On Tue, Jan 12, 2010 at 5:35 AM, Wu Fengguang <fengguang.wu@intel.com> wrote:
> On Tue, Jan 12, 2010 at 10:39:03AM +0800, KAMEZAWA Hiroyuki wrote:
>> On Tue, 12 Jan 2010 10:33:08 +0800
>> Wu Fengguang <fengguang.wu@intel.com> wrote:
>>
>> > Sure, here it is :)
>> > ---
>> > x86: use the generic page_is_ram()
>> >
>> > The generic resource based page_is_ram() works better with memory
>> > hotplug/hotremove. So switch the x86 e820map based code to it.
>> >
>> > CC: Andi Kleen <andi@firstfloor.org>
>> > CC: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
>> > Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
>>
>> Ack.
>
> Thank you.
>
>>
>> > +#ifdef CONFIG_X86
>> > +   /*
>> > +    * A special case is the first 4Kb of memory;
>> > +    * This is a BIOS owned area, not kernel ram, but generally
>> > +    * not listed as such in the E820 table.
>> > +    */
>> > +   if (pfn == 0)
>> > +           return 0;
>> > +
>> > +   /*
>> > +    * Second special case: Some BIOSen report the PC BIOS
>> > +    * area (640->1Mb) as ram even though it is not.
>> > +    */
>> > +   if (pfn >= (BIOS_BEGIN >> PAGE_SHIFT) &&
>> > +       pfn <  (BIOS_END   >> PAGE_SHIFT))
>> > +           return 0;
>> > +#endif
>>
>> I'm glad if this part is sorted out in clean way ;)
>
> Two possible solutions are:
>
> - to exclude the above two ranges directly in e820 map;
> - to not add the above two ranges into iomem_resource.
>
> Yinghai, do you have any suggestions?
> We want to get rid of the two explicit tests from page_is_ram().

please check attached patch.

YH