX interrupt missed.

Quoted 282 lines

Anyway, It is the defect of the fusb302 driver and should be fixed in it . Hope my analysis is helpful to you.

BR,
Frank

Quoted 273 lines · Jagan Teki

+ Frank
+ Guenter
+ linux-usb
(for any suggestions)

Quoted 30 lines · Heiko Stuebner

Am Dienstag, 25. Juli 2023, 10:02:21 CEST schrieb Christopher Obbard:

Quoted 29 lines

Hi Da, Jagan,

Quoted 16 lines · Da Xue
Quoted 14 lines · Christopher Obbard
Quoted 13 lines · Jagan Teki

The power supply circuit in ROC-RK3399-PC is

Power Supply (or PPS) => FUSB => MP8859

VUBS pin of FUSB and IN pin of MP8859 is supplied via TYPEC_IN.

The MP8859 operated with 2.8V to 22V supply voltage and typical applications this supply can be 12V.

This patch is trying to support the PD by changing the FUSB VBUS supply to 12V and tune the I2C7 timings from downstream kernel.

Tested with PD3.0 PPS with supply voltages of 12V/3A and 20V/5A.

Quoted 8 lines
Quoted 5 lines

Hi Jagan,

This series works fine with a "dumb" (no PD negotiation) 5.1V Raspberry Pi PSU.

It also works fine with a Dell 45W USB-C Laptop Power Supply (model AA45NM170) which provides 5V@3A,9V@3A,15V@3A,[email protected], where Linux master fails and just tells the USB-PD PSU to power-off.

I think this depends on the recent Rockchip TCPM changes. FUSB302 has been a pain in the USB or else this platform would have launched a lot longer ago.

Sorry, I was testing this patch on top of next-20230724 which includes https://patchwork.kernel.org/project/linux-rockchip/list/?series=757752:

Quoted 7 lines
Quoted 4 lines

8be558dcffe69b078b34b1fa93b82acaf4ce4957 ("usb: typec: tcpm: add get max power support") 1e35f074399dece73d5df11847d4a0d7a6f49434 ("usb: typec: tcpm: fix cc role at port reset")

Can you check if I am missing any other patches ?

so I guess the question is, are there any new clues and/or does this patch make the situaton better or worse for the roc-rk3399-pc board as it stands now.

It is working with PD3.0 (EHO 100W GaN) but seems to be inconsistent. For the non-working cases, I did check in RK3399 and RK3588 designs the PD run machine setup in tcpm with respect to the fusb302 chip receiving hard reset due to BC_LVL, handler pending.

Here, is the sample log (please don't mind, the log combined with other functions as I've enabled the static tcpm logs)

