[v1] Revert "usb: typec: tcpm: fix cc role at port reset"

Message ID 20240117113806.2584341-1-badhri@google.com
State New
Headers
Series [v1] Revert "usb: typec: tcpm: fix cc role at port reset" |

Commit Message

Badhri Jagan Sridharan Jan. 17, 2024, 11:38 a.m. UTC
  This reverts commit 1e35f074399dece73d5df11847d4a0d7a6f49434.

Given that ERROR_RECOVERY calls into PORT_RESET for Hi-Zing
the CC pins, setting CC pins to default state during PORT_RESET
breaks error recovery.

4.5.2.2.2.1 ErrorRecovery State Requirements
The port shall not drive VBUS or VCONN, and shall present a
high-impedance to ground (above zOPEN) on its CC1 and CC2 pins.

Hi-Zing the CC pins is the inteded behavior for PORT_RESET.
CC pins are set to default state after tErrorRecovery in
PORT_RESET_WAIT_OFF.

4.5.2.2.2.2 Exiting From ErrorRecovery State
A Sink shall transition to Unattached.SNK after tErrorRecovery.
A Source shall transition to Unattached.SRC after tErrorRecovery.

Cc: stable@kernel.org
Fixes: 1e35f074399d ("usb: typec: tcpm: fix cc role at port reset")
Signed-off-by: Badhri Jagan Sridharan <badhri@google.com>
---
 drivers/usb/typec/tcpm/tcpm.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)


base-commit: 933bb7b878ddd0f8c094db45551a7daddf806e00
  

Comments

Badhri Jagan Sridharan Jan. 17, 2024, 11:50 a.m. UTC | #1
Hi all,

Please ignore this patch as I sent a V2 which fixes the commit message.

Thanks,
Badhri


On Wed, Jan 17, 2024 at 3:38 AM Badhri Jagan Sridharan
<badhri@google.com> wrote:
>
> This reverts commit 1e35f074399dece73d5df11847d4a0d7a6f49434.
>
> Given that ERROR_RECOVERY calls into PORT_RESET for Hi-Zing
> the CC pins, setting CC pins to default state during PORT_RESET
> breaks error recovery.
>
> 4.5.2.2.2.1 ErrorRecovery State Requirements
> The port shall not drive VBUS or VCONN, and shall present a
> high-impedance to ground (above zOPEN) on its CC1 and CC2 pins.
>
> Hi-Zing the CC pins is the inteded behavior for PORT_RESET.
> CC pins are set to default state after tErrorRecovery in
> PORT_RESET_WAIT_OFF.
>
> 4.5.2.2.2.2 Exiting From ErrorRecovery State
> A Sink shall transition to Unattached.SNK after tErrorRecovery.
> A Source shall transition to Unattached.SRC after tErrorRecovery.
>
> Cc: stable@kernel.org
> Fixes: 1e35f074399d ("usb: typec: tcpm: fix cc role at port reset")
> Signed-off-by: Badhri Jagan Sridharan <badhri@google.com>
> ---
>  drivers/usb/typec/tcpm/tcpm.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c
> index 5945e3a2b0f7..9d410718eaf4 100644
> --- a/drivers/usb/typec/tcpm/tcpm.c
> +++ b/drivers/usb/typec/tcpm/tcpm.c
> @@ -4876,8 +4876,7 @@ static void run_state_machine(struct tcpm_port *port)
>                 break;
>         case PORT_RESET:
>                 tcpm_reset_port(port);
> -               tcpm_set_cc(port, tcpm_default_state(port) == SNK_UNATTACHED ?
> -                           TYPEC_CC_RD : tcpm_rp_cc(port));
> +               tcpm_set_cc(port, TYPEC_CC_OPEN);
>                 tcpm_set_state(port, PORT_RESET_WAIT_OFF,
>                                PD_T_ERROR_RECOVERY);
>                 break;
>
> base-commit: 933bb7b878ddd0f8c094db45551a7daddf806e00
> --
> 2.43.0.429.g432eaa2c6b-goog
>
  
Gábor Stefanik Jan. 17, 2024, 4:07 p.m. UTC | #2
This will break operation of batteryless devices relying on a USB
Type-C port for their power needs, as the port reset upon controller
initialization will cause power to be cut to the device, resulting in
a boot loop.
Devices using the FUSB302C port controller are especially severely
affected, as upon losing power, this controller can retain CC states
for a very long time (potentially forever if some parasitic source of
power is present), requiring a full mechanical disconnect-reconnect
cycle before the device receives power again.

While the USB Type C specification does require this behavior, I would
consider this an oversight in the standard (perhaps left over from
when USB Power Delivery was still going to be USB Battery Charging
2.0).

