Re: 88e6161 failed create slave phy

Re: 88e6161 failed create slave phy

[Andrew Lunn]

On Thu, Jan 25, 2018 at 09:15:16PM +0800, ?????? wrote:
> hello, I am compiling kernel for 88f6281 with 88e6161 switch for 4.9.73.but the dsa does not work correctly.I think it is a bug because 4.4 and 4.14 works fine.
> both two bindings get the same result (port uninitialized, could not setup slave phy -19). thanks
> this it the log

Please can you send us your .dts file.

       Thanks
	  Andrew

Re: 88e6161 failed create slave phy

[Andrew Lunn]

On Thu, Jan 25, 2018 at 09:15:16PM +0800, ?????? wrote:

> hello, I am compiling kernel for 88f6281 with 88e6161 switch for
> 4.9.73.but the dsa does not work correctly.I think it is a bug
> because 4.4 and 4.14 works fine.

Sorry it has taken so long to get around to this. I've done some
testing with 4.9.78:

VF610 based ZII devel B with 88E6352 and 88E6185 switches: Works fine
Kirkwood based DIR665 with 88E6171: Works fine.
Armada XP based WRT1900AC with a 88E6172 switch: Works fine

The 6161 is a member of the 6165 family. That family contains 6123,
6161, and 6165. I don't have any of them. So i cannot specifically
test this switch family.

I think to make progress, you are going to need to perform a git bisect
to work out what makes it work in later kernels.

   Andrew

Re: 88e6161 failed create slave phy

[Andrew Lunn]

On Thu, Jan 25, 2018 at 09:15:16PM +0800, ?????? wrote:
> hello, I am compiling kernel for 88f6281 with 88e6161 switch for 4.9.73.but the dsa does not work correctly.I think it is a bug because 4.4 and 4.14 works fine.
> both two bindings get the same result (port uninitialized, could not setup slave phy -19). thanks
> this it the log

Please could you try cherry-picking:

ec8378bb4d8e ("net: dsa: mv88e6xxx: 6161 uses global 2 for PHY access")
0ac64c394900 ("net: dsa: mv88e6xxx: mv88e6161 uses mv88e6320 stats snapshot")
5ebe31d7b20f ("net: dsa: mv88e6xxx: Have 6161/6123 use EDSA tags")

The first one probably fixes your PHY issue. The second will fix
ethtool -S. The last one is just nice to have, helps when using
tcpdump on the master interface.

Thanks go to Chris Healy for pointing out these patches.

       Andrew