struct timex and Linux adjtimex()

[Poul-Henning Kamp]

--------
Konstantin Belousov writes:

> 1. Implement new syscall, which would take extended struct timex.
>    ntp_adjtimex() perhaps should be kept for backward compatibility.
>    [It does not matter where struct timeval is placed in the updated
>    struct timex, see below].

That would break all ports with timekeeping software.

Given the role of keeping the system-clock wrangled is single-user
and not something you would normally run with old-version software,
the simplest solution *if* we want to do this, is to simply extend
struct timex for 13-R.

BTW: Does anybody know what (is supposed to) happens of ADJ_SETOFFSET
is active with other modes in struct timex, in particular MODE_OFFSET ?

Is there a document describing how that is supposed to work ?

> Add me to the review anyway.

Me to.

> NB. compat32 for ntp_adjtime(2) seems to be completely broken.  It is probabl
> y
> a good moment to fix old syscall, since you would need to implement compat32
> shims for the new one anyway.

As per above:  Compat'ing this syscall makes very little sense in practice.

-- 
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.