Messages in this thread | | | Subject | Re: BUG() at boot in __phys_addr with DEBUG_VIRTUAL | From | Matt Fleming <> | Date | Thu, 13 Nov 2014 13:08:03 +0000 |
| |
On Wed, 2014-11-12 at 07:20 -0800, Dave Hansen wrote: > > The path that we're actually hitting this from is: > > > __split_large_page(struct cpa_data *cpa, pte_t *kpte, unsigned long > ... > > if (pfn_range_is_mapped(PFN_DOWN(__pa(address)), > > PFN_DOWN(__pa(address)) + 1)) > > split_page_count(level);
I'm sure I'm just being dumb, but I can't for the life of me work out why we need to calculate the physical address from the virtual address at this point - we should already have that information.
> So perhaps efi_map_region() is handing an address from the EFI identity > map down in here. __pa() gets called on it, but that fails since __pa() > only works on the *KERNEL* identity map.
Yep, sounds right.
| |