Hi Prashanth,
On Fri, Sep 15, 2023 at 07:10:25PM +0530, Prashanth K wrote:
> On 15-09-23 06:02 pm, Heikki Krogerus wrote:
> > On Tue, Sep 12, 2023 at 04:37:47PM +0530, Prashanth K wrote:
> > >
> > >
> > > On 11-09-23 06:19 pm, Heikki Krogerus wrote:
> > > > On Mon, Sep 11, 2023 at 02:34:15PM +0530, Prashanth K wrote:
> > > > > Currently if ucsi_send_command() fails, then we bail out without
> > > > > clearing EVENT_PENDING flag. So when the next connector change
> > > > > event comes, ucsi_connector_change() won't queue the con->work,
> > > > > because of which none of the new events will be processed.
> > > > >
> > > > > Fix this by clearing EVENT_PENDING flag if ucsi_send_command()
> > > > > fails.
> > > > >
> > > > > Cc: <[email protected]> # 5.16
> > > > > Fixes: 512df95b9432 ("usb: typec: ucsi: Better fix for missing unplug events issue")
> > > > > Signed-off-by: Prashanth K <[email protected]>
> > > > > ---
> > > > > drivers/usb/typec/ucsi/ucsi.c | 1 +
> > > > > 1 file changed, 1 insertion(+)
> > > > >
> > > > > diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c
> > > > > index c6dfe3d..509c67c 100644
> > > > > --- a/drivers/usb/typec/ucsi/ucsi.c
> > > > > +++ b/drivers/usb/typec/ucsi/ucsi.c
> > > > > @@ -884,6 +884,7 @@ static void ucsi_handle_connector_change(struct work_struct *work)
> > > > > if (ret < 0) {
> > > > > dev_err(ucsi->dev, "%s: GET_CONNECTOR_STATUS failed (%d)\n",
> > > > > __func__, ret);
> > > > > + clear_bit(EVENT_PENDING, &con->ucsi->flags);
> > > > > goto out_unlock;
> > > > > }
> > > >
> > > > I think it would be better to just move that label (out_unlock) above
> > > > the point where clear_bit() is already called instead of separately
> > > > calling it like that. That way the Connector Change Event will
> > > > also get acknowledged.
> > > Do we really need to ACK in this case since we didn't process the current
> > > connector change event
> >
> > You won't get the next event before the first one was ACK'd, right?
> >
>
> The spec says that we need to ACK if we received AND processed a CCI
>
> "4.5.4 Acknowledge Command Completion and/or Change Indication (R)
> This command is used to acknowledge to the PPM that the OPM received and
> processed a Command Completion and/or a Connector Change Indication."
>
> And here in this case, we have received, but not processed the event.
> So I'm not really sure what to do here in this case. If we don't send an
> ACK, then would the PPM think that OPM is not responding and reset it?
> OR would it resend the previous event again since we didn't ACK?
Every PPM behaves differently.
Did you actually see that happening - GET_CONNECTOR_STATUS failed? Can
you reproduce it?
thanks,
--
heikki
On 15-09-23 07:27 pm, Heikki Krogerus wrote:
> Hi Prashanth,
>
> On Fri, Sep 15, 2023 at 07:10:25PM +0530, Prashanth K wrote:
>> On 15-09-23 06:02 pm, Heikki Krogerus wrote:
>>> On Tue, Sep 12, 2023 at 04:37:47PM +0530, Prashanth K wrote:
>>>>
>>>>
>>>> On 11-09-23 06:19 pm, Heikki Krogerus wrote:
>>>>> On Mon, Sep 11, 2023 at 02:34:15PM +0530, Prashanth K wrote:
>>>>>> Currently if ucsi_send_command() fails, then we bail out without
>>>>>> clearing EVENT_PENDING flag. So when the next connector change
>>>>>> event comes, ucsi_connector_change() won't queue the con->work,
>>>>>> because of which none of the new events will be processed.
>>>>>>
>>>>>> Fix this by clearing EVENT_PENDING flag if ucsi_send_command()
>>>>>> fails.
>>>>>>
>>>>>> Cc: <[email protected]> # 5.16
>>>>>> Fixes: 512df95b9432 ("usb: typec: ucsi: Better fix for missing unplug events issue")
>>>>>> Signed-off-by: Prashanth K <[email protected]>
>>>>>> ---
>>>>>> drivers/usb/typec/ucsi/ucsi.c | 1 +
>>>>>> 1 file changed, 1 insertion(+)
>>>>>>
>>>>>> diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c
>>>>>> index c6dfe3d..509c67c 100644
>>>>>> --- a/drivers/usb/typec/ucsi/ucsi.c
>>>>>> +++ b/drivers/usb/typec/ucsi/ucsi.c
>>>>>> @@ -884,6 +884,7 @@ static void ucsi_handle_connector_change(struct work_struct *work)
>>>>>> if (ret < 0) {
>>>>>> dev_err(ucsi->dev, "%s: GET_CONNECTOR_STATUS failed (%d)\n",
>>>>>> __func__, ret);
>>>>>> + clear_bit(EVENT_PENDING, &con->ucsi->flags);
>>>>>> goto out_unlock;
>>>>>> }
>>>>>
>>>>> I think it would be better to just move that label (out_unlock) above
>>>>> the point where clear_bit() is already called instead of separately
>>>>> calling it like that. That way the Connector Change Event will
>>>>> also get acknowledged.
