Skip to contentSkip to conversation

efi/mokvar: Reserve the table only if it is in boot services data

RFC PATCH In review Borislav Petkov 2 participants 4 messages Jun 30, 2021 – Jul 19, 2021 EFI

lore
Share & tools
lore
#0 lore raw

RFC PATCH 1/1 efi/mokvar: Reserve the table only if it is in boot services data

Hi guys,

so below is what we've been staring at recently, please doublecheck me whether I'm even making sense here.

Thx!

---
From: Borislav Petkov <[email protected]>

One of the SUSE QA tests triggered:

localhost kernel: efi: Failed to lookup EFI memory descriptor for 0x000000003dcf8000

which comes from x86's version of efi_arch_mem_reserve() trying to reserve a memory region. Usually, that function expects EFI_BOOT_SERVICES_DATA memory descriptors but the above case is for the MOKvar table which is allocated in the EFI shim as runtime services.

That lead to a fix changing the allocation of that table to boot services.

However, that fix broke booting SEV guests with that shim leading to this kernel fix

8d651ee9c71b ("x86/ioremap: Map EFI-reserved memory as encrypted for SEV")

which extended the ioremap hint to map reserved EFI boot services as decrypted too.

However, all that wasn't needed, IMO, because that error message in efi_arch_mem_reserve() was innocuous in this case - if the MOKvar table is not in boot services, then it doesn't need to be reserved in the first place because it is, well, in runtime services which *should* be reserved anyway.

So do that reservation for the MOKvar table only if it is allocated in boot services data. I couldn't find any requirement about where that table should be allocated in, unlike the ESRT which allocation is mandated to be done in boot services data by the UEFI spec.

---
 drivers/firmware/efi/mokvar-table.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/firmware/efi/mokvar-table.c +4 -1
index d8bc01340686..38722d2009e2 100644
@@ -180,7 +180,10 @@ void __init efi_mokvar_table_init(void)
 		pr_err("EFI MOKvar config table is not valid\n");
 		return;
 	}
-	efi_mem_reserve(efi.mokvar_table, map_size_needed);
+
+	if (md.type == EFI_BOOT_SERVICES_DATA)
+		efi_mem_reserve(efi.mokvar_table, map_size_needed);
+
 	efi_mokvar_table_size = map_size_needed;
 }
 
Signature · 8 lines
2.29.2


-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
maintainer Acked-by on v1 1/1
#1 lore raw
Quoted 36 lines · Borislav Petkov

Hi guys,

so below is what we've been staring at recently, please doublecheck me whether I'm even making sense here.

Thx!

---
From: Borislav Petkov <[email protected]>

One of the SUSE QA tests triggered:

localhost kernel: efi: Failed to lookup EFI memory descriptor for 0x000000003dcf8000

which comes from x86's version of efi_arch_mem_reserve() trying to reserve a memory region. Usually, that function expects EFI_BOOT_SERVICES_DATA memory descriptors but the above case is for the MOKvar table which is allocated in the EFI shim as runtime services.

That lead to a fix changing the allocation of that table to boot services.

However, that fix broke booting SEV guests with that shim leading to this kernel fix

8d651ee9c71b ("x86/ioremap: Map EFI-reserved memory as encrypted for SEV")

which extended the ioremap hint to map reserved EFI boot services as decrypted too.

However, all that wasn't needed, IMO, because that error message in efi_arch_mem_reserve() was innocuous in this case - if the MOKvar table is not in boot services, then it doesn't need to be reserved in the first place because it is, well, in runtime services which *should* be reserved anyway.

Quoted 6 lines

So do that reservation for the MOKvar table only if it is allocated in boot services data. I couldn't find any requirement about where that table should be allocated in, unlike the ESRT which allocation is mandated to be done in boot services data by the UEFI spec.

Would you like me to queue this as a fix?

Quoted 29 lines
---
 drivers/firmware/efi/mokvar-table.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/firmware/efi/mokvar-table.c +4 -2
index d8bc01340686..38722d2009e2 100644
@@ -180,7 +180,10 @@ void __init efi_mokvar_table_init(void)
                pr_err("EFI MOKvar config table is not valid\n");
                return;
        }
-       efi_mem_reserve(efi.mokvar_table, map_size_needed);
+
+       if (md.type == EFI_BOOT_SERVICES_DATA)
+               efi_mem_reserve(efi.mokvar_table, map_size_needed);
+
        efi_mokvar_table_size = map_size_needed;
 }
 
--

2.29.2

--
Regards/Gruss,
Boris.

https://people.kernel.org/tglx/notes-about-netiquette

#2 lore raw
Quoted 1 line · Ard Biesheuvel

Thanks for checking me here.

Quoted 1 line

Would you like me to queue this as a fix?

Well, I wanna say it is unnecessary to send it