RE: [PATCH V2 1/2] x86/efi: Unmap EFI boot services code/data regions from efi_pgd
From: Prakhya, Sai Praneeth
Date: Mon Oct 29 2018 - 14:32:40 EST
> > In general it's good to do on list because others can learn from the
> > answers as well.
> So you somehow managed to keep everyone and the lists in CC nevertheless.
> Did not pay attention to the CC list until I got my reply on the list :)
Sorry! my bad.. yes, the conversation did happen on the list and it's by accident.
> > > > > > + retval = __change_page_attr_set_clr(&cpa, 0);
> > > So some questions I have are,
> > > 1. What did Peter Z mean here? "How is that not a TLB invalidation bug ?"
> > __flush_tlb_all() flushes only the mapping on the current CPU. So in a
> > SMP environment it's not sufficient.
Thanks for the explanation. The issue makes sense to me now.
> > I'll have a second look at the whole thing and reply on list.
> So actually for the problem at hand __flush_tlb_all() works, because that is
> called way before the secondary CPUs are up.
> Ditto for kernel_map_pages_in_pgd().
Yes, you are right. Both the functions are *always* called before SMP initialization
and as you rightly pointed out they should be marked with __init.
> But both functions want to be marked
> __init and aside of that they both should contain a warning when they are called
> after the secondary CPUs have been brought up.
Makes sense too.. I will include these changes in V3.