Hi, On 23.05.22 13:04, Sascha Hauer wrote: > On Mon, May 23, 2022 at 11:25:23AM +0200, Ahmad Fatoum wrote: >> Allow barebox compute a suitable earlycon parameter. >> >> Signed-off-by: Ahmad Fatoum <a.fatoum@xxxxxxxxxxxxxx> >> --- >> drivers/serial/serial_imx.c | 2 ++ >> 1 file changed, 2 insertions(+) >> >> diff --git a/drivers/serial/serial_imx.c b/drivers/serial/serial_imx.c >> index 6a933c930e9b..cf9c3f02a7ef 100644 >> --- a/drivers/serial/serial_imx.c >> +++ b/drivers/serial/serial_imx.c >> @@ -234,6 +234,8 @@ static int imx_serial_probe(struct device_d *dev) >> cdev->flush = imx_serial_flush; >> cdev->setbrg = imx_serial_setbaudrate; >> cdev->linux_console_name = "ttymxc"; >> + cdev->linux_earlycon_name = "ec_imx6q"; > > "ec_imx6q" sounds wrong for the older i.MX SoCs. I don't know how > OF_EARLYCON_DECLARE works, but this: > > OF_EARLYCON_DECLARE(ec_imx6q, "fsl,imx6q-uart", imx_console_early_setup); > OF_EARLYCON_DECLARE(ec_imx21, "fsl,imx21-uart", imx_console_early_setup); > > looks like it matches a device tree compatible entry. For the older SoCs > you should likely use "ec_imx21". Oh, I can fix this for v2. > Anyway, plain "earlycon" just works, why the hassle? Not all boards have a DT with stdout-path or probe from device tree. I ran into this issue three times now (twice DT-enabled boards, but no stdout-path, once on x86 with no DT) and figured it would be nifty if barebox could just optionally generate it. Cheers, Ahmad > > Sascha > -- Pengutronix e.K. | | Steuerwalder Str. 21 | http://www.pengutronix.de/ | 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 | _______________________________________________ barebox mailing list barebox@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/barebox
