* [QUESTION] Device tree for gmii-to-rgmii phy driver
@ 2023-03-16 16:52 Stefan Herbrechtsmeier
2023-03-17 7:58 ` Michal Simek
0 siblings, 1 reply; 2+ messages in thread
From: Stefan Herbrechtsmeier @ 2023-03-16 16:52 UTC (permalink / raw)
To: Siva Durga Prasad Paladugu; +Cc: U-Boot Mailing List, Michal Simek
Hi Siva Durga Prasad Paladugu,
we want to add rgmii-id support to the gmiitorgmii phy driver. How does
the correct device tree for the gmiitorgmii looks like? The converter
sits between the MAC and the external PHY (MAC <==> GMII2RGMII <==>
RGMII_PHY) which would mean that the phy-handle of the MAC should point
to the GMII2RGMII and the GMII2RGMII to RGMII_PHY:
&gem0 {
phy-handle = <&gmiitorgmii0>;
phy-mode = "gmii";
mdio {
ethphy0: ethernet-phy@1 {
};
gmiitorgmii0: gmiitorgmii@8 {
compatible = "xlnx,gmii-to-rgmii-1.0";
phy-handle = <ðphy0>;
phy-mode = "rgmii-id";
};
};
};
This device tree snippet does not work in the kernel. The kernel
gmii2rgmii driver does not use a phy driver as base and manipulates the
functions of the external phy object instead. It requires the following
device tree snippet:
&gem0 {
phy-handle = <ðphy0>;
phy-mode = "rgmii-id";
mdio {
ethphy0: ethernet-phy@1 {
};
gmiitorgmii0: gmiitorgmii@8 {
compatible = "xlnx,gmii-to-rgmii-1.0";
phy-handle = <ðphy0>;
phy-mode = "rgmii-id";
};
};
};
This does not work in u-boot because the phy framework ignores the
phy-handle of the MAC for the gmii2rgmii but the gmii2rgmii driver
itself checks the phy-mode of the MAC.
What is the correct device tree and thereby proper implementation?
Regards
Stefan
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [QUESTION] Device tree for gmii-to-rgmii phy driver
2023-03-16 16:52 [QUESTION] Device tree for gmii-to-rgmii phy driver Stefan Herbrechtsmeier
@ 2023-03-17 7:58 ` Michal Simek
0 siblings, 0 replies; 2+ messages in thread
From: Michal Simek @ 2023-03-17 7:58 UTC (permalink / raw)
To: Stefan Herbrechtsmeier, Siva Durga Prasad Paladugu, Katakam, Harini
Cc: U-Boot Mailing List
Hi,
On 3/16/23 17:52, Stefan Herbrechtsmeier wrote:
> Hi Siva Durga Prasad Paladugu,
>
> we want to add rgmii-id support to the gmiitorgmii phy driver. How does the
> correct device tree for the gmiitorgmii looks like? The converter sits between
> the MAC and the external PHY (MAC <==> GMII2RGMII <==> RGMII_PHY) which would
> mean that the phy-handle of the MAC should point to the GMII2RGMII and the
> GMII2RGMII to RGMII_PHY:
>
> &gem0 {
> phy-handle = <&gmiitorgmii0>;
> phy-mode = "gmii";
>
> mdio {
> ethphy0: ethernet-phy@1 {
> };
>
> gmiitorgmii0: gmiitorgmii@8 {
> compatible = "xlnx,gmii-to-rgmii-1.0";
> phy-handle = <ðphy0>;
> phy-mode = "rgmii-id";
> };
> };
> };
>
> This device tree snippet does not work in the kernel. The kernel gmii2rgmii
> driver does not use a phy driver as base and manipulates the functions of the
> external phy object instead. It requires the following device tree snippet:
>
> &gem0 {
> phy-handle = <ðphy0>;
> phy-mode = "rgmii-id";
>
> mdio {
> ethphy0: ethernet-phy@1 {
> };
>
> gmiitorgmii0: gmiitorgmii@8 {
> compatible = "xlnx,gmii-to-rgmii-1.0";
> phy-handle = <ðphy0>;
> phy-mode = "rgmii-id";
> };
> };
> };
>
> This does not work in u-boot because the phy framework ignores the phy-handle of
> the MAC for the gmii2rgmii but the gmii2rgmii driver itself checks the phy-mode
> of the MAC.
>
> What is the correct device tree and thereby proper implementation?
It is known issue for some time and U-Boot description is correct. Linux needs
to be fixed. I remember there were any issue with framework not supporting it
that's why unfortunately different description needs to be used.
I am CCing Harini. Here team should work on getting this fixed.
Thanks,
Michal
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2023-03-17 7:59 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-03-16 16:52 [QUESTION] Device tree for gmii-to-rgmii phy driver Stefan Herbrechtsmeier
2023-03-17 7:58 ` Michal Simek
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.