Re: [gnumach] /dev/cons/kbd rejects kbd_drv.so set-event-mode ioctl with EBADF; please implement KDSKBMODE or expose /dev/input/event*

Hi Samuel,

Re-probed with `fsysopts`, you were right on both points:

  fsysopts /dev/vcs   -> /hurd/console  (directory, empty)
  fsysopts /dev/cons  -> ext2fs         (regular file on rootfs)
  /etc/default/hurd-console first line -> ENABLE='true'

So /dev/vcs is the directory with the active translator, and my
original probe was looking at the wrong node.  Premise of the
report does not hold on this image; withdrawing the gnumach
KDSKBMODE ask.

If hurd-i386 surfaces a real amd64-specific issue later, I will
file fresh against the right layer rather than reviving this
thread.

Thanks for walking me through the diagnostics.

Borja Tarraso
[email protected]

El dom, 31 may 2026 a las 19:58, Samuel Thibault
(<[email protected]>) escribió:
>
> Samuel Thibault, le dim. 31 mai 2026 18:55:29 +0200, a ecrit:
> > Firstly, the 64bit port of the Hurd is very recent, so expect things not
> > to be completely done.
>
> Meaning: do test on hurd-i386 before assuming there is some
> documentation changes to be made. If it's amd64-specific, better just
> fix the issue.
>
> Samuel