Hi Uwe, Am Montag, 4. März 2024, 10:15:57 CET schrieb Uwe Kleine-König: > On 04.03.24 10:07, Heiko Stübner wrote: > > Am Montag, 4. März 2024, 09:46:11 CET schrieb Uwe Kleine-König: > >> While it requires to have the right phy driver loaded (i.e. motorcomm) > >> to make the phy asserting the right delays, this is generally the > >> preferred way to define the MAC <-> PHY connection. > >> > >> Signed-off-by: Uwe Kleine-König <ukleinek@xxxxxxxxxx> > >> --- > >> Hello, > >> > >> Andrew already pointed out when I posted the patch introducing the gmac0 node > >> that rgmii-id would be the preferred way to setup things. Back then this didn't > >> happen because this change broke reception of network packets. However this > >> only happend because I didn't have the right phy driver loaded. > > > > trying to understand how the (not) loaded module fits into this :-) > > The mdio-bus is supposed to probe the phy and load the appropriate module. > > > > From your description it sounds like the correct phy module needs to be > > actually loaded? Or was that meant to be a "requires to have the right phy > > driver compiled" instead? > > The latter. i.e. with MOTORCOMM_PHY=n it's broken, but works fine with > MOTORCOMM_PHY=y or =m. ah great, then it's really fine. If it's ok with you I'll change the "loaded" to "available" then (for compiled in or as module) Heiko