[    0.342297] FUSB: sw reset
[    0.346084] FUSB: fusb302 device ID: 0x91
[    0.346331] TCPM: Setting usb_comm capable false
[    0.351365] FUSB: pd := off
[    0.351772] FUSB: vbus is already Off
[    0.352019] FUSB: charge is already Off
[    0.352342] FUSB: vconn is already Off
[    0.352680] TCPM: Setting voltage/current limit 0 mV 0 mA
[    0.353023] TCPM: polarity 0
[    0.353497] TCPM: Requesting mux state 0, usb-role 0, orientation 0
[    0.354544] FUSB: pd header := Sink, Device
[    0.355097] TCPM: state change INVALID_STATE -> SNK_UNATTACHED
[rev1 NONE_AMS]
[    0.355469] FUSB: cc1=Open, cc2=Open
[    0.356107] TCPM: state change SNK_UNATTACHED -> PORT_RESET [rev1 NONE_AMS]
[    0.356424] TCPM: 1-0022: registered
[    0.356428] TCPM: Setting usb_comm capable false
[    0.357964] i2c 3-0011: Fixed dependency cycle(s) with
/i2s@fe470000/port/endpoint
[    0.362173] FUSB: pd := off
[    0.362182] FUSB: vbus is already Off
[    0.362445] FUSB: charge is already Off
[    0.362778] FUSB: vconn is already Off
[    0.363126] TCPM: Setting voltage/current limit 0 mV 0 mA
[    0.363479] TCPM: polarity 0
[    0.363970] TCPM: Requesting mux state 0, usb-role 0, orientation 0
[    0.364339] sdhci: Secure Digital Host Controller Interface driver
[    0.365298] FUSB: pd header := Sink, Device
[    0.365440] sdhci: Copyright(c) Pierre Ossman
[    0.366110] Synopsys Designware Multimedia Card Interface Driver
[    0.366222] TCPM: cc:=2
[    0.366758] FUSB: cc := Rd
[    0.367564] sdhci-pltfm: SDHCI platform and OF driver helper
[    0.370341] ledtrig-cpu: registered to indicate activity on CPUs
[    0.371295] scmi_protocol scmi_dev.1: Enabled polling mode TX
channel - prot_id:16
[    0.372067] arm-scmi firmware:scmi: SCMI Notifications - Core Enabled.
[    0.372670] arm-scmi firmware:scmi: SCMI Protocol v2.0 'rockchip:'
Firmware version 0x0
[    0.374620] TCPM: pending state change PORT_RESET ->
PORT_RESET_WAIT_OFF @ 100 ms [rev1 NONE_AMS]
[    0.374638] TCPM: state change PORT_RESET -> PORT_RESET_WAIT_OFF
[delayed 100 ms]
[    0.375435] TCPM: pending state change PORT_RESET_WAIT_OFF ->
SNK_UNATTACHED @ 920 ms [rev1 NONE_AMS]
[    0.379490] SMCCC: SOC_ID: ARCH_SOC_ID not implemented, skipping ....
[    0.381481] usbcore: registered new interface driver usbhid
[    0.381972] usbhid: USB HID core driver
[    0.385775] hw perfevents: enabled with armv8_cortex_a55 PMU
driver, 7 counters available
[    0.387875] hw perfevents: enabled with armv8_cortex_a76 PMU
driver, 7 counters available
[    0.390529] optee: probing for conduit method.
[    0.390926] optee: api uid mismatch
[    0.391234] optee: probe of firmware:optee failed with error -22
[    0.394071] NET: Registered PF_PACKET protocol family
[    0.394571] 9pnet: Installing 9P2000 support
[    0.394986] Key type dns_resolver registered
[    0.399904] registered taskstats version 1
[    0.400365] Loading compiled-in X.509 certificates
[    0.400933] mmc0: SDHCI controller on fe2e0000.mmc [fe2e0000.mmc] using ADMA
[    0.407446] mmc0: Failed to initialize a non-removable card
[    0.442034] rk808-regulator rk808-regulator.1.auto: there is no dvs0 gpio
[    0.443181] rk808-regulator rk808-regulator.1.auto: there is no dvs1 gpio
[    0.448771] vdd_2v0_pldo_s3: Bringing 1975000uV into 2000000-2000000uV
[    0.450335] vcc_3v3_s3: Bringing 3275000uV into 3300000-3300000uV
[    0.452301] vcc_1v8_s3: Bringing 1775000uV into 1800000-1800000uV
[    0.475134] ehci-platform fc800000.usb: EHCI Host Controller
[    0.475656] ehci-platform fc800000.usb: new USB bus registered,
assigned bus number 1
[    0.476412] ehci-platform fc800000.usb: irq 67, io mem 0xfc800000
[    0.476870] ohci-platform fc840000.usb: Generic Platform OHCI controller
[    0.477563] ohci-platform fc840000.usb: new USB bus registered,
assigned bus number 2
[    0.478298] ohci-platform fc840000.usb: irq 68, io mem 0xfc840000
[    0.480664] fan53555-regulator 0-0042: FAN53555 Option[10] Rev[1] Detected!
[    0.481424] fan53555-regulator 0-0043: FAN53555 Option[10] Rev[1] Detected!
[    0.484586] dwmmc_rockchip fe2c0000.mmc: IDMAC supports 32-bit address mode.
[    0.485274] dwmmc_rockchip fe2c0000.mmc: Using internal DMA controller.
[    0.485875] dwmmc_rockchip fe2c0000.mmc: Version ID is 270a
[    0.486417] dwmmc_rockchip fe2c0000.mmc: DW MMC controller at irq
69,32 bit host data width,256 deep fifo
[    0.488188] dwmmc_rockchip fe2c0000.mmc: Got CD GPIO
[    0.491377] cfg80211: Loading compiled-in X.509 certificates for
regulatory database
[    0.492171] ehci-platform fc880000.usb: EHCI Host Controller
[    0.492179] ohci-platform fc8c0000.usb: Generic Platform OHCI controller
[    0.492680] ehci-platform fc880000.usb: new USB bus registered,
assigned bus number 3
[    0.493271] ohci-platform fc8c0000.usb: new USB bus registered,
assigned bus number 4
[    0.493350] ehci-platform fc800000.usb: USB 2.0 started, EHCI 1.00
[    0.494008] ehci-platform fc880000.usb: irq 70, io mem 0xfc880000
[    0.494520] hub 1-0:1.0: USB hub found
[    0.494542] hub 1-0:1.0: 1 port detected
[    0.494745] ohci-platform fc8c0000.usb: irq 71, io mem 0xfc8c0000
[    0.501157] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req
400000Hz, actual 400000HZ div = 0)
[    0.508986] ehci-platform fc880000.usb: USB 2.0 started, EHCI 1.00
[    0.510401] hub 3-0:1.0: USB hub found
[    0.510779] hub 3-0:1.0: 1 port detected
[    0.542084] hub 2-0:1.0: USB hub found
[    0.542438] hub 2-0:1.0: 1 port detected
[    0.562045] hub 4-0:1.0: USB hub found
[    0.562506] hub 4-0:1.0: 1 port detected
[    0.595247] mmc_host mmc1: Bus speed (slot 0) = 198000000Hz (slot
req 200000000Hz, actual 198000000HZ div = 0)
[    0.748985] usb 1-1: new high-speed USB device number 2 using ehci-platform
[    0.906736] hub 1-1:1.0: USB hub found
[    0.907309] hub 1-1:1.0: 4 ports detected
[    0.923345] dwmmc_rockchip fe2c0000.mmc: Successfully tuned phase to 51
[    0.923962] mmc1: new ultra high speed SDR104 SDHC card at address aaaa
[    0.925357] mmcblk1: mmc1:aaaa SD32G 29.7 GiB
[    0.932234]  mmcblk1: p1 p2 p3
[    0.986929] Freeing initrd memory: 24648K
[    1.000498] Loaded X.509 cert 'sforshee: 00b28ddf47aef9cea7'
[    1.001061] clk: Disabling unused clocks
[    1.001925] ALSA device list:
[    1.002190]   No soundcards found.
[    1.002595] dw-apb-uart feb50000.serial: forbid DMA for kernel console
[    1.004969] platform regulatory.0: Direct firmware load for
regulatory.db failed with error -2
[    1.005732] cfg80211: failed to load regulatory.db
[    1.007117] Freeing unused kernel memory: 9216K
[    1.029098] Run /init as init process
Loading, please wait...
Starting version 247.3-7+deb11u1
[    1.296214] TCPM: state change PORT_RESET_WAIT_OFF ->
SNK_UNATTACHED [delayed 920 ms]
[    1.301978] FUSB: start drp toggling
[    1.302699] TCPM: state change SNK_UNATTACHED -> TOGGLING [rev1 NONE_AMS]
[    1.304531] FUSB: IRQ: VBUS_OK, vbus=On
[    1.307751] FUSB: IRQ: TOGDONE
[    1.314498] FUSB: detected cc1=Rp-3.0, cc2=Open
[    1.314880] FUSB: cc1=Rp-3.0, cc2=Open
[    1.315309] TCPM: state change TOGGLING -> SNK_ATTACH_WAIT [rev1 NONE_AMS]

