Le Tue, 28 Jun 2022 17:42:58 +0100, "Russell King (Oracle)" <linux@xxxxxxxxxxxxxxx> a écrit : > > + break; > > + default: > > + return -EOPNOTSUPP; > > + } > > + > > + val = FIELD_PREP(MIIC_CONVCTRL_CONV_MODE, conv_mode) | > > + FIELD_PREP(MIIC_CONVCTRL_CONV_SPEED, speed); > > + > > + miic_reg_rmw(miic, MIIC_CONVCTRL(port), > > + MIIC_CONVCTRL_CONV_MODE | MIIC_CONVCTRL_CONV_SPEED, val); > > + miic_converter_enable(miic_port->miic, miic_port->port, 1); > > + > > + return 0; > > +} > > the stting of the speed here. As this function can be called as a result > of ethtool setting the configuration while the link is up, this could > have disasterous effects on the link. This will only happen if there is > no PHY present and we aren't using fixed-link mode. > > Therefore, I'm willing to get this pass, but I think it would be better > if the speed was only updated if the interface setting is actually > being changed. So: Hi Russell, Ok, I'll make a follow-up patch to handle that properly. -- Clément Léger, Embedded Linux and Kernel engineer at Bootlin https://bootlin.com
