* create_periodic_time(): why forced to 0.9ms ?
@ 2007-06-29 22:41 Russ Blaine
2007-07-01 20:44 ` Keir Fraser
0 siblings, 1 reply; 3+ messages in thread
From: Russ Blaine @ 2007-06-29 22:41 UTC (permalink / raw)
To: xen-devel
create_periodic_time() forces the period to be at least 0.9 ms. Can anyone
explain the history behind this? Why is this done, and how was this constant
chosen?
-----------------------------------------------------
Russ Blaine | Solaris Kernel | russell.blaine@sun.com
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: create_periodic_time(): why forced to 0.9ms ?
2007-06-29 22:41 create_periodic_time(): why forced to 0.9ms ? Russ Blaine
@ 2007-07-01 20:44 ` Keir Fraser
2007-07-02 5:33 ` Zhai, Edwin
0 siblings, 1 reply; 3+ messages in thread
From: Keir Fraser @ 2007-07-01 20:44 UTC (permalink / raw)
To: Russ Blaine, xen-devel
That's a question for whoever at Intel originally wrote that code. I'm sure
the answer is that anything smaller is too small. :-) Which is kind of
reasonable...
-- Keir
On 29/6/07 23:41, "Russ Blaine" <russell.blaine@sun.com> wrote:
> create_periodic_time() forces the period to be at least 0.9 ms. Can anyone
> explain the history behind this? Why is this done, and how was this constant
> chosen?
>
> -----------------------------------------------------
> Russ Blaine | Solaris Kernel | russell.blaine@sun.com
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: create_periodic_time(): why forced to 0.9ms ?
2007-07-01 20:44 ` Keir Fraser
@ 2007-07-02 5:33 ` Zhai, Edwin
0 siblings, 0 replies; 3+ messages in thread
From: Zhai, Edwin @ 2007-07-02 5:33 UTC (permalink / raw)
To: Keir Fraser; +Cc: Russ Blaine, xen-devel
I'm not the original writer, but know some reason for this:)
At the beginning, this logic is used for safe guard in case bad guest set too
low freq in PIT channel.
The lowest PIT freq we known that time is 1ms (linux 2.6 kernel) and allow some
inaccuracy because of the round divide(%10, i.e 0.9 ms as you can see).
On Sun, Jul 01, 2007 at 09:44:43PM +0100, Keir Fraser wrote:
> That's a question for whoever at Intel originally wrote that code. I'm sure
> the answer is that anything smaller is too small. :-) Which is kind of
> reasonable...
>
> -- Keir
>
> On 29/6/07 23:41, "Russ Blaine" <russell.blaine@sun.com> wrote:
>
> > create_periodic_time() forces the period to be at least 0.9 ms. Can anyone
> > explain the history behind this? Why is this done, and how was this constant
> > chosen?
> >
> > -----------------------------------------------------
> > Russ Blaine | Solaris Kernel | russell.blaine@sun.com
> >
> > _______________________________________________
> > Xen-devel mailing list
> > Xen-devel@lists.xensource.com
> > http://lists.xensource.com/xen-devel
>
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
>
--
best rgds,
edwin
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2007-07-02 5:33 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-06-29 22:41 create_periodic_time(): why forced to 0.9ms ? Russ Blaine
2007-07-01 20:44 ` Keir Fraser
2007-07-02 5:33 ` Zhai, Edwin
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.