Begin: Loading essential drivers ... done.
Begin: Running /scripts/init-premount ... done.
Begin: Mounting root file system ... Begin: Running /scripts/local-top ... done.
Begin: Running /scripts/local-premount ... done.
Begin: Will now check root file system ... fsck from util-linux 2.36.1 [/sbin/fsck.ext4 (1) -- /dev/mmcblk1p3] fsck.ext4 -a -C0 /dev/mmcblk1p3

[    1.315655] TCPM: pending state change SNK_ATTACH_WAIT ->
SNK_DEBOUNCED @ 200 ms [rev1 NONE_AMS]
[    1.516302] TCPM: state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED
[delayed 200 ms]
[    1.517159] TCPM: state change SNK_DEBOUNCED -> SNK_ATTACHED [rev1 NONE_AMS]
[    1.517824] TCPM: polarity 0
[    1.518454] TCPM: Requesting mux state 1, usb-role 2, orientation 1
[    1.519744] FUSB: pd header := Sink, Device
[    1.520320] TCPM: state change SNK_ATTACHED -> SNK_STARTUP [rev1 NONE_AMS]
[    1.520745] TCPM: state change SNK_STARTUP -> SNK_DISCOVERY [rev3 NONE_AMS]
[    1.521388] TCPM: Setting voltage/current limit 5000 mV 3000 mA
[    1.522016] TCPM: vbus=0 charge:=1
[    1.522547] FUSB: vbus is already Off
ram: recovering journal
[    1.522857] TCPM: state change SNK_DISCOVERY ->
SNK_WAIT_CAPABILITIES [rev3 NONE_AMS]
[    1.528181] FUSB: pd := on
ram: clean, 1521/594512 files, 74956/2374144 blocks
done.
[    1.528977] TCPM: pending state change SNK_WAIT_CAPABILITIES ->
SNK_SOFT_RESET @ 310 ms [rev3 NONE_AMS]
[    1.568848] EXT4-fs (mmcblk1p3): mounted filesystem

b550eef1-48e4-4bac-9312-2ea23c77a372 r/w with ordered data mode. Quota
mode: none.
done.
Begin: Running /scripts/local-bottom ... done.
Begin: Running /scripts/init-bottom ... done.

[    1.650560] EXT4-fs (mmcblk1p3): re-mounted

b550eef1-48e4-4bac-9312-2ea23c77a372 r/w. Quota mode: none. Starting syslogd: OK
Starting klogd: OK
Running sysctl: OK
Initializing random number generator: OK
Saving random seed: [ 1.839341] TCPM: state change
SNK_WAIT_CAPABILITIES -> SNK_SOFT_RESET [delayed 310 ms]