Badhri Jagan Sridharan <badhri@google.com> ezt írta (időpont: 2024.
jan. 17., Sze, 12:38):
>
> This reverts commit 1e35f074399dece73d5df11847d4a0d7a6f49434.
>
> Given that ERROR_RECOVERY calls into PORT_RESET for Hi-Zing
> the CC pins, setting CC pins to default state during PORT_RESET
> breaks error recovery.
>
> 4.5.2.2.2.1 ErrorRecovery State Requirements
> The port shall not drive VBUS or VCONN, and shall present a
> high-impedance to ground (above zOPEN) on its CC1 and CC2 pins.
>
> Hi-Zing the CC pins is the inteded behavior for PORT_RESET.
> CC pins are set to default state after tErrorRecovery in
> PORT_RESET_WAIT_OFF.
>
> 4.5.2.2.2.2 Exiting From ErrorRecovery State
> A Sink shall transition to Unattached.SNK after tErrorRecovery.
> A Source shall transition to Unattached.SRC after tErrorRecovery.
>
> Cc: stable@kernel.org
> Fixes: 1e35f074399d ("usb: typec: tcpm: fix cc role at port reset")
> Signed-off-by: Badhri Jagan Sridharan <badhri@google.com>
> ---
>  drivers/usb/typec/tcpm/tcpm.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c
> index 5945e3a2b0f7..9d410718eaf4 100644
> --- a/drivers/usb/typec/tcpm/tcpm.c
> +++ b/drivers/usb/typec/tcpm/tcpm.c
> @@ -4876,8 +4876,7 @@ static void run_state_machine(struct tcpm_port *port)
>                 break;
>         case PORT_RESET:
>                 tcpm_reset_port(port);
> -               tcpm_set_cc(port, tcpm_default_state(port) == SNK_UNATTACHED ?
> -                           TYPEC_CC_RD : tcpm_rp_cc(port));
> +               tcpm_set_cc(port, TYPEC_CC_OPEN);
>                 tcpm_set_state(port, PORT_RESET_WAIT_OFF,
>                                PD_T_ERROR_RECOVERY);
>                 break;
>
> base-commit: 933bb7b878ddd0f8c094db45551a7daddf806e00
> --
> 2.43.0.429.g432eaa2c6b-goog
>
>
  
Badhri Jagan Sridharan Jan. 23, 2024, 11:58 a.m. UTC | #3
Hi Gabor,

While HI-Zing CC pins disrupts power for batteryless devices, not
Hi-Zing CC pins would prevent clean error recovery for self powered
devices.
Hi-Zing CC pins would make the port partner recognize it as disconnect
and will result in bringup the connection back cleanly.

How about leveraging "self-powered" device tree property and Hi-Zing
CC pins only when using "self-powered" ?

Regards,
Badhri

On Wed, Jan 17, 2024 at 8:07 AM Gábor Stefanik <netrolller.3d@gmail.com> wrote:
>
> This will break operation of batteryless devices relying on a USB
> Type-C port for their power needs, as the port reset upon controller
> initialization will cause power to be cut to the device, resulting in
> a boot loop.
> Devices using the FUSB302C port controller are especially severely
> affected, as upon losing power, this controller can retain CC states
> for a very long time (potentially forever if some parasitic source of
> power is present), requiring a full mechanical disconnect-reconnect
> cycle before the device receives power again.
>
> While the USB Type C specification does require this behavior, I would
> consider this an oversight in the standard (perhaps left over from
> when USB Power Delivery was still going to be USB Battery Charging
> 2.0).
>
> Badhri Jagan Sridharan <badhri@google.com> ezt írta (időpont: 2024.
> jan. 17., Sze, 12:38):
> >
> > This reverts commit 1e35f074399dece73d5df11847d4a0d7a6f49434.
> >
> > Given that ERROR_RECOVERY calls into PORT_RESET for Hi-Zing
> > the CC pins, setting CC pins to default state during PORT_RESET
> > breaks error recovery.
> >
> > 4.5.2.2.2.1 ErrorRecovery State Requirements
> > The port shall not drive VBUS or VCONN, and shall present a
> > high-impedance to ground (above zOPEN) on its CC1 and CC2 pins.
> >
> > Hi-Zing the CC pins is the inteded behavior for PORT_RESET.
> > CC pins are set to default state after tErrorRecovery in
> > PORT_RESET_WAIT_OFF.
> >
> > 4.5.2.2.2.2 Exiting From ErrorRecovery State
> > A Sink shall transition to Unattached.SNK after tErrorRecovery.
> > A Source shall transition to Unattached.SRC after tErrorRecovery.
> >
> > Cc: stable@kernel.org
> > Fixes: 1e35f074399d ("usb: typec: tcpm: fix cc role at port reset")
> > Signed-off-by: Badhri Jagan Sridharan <badhri@google.com>
> > ---
> >  drivers/usb/typec/tcpm/tcpm.c | 3 +--
> >  1 file changed, 1 insertion(+), 2 deletions(-)
> >
> > diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c
> > index 5945e3a2b0f7..9d410718eaf4 100644
> > --- a/drivers/usb/typec/tcpm/tcpm.c
> > +++ b/drivers/usb/typec/tcpm/tcpm.c
> > @@ -4876,8 +4876,7 @@ static void run_state_machine(struct tcpm_port *port)
> >                 break;
> >         case PORT_RESET:
> >                 tcpm_reset_port(port);
> > -               tcpm_set_cc(port, tcpm_default_state(port) == SNK_UNATTACHED ?
> > -                           TYPEC_CC_RD : tcpm_rp_cc(port));
> > +               tcpm_set_cc(port, TYPEC_CC_OPEN);
> >                 tcpm_set_state(port, PORT_RESET_WAIT_OFF,
> >                                PD_T_ERROR_RECOVERY);
> >                 break;
> >
> > base-commit: 933bb7b878ddd0f8c094db45551a7daddf806e00
> > --
> > 2.43.0.429.g432eaa2c6b-goog
> >
> >
  

Patch

diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c
index 5945e3a2b0f7..9d410718eaf4 100644
--- a/drivers/usb/typec/tcpm/tcpm.c
+++ b/drivers/usb/typec/tcpm/tcpm.c
@@ -4876,8 +4876,7 @@  static void run_state_machine(struct tcpm_port *port)
 		break;
 	case PORT_RESET:
 		tcpm_reset_port(port);
-		tcpm_set_cc(port, tcpm_default_state(port) == SNK_UNATTACHED ?
-			    TYPEC_CC_RD : tcpm_rp_cc(port));
+		tcpm_set_cc(port, TYPEC_CC_OPEN);
 		tcpm_set_state(port, PORT_RESET_WAIT_OFF,
 			       PD_T_ERROR_RECOVERY);
 		break;