| Message ID | 20210223161906.9177-3-michael@walle.cc |
|---|---|
| State | Superseded |
| Delegated to: | Priyanka Jain |
| Headers | show |
| Series | net: dsa: various fixes | expand |
On Tue, Feb 23, 2021 at 05:19:06PM +0100, Michael Walle wrote: > DSA needs to have the master device probed first for MAC inheritance. > Until now, it only works by chance because the only user (LS1028A SoC) > will probe the master device first. The probe order is given by the PCI > device ordering, thus it works because the master device has a "smaller" > BDF then the switch device. > > Explicitly probe the master device in dsa_port_probe(). > > Fixes: fc054d563bfb ("net: Introduce DSA class for Ethernet switches") > Signed-off-by: Michael Walle <michael@walle.cc> > --- Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com> By the way we had this in the old driver that marinated too much and never got merged, I am not sure why we removed it during the second submission process: https://github.com/openil/u-boot/commit/2544ed8051d3dce55b12e13b6c2b476733d19c05
Am 2021-02-23 17:19, schrieb Michael Walle: > DSA needs to have the master device probed first for MAC inheritance. > Until now, it only works by chance because the only user (LS1028A SoC) > will probe the master device first. The probe order is given by the PCI > device ordering, thus it works because the master device has a > "smaller" > BDF then the switch device. > > Explicitly probe the master device in dsa_port_probe(). > > Fixes: fc054d563bfb ("net: Introduce DSA class for Ethernet switches") > Signed-off-by: Michael Walle <michael@walle.cc> > --- > net/dsa-uclass.c | 8 ++++++++ > 1 file changed, 8 insertions(+) > > diff --git a/net/dsa-uclass.c b/net/dsa-uclass.c > index 88a8ea9352..242e2be035 100644 > --- a/net/dsa-uclass.c > +++ b/net/dsa-uclass.c > @@ -284,6 +284,14 @@ static int dsa_port_probe(struct udevice *pdev) > if (!master) > return -ENODEV; > > + /* > + * Probe the master device. We depend on the master device for proper > + * operation and we also need it for MAC inheritance below. > + */ > + ret = device_probe(master); btw, there is a "int ret" missing above. Will be fixed in the non-RFC version. > + if (ret) > + return ret; > + > /* > * Inherit port's hwaddr from the DSA master, unless the port already > * has a unique MAC address specified in the environment.
diff --git a/net/dsa-uclass.c b/net/dsa-uclass.c index 88a8ea9352..242e2be035 100644 --- a/net/dsa-uclass.c +++ b/net/dsa-uclass.c @@ -284,6 +284,14 @@ static int dsa_port_probe(struct udevice *pdev) if (!master) return -ENODEV; + /* + * Probe the master device. We depend on the master device for proper + * operation and we also need it for MAC inheritance below. + */ + ret = device_probe(master); + if (ret) + return ret; + /* * Inherit port's hwaddr from the DSA master, unless the port already * has a unique MAC address specified in the environment.
DSA needs to have the master device probed first for MAC inheritance. Until now, it only works by chance because the only user (LS1028A SoC) will probe the master device first. The probe order is given by the PCI device ordering, thus it works because the master device has a "smaller" BDF then the switch device. Explicitly probe the master device in dsa_port_probe(). Fixes: fc054d563bfb ("net: Introduce DSA class for Ethernet switches") Signed-off-by: Michael Walle <michael@walle.cc> --- net/dsa-uclass.c | 8 ++++++++ 1 file changed, 8 insertions(+)