[    1.839362] TCPM: AMS SOFT_RESET_AMS start
[    1.840087] TCPM: state change SNK_SOFT_RESET -> AMS_START [rev3
SOFT_RESET_AMS]
[    1.840461] TCPM: state change AMS_START -> SOFT_RESET_SEND [rev3
SOFT_RESET_AMS]
[    1.841161] TCPM: PD TX, header: 0x8d
[    1.844492] FUSB: sending PD message header: 8d
[    1.844836] FUSB: sending PD message len: 0
[    1.847576] FUSB: IRQ: BC_LVL, handler pending
[    1.847964] FUSB: IRQ: PD tx success
[    1.850111] FUSB: PD message header: 161
[    1.850445] FUSB: PD message len: 0
[    1.850802] TCPM: PD TX complete, status: 0
[    1.851211] TCPM: pending state change SOFT_RESET_SEND ->
HARD_RESET_SEND @ 60 ms [rev3 SOFT_RESET_AMS]
[    1.853070] FUSB: IRQ: BC_LVL, handler pending
[    1.853914] FUSB: IRQ: PD sent good CRC
[    1.856233] FUSB: PD message header: 1a3
[    1.856590] FUSB: PD message len: 0
[    1.857065] TCPM: PD RX, header: 0x1a3 [1]
[    1.857402] TCPM: AMS SOFT_RESET_AMS finished
[    1.857774] TCPM: state change SOFT_RESET_SEND ->
SNK_WAIT_CAPABILITIES [rev3 NONE_AMS]
[    1.864223] FUSB: pd := on
[    1.864967] TCPM: pending state change SNK_WAIT_CAPABILITIES ->
HARD_RESET_SEND @ 310 ms [rev3 NONE_AMS]
[    1.877754] FUSB: IRQ: BC_LVL, handler pending
[    1.878606] FUSB: IRQ: PD received hardreset
[    1.879902] TCPM: state change SNK_WAIT_CAPABILITIES ->
HARD_RESET_START [rev3 HARD_RESET]
[    1.887331] FUSB: pd := off
[    1.888078] TCPM: state change HARD_RESET_START ->
SNK_HARD_RESET_SINK_OFF [rev3 HARD_RESET]
[    1.888339] TCPM: vconn:=0
[    1.889118] FUSB: vconn is already Off
[    1.889370] TCPM: Requesting mux state 1, usb-role 2, orientation 1
[    1.890621] FUSB: pd header := Sink, Device

Look like Frank sent some patch related(not 100% sure) to this in linux-usb [1], but the result is the same even with this change. I'm still debugging tcpm, any pointers on this would be useful.

[1] https://patchwork.kernel.org/project/linux-usb/patch/[email protected]/

Thanks,
Jagan.

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

#8 lore raw

Just to update you guys with my findings. I spent close to two weeks debugging this with a half dozen power supplies.

fusb302 needs a delay of 200us between setting the power registers and checking vbus. Anything less results in invalid initialization.

Quoted 8 lines · Da Xue
Quoted 6 lines · Frank Wang

Hi Jagan,

In my view, it is the problem that the fusb302 missed the Source caps of PD charger (I assume the charger is normal), maybe the RX interrupt was missed or its FIFO had been touched.

The root of the problem is that some power supply MCUs will crap out if not negotiated within a certain time frame and return invalid data.

Req for source caps do not work. Soft reset do not work. Only a hard reset can fix it. So this automatically rule out powering non-battery devices with these "broken" PD PSUs unless someone have the exact code for those state machines to see if there's a workaround.

The TCPM code does not request source caps so I added a stage for doing that as a backup for WAIT_CAPS. This reduces the necessity of SOFT_RESET.
Some PSUs will return bad messages that cause the state machine to jump to error recovery. I had to filter out these specific messages or else hit warnings since source capabilities gets redefined without being cleared.
I had to clear capabilities when state jumps from NEGO_CAPS to HARD_RESET_START.

Quoted 6 lines

I can confirm these findings. After logging the flushes, I see multiple tcpm states that experience flushes where relevant PDOs sent by the chargers are flushed instead of acted upon due to the timing. This causes multiple failures in the state machine resulting in HARD_RESETs.

So the only way I managed to get things semi-working was to remove the RX flush when turning RX on. Without removing the RX flush, the PD profiles get flushed by WAIT_CAP.

Hope this helps someone.

Quoted 294 lines

Since the driver is not privy to the tcpm_states, do we add explicit flush functions to TCPM and split them from set_pd_rx?

Or is there any implementation suggestions instead of hacking away at TCPM before set_pd_rx to check state.

Quoted 287 lines

The patch you mentioned ([1] in your mail) may be able to fix your issue if the FIFO of the fusb302 is flushed by set_pd_rx() invoked in TCPM, but it has no effort for RX interrupt missed.

Anyway, It is the defect of the fusb302 driver and should be fixed in it . Hope my analysis is helpful to you.

BR,
Frank

Quoted 273 lines · Jagan Teki

+ Frank
+ Guenter
+ linux-usb
(for any suggestions)

Quoted 30 lines · Heiko Stuebner

Am Dienstag, 25. Juli 2023, 10:02:21 CEST schrieb Christopher Obbard:

Quoted 29 lines

Hi Da, Jagan,

Quoted 16 lines · Da Xue
Quoted 14 lines · Christopher Obbard
Quoted 13 lines · Jagan Teki

The power supply circuit in ROC-RK3399-PC is

Power Supply (or PPS) => FUSB => MP8859

VUBS pin of FUSB and IN pin of MP8859 is supplied via TYPEC_IN.

The MP8859 operated with 2.8V to 22V supply voltage and typical applications this supply can be 12V.

