[PATCH v5] usb: typec: ucsi: Fix busy loop on ASUS VivoBooks

[PATCH v5] usb: typec: ucsi: Fix busy loop on ASUS VivoBooks

[Christian A. Ehrhardt]

If the busy indicator is set, all other fields in CCI should be
clear according to the spec. However, some UCSI implementations do
not follow this rule and report bogus data in CCI along with the
busy indicator. Ignore the contents of CCI if the busy indicator is
set.

If a command timeout is hit it is possible that the EVENT_PENDING
bit is cleared while connector work is still scheduled which can
cause the EVENT_PENDING bit to go out of sync with scheduled connector
work. Check and set the EVENT_PENDING bit on entry to
ucsi_handle_connector_change() to fix this.

Finally, check UCSI_CCI_BUSY before the return code of ->sync_control.
This ensures that the command is cancelled even if ->sync_control
returns an error (most likely -ETIMEDOUT).

Reported-by: Anurag Bijea <icaliberdev@gmail.com>
Closes: https://bugzilla.kernel.org/show_bug.cgi?id=219108
Bisected-by: Christian Heusel <christian@heusel.eu>
Tested-by: Anurag Bijea <icaliberdev@gmail.com>
Fixes: de52aca4d9d5 ("usb: typec: ucsi: Never send a lone connector change ack")
Cc: stable@vger.kernel.org
Signed-off-by: Christian A. Ehrhardt <lk@c--e.de>
---
 V1 -> V2: Split out the ucsi_acpi.c part
 V2 -> V3: Dropped the ucsi_acpi.c part due to conflicts
 V3 -> V4: Additional fix for command cancelling
 V4 -> V5: Rebased onto usb-next

 drivers/usb/typec/ucsi/ucsi.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c
index 35dce4057c25..e0f3925e401b 100644
--- a/drivers/usb/typec/ucsi/ucsi.c
+++ b/drivers/usb/typec/ucsi/ucsi.c
@@ -38,6 +38,10 @@
 
 void ucsi_notify_common(struct ucsi *ucsi, u32 cci)
 {
+	/* Ignore bogus data in CCI if busy indicator is set. */
+	if (cci & UCSI_CCI_BUSY)
+		return;
+
 	if (UCSI_CCI_CONNECTOR(cci))
 		ucsi_connector_change(ucsi, UCSI_CCI_CONNECTOR(cci));
 
@@ -103,15 +107,13 @@ static int ucsi_run_command(struct ucsi *ucsi, u64 command, u32 *cci,
 		return -EINVAL;
 
 	ret = ucsi->ops->sync_control(ucsi, command);
-	if (ret)
-		return ret;
-
-	ret = ucsi->ops->read_cci(ucsi, cci);
-	if (ret)
-		return ret;
+	if (ucsi->ops->read_cci(ucsi, cci))
+		return -EIO;
 
 	if (*cci & UCSI_CCI_BUSY)
 		return ucsi_run_command(ucsi, UCSI_CANCEL, cci, NULL, 0, false) ?: -EBUSY;
+	if (ret)
+		return ret;
 
 	if (!(*cci & UCSI_CCI_COMMAND_COMPLETE))
 		return -EIO;
@@ -1197,6 +1199,10 @@ static void ucsi_handle_connector_change(struct work_struct *work)
 
 	mutex_lock(&con->lock);
 
+	if (!test_and_set_bit(EVENT_PENDING, &ucsi->flags))
+		dev_err_once(ucsi->dev, "%s entered without EVENT_PENDING\n",
+			     __func__);
+
 	command = UCSI_GET_CONNECTOR_STATUS | UCSI_CONNECTOR_NUMBER(con->num);
 
 	ret = ucsi_send_command_common(ucsi, command, &con->status,
-- 
2.43.0

Re: [PATCH v5] usb: typec: ucsi: Fix busy loop on ASUS VivoBooks

[Heikki Krogerus]

