Re: state of some x86 acpi patches

[Yinghai Lu <yinghai@kernel.org>]

Tony Luck wrote:
> On Sun, Feb 15, 2009 at 3:10 PM, Yinghai Lu <yinghai@kernel.org> wrote:
>> Index: linux-2.6/arch/ia64/kernel/acpi.c
>> ===================================================================
>> --- linux-2.6.orig/arch/ia64/kernel/acpi.c
>> +++ linux-2.6/arch/ia64/kernel/acpi.c
>> @@ -199,7 +199,7 @@ char *__init __acpi_map_table(unsigned l
>>        return __va(phys_addr);
>>  }
>>
>> -char *__init __acpi_unmap_table(unsigned long virt_addr, unsigned long size)
>> +void __init __acpi_unmap_table(char *map, unsigned long size)
>>  {
>>  }
> 
> I'm seeing some section mismatch warnings for this function (when building
> linux-next tag: next-20090220) as it is called by some non __init functions:
> 
> WARNING: vmlinux.o(.text+0x498912): Section mismatch in reference from
> the function early_acpi_os_unmap_memory() to the function
> .init.text:__acpi_unmap_table()
> 
> WARNING: vmlinux.o(.text+0x4989a2): Section mismatch in reference from
> the function acpi_os_unmap_memory() to the function
> .init.text:__acpi_unmap_table()
> 
> Simplest solution would seem to be dropping the "__init" (since the
> function is empty, it won't bloat the kernel by much if we don't manage
> to free up the memory - just 32 bytes :-)

it seems Jeremy has one patch for that...not sure if Ingo put that in tip or not..

YH