linux-kernel - Re: [PATCH net-next] net: phy: micrel: Add support for PTP_PF

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <c63232b1-2ee5-487a-b987-7cc6b47d3da3@linux.dev>
Date: Thu, 25 Apr 2024 01:10:01 +0100
From: Vadim Fedorenko <vadim.fedorenko@...ux.dev>
To: Horatiu Vultur <horatiu.vultur@...rochip.com>
Cc: andrew@...n.ch, hkallweit1@...il.com, linux@...linux.org.uk,
 davem@...emloft.net, edumazet@...gle.com, kuba@...nel.org,
 pabeni@...hat.com, richardcochran@...il.com, netdev@...r.kernel.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next] net: phy: micrel: Add support for PTP_PF_EXTTS
 for lan8814

On 24/04/2024 20:12, Horatiu Vultur wrote:
> The 04/24/2024 11:57, Vadim Fedorenko wrote:
> 
> Hi Vadim,
> 
>>
>> On 23/04/2024 20:57, Horatiu Vultur wrote:
>>> Extend the PTP programmable gpios to implement also PTP_PF_EXTTS
>>> function. The pins can be configured to capture both of rising
>>> and falling edge. Once the event is seen, then an interrupt is
>>> generated and the LTC is saved in the registers.
>>> On lan8814 only GPIO 3 can be configured for this.
>>>
>>> This was tested using:
>>> ts2phc -m -l 7 -s generic -f ts2phc.cfg
>>>
>>> Where the configuration was the following:
>>> ---
>>> [global]
>>> ts2phc.pin_index  3
>>>
>>> [eth0]
>>> ---
>>>
>>> Signed-off-by: Horatiu Vultur <horatiu.vultur@...rochip.com>
>>
>> I'm not sure what happened to (fac63186f116 net: phy: micrel: Add
>> support for PTP_PF_EXTTS for lan8841), looks like this patch is the
>> rework previous with the limit to GPIO 3 only. In this case comments
>> below are applicable.
> 
> These are two different PHYs:
> 1. lan8814 which is a quad PHY and the patch is this PHY
> 2. lan8841 which is a single PHY. And the commit that you mention it was
>     for that PHY.
> So this commit is not rework of the commit that you mention.

Ah, I see, sorry for the mess..

> 
> ...
> 
>>
>>> +static int lan8814_ptp_extts(struct ptp_clock_info *ptpci,
>>> +                          struct ptp_clock_request *rq, int on)
>>> +{
>>> +     struct lan8814_shared_priv *shared = container_of(ptpci, struct lan8814_shared_priv,
>>> +                                                       ptp_clock_info);
>>> +     struct phy_device *phydev = shared->phydev;
>>> +     int pin;
>>> +
>>> +     if (rq->extts.flags & ~(PTP_ENABLE_FEATURE |
>>> +                             PTP_EXTTS_EDGES |
>>> +                             PTP_STRICT_FLAGS))
>>> +             return -EOPNOTSUPP;
>>> +
>>> +     pin = ptp_find_pin(shared->ptp_clock, PTP_PF_EXTTS,
>>> +                        rq->extts.index);
>>> +     if (pin == -1 || pin != LAN8814_PTP_EXTTS_NUM)
>>> +             return -EINVAL;
>>
>> I'm not sure how will enable request pass this check?
>> In lan8814_ptp_probe_once pins are initialized with PTP_PF_NONE,
>> and ptp_find_pin will always return -1, which will end up with
>> -EINVAL here and never hit lan8814_ptp_extts_on/lan8814_ptp_extts_off
>>
> 
> Why ptp_find_pin will always return -1? Because we can set the function
> of the pin.

ah, I see, PTP_PIN_SETFUNC + PTP_EXTTS_REQUEST ioctls will do the
configuration. Maybe make GPIO 3 as PTP_PF_EXTTS function by default?

> ...
> 
>   >       }
>>> @@ -3148,6 +3263,10 @@ static int lan8814_ptpci_verify(struct ptp_clock_info *ptp, unsigned int pin,
>>>                if (pin >= LAN8814_PTP_PEROUT_NUM || pin != chan)
>>>                        return -1;
>>>                break;
>>> +     case PTP_PF_EXTTS:
>>> +             if (pin != LAN8814_PTP_EXTTS_NUM)
>>
>> Here the check states that exactly GPIO 3 can have EXTTS function, but
>> later in the config...
> 
> ...
>>
>>> +                     return -1;
>>> +             break;
>>>        default:
>>>                return -1;
>>>        }
>>>
>>> @@ -3541,7 +3721,7 @@ static int lan8814_ptp_probe_once(struct phy_device *phydev)
>>>        snprintf(shared->ptp_clock_info.name, 30, "%s", phydev->drv->name);
>>>        shared->ptp_clock_info.max_adj = 31249999;
>>>        shared->ptp_clock_info.n_alarm = 0;
>>> -     shared->ptp_clock_info.n_ext_ts = 0;
>>> +     shared->ptp_clock_info.n_ext_ts = LAN8814_PTP_EXTTS_NUM;
>>
>> Here ptp_clock is configured to have 3 pins supporting EXTTS.
>> Looks like it should be n_ext_ts = 1;
> 
> Good point, let me have a look at this.

I have checked it while checking enable part. Conditions in ptp_ioctl
give no options to limit lowest number of pin which supports EXTTS.

I think that the ptp_clock_info documentation is misleading here:

* @n_ext_ts:  The number of external time stamp channels.

should be replaced to something like "max index of external time
stamp channel".

With all above the patch LGTM!

Reviewed-by: Vadim Fedorenko <vadim.fedorenko@...ux.dev>

> 
>>
>>>        shared->ptp_clock_info.n_pins = LAN8814_PTP_GPIO_NUM;
>>>        shared->ptp_clock_info.pps = 0;
>>>        shared->ptp_clock_info.pin_config = shared->pin_config;
>>
>>
>