This patch is trying to support the PD by changing the FUSB VBUS supply to 12V and tune the I2C7 timings from downstream kernel.

Tested with PD3.0 PPS with supply voltages of 12V/3A and 20V/5A.

Quoted 8 lines
Quoted 5 lines

Hi Jagan,

This series works fine with a "dumb" (no PD negotiation) 5.1V Raspberry Pi PSU.

It also works fine with a Dell 45W USB-C Laptop Power Supply (model AA45NM170) which provides 5V@3A,9V@3A,15V@3A,[email protected], where Linux master fails and just tells the USB-PD PSU to power-off.

I think this depends on the recent Rockchip TCPM changes. FUSB302 has been a pain in the USB or else this platform would have launched a lot longer ago.

Sorry, I was testing this patch on top of next-20230724 which includes https://patchwork.kernel.org/project/linux-rockchip/list/?series=757752:

Quoted 7 lines
Quoted 4 lines

8be558dcffe69b078b34b1fa93b82acaf4ce4957 ("usb: typec: tcpm: add get max power support") 1e35f074399dece73d5df11847d4a0d7a6f49434 ("usb: typec: tcpm: fix cc role at port reset")

Can you check if I am missing any other patches ?

so I guess the question is, are there any new clues and/or does this patch make the situaton better or worse for the roc-rk3399-pc board as it stands now.

It is working with PD3.0 (EHO 100W GaN) but seems to be inconsistent. For the non-working cases, I did check in RK3399 and RK3588 designs the PD run machine setup in tcpm with respect to the fusb302 chip receiving hard reset due to BC_LVL, handler pending.

Here, is the sample log (please don't mind, the log combined with other functions as I've enabled the static tcpm logs)

