2.6.22 rebase.

[Dave Jones]

On Wed, Jul 11, 2007 at 01:50:45PM +1000, Nigel Cunningham wrote:

 > > actually x86-64 tickless has been in rawhide about 2 weeks, and is now
 > > in this rebase for FC6/F7.  Whether we roll with it for a proper update
 > > depends on how it fares during it's trial in updates-testing.
 > > So far the forced-hpet part was the only bit that's caused trouble
 > > afaics, so with that disabled, it should be (hopefully) boring.
 > 
 > Ah. Now I see it. I was looking for 'hz', not 'hires-timers'. I haven't tried 
 > forced hpet support yet. Am I right in thinking that's only for Intels? 
 > (Mine's an AMD based Mitac).

So far, yes. 

+//DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ESB2_0,
+//                         ich_force_enable_hpet);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_1,
+                         ich_force_enable_hpet);
+//DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_1,
+//                         ich_force_enable_hpet);
+//DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_31,
+//                         ich_force_enable_hpet);
+//DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_1,
+//                         ich_force_enable_hpet);


however on some other chipsets, mainline already does force-enabling..
For example on my VIA EPIA ...

[   28.152336] Force enabled HPET at base address 0xfed00000


	Dave

-- 
http://www.codemonkey.org.uk