On 01/30/2012 05:59 AM, Paolo Bonzini wrote:
On 01/30/2012 12:56 PM, Stefano Stabellini wrote:
> Depending on "-rtc clock=vm" or "-rtc clock=rt", this may not be true.
Good point.
I should check for rtc_clock == host_clock.
> Why do you need to instantiate an RTC at all?
I don't, in fact in previous versions of this patch series I tried to do
that but it is difficult and makes the common code worse, so I followed
Anthony's suggestion of just disabling the rtc_clock.
I see. It shouldn't surprise you that I completely disagree with Anthony on
this. :)
Give me some credit at least...
The original patches didn't disable the RTC, it introduced a half-neutered Xen
specific RTC.
My original suggestion (which I still think is the best approach) is to change
the RTC emulation to not use a second timer at all.
Regards,
Anthony Liguori
Paolo