>>>> Do we really need to ACK in this case since we didn't process the current
>>>> connector change event
>>>
>>> You won't get the next event before the first one was ACK'd, right?
>>>
>>
>> The spec says that we need to ACK if we received AND processed a CCI
>>
>> "4.5.4 Acknowledge Command Completion and/or Change Indication (R)
>> This command is used to acknowledge to the PPM that the OPM received and
>> processed a Command Completion and/or a Connector Change Indication."
>>
>> And here in this case, we have received, but not processed the event.
>> So I'm not really sure what to do here in this case. If we don't send an
>> ACK, then would the PPM think that OPM is not responding and reset it?
>> OR would it resend the previous event again since we didn't ACK?
>
> Every PPM behaves differently.
>
> Did you actually see that happening - GET_CONNECTOR_STATUS failed? Can
> you reproduce it?
>
Yea we actually hit the issue once where GET_CONNECTOR_STATUS failed and
subsequent events didn't get queued since EVENT_PENDING wasn't cleared.
Its not easily reproducible (<1%) though.
[4948:kworker/0:3]UCSI: ucsi_qti_glink_write: timed out
[4948:kworker/0:3]ucsi_glink soc:qcom,pmic_glink:qcom,ucsi:
ucsi_handle_connector_change: GET_CONNECTOR_STATUS failed (-110)
Regards,
Prashanth K
On Sat, Sep 16, 2023 at 01:58:30PM +0530, Prashanth K wrote:
>
>
> On 15-09-23 07:27 pm, Heikki Krogerus wrote:
> > Hi Prashanth,
> >
> > On Fri, Sep 15, 2023 at 07:10:25PM +0530, Prashanth K wrote:
> > > On 15-09-23 06:02 pm, Heikki Krogerus wrote:
> > > > On Tue, Sep 12, 2023 at 04:37:47PM +0530, Prashanth K wrote:
> > > > >
> > > > >
> > > > > On 11-09-23 06:19 pm, Heikki Krogerus wrote:
> > > > > > On Mon, Sep 11, 2023 at 02:34:15PM +0530, Prashanth K wrote:
> > > > > > > Currently if ucsi_send_command() fails, then we bail out without
> > > > > > > clearing EVENT_PENDING flag. So when the next connector change
> > > > > > > event comes, ucsi_connector_change() won't queue the con->work,
> > > > > > > because of which none of the new events will be processed.
> > > > > > >
> > > > > > > Fix this by clearing EVENT_PENDING flag if ucsi_send_command()
> > > > > > > fails.
> > > > > > >
> > > > > > > Cc: <[email protected]> # 5.16
> > > > > > > Fixes: 512df95b9432 ("usb: typec: ucsi: Better fix for missing unplug events issue")
> > > > > > > Signed-off-by: Prashanth K <[email protected]>
> > > > > > > ---
> > > > > > > drivers/usb/typec/ucsi/ucsi.c | 1 +
> > > > > > > 1 file changed, 1 insertion(+)
> > > > > > >
> > > > > > > diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c
> > > > > > > index c6dfe3d..509c67c 100644
> > > > > > > --- a/drivers/usb/typec/ucsi/ucsi.c
> > > > > > > +++ b/drivers/usb/typec/ucsi/ucsi.c
> > > > > > > @@ -884,6 +884,7 @@ static void ucsi_handle_connector_change(struct work_struct *work)
> > > > > > > if (ret < 0) {
> > > > > > > dev_err(ucsi->dev, "%s: GET_CONNECTOR_STATUS failed (%d)\n",
> > > > > > > __func__, ret);
> > > > > > > + clear_bit(EVENT_PENDING, &con->ucsi->flags);
> > > > > > > goto out_unlock;
> > > > > > > }
> > > > > >
> > > > > > I think it would be better to just move that label (out_unlock) above
> > > > > > the point where clear_bit() is already called instead of separately
> > > > > > calling it like that. That way the Connector Change Event will
> > > > > > also get acknowledged.
> > > > > Do we really need to ACK in this case since we didn't process the current
> > > > > connector change event
> > > >
> > > > You won't get the next event before the first one was ACK'd, right?
> > > >
> > >
> > > The spec says that we need to ACK if we received AND processed a CCI
> > >
> > > "4.5.4 Acknowledge Command Completion and/or Change Indication (R)
> > > This command is used to acknowledge to the PPM that the OPM received and
> > > processed a Command Completion and/or a Connector Change Indication."
> > >
> > > And here in this case, we have received, but not processed the event.
> > > So I'm not really sure what to do here in this case. If we don't send an
> > > ACK, then would the PPM think that OPM is not responding and reset it?
> > > OR would it resend the previous event again since we didn't ACK?
> >
> > Every PPM behaves differently.
> >
> > Did you actually see that happening - GET_CONNECTOR_STATUS failed? Can
> > you reproduce it?
> >
>
> Yea we actually hit the issue once where GET_CONNECTOR_STATUS failed and
> subsequent events didn't get queued since EVENT_PENDING wasn't cleared. Its
> not easily reproducible (<1%) though.
>
> [4948:kworker/0:3]UCSI: ucsi_qti_glink_write: timed out
> [4948:kworker/0:3]ucsi_glink soc:qcom,pmic_glink:qcom,ucsi:
> ucsi_handle_connector_change: GET_CONNECTOR_STATUS failed (-110)
Okay. It would be really interesting to know why is it failing.
But let's just go with this for now.
thanks,
--
heikki