Xen devel - Jun 2006 - [PATCH]HVM acpi guest OS suppot in piix4 ACPI event logical model-part 2 of 4

Attached please see the hvm guest os acpi patch part 2 of
4:piix4acpi_timer
ACPI timer is required during guest windows installation and boot.

Signed-off-by: Tang Liang <tangliang@lenovo.com>
Signed-off-by: Winston Wang <winston.l.wang@intel.com>



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Hi,

On 6/17/06, Wang, Winston L <winston.l.wang@intel.com>
wrote:
> Attached please see the hvm guest os acpi patch part 2 of
> 4:piix4acpi_timer
> ACPI timer is required during guest windows installation and boot.
The qemu timer used to implement the ACPI timer doesn''t seem to work
quite right, the time it is set to expire becomes immediately out of
syn with qemu''s vm_clock and this then causes the timer to pretty much
fire all the time, resulting in qemu-dm using between 20 and 30% of
CPU on my machine.

How about the following change:
--- tools/ioemu/hw/piix4acpi.c	2006-06-27 11:12:20.000000000 +0100
+++ tools/ioemu.hg/hw/piix4acpi.c	2006-06-29 09:54:56.513574005 +0100
@@ -111,7 +110,8 @@
 static void pm_timer_update(void *opaque)
 {
     PMTState *s = opaque;
-    s->next_pm_time += muldiv64(1, ticks_per_sec,FREQUENCE_PMTIMER);
+    s->next_pm_time = qemu_get_clock(vm_clock) +
+        muldiv64(1, ticks_per_sec,FREQUENCE_PMTIMER);
     qemu_mod_timer(s->pm_timer, s->next_pm_time);
     acpi_state->pm1_timer ++;

I''m not convinced if vm_clock actually works at all for us.

     christian

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel