Messages in this thread | | | Date | Thu, 4 Oct 2012 09:19:08 -0700 | Subject | Re: [PATCH 04/13] x86, mm: Revert back good_end setting for 64bit | From | Yinghai Lu <> |
| |
On Wed, Oct 3, 2012 at 9:51 AM, Jacob Shin <jacob.shin@amd.com> wrote: > Any comments, thoughts? hpa? Yinghai? > > So it seems that during init_memory_mapping Xen needs to modify page table > bits and the memory where the page tables live needs to be direct mapped at > that time. > > Since we now call init_memory_mapping for every E820_RAM range sequencially, > the only way to satisfy Xen is to find_early_page_table_space (good_end needs > to be within memory already mapped at the time) for every init_memory_mapping > call. > > What do you think Yinghai?
that may put the page table on near end of every ram range for next memory range.
then kdump may have problem get big range again.
Yinghai
| |