Re: [Xen-devel] CONFIG_SCRUB_DEBUG=y + arm64 + livepatch = Xen BUG at page_alloc.c:738

[Julien Grall <julien.grall@xxxxxxx>]

Hi,

On 15/09/2017 20:50, Konrad Rzeszutek Wilk wrote:
> .snip..
> > (XEN) Loading dom0 DTB to 0x0000000017e00000-0x0000000017e08265
> > (XEN) init_domheap_pages: 0xb87b1->0xb87bc
> > (XEN) init_heap_pages: 0xb87b1 -> 0xb87bc
> > (XEN) init_domheap_pages: 0xb88f1->0xb98ae
> > (XEN) init_heap_pages: 0xb88f1 -> 0xb98ae    <- so the memory is from here
> >
> > (XEN) Scrubbing Free RAM on 1 nodes using 8 CPUs
> > (XEN) Initial low memory virq threshold set at 0x4000 pages.
> > (XEN) Scrubbing Free RAM on 1 nodes using 8 CPUs
> I expanded this a bit to include:
>
> (XEN) Initial low memory virq threshold set at 0x4000 pages.
> (XEN) NODE0 start=0x0 pages=579306, max_page=0xc0000
> (XEN) NODE0:  #0  #1  #2  #3  #4  #5  #6  #7  (total=8)
> (XEN) NODE0 start=0x0, per_cpu=72413, rem=2, end=0x8d6ea
>                                                  ^^^^^^^^  that is the end of 
> memory, but
> 0xb98ae > 0x8d6ae.
>
> end = min(node_start_pfn(i) + node_spanned_pages(i), max_page);
>           0                   0x8D6EA                0xc0000
>
>
> Oh:
>
> /* XXX: implement NUMA support */
> #define node_spanned_pages(nid) (total_pages)
If I understand correctly, this would be wrong here because the memory 
could be in multiple banks.
So, I think, what we would want to return here is bank[n]->end - 
bank[0]->start where n is the last bank.
Cheers,

--
Julien Grall

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel