Re: AMD GPU driver load hitting BUG_ON in sync_global_pgds_l5()

[Bharata B Rao <bharata@amd.com>]

On 23-Apr-25 9:31 PM, Dave Hansen wrote:
> On 4/23/25 02:30, Bharata B Rao wrote:
>> On 22-Apr-25 8:43 PM, Dave Hansen wrote:
>>> On 4/21/25 23:34, Bharata B Rao wrote:
>>>> At the outset, it appears that the selection of vmemmap_base doesn't
>>>> seem to consider if there is going to be enough room of accommodating
>>>> future hot plugged pages.
>>>
>>> Is this future hotplug area in the memory map at boot?
>>
>> The KVM guest isn't using any -m maxmem option if that's what you are
>> hinting at.
> 
> How could vmemmap_base consider future hotplug areas if it isn't told
> where they will be?

This is device private memory which means only struct pages need to be 
mapped. What's the way by which kernel will know about the future growth 
in the number of struct pages to accommodate the incoming device private 
memory?

In any case, how can kaslr put vmemmap_base completely out of the range 
earmarked for it in mm.rst?

Regards,
Bharata.