Subject: Re: Updating Xen's time-of-day clock
To: None <port-xen@NetBSD.org>
From: Jed Davis <jdev@panix.com>
List: port-xen
Date: 05/05/2006 15:09:46
Greg Troxel <gdt@ir.bbn.com> writes:
> I think that periodic resets of the xen clock should not cause the
> printf, at least if the value is "low", and low might be < 200 ppm.
That printf isn't on by default, even with options DEBUG. I think I
need to rename a #define or two to make that clearer.
> Perhaps the hypervisor needs to grow an adjtime call :-)
That was my thought when I saw what Linux is doing. Alternately, the
hypervisor could change the semantics of the settime call -- which,
after all, it needs to have called periodically anyway -- to try to
synchronize with the dom0 in a more gradual manner. (Note that what's
being passed to Xen is both a wall time and an associated system time,
so it has enough info for accurate drift correction.)
--
(let ((C call-with-current-continuation)) (apply (lambda (x y) (x y)) (map
((lambda (r) ((C C) (lambda (s) (r (lambda l (apply (s s) l)))))) (lambda
(f) (lambda (l) (if (null? l) C (lambda (k) (display (car l)) ((f (cdr l))
(C k))))))) '((#\J #\d #\D #\v #\s) (#\e #\space #\a #\i #\newline)))))