Re: [PATCH 4/4] x86, efi: Map runtime services 1:1

[Matt Fleming <matt-HNK1S37rvNbeXh+fF434Mdi2O/JbrIOy@public.gmane.org>]

On Sun, 02 Jun, at 02:56:10PM, Borislav Petkov wrote:
> From: Borislav Petkov <bp-l3A5Bk7waGM@public.gmane.org>
> 
> Due to the braindead design of EFI, we cannot map runtime services more
> than once for the duration of a booted system. Thus, if we want to use
> EFI runtime services in a kexec'ed kernel, maybe the only possible and
> sensible approach would be to map them 1:1 so that when the kexec kernel
> loads, it can simply call those addresses without the need for remapping
> (which doesn't work anyway).
> 
> Furthermore, this mapping approach could be of help with b0rked EFI
> implementations for a different set of reasons.
> 
> This implementation is 64-bit only for now and it boots fine in kvm with
> OVMF BIOS.
> 
> Signed-off-by: Borislav Petkov <bp-l3A5Bk7waGM@public.gmane.org>
> ---
>  arch/x86/include/asm/efi.h          |   2 +
>  arch/x86/platform/efi/efi.c         | 161 +++++++++++++++++++++++++++++-------
>  arch/x86/platform/efi/efi_stub_64.S |  48 +++++++++++
>  3 files changed, 180 insertions(+), 31 deletions(-)
 
This patch makes my TunnelMountain machine spin in handle_pte_fault()
when triggering one of the "firmware makes references to physical
addresses" code paths.

I'll try and dig into this tomorrow to figure out what's going on.

-- 
Matt Fleming, Intel Open Source Technology Center