On Sat, Aug 07, 2021 at 05:40:35PM +0200, Andrew Lunn wrote: > I suspect it is a 50/50 roll of a dice what rx and tx actually > mean. Is it from the perspective of the MAC or the PHY? Luckily, > rgmii-rxid and rgmii-txid don't appear in DT very often. I checked an NXP board schematic which has both, and: When you connect a MAC to a PHY using RGMII, the RXD[3:0] pins of the MAC connect to the RXD[3:0] pins of the PHY, and the TXD[3:0] goes to TXD[3:0]. So it is neither the perspective of the MAC nor of the PHY. When you connect a MAC to a MAC using RGMII, the RXD[3:0] of one MAC goes to the TXD[3:0] of the other, and vice versa. Nonetheless, a phy-mode of "rgmii-rxid" always means to the local MAC that "the RX delays have been dealt with" - either by PCB traces, or a remote MAC applying TX delay, or the PHY applying RX delay.