[    0.342297] FUSB: sw reset
[    0.346084] FUSB: fusb302 device ID: 0x91
[    0.346331] TCPM: Setting usb_comm capable false
[    0.351365] FUSB: pd := off
[    0.351772] FUSB: vbus is already Off
[    0.352019] FUSB: charge is already Off
[    0.352342] FUSB: vconn is already Off
[    0.352680] TCPM: Setting voltage/current limit 0 mV 0 mA
[    0.353023] TCPM: polarity 0
[    0.353497] TCPM: Requesting mux state 0, usb-role 0, orientation 0
[    0.354544] FUSB: pd header := Sink, Device
[    0.355097] TCPM: state change INVALID_STATE -> SNK_UNATTACHED
[rev1 NONE_AMS]
[    0.355469] FUSB: cc1=Open, cc2=Open
[    0.356107] TCPM: state change SNK_UNATTACHED -> PORT_RESET [rev1 NONE_AMS]
[    0.356424] TCPM: 1-0022: registered
[    0.356428] TCPM: Setting usb_comm capable false
[    0.357964] i2c 3-0011: Fixed dependency cycle(s) with
/i2s@fe470000/port/endpoint
[    0.362173] FUSB: pd := off
[    0.362182] FUSB: vbus is already Off
[    0.362445] FUSB: charge is already Off
[    0.362778] FUSB: vconn is already Off
[    0.363126] TCPM: Setting voltage/current limit 0 mV 0 mA
[    0.363479] TCPM: polarity 0
[    0.363970] TCPM: Requesting mux state 0, usb-role 0, orientation 0
[    0.364339] sdhci: Secure Digital Host Controller Interface driver
[    0.365298] FUSB: pd header := Sink, Device
[    0.365440] sdhci: Copyright(c) Pierre Ossman
[    0.366110] Synopsys Designware Multimedia Card Interface Driver
[    0.366222] TCPM: cc:=2
[    0.366758] FUSB: cc := Rd
[    0.367564] sdhci-pltfm: SDHCI platform and OF driver helper
[    0.370341] ledtrig-cpu: registered to indicate activity on CPUs
[    0.371295] scmi_protocol scmi_dev.1: Enabled polling mode TX
channel - prot_id:16
[    0.372067] arm-scmi firmware:scmi: SCMI Notifications - Core Enabled.
[    0.372670] arm-scmi firmware:scmi: SCMI Protocol v2.0 'rockchip:'
Firmware version 0x0
[    0.374620] TCPM: pending state change PORT_RESET ->
PORT_RESET_WAIT_OFF @ 100 ms [rev1 NONE_AMS]
[    0.374638] TCPM: state change PORT_RESET -> PORT_RESET_WAIT_OFF
[delayed 100 ms]
[    0.375435] TCPM: pending state change PORT_RESET_WAIT_OFF ->
SNK_UNATTACHED @ 920 ms [rev1 NONE_AMS]
[    0.379490] SMCCC: SOC_ID: ARCH_SOC_ID not implemented, skipping ....
[    0.381481] usbcore: registered new interface driver usbhid
[    0.381972] usbhid: USB HID core driver
[    0.385775] hw perfevents: enabled with armv8_cortex_a55 PMU
driver, 7 counters available
[    0.387875] hw perfevents: enabled with armv8_cortex_a76 PMU
driver, 7 counters available
[    0.390529] optee: probing for conduit method.
[    0.390926] optee: api uid mismatch
[    0.391234] optee: probe of firmware:optee failed with error -22
[    0.394071] NET: Registered PF_PACKET protocol family
[    0.394571] 9pnet: Installing 9P2000 support
[    0.394986] Key type dns_resolver registered
[    0.399904] registered taskstats version 1
[    0.400365] Loading compiled-in X.509 certificates
[    0.400933] mmc0: SDHCI controller on fe2e0000.mmc [fe2e0000.mmc] using ADMA
[    0.407446] mmc0: Failed to initialize a non-removable card
[    0.442034] rk808-regulator rk808-regulator.1.auto: there is no dvs0 gpio
[    0.443181] rk808-regulator rk808-regulator.1.auto: there is no dvs1 gpio
[    0.448771] vdd_2v0_pldo_s3: Bringing 1975000uV into 2000000-2000000uV
[    0.450335] vcc_3v3_s3: Bringing 3275000uV into 3300000-3300000uV
[    0.452301] vcc_1v8_s3: Bringing 1775000uV into 1800000-1800000uV
[    0.475134] ehci-platform fc800000.usb: EHCI Host Controller
[    0.475656] ehci-platform fc800000.usb: new USB bus registered,
assigned bus number 1
[    0.476412] ehci-platform fc800000.usb: irq 67, io mem 0xfc800000
[    0.476870] ohci-platform fc840000.usb: Generic Platform OHCI controller
[    0.477563] ohci-platform fc840000.usb: new USB bus registered,
assigned bus number 2
[    0.478298] ohci-platform fc840000.usb: irq 68, io mem 0xfc840000
[    0.480664] fan53555-regulator 0-0042: FAN53555 Option[10] Rev[1] Detected!
[    0.481424] fan53555-regulator 0-0043: FAN53555 Option[10] Rev[1] Detected!
[    0.484586] dwmmc_rockchip fe2c0000.mmc: IDMAC supports 32-bit address mode.
[    0.485274] dwmmc_rockchip fe2c0000.mmc: Using internal DMA controller.
[    0.485875] dwmmc_rockchip fe2c0000.mmc: Version ID is 270a
[    0.486417] dwmmc_rockchip fe2c0000.mmc: DW MMC controller at irq
69,32 bit host data width,256 deep fifo
[    0.488188] dwmmc_rockchip fe2c0000.mmc: Got CD GPIO
[    0.491377] cfg80211: Loading compiled-in X.509 certificates for
regulatory database
[    0.492171] ehci-platform fc880000.usb: EHCI Host Controller
[    0.492179] ohci-platform fc8c0000.usb: Generic Platform OHCI controller
[    0.492680] ehci-platform fc880000.usb: new USB bus registered,
assigned bus number 3
[    0.493271] ohci-platform fc8c0000.usb: new USB bus registered,
assigned bus number 4
[    0.493350] ehci-platform fc800000.usb: USB 2.0 started, EHCI 1.00
[    0.494008] ehci-platform fc880000.usb: irq 70, io mem 0xfc880000
[    0.494520] hub 1-0:1.0: USB hub found
[    0.494542] hub 1-0:1.0: 1 port detected
[    0.494745] ohci-platform fc8c0000.usb: irq 71, io mem 0xfc8c0000
[    0.501157] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req
400000Hz, actual 400000HZ div = 0)
[    0.508986] ehci-platform fc880000.usb: USB 2.0 started, EHCI 1.00
[    0.510401] hub 3-0:1.0: USB hub found
[    0.510779] hub 3-0:1.0: 1 port detected
[    0.542084] hub 2-0:1.0: USB hub found
[    0.542438] hub 2-0:1.0: 1 port detected
[    0.562045] hub 4-0:1.0: USB hub found
[    0.562506] hub 4-0:1.0: 1 port detected
[    0.595247] mmc_host mmc1: Bus speed (slot 0) = 198000000Hz (slot
req 200000000Hz, actual 198000000HZ div = 0)
[    0.748985] usb 1-1: new high-speed USB device number 2 using ehci-platform
[    0.906736] hub 1-1:1.0: USB hub found
[    0.907309] hub 1-1:1.0: 4 ports detected
[    0.923345] dwmmc_rockchip fe2c0000.mmc: Successfully tuned phase to 51
[    0.923962] mmc1: new ultra high speed SDR104 SDHC card at address aaaa
[    0.925357] mmcblk1: mmc1:aaaa SD32G 29.7 GiB
[    0.932234]  mmcblk1: p1 p2 p3
[    0.986929] Freeing initrd memory: 24648K
[    1.000498] Loaded X.509 cert 'sforshee: 00b28ddf47aef9cea7'
[    1.001061] clk: Disabling unused clocks
[    1.001925] ALSA device list:
[    1.002190]   No soundcards found.
[    1.002595] dw-apb-uart feb50000.serial: forbid DMA for kernel console
[    1.004969] platform regulatory.0: Direct firmware load for
regulatory.db failed with error -2
[    1.005732] cfg80211: failed to load regulatory.db
[    1.007117] Freeing unused kernel memory: 9216K
[    1.029098] Run /init as init process
Loading, please wait...
Starting version 247.3-7+deb11u1
[    1.296214] TCPM: state change PORT_RESET_WAIT_OFF ->
SNK_UNATTACHED [delayed 920 ms]
[    1.301978] FUSB: start drp toggling
[    1.302699] TCPM: state change SNK_UNATTACHED -> TOGGLING [rev1 NONE_AMS]
[    1.304531] FUSB: IRQ: VBUS_OK, vbus=On
[    1.307751] FUSB: IRQ: TOGDONE
[    1.314498] FUSB: detected cc1=Rp-3.0, cc2=Open
[    1.314880] FUSB: cc1=Rp-3.0, cc2=Open
[    1.315309] TCPM: state change TOGGLING -> SNK_ATTACH_WAIT [rev1 NONE_AMS]

