[Crash-utility] [PATCH 6/7] x86, kaslr: report kernel offset on panic

H. Peter Anvin hpa at zytor.com
Wed Oct 9 14:13:26 UTC 2013


On 10/09/2013 03:04 AM, HATAYAMA Daisuke wrote:
>>>
>>> Sorry, I don't understand why relocation size cannot be calculated when
>>> CONFIG_PHYSICALSTART > CONFIG_PHYSICAL_ALIGN. Could you explain that?
>>
>> I just meant that when CONFIG_PHYSICAL_START > CONFIG_PHYSICAL_ALIGN,
>> the 32-bit x86 kernel
>> gets relocated (like the secondary kdump kernel), but that information
>> is not readily available
>> from the vmlinux/vmcore pair.
>>
> 
> My understanding on CONFIG_PHYSICAL_ALIGN was that starting address of
> kernel text area
> is always rounded up to CONFIG_PHYSICAL_ALIGN, only. Your explanation
> would be part I don't
> understand well. I'll reconsider it locally...
> 

If CONFIG_PHYSICAL_START == CONFIG_PHYSICAL_ALIGN, then it is very
likely that the kernel (in the absence of kASLR) will be run at the
CONFIG_PHYSICAL_START address, as the initial loading address, usually 1
MB, will be rounded up to CONFIG_PHYSICAL_ALIGN.

Since CONFIG_PHYSICAL_START is the unrelocated linking address, they end
up matching.

	-hpa





More information about the Crash-utility mailing list