tt_ioctl

[Poul-Henning Kamp]

In message <20080409.101629.-146244298.imp at bsdimp.com>, "M. Warner Losh" writes
:
>In message: <50325.1207757374 at critter.freebsd.dk>
>            "Poul-Henning Kamp" <phk at phk.freebsd.dk> writes:
>: In message <18428.59782.318085.53492 at gromit.timing.com>, John E Hein writes:
>: 
>: >That said... if we do decide to _not_ hook up t_ioctl, then we should
>: >just remove it entirely since it's misleading that it's there but not
>: >hooked up.
>: 
>: Well, if we agree that you should use a non-tty device, we may still
>: miss the poster child ioctl to make the decision...
>
>I don't think we agreed that should use a non-tty device.  There's
>good reasons to allow ioctls passed through to the driver.

Yes, but these particular ioctls don't sound anything like that
reason.

If you make them available, you either have to make them unavailable
when the tty is used as a login terminal (either by the documentation
wise laborious cua/tty split or by other worse means)

If you make them available on a separate device, that entire issue
disappears.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.