Begin: Loading essential drivers ... done.
Begin: Running /scripts/init-premount ... done.
Begin: Mounting root file system ... Begin: Running /scripts/local-top ... done.
Begin: Running /scripts/local-premount ... done.
Begin: Will now check root file system ... fsck from util-linux 2.36.1 [/sbin/fsck.ext4 (1) -- /dev/mmcblk1p3] fsck.ext4 -a -C0 /dev/mmcblk1p3

[    1.315655] TCPM: pending state change SNK_ATTACH_WAIT ->
SNK_DEBOUNCED @ 200 ms [rev1 NONE_AMS]
[    1.516302] TCPM: state change SNK_ATTACH_WAIT -> SNK_DEBOUNCED
[delayed 200 ms]
[    1.517159] TCPM: state change SNK_DEBOUNCED -> SNK_ATTACHED [rev1 NONE_AMS]
[    1.517824] TCPM: polarity 0
[    1.518454] TCPM: Requesting mux state 1, usb-role 2, orientation 1
[    1.519744] FUSB: pd header := Sink, Device
[    1.520320] TCPM: state change SNK_ATTACHED -> SNK_STARTUP [rev1 NONE_AMS]
[    1.520745] TCPM: state change SNK_STARTUP -> SNK_DISCOVERY [rev3 NONE_AMS]
[    1.521388] TCPM: Setting voltage/current limit 5000 mV 3000 mA
[    1.522016] TCPM: vbus=0 charge:=1
[    1.522547] FUSB: vbus is already Off
ram: recovering journal
[    1.522857] TCPM: state change SNK_DISCOVERY ->
SNK_WAIT_CAPABILITIES [rev3 NONE_AMS]
[    1.528181] FUSB: pd := on
ram: clean, 1521/594512 files, 74956/2374144 blocks
done.
[    1.528977] TCPM: pending state change SNK_WAIT_CAPABILITIES ->
SNK_SOFT_RESET @ 310 ms [rev3 NONE_AMS]
[    1.568848] EXT4-fs (mmcblk1p3): mounted filesystem

b550eef1-48e4-4bac-9312-2ea23c77a372 r/w with ordered data mode. Quota
mode: none.
done.
Begin: Running /scripts/local-bottom ... done.
Begin: Running /scripts/init-bottom ... done.

[    1.650560] EXT4-fs (mmcblk1p3): re-mounted

b550eef1-48e4-4bac-9312-2ea23c77a372 r/w. Quota mode: none. Starting syslogd: OK
Starting klogd: OK
Running sysctl: OK
Initializing random number generator: OK
Saving random seed: [ 1.839341] TCPM: state change
SNK_WAIT_CAPABILITIES -> SNK_SOFT_RESET [delayed 310 ms]

[    1.839362] TCPM: AMS SOFT_RESET_AMS start
[    1.840087] TCPM: state change SNK_SOFT_RESET -> AMS_START [rev3
SOFT_RESET_AMS]
[    1.840461] TCPM: state change AMS_START -> SOFT_RESET_SEND [rev3
SOFT_RESET_AMS]
[    1.841161] TCPM: PD TX, header: 0x8d
[    1.844492] FUSB: sending PD message header: 8d
[    1.844836] FUSB: sending PD message len: 0
[    1.847576] FUSB: IRQ: BC_LVL, handler pending
[    1.847964] FUSB: IRQ: PD tx success
[    1.850111] FUSB: PD message header: 161
[    1.850445] FUSB: PD message len: 0
[    1.850802] TCPM: PD TX complete, status: 0
[    1.851211] TCPM: pending state change SOFT_RESET_SEND ->
HARD_RESET_SEND @ 60 ms [rev3 SOFT_RESET_AMS]
[    1.853070] FUSB: IRQ: BC_LVL, handler pending
[    1.853914] FUSB: IRQ: PD sent good CRC
[    1.856233] FUSB: PD message header: 1a3
[    1.856590] FUSB: PD message len: 0
[    1.857065] TCPM: PD RX, header: 0x1a3 [1]
[    1.857402] TCPM: AMS SOFT_RESET_AMS finished
[    1.857774] TCPM: state change SOFT_RESET_SEND ->
SNK_WAIT_CAPABILITIES [rev3 NONE_AMS]
[    1.864223] FUSB: pd := on
[    1.864967] TCPM: pending state change SNK_WAIT_CAPABILITIES ->
HARD_RESET_SEND @ 310 ms [rev3 NONE_AMS]
[    1.877754] FUSB: IRQ: BC_LVL, handler pending
[    1.878606] FUSB: IRQ: PD received hardreset
[    1.879902] TCPM: state change SNK_WAIT_CAPABILITIES ->
HARD_RESET_START [rev3 HARD_RESET]
[    1.887331] FUSB: pd := off
[    1.888078] TCPM: state change HARD_RESET_START ->
SNK_HARD_RESET_SINK_OFF [rev3 HARD_RESET]
[    1.888339] TCPM: vconn:=0
[    1.889118] FUSB: vconn is already Off
[    1.889370] TCPM: Requesting mux state 1, usb-role 2, orientation 1
[    1.890621] FUSB: pd header := Sink, Device

