On Tue, Sep 30, 2014 at 09:04:16AM +0200, Geert Uytterhoeven wrote: > Hi Simon, > > On Tue, Sep 30, 2014 at 8:30 AM, Simon Horman <horms@xxxxxxxxxxxx> wrote: > > On Tue, Sep 30, 2014 at 11:07:56AM +0900, Simon Horman wrote: > >> On Thu, Sep 25, 2014 at 06:13:16PM +0200, Geert Uytterhoeven wrote: > >> > This patch series adds chosen/stdout-path to the various shmobile DTS files > >> > that are used for reference or multi-platform builds. > >> > > >> > This serves two purposes: > >> > 1. It makes automatic console selection work in the absence of a "console=" > >> > parameter on the kernel command line, allowing to remove "console=" > >> > parameters from chosen/bootargs on many platforms. > >> > Note that these "console=" parameters are still needed in some cases: > >> > a. If console output is wanted on two consoles devices (graphical and > >> > serial consoles), like on armadillo and kzm9g, > >> > b. When using a non-default baud for the serial console, like on > >> > henninger and alt, > >> > 2. It will allow the PM domain code to find the PM domain for the console > >> > device, to make "no_console_suspend" work in the DT case. > >> > For now, this will matter on armadillo. > >> > > >> > The first two patches are small cleanups for emev2 dtsi, on top of which the > >> > change for kzm9d is built. > >> > >> Thanks, I have queued these up for v3.19. > > > > I spoke a little too soon. > > > > While testing these patches I noticed that the koelsch patch > > cases there to be no kernel output to the console (during boot) > > when booting using legacy C. > > > > I assume that this is due to removing > > console=ttySC6,115200 from bootargs. > > Sorry, in my mind koelsch legacy is already gone ;-) > > > I also assume this affects all boards that can still > > be booted using legacy C. > > Only the ones where the DTS is shared between legacy and > reference/multi-platform. > Armadillo is not affected, as the console parameters are kept to have > output on both graphical and serial console. > So that leaves us with lager and koelsch only. > > > I have dropped this series for now. > > Do you want me to resend with the console= parameter kept for koelsch > and lager? > Or should we delay this whole series until koelsch and lager legacy > have been dropped? I think it would be best if you reposted the series. > IIUC, this is going to happen soon for koelsch, but not for lager? Yes, that is my understanding too. -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html