Re: [PATCH net-next v3 06/12] net: ethernet: oa_tc6: implement internal PHY initialization

[<Parthiban.Veerasooran@microchip.com>]

Hi Andrew,

On 18/03/24 4:31 pm, Parthiban.Veerasooran@microchip.com wrote:
> Hi Andrew,
> 
> On 08/03/24 7:03 pm, Andrew Lunn wrote:
>> EXTERNAL EMAIL: Do not click links or open attachments unless you know the content is safe
>>
>>> Ok, as per the table 6 in the spec, PHY C45 registers are mapped in the
>>> MMS like below,
>>>
>>> PHY – PCS Registers (MMD 3)  --->  MMS 2
>>> PHY – PMA/PMD Registers (MMD 1)  --->   MMS 3
>>> PHY – Vendor Specific and PLCA Registers (MMD 31)  --->  MMS 4
>>> PHY – Auto-Negotiation Registers (MMD 7)  --->  MMS 5
>>> PHY – Power Unit (MMD 13)  --->  MMS 6
>>>
>>> MMD 13 for PHY - Power Unit is not defined in the mdio.h. So in the
>>> below code I have defined it locally (MDIO_MMD_POWER_UNIT). May be
>>> needed to do this in the mdio.h file when coming to this patch.
>>
>> Yes, please add it to mdio.h
> Sure will add it in the mdio.h file.
>>
>>> /* PHY – Clause 45 registers memory map selector (MMS) as per table 6 in
>>> the OPEN Alliance specification.
>>>     */
>>> #define OA_TC6_PHY_PCS_MMS2                     2       /* MMD 3 */
>>> #define OA_TC6_PHY_PMA_PMD_MMS3                 3       /* MMD 1 */
>>> #define OA_TC6_PHY_VS_PLCA_MMS4                 4       /* MMD 31 */
>>> #define OA_TC6_PHY_AUTO_NEG_MMS5                5       /* MMD 7 */
>>> #define OA_TC6_PHY_POWER_UNIT_MMS6              6       /* MMD 13 */
>>>
>>> /* MDIO Manageable Device (MMD) for PHY Power Unit */
>>> #define MDIO_MMD_POWER_UNIT                     13      /* PHY Power Unit */
>>>
>>> static int oa_tc6_mdiobus_read_c45(struct mii_bus *bus, int addr, int
>>> devnum, int regnum)
>>> {
>>>
>>>            struct oa_tc6 *tc6 = bus->priv;
>>>
>>>            u32 regval;
>>>
>>>            bool ret;
>>>
>>>            u32 mms;
>>>
>>>
>>>
>>>            if (devnum == MDIO_MMD_PCS)
>>>
>>>                    mms = OA_TC6_PHY_PCS_MMS2;
>>>
>>>            else if (devnum == MDIO_MMD_PMAPMD)
>>>
>>>                    mms = OA_TC6_PHY_PMA_PMD_MMS3;
>>>
>>>            else if (devnum == MDIO_MMD_VEND2)
>>>
>>>                    mms = OA_TC6_PHY_VS_PLCA_MMS4;
>>>
>>>            else if (devnum == MDIO_MMD_AN)
>>>
>>>                    mms = OA_TC6_PHY_AUTO_NEG_MMS5;
>>>
>>>            else if (devnum == MDIO_MMD_POWER_UNIT)
>>>
>>>                    mms = OA_TC6_PHY_POWER_UNIT_MMS6;
>>
>> I would probably use a switch statement.
> Ah ok, I will use switch here.
>>
>>>
>>>            else
>>>
>>>                    return -ENOTSUPP;
>>
>> 802.3 says:
>>
>>     If a device supports the MDIO interface it shall respond to all
>>     possible register addresses for the device and return a value of
>>     zero for undefined and unsupported registers. Writes to undefined
>>     registers and read-only registers shall have no effect. The
>>     operation of an MMD shall not be affected by writes to reserved and
>>     unsupported register bits, and such register bits shall return a
>>     value of zero when read.
>>
>> So maybe return 0. ENOTSUPP is wrong, that is an NFS only error
>> code. The generic one is EOPNOTSUPP. I would say -EOPNOTSUPP is also
>> O.K.
> Sure, I will use -EOPNOTSUPP in the next version instead of -ENOTSUPP.
>>
>>>            ret = oa_tc6_read_register(tc6, (mms << 16) | regnum, &regval);
>>>
>>>            if (ret)
>>>
>>>                    return -ENODEV;
>>
>> oa_tc6_read_register() should return an error code, so return whatever
>> is returns. Don't overwrite error codes. It makes it harder to track
>> errors through the call stack.
> Ah ok, will return the error code from oa_tc6_read_register() in the
> next version.
After implementing the new APIs oa_tc6_mdiobus_read_c45() and 
oa_tc6_mdiobus_write_c45(), I tried testing. But, for some reason these 
APIs are never get called by phy_read_mmd() or phy_write_mmd() as those 
APIs are checking for phydev->is_c45 flag for calling this new c45 APIs 
which is not set in this case.

If the phydev is found via c22, phylib does not set phydev->is_c45, and 
everything ends up going indirect.

To verify the APIs, I just called them locally withing the oa_tc6_init() 
function, there I got the expected results but accessing them through 
phylib is not working because of the above reason.

Do you have any idea or suggestion to test this APIs or do I miss 
anything here?

Best regards,
Parthiban V
> 
> Best regards,
> Parthiban V
>>
>>          Andrew
>>
>