Look like Frank sent some patch related(not 100% sure) to this in linux-usb [1], but the result is the same even with this change. I'm still debugging tcpm, any pointers on this would be useful.

[1] https://patchwork.kernel.org/project/linux-usb/patch/[email protected]/

Thanks,
Jagan.

author
#9 lore raw
In reply to Christopher Obbard

Hi Christopher,

Quoted 14 lines · Christopher Obbard
Quoted 12 lines · Jagan Teki

The power supply circuit in ROC-RK3399-PC is

Power Supply (or PPS) => FUSB => MP8859

VUBS pin of FUSB and IN pin of MP8859 is supplied via TYPEC_IN.

The MP8859 operated with 2.8V to 22V supply voltage and typical applications this supply can be 12V.

This patch is trying to support the PD by changing the FUSB VBUS supply to 12V and tune the I2C7 timings from downstream kernel.

Quoted 7 lines
Quoted 1 line

Tested with PD3.0 PPS with supply voltages of 12V/3A and 20V/5A.

Hi Jagan,

This series works fine with a "dumb" (no PD negotiation) 5.1V Raspberry Pi PSU.

It also works fine with a Dell 45W USB-C Laptop Power Supply (model AA45NM170) which provides 5V@3A,9V@3A,15V@3A,[email protected], where Linux master fails and just tells the USB-PD PSU to power-off.

Thanks for testing.

Quoted 3 lines

It doesn't work with a Lenovo 65W PSU (model ADLX65YLC3D) which provides 5V@2A,9V@2A,15V@3A,[email protected], after negotiation the driver turns the PD PSU off and on again, resetting the board. So it'd be great to get this fixed, but it seems like the sink-pdos should already support this PSU ?

Yes, it is already supported. Seems like some issues on tcpm+fusb302.

Quoted 26 lines

As a side note for full transparency, another issue is that with Linux master or with this patch applied, applying power to the DC_12V header without a USB-C PSU connected (e.g. powering from the POE expansion shield), boot hangs and dumps to an initramfs shell with:

  [    7.411798] random: crng init done
  [   12.568138] platform fe3a0000.usb: deferred probe pending
  [   12.568673] platform sdio-pwrseq: deferred probe pending
  [   12.569162] platform fe3e0000.usb: deferred probe pending
  [   12.569658] platform adc-keys: deferred probe pending
  [   12.570123] i2c 7-0022: deferred probe pending
  [   12.570533] i2c 4-0022: deferred probe pending
  [   12.570944] platform ff940000.hdmi: deferred probe pending
  [   12.571448] platform vcc3v0-sd: deferred probe pending
  [   12.572000] platform vcc1v8-s3: deferred probe pending
  [   12.572475] platform sys-12v: deferred probe pending
  [   12.572933] platform vcc3v3-sys: deferred probe pending
  [   12.573412] platform fe320000.mmc: deferred probe pending
  [   12.573907] platform vcca-0v9: deferred probe pending
  [   12.574371] platform vcc5v0-host-regulator: deferred probe pending
  [   12.574935] platform ff770000.syscon:usb2phy@e450: deferred probe pending
  [   12.575552] platform vcc-vbus-typec1: deferred probe pending
  [   12.576090] platform fe300000.ethernet: deferred probe pending
  [   12.576623] platform vcc-sys: deferred probe pending
  [   12.577080] platform ff770000.syscon:usb2phy@e460: deferred probe pending
  [   12.577697] platform ff320000.syscon:io-domains: deferred probe pending
Quoted 8 lines
  [   12.578298] platform ff770000.syscon:io-domains: deferred probe pending
  [   12.578901] platform fe800000.usb: deferred probe pending
  [   12.579395] platform fe900000.usb: deferred probe pending
  [   12.579904] platform vdd-log: deferred probe pending
  [   12.580362] i2c 0-001b: deferred probe pending
  [   12.580772] i2c 0-0040: deferred probe pending
  [   12.581182] platform cpufreq-dt: deferred probe pending
  [   12.581663] i2c 0-0041: deferred probe pending

I didn't test POE with this change, I will check this.

Thanks,
Jagan.