Re: [PATCH] igc: Remove _I_PHY_ID check for i225 devices

[Linjun Bao <meljbao@gmail.com>]

On 2022/8/10 上午1:32, Tony Nguyen wrote:
> On 8/9/2022 6:35 AM, Linjun Bao wrote:
>> Source commit 7c496de538ee ("igc: Remove _I_PHY_ID checking"),
>> remove _I_PHY_ID check for i225 device, since i225 devices only
>> have one PHY vendor.
>
> What are you trying to do with this patch? You're referencing the original commit so you know it's committed, but it's not clear to me why you are re-sending it.
>
I'm new here, please correct me if I am doing things in the wrong way.


Yes this commit was committed to mainline about one year ago. But this commit has not been included into kernel 5.4 yet, and I encountered the probe failure when using alderlake-s with Ethernet adapter i225-LM. Since I could not directly apply the patch 7c496de538ee to kernel 5.4, so I generated this patch for kernel 5.4 usage.


Looks like sending a duplicated patch is not expected. Would you please advise what is the proper action when encountering such case? I would like this fix to be implemented into LTS kernel 5.4, I also wrote a ticket on https://bugzilla.kernel.org/show_bug.cgi?id=216261, but no response.


Regards

Joseph


> Thanks,
> Tony
>
>> Signed-off-by: Linjun Bao <meljbao@gmail.com>
>> ---
>>   drivers/net/ethernet/intel/igc/igc_base.c | 10 +---------
>>   drivers/net/ethernet/intel/igc/igc_main.c |  3 +--
>>   drivers/net/ethernet/intel/igc/igc_phy.c  |  6 ++----
>>   3 files changed, 4 insertions(+), 15 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/intel/igc/igc_base.c b/drivers/net/ethernet/intel/igc/igc_base.c
>> index db289bcce21d..d66429eb14a5 100644
>> --- a/drivers/net/ethernet/intel/igc/igc_base.c
>> +++ b/drivers/net/ethernet/intel/igc/igc_base.c
>> @@ -187,15 +187,7 @@ static s32 igc_init_phy_params_base(struct igc_hw *hw)
>>         igc_check_for_copper_link(hw);
>>   -    /* Verify phy id and set remaining function pointers */
>> -    switch (phy->id) {
>> -    case I225_I_PHY_ID:
>> -        phy->type    = igc_phy_i225;
>> -        break;
>> -    default:
>> -        ret_val = -IGC_ERR_PHY;
>> -        goto out;
>> -    }
>> +    phy->type = igc_phy_i225;
>>     out:
>>       return ret_val;
>> diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
>> index 9ba05d9aa8e0..b8297a63a7fd 100644
>> --- a/drivers/net/ethernet/intel/igc/igc_main.c
>> +++ b/drivers/net/ethernet/intel/igc/igc_main.c
>> @@ -2884,8 +2884,7 @@ bool igc_has_link(struct igc_adapter *adapter)
>>           break;
>>       }
>>   -    if (hw->mac.type == igc_i225 &&
>> -        hw->phy.id == I225_I_PHY_ID) {
>> +    if (hw->mac.type == igc_i225) {
>>           if (!netif_carrier_ok(adapter->netdev)) {
>>               adapter->flags &= ~IGC_FLAG_NEED_LINK_UPDATE;
>>           } else if (!(adapter->flags & IGC_FLAG_NEED_LINK_UPDATE)) {
>> diff --git a/drivers/net/ethernet/intel/igc/igc_phy.c b/drivers/net/ethernet/intel/igc/igc_phy.c
>> index 6156c76d765f..1be112ce6774 100644
>> --- a/drivers/net/ethernet/intel/igc/igc_phy.c
>> +++ b/drivers/net/ethernet/intel/igc/igc_phy.c
>> @@ -235,8 +235,7 @@ static s32 igc_phy_setup_autoneg(struct igc_hw *hw)
>>               return ret_val;
>>       }
>>   -    if ((phy->autoneg_mask & ADVERTISE_2500_FULL) &&
>> -        hw->phy.id == I225_I_PHY_ID) {
>> +    if (phy->autoneg_mask & ADVERTISE_2500_FULL) {
>>           /* Read the MULTI GBT AN Control Register - reg 7.32 */
>>           ret_val = phy->ops.read_reg(hw, (STANDARD_AN_REG_MASK <<
>>                           MMD_DEVADDR_SHIFT) |
>> @@ -376,8 +375,7 @@ static s32 igc_phy_setup_autoneg(struct igc_hw *hw)
>>           ret_val = phy->ops.write_reg(hw, PHY_1000T_CTRL,
>>                            mii_1000t_ctrl_reg);
>>   -    if ((phy->autoneg_mask & ADVERTISE_2500_FULL) &&
>> -        hw->phy.id == I225_I_PHY_ID)
>> +    if (phy->autoneg_mask & ADVERTISE_2500_FULL)
>>           ret_val = phy->ops.write_reg(hw,
>>                            (STANDARD_AN_REG_MASK <<
>>                            MMD_DEVADDR_SHIFT) |