Hi Russell, thanks for review On 5/19/23 16:43, Russell King (Oracle) wrote: > On Fri, May 19, 2023 at 04:13:03PM +0200, alexis.lothore@xxxxxxxxxxx wrote: >> From: Alexis Lothoré <alexis.lothore@xxxxxxxxxxx> >> >> Marvell 88E6361 is an 8-port switch derived from the >> 88E6393X/88E9193X/88E6191X switches family. It can benefit from the >> existing mv88e6xxx driver by simply adding the proper switch description in >> the driver. Main differences with other switches from this >> family are: >> - 8 ports exposed (instead of 11): ports 1, 2 and 8 not available >> - No 5GBase-x nor SFI/USXGMII support >> >> --- >> Changes since v1: >> - define internal phys offset >> - enforce 88e6361 features in mv88e6393x_phylink_get_caps >> - enforce 88e6361 features in mv88e6393x_port_set_speed_duplex >> - enforce 88e6361 features in mv88e6393x_port_max_speed_mode > > Not exactly related to this patch, but please do not rely on this "max > speed mode" - please always ensure that you specify the phy-mode and > fixed-link settings for CPU and DSA ports in firmware. Thanks. I would like to make sure to fully understand your point: - when telling so specify phy-mode and fixed-link in firmware, you mean device-tree, right ? - when checking for code and execution flow, I observe that port_max_speed is always called and its output is always used to configure shared ports mode in mv88e6xxx driver. Are you telling that eventually, the whole mv88e6xxx driver should stop relying on port_max_speed_mode for shared ports ? Kind regards, -- Alexis Lothoré, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