On Thu, Sep 12, 2024 at 09:41:32AM +0200, Christian A. Ehrhardt wrote:
> If the busy indicator is set, all other fields in CCI should be
> clear according to the spec. However, some UCSI implementations do
> not follow this rule and report bogus data in CCI along with the
> busy indicator. Ignore the contents of CCI if the busy indicator is
> set.
> 
> If a command timeout is hit it is possible that the EVENT_PENDING
> bit is cleared while connector work is still scheduled which can
> cause the EVENT_PENDING bit to go out of sync with scheduled connector
> work. Check and set the EVENT_PENDING bit on entry to
> ucsi_handle_connector_change() to fix this.
> 
> Finally, check UCSI_CCI_BUSY before the return code of ->sync_control.
> This ensures that the command is cancelled even if ->sync_control
> returns an error (most likely -ETIMEDOUT).
> 
> Reported-by: Anurag Bijea <icaliberdev@gmail.com>
> Closes: https://bugzilla.kernel.org/show_bug.cgi?id=219108
> Bisected-by: Christian Heusel <christian@heusel.eu>
> Tested-by: Anurag Bijea <icaliberdev@gmail.com>
> Fixes: de52aca4d9d5 ("usb: typec: ucsi: Never send a lone connector change ack")
> Cc: stable@vger.kernel.org
> Signed-off-by: Christian A. Ehrhardt <lk@c--e.de>

Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>

> ---
>  V1 -> V2: Split out the ucsi_acpi.c part
>  V2 -> V3: Dropped the ucsi_acpi.c part due to conflicts
>  V3 -> V4: Additional fix for command cancelling
>  V4 -> V5: Rebased onto usb-next
> 
>  drivers/usb/typec/ucsi/ucsi.c | 18 ++++++++++++------
>  1 file changed, 12 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c
> index 35dce4057c25..e0f3925e401b 100644
> --- a/drivers/usb/typec/ucsi/ucsi.c
> +++ b/drivers/usb/typec/ucsi/ucsi.c
> @@ -38,6 +38,10 @@
>  
>  void ucsi_notify_common(struct ucsi *ucsi, u32 cci)
>  {
> +	/* Ignore bogus data in CCI if busy indicator is set. */
> +	if (cci & UCSI_CCI_BUSY)
> +		return;
> +
>  	if (UCSI_CCI_CONNECTOR(cci))
>  		ucsi_connector_change(ucsi, UCSI_CCI_CONNECTOR(cci));
>  
> @@ -103,15 +107,13 @@ static int ucsi_run_command(struct ucsi *ucsi, u64 command, u32 *cci,
>  		return -EINVAL;
>  
>  	ret = ucsi->ops->sync_control(ucsi, command);
> -	if (ret)
> -		return ret;
> -
> -	ret = ucsi->ops->read_cci(ucsi, cci);
> -	if (ret)
> -		return ret;
> +	if (ucsi->ops->read_cci(ucsi, cci))
> +		return -EIO;
>  
>  	if (*cci & UCSI_CCI_BUSY)
>  		return ucsi_run_command(ucsi, UCSI_CANCEL, cci, NULL, 0, false) ?: -EBUSY;
> +	if (ret)
> +		return ret;
>  
>  	if (!(*cci & UCSI_CCI_COMMAND_COMPLETE))
>  		return -EIO;
> @@ -1197,6 +1199,10 @@ static void ucsi_handle_connector_change(struct work_struct *work)
>  
>  	mutex_lock(&con->lock);
>  
> +	if (!test_and_set_bit(EVENT_PENDING, &ucsi->flags))
> +		dev_err_once(ucsi->dev, "%s entered without EVENT_PENDING\n",
> +			     __func__);
> +
>  	command = UCSI_GET_CONNECTOR_STATUS | UCSI_CONNECTOR_NUMBER(con->num);
>  
>  	ret = ucsi_send_command_common(ucsi, command, &con->status,
> -- 
> 2.43.0

-- 
heikki