[Macchiato] macchiatobin problem

[Baruch Siach]

Hi Alpha,

alpha_one_x86 writes:
> *Your patch worked, some body can push it in upstream?*

I'll run a few more tests with this patch, and push upstream as time
permits. Hopefully this week.

Note that we will need a separate .dts file for the single-shot. So my
patch will not go in as is.

> I remain blocked on:
>
> ethtool -m eth3
> Cannot get module EEPROM information: Operation not supported
>
> [ 22.265548] mvpp2 f4000000.ethernet eth3: configuring for
> inband/2500base-x link mode
> [ 22.266897] IPv6: ADDRCONF(NETDEV_UP): eth3: link is not ready

Do you see your SFP module detected?

Please send your full kernel log.

> I try do work:
>
> https://www.fs.com/products/48929.html Cisco GLC-LH-SM-20 Compatible
> 1000BASE-LX/LH SFP 1310nm 20km DOM Transceiver Module
>
> But failed, maybe not dynamically down to 1G.

Recent mainline kernel can configure Ethernet port speed dynamically. I
tested it here with 1G optic modules. The trouble is that SFP
configuration is not always correct.

baruch

> On 11/01/18 07:23, Baruch Siach wrote:
>> Hi Alpha,
>>
>> Please keep the Macchiato list on Cc
>>
>> alpha_one_x86 writes:
>>>> If this is the case, what DT are you using? Note that mainline kernel 
>>>> armada-8040-mcbin.dts is meant for the DoubleShot, which has 88x3310 PHYs on 
>>>> eth0 and eth1 (marked U2 and U3 on the PCB). These PHYs are not assembled on 
>>>> the SingleShot. Instead, both interfaces are directly connected to SFP cages. 
>>>> So the error message (-19 == -ENODEV) is correct. You need to remove the PHYs 
>>>> description from the DT for the SingleShot.
>>> I have Single Shot. And for vanilla I use armada-8040-mcbin.dts/dtb
>>> Can you indicate me how change the dts? (maybe send me a patch), I see
>>> multiple eth0, eth1, ... ref.
>> The ethX reference is quite confusing indeed. When you see cpX_ethY it
>> refers to different Ethernet MAC instances on each CP. There are two
>> CP110 (cp0 and cp1), each with three Ethernet MACs.
>>
>> For the Single Shot you can try the following completely untested patch
>> against v4.19.
>>
>> diff --git a/arch/arm64/boot/dts/marvell/armada-8040-mcbin.dts b/arch/arm64/boot/dts/marvell/armada-8040-mcbin.dts
>> index 56fa44860909..3da9b4cd78a4 100644
>> --- a/arch/arm64/boot/dts/marvell/armada-8040-mcbin.dts
>> +++ b/arch/arm64/boot/dts/marvell/armada-8040-mcbin.dts
>> @@ -227,22 +227,6 @@
>>  	};
>>  };
>>  
>> -&cp0_xmdio {
>> -	status = "okay";
>> -
>> -	phy0: ethernet-phy at 0 {
>> -		compatible = "ethernet-phy-ieee802.3-c45";
>> -		reg = <0>;
>> -		sfp = <&sfp_eth0>;
>> -	};
>> -
>> -	phy8: ethernet-phy at 8 {
>> -		compatible = "ethernet-phy-ieee802.3-c45";
>> -		reg = <8>;
>> -		sfp = <&sfp_eth1>;
>> -	};
>> -};
>> -
>>  &cp0_ethernet {
>>  	status = "okay";
>>  };
>> @@ -250,10 +234,11 @@
>>  &cp0_eth0 {
>>  	status = "okay";
>>  	/* Network PHY */
>> -	phy = <&phy0>;
>>  	phy-mode = "10gbase-kr";
>>  	/* Generic PHY, providing serdes lanes */
>>  	phys = <&cp0_comphy4 0>;
>> +	managed = "in-band-status";
>> +	sfp = <&sfp_eth0>;
>>  };
>>  
>>  &cp0_sata0 {
>> @@ -288,10 +273,11 @@
>>  &cp1_eth0 {
>>  	status = "okay";
>>  	/* Network PHY */
>> -	phy = <&phy8>;
>>  	phy-mode = "10gbase-kr";
>>  	/* Generic PHY, providing serdes lanes */
>>  	phys = <&cp1_comphy4 0>;
>> +	managed = "in-band-status";
>> +	sfp = <&sfp_eth1>;
>>  };
>>  
>>  &cp1_eth1 {
>>
>> Hope this helps,
>> baruch

-- 
     http://baruch.siach.name/blog/                  ~. .~   Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
   - baruch at tkos.co.il - tel: +972.52.368.4656, http://www.tkos.co.il -