Bizarre clone attempt failures on Raspberry Pi2...

[Luiz Otavio O Souza]

On 15 July 2016 at 14:27, bob prohaska wrote:
> On Fri, Jul 15, 2016 at 12:33:14PM -0400, Paul Mather wrote:
>>
>> Incidentally, does setting console="vidconsole" in /boot/loader.conf fix the problem of a lack of /etc/rc messages for those who are using an HDMI monitor as their primary/only console?
>
> Not with that particular syntax. It just produces a message about "bad definition". HDMI
> output ends with the kernel probe, resuming with a login prompt.

No, setting console to vidconsole doesn't work.

See conscontrol(8), the first high level console is always used and in
this case, it is the serial console as the vt console comes too late
in boot (there is another kind of driver that implements the vt_early
support to cope with that, but I can't say if that is a doable
solution for RPi - IIRC I once thought it was possible...).

That said, I have this _hack_ that changes the high level console to
vt once it is attached (not a real solution, just quick test for debug
purposes):

Index: sys/dev/vt/vt_core.c
===================================================================
--- sys/dev/vt/vt_core.c        (revision 279665)
+++ sys/dev/vt/vt_core.c        (working copy)
@@ -2686,6 +2686,7 @@
         * it.
         */
        termcn_cnregister(vd->vd_windows[VT_CONSWINDOW]->vw_terminal);
+       cnselect(vd->vd_windows[VT_CONSWINDOW]->vw_terminal->consdev);
 }

 static void


After the boot you can always change your console device back and
forth with the sysctl kern.console.

The real solution is implement what Warner said, a real multiple
device support for high level console.


Luiz