linux-kernel - Re: [PATCH v1] usb: typec: tcpm: Fix TDA 2.2.1.1 and TDA 2.2.1.2 failures

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5461f039-ca37-daaa-78c1-d6fc69d17004@roeck-us.net>
Date:   Tue, 11 Aug 2020 18:44:25 -0700
From:   Guenter Roeck <linux@...ck-us.net>
To:     Badhri Jagan Sridharan <badhri@...gle.com>
Cc:     Heikki Krogerus <heikki.krogerus@...ux.intel.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        USB <linux-usb@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v1] usb: typec: tcpm: Fix TDA 2.2.1.1 and TDA 2.2.1.2
 failures

On 8/11/20 6:08 PM, Badhri Jagan Sridharan wrote:
> Yes I did ! https://lkml.org/lkml/2020/8/11/788 !
> 
> For me gmail is grouping V1 and V2 patches in the same thread so
> thought you saw the v2 patch :P
> 

You copied me on v1, but not on v2. I am not on any of the mailing lists
you sent v2 to, so I did not see it.

Guenter

> Thanks,
> Badhri
> 
> On Tue, Aug 11, 2020 at 5:36 PM Guenter Roeck <linux@...ck-us.net> wrote:
>>
>> On 8/11/20 4:21 PM, Badhri Jagan Sridharan wrote:
>>> Thanks Guenter ! However I don't see a reviewed-by tag :)
>>>
>>
>> Confused. Did you send a v2 with the changes we discussed ?
>> I didn't see that.
>>
>> Thanks,
>> Guenter
>>
>>> On Tue, Aug 11, 2020 at 1:18 PM Guenter Roeck <linux@...ck-us.net> wrote:
>>>>
>>>> On 8/11/20 12:39 PM, Badhri Jagan Sridharan wrote:
>>>>> On Tue, Aug 11, 2020 at 11:45 AM Guenter Roeck <linux@...ck-us.net> wrote:
>>>>>>
>>>>>> On Tue, Aug 11, 2020 at 11:24:07AM -0700, Badhri Jagan Sridharan wrote:
>>>>>>> On Mon, Aug 10, 2020 at 6:51 PM Guenter Roeck <linux@...ck-us.net> wrote:
>>>>>>>>
>>>>>>>> On 8/10/20 6:11 PM, Badhri Jagan Sridharan wrote:
>>>>>>>>> >From the spec:
>>>>>>>>> "7.1.5 Response to Hard Resets
>>>>>>>>> Hard Reset Signaling indicates a communication failure has occurred and
>>>>>>>>> the Source Shall stop driving VCONN, Shall remove Rp from the VCONN pin
>>>>>>>>> and Shall drive VBUS to vSafe0V as shown in Figure 7-9. The USB connection
>>>>>>>>> May reset during a Hard Reset since the VBUS voltage will be less than
>>>>>>>>> vSafe5V for an extended period of time. After establishing the vSafe0V
>>>>>>>>> voltage condition on VBUS, the Source Shall wait tSrcRecover before
>>>>>>>>> re-applying VCONN and restoring VBUS to vSafe5V. A Source Shall conform
>>>>>>>>> to the VCONN timing as specified in [USB Type-C 1.3]."
>>>>>>>>>
>>>>>>>>> Signed-off-by: Badhri Jagan Sridharan <badhri@...gle.com>
>>>>>>>>> ---
>>>>>>>>>  drivers/usb/typec/tcpm/tcpm.c | 16 +++++++++++++---
>>>>>>>>>  1 file changed, 13 insertions(+), 3 deletions(-)
>>>>>>>>>
>>>>>>>>> diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c
>>>>>>>>> index 3ef37202ee37..e41c4e5d3c71 100644
>>>>>>>>> --- a/drivers/usb/typec/tcpm/tcpm.c
>>>>>>>>> +++ b/drivers/usb/typec/tcpm/tcpm.c
>>>>>>>>> @@ -3372,13 +3372,19 @@ static void run_state_machine(struct tcpm_port *port)
>>>>>>>>>                       tcpm_set_state(port, SNK_HARD_RESET_SINK_OFF, 0);
>>>>>>>>>               break;
>>>>>>>>>       case SRC_HARD_RESET_VBUS_OFF:
>>>>>>>>> -             tcpm_set_vconn(port, true);
>>>>>>>>> +             /*
>>>>>>>>> +              * 7.1.5 Response to Hard Resets
>>>>>>>>> +              * Hard Reset Signaling indicates a communication failure has occurred and the
>>>>>>>>> +              * Source Shall stop driving VCONN, Shall remove Rp from the VCONN pin and Shall
>>>>>>>>> +              * drive VBUS to vSafe0V as shown in Figure 7-9.
>>>>>>>>> +              */
>>>>>>>>> +             tcpm_set_vconn(port, false);
>>>>>>>>>               tcpm_set_vbus(port, false);
>>>>>>>>>               tcpm_set_roles(port, port->self_powered, TYPEC_SOURCE,
>>>>>>>>>                              tcpm_data_role_for_source(port));
>>>>>>>>> -             tcpm_set_state(port, SRC_HARD_RESET_VBUS_ON, PD_T_SRC_RECOVER);
>>>>>>>>
>>>>>>>> I am a bit concerned about this. If I understand correctly, it means that
>>>>>>>> we won't turn VBUS back on unless a SRC_HARD_RESET_VBUS_OFF PD event is received.
>>>>>>>> Is that correct ? What happens if that event is never received ?
>>>>>>>>
>>>>>>>> Thanks,
>>>>>>>> Guenter
>>>>>>>
>>>>>>> The term PD event is a little ambiguous to me. Trying to summarize the workflow.
>>>>>>> Lower level tcpc driver would have to call tcpm_vbus_change which
>>>>>>> would in-turn trigger TCPM_VBUS_EVENT
>>>>>>> and queries port->tcpc->get_vbus to get the vbus status. It is not
>>>>>>> really a PD protocol driven event hence the
>>>>>>> confusion.
>>>>>>>
>>>>>>> "What happens if that event is never received ?"
>>>>>>> Yeah TCPM would be in SRC_HARD_RESET_VBUS_OFF till the tcpc calls the
>>>>>>> tcpm_vbus_change.
>>>>>>> Do you suspect that existing tcpc would not have the capability to
>>>>>>> monitor vbus status while sourcing and call tcpm_vbus_change?
>>>>>>>
>>>>>> That, or the driver might be buggy, or the hardware does't signal a status
>>>>>> update, or the update gets lost. I think we should have some backup,
>>>>>> to trigger if the event is not received in a reasonable amout of time.
>>>>>> I don't know if the specification has some kind of maximum limit. If
>>>>>> not, we should still have something
>>>>>>
>>>>>> Thanks,
>>>>>> Guenter
>>>>>
>>>>> Got it ! The specification actually has a bound for vbus off.
>>>>> tSafe0V  - Time to reach vSafe0V max - 650ms. (PD_T_SAFE_0V).
>>>>> So I will bound it to that.
>>>>>
>>>>
>>>> Excellent. Thanks a lot for looking into this!
>>>>
>>>> Guenter
>>>>
>>>>> >From Table 7-12 Sequence Description for a Source Initiated Hard Reset:
>>>>> 4.Policy Engine waits tPSHardReset after sending Hard Reset Signaling
>>>>> and then tells the Device Policy Manager to instruct the power supply
>>>>> to perform a Hard Reset. The transition to vSafe0V Shall occur within
>>>>> tSafe0V (t2).
>>>>> 5 After tSrcRecover the Source applies power to VBUS in an attempt to
>>>>> re-establish communication with the Sink and resume USB Default
>>>>> Operation. The transition to vSafe5V Shall occur within tSrcTurnOn
>>>>> (t4).
>>>>>
>>>>> Thanks,
>>>>> Badhri
>>>>>
>>>>>>
>>>>>>> Thanks,
>>>>>>> Badhri
>>>>>>>
>>>>>>>
>>>>>>>>>               break;
>>>>>>>>>       case SRC_HARD_RESET_VBUS_ON:
>>>>>>>>> +             tcpm_set_vconn(port, true);
>>>>>>>>>               tcpm_set_vbus(port, true);
>>>>>>>>>               port->tcpc->set_pd_rx(port->tcpc, true);
>>>>>>>>>               tcpm_set_attached_state(port, true);
>>>>>>>>> @@ -3944,7 +3950,11 @@ static void _tcpm_pd_vbus_off(struct tcpm_port *port)
>>>>>>>>>               tcpm_set_state(port, SNK_HARD_RESET_WAIT_VBUS, 0);
>>>>>>>>>               break;
>>>>>>>>>       case SRC_HARD_RESET_VBUS_OFF:
>>>>>>>>> -             tcpm_set_state(port, SRC_HARD_RESET_VBUS_ON, 0);
>>>>>>>>> +             /*
>>>>>>>>> +              * After establishing the vSafe0V voltage condition on VBUS, the Source Shall wait
>>>>>>>>> +              * tSrcRecover before re-applying VCONN and restoring VBUS to vSafe5V.
>>>>>>>>> +              */
>>>>>>>>> +             tcpm_set_state(port, SRC_HARD_RESET_VBUS_ON, PD_T_SRC_RECOVER);
>>>>>>>>>               break;
>>>>>>>>>       case HARD_RESET_SEND:
>>>>>>>>>               break;
>>>>>>>>>
>>>>>>>>
>>>>
>>