Andi Kleen wrote:
> On Tuesday 17 January 2006 22:32, Tim Chen wrote:
>> Hi Andi,
>>
>> With this patch in 2.6.15-git12,
>>
>>
(http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a
>> =com mit;h=6c5acd160a10c76e8debf4f8fa8256d7c914f290), I found that
>> kernel could no longer boot when I include "mem=1G" boot option on a
>> dual Xeon machine that has more than 4G of memory. The machine
>> boots normally without "mem=1G" specified or when the patch is
>> backed out. I have included the boot log of the failed boot and the
>> normal boot for your reference. The table_start was relocated from
>> 0x8000 to 0x100000000 with the patch applied.
>
> Still don't understand why 4GB doesn't work with mem=, but git13 will
> not relocate the tables there anymore and shouldn't have this
> problem.
>
> -Andi
Thanks. I have verified that git13 code no longer have this issue.
Tim