netdev - Re: [PATCH net-next 3/3] net: dsa: mt7530: Add support for port 5

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <20190727212651.Horde.WrWPsZ4JpBQzJT2cFdGafCP@www.vdorst.com>
Date:   Sat, 27 Jul 2019 21:26:51 +0000
From:   René van Dorst <opensource@...rst.com>
To:     Russell King - ARM Linux admin <linux@...linux.org.uk>
Cc:     netdev@...r.kernel.org, frank-w@...lic-files.de,
        sean.wang@...iatek.com, f.fainelli@...il.com, davem@...emloft.net,
        matthias.bgg@...il.com, andrew@...n.ch, vivien.didelot@...il.com,
        john@...ozen.org, linux-mediatek@...ts.infradead.org,
        linux-mips@...r.kernel.org, robh+dt@...nel.org,
        devicetree@...r.kernel.org
Subject: Re: [PATCH net-next 3/3] net: dsa: mt7530: Add support for port 5

Quoting Russell King - ARM Linux admin <linux@...linux.org.uk>:

> On Wed, Jul 24, 2019 at 09:25:49PM +0200, René van Dorst wrote:
>> Adding support for port 5.
>>
>> Port 5 can muxed/interface to:
>> - internal 5th GMAC of the switch; can be used as 2nd CPU port or as
>>   extra port with an external phy for a 6th ethernet port.
>> - internal PHY of port 0 or 4; Used in most applications so that port 0
>>   or 4 is the WAN port and interfaces with the 2nd GMAC of the SOC.
>
> ...
>
>> @@ -1381,15 +1506,19 @@ static void mt7530_phylink_validate(struct  
>> dsa_switch *ds, int port,
>>  	phylink_set_port_modes(mask);
>>  	phylink_set(mask, Autoneg);
>>
>> -	if (state->interface != PHY_INTERFACE_MODE_TRGMII) {
>> +	if (state->interface == PHY_INTERFACE_MODE_TRGMII) {
>> +		phylink_set(mask, 1000baseT_Full);
>> +	} else {
>>  		phylink_set(mask, 10baseT_Half);
>>  		phylink_set(mask, 10baseT_Full);
>>  		phylink_set(mask, 100baseT_Half);
>>  		phylink_set(mask, 100baseT_Full);
>> -		phylink_set(mask, 1000baseT_Half);
>> -	}
>>
>> -	phylink_set(mask, 1000baseT_Full);
>> +		if (state->interface != PHY_INTERFACE_MODE_MII) {
>> +			phylink_set(mask, 1000baseT_Half);
>> +			phylink_set(mask, 1000baseT_Full);
>> +		}
>> +	}
>

Hi Russell,

Thanks for your review and many useful comments and explanations.

> As port 5 could use an external PHY, and it supports gigabit speeds,
> consider that the PHY may provide not only copper but also fiber
> connectivity, so port 5 should probably also have 1000baseX modes
> too, which would allow such a PHY to bridge the switch to fiber.

I shall add the 1000baseX modes.

My device, Ubiquiti EdgeRouter X SFP, has this setup.
Port 5 is connected to a at8033 phy which acts as a RGMII-SerDes converter for
the SFP cage. According to the datasheet it only support 100BASE-FX and
1000BASE-X. With bootstrap resistors the PHY is put in RGMII-SerDes 1000BASE-X
mode.

The problem I had is that the current mainline driver doesn't support  
this mode
so I had to hack it in myself [0][1]. I probably doing the wrong thing with my
phy driver. Driver works for me, it detects a link and sets-up a 1gbit link.
So I can test port 5. But the driver may report all the wrong values to
PHYLIB/PHYLINK. But now that I learned more about it I can revise the driver.


By reading your previous emails, my setup could official not support the
FiberStore SFP-GB-GE-T module, because it requests a SGMII interface.
But my PHY only supports 1000BaseX and my code currently doesn't error out.

dmesg output of this module:
[    3.382637] sfp sfp_lan5: module FiberStore       SFP-GB-GE-T       
rev B    sn <snip>      dc 19-12-17
[    3.402048] sfp sfp_lan5:   unknown/unspecified connector, encoding  
8b10b, nominal bitrate 1.3Gbps +0% -0%
[    3.421268] sfp sfp_lan5:   1000BaseSX- 1000BaseLX- 1000BaseCX-  
1000BaseT+ 100BaseLX- 100BaseFX- BaseBX10- BasePX-
[    3.441867] sfp sfp_lan5:   10GBaseSR- 10GBaseLR- 10GBaseLRM- 10GBaseER-
[    3.455208] sfp sfp_lan5:   Copper length: 100m
[    3.464225] sfp sfp_lan5:   Options: txdisable
[    3.473066] sfp sfp_lan5:   Diagnostics:
[    3.481034] sfp sfp_lan5: Unknown/unsupported extended compliance  
code: 0x01
[    3.495069] Atheros 8031 ethernet mdio-bus:07: SFP interface sgmii

What is the best way to do it in case of SGMII interface request?
Return that we don't support SGMII or report that we only support 1  
mode and no
auto-negotiation?

Greats,

René

>

> --
> RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
> FTTC broadband for 0.8mile line in suburbia: sync at 12.1Mbps down 622kbps up
> According to speedtest.net: 11.9Mbps down 500kbps up


[0]:  
https://github.com/vDorst/linux-1/commit/dad5d6ec65cfa99c204e9756b3fc234071709292
[1]:  
https://github.com/vDorst/linux-1/commit/a3aa74e84796604ab8619cfaf1c299c115a8736f