Re: [PATCH v2 3/3] net: stmmac: dwmac-qcom-ethqos: add support for EMAC on qcs615 platforms

[Yijie Yang <quic_yijiyang@quicinc.com>]

On 2024-12-24 13:05, Dmitry Baryshkov wrote:
> On Tue, Dec 24, 2024 at 12:36:29PM +0800, Yijie Yang wrote:
>>
>>
>> On 2024-12-24 12:18, Dmitry Baryshkov wrote:
>>> On Tue, Dec 24, 2024 at 11:07:03AM +0800, Yijie Yang wrote:
>>>> qcs615 uses EMAC version 2.3.1, add the relevant defines and add the new
>>>> compatible.
>>>>
>>>> Signed-off-by: Yijie Yang <quic_yijiyang@quicinc.com>
>>>> ---
>>>>    drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c | 17 +++++++++++++++++
>>>>    1 file changed, 17 insertions(+)
>>>>
>>>> diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
>>>> index 901a3c1959fa57efb078da795ad4f92a8b6f71e1..8c76beaee48821eb2853f4e3f8bfd37db8cadf78 100644
>>>> --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
>>>> +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c
>>>> @@ -249,6 +249,22 @@ static const struct ethqos_emac_driver_data emac_v2_1_0_data = {
>>>>    	.has_emac_ge_3 = false,
>>>>    };
>>>> +static const struct ethqos_emac_por emac_v2_3_1_por[] = {
>>>> +	{ .offset = RGMII_IO_MACRO_CONFIG,	.value = 0x00C01343 },
>>>> +	{ .offset = SDCC_HC_REG_DLL_CONFIG,	.value = 0x2004642C },
>>>
>>> lowercase the hex, please.
>>
>> I will take care of it.
>>
>>>
>>>> +	{ .offset = SDCC_HC_REG_DDR_CONFIG,	.value = 0x00000000 },
>>>> +	{ .offset = SDCC_HC_REG_DLL_CONFIG2,	.value = 0x00200000 },
>>>> +	{ .offset = SDCC_USR_CTL,		.value = 0x00010800 },
>>>> +	{ .offset = RGMII_IO_MACRO_CONFIG2,	.value = 0x00002060 },
>>>> +};
>>>> +
>>>> +static const struct ethqos_emac_driver_data emac_v2_3_1_data = {
>>>> +	.por = emac_v2_3_1_por,
>>>> +	.num_por = ARRAY_SIZE(emac_v2_3_1_por),
>>>> +	.rgmii_config_loopback_en = true,
>>>> +	.has_emac_ge_3 = false,
>>>> +};
>>>
>>> Modulo emac_v2_3_1_por vs emac_v2_3_0_por, this is the same as
>>> emac_v2_3_0_data. Which means that bindings for qcs615-ethqos should be
>>> corrected to use qcom,qcs404-ethqos as as fallback entry, making this
>>> patch unused. Please correct the bindings instead.
>>
>> Although they currently share the same data, they are actually two different
>> versions. Their differences are not apparent now but will become evident
>> once new features are uploaded. If I revert to qcom,qcs404-ethqos now, it
>> will be challenging to distinguish between them in the future.
> 
> Which features? Moreover, note, the use of the fallback doesn't preclude
> you from addign a new compat entry later on. By having a fallback you
> simply declare that the device A is also compatible with the device B.
> 

Yes, you're right. I confirmed with the hardware team that the EMAC of 
qcs615-ride is the same as that of qcs404. I will fall back the 
compatible of qcs615 to the latter and drop the unnecessary patches.

>>
>>>
>>>> +
>>>>    static const struct ethqos_emac_por emac_v3_0_0_por[] = {
>>>>    	{ .offset = RGMII_IO_MACRO_CONFIG,	.value = 0x40c01343 },
>>>>    	{ .offset = SDCC_HC_REG_DLL_CONFIG,	.value = 0x2004642c },
>>>> @@ -898,6 +914,7 @@ static int qcom_ethqos_probe(struct platform_device *pdev)
>>>>    static const struct of_device_id qcom_ethqos_match[] = {
>>>>    	{ .compatible = "qcom,qcs404-ethqos", .data = &emac_v2_3_0_data},
>>>> +	{ .compatible = "qcom,qcs615-ethqos", .data = &emac_v2_3_1_data},
>>>>    	{ .compatible = "qcom,sa8775p-ethqos", .data = &emac_v4_0_0_data},
>>>>    	{ .compatible = "qcom,sc8280xp-ethqos", .data = &emac_v3_0_0_data},
>>>>    	{ .compatible = "qcom,sm8150-ethqos", .data = &emac_v2_1_0_data},
>>>>
>>>> -- 
>>>> 2.34.1
>>>>
>>>
>>
>> -- 
>> Best Regards,
>> Yijie
>>
> 

-- 
Best Regards,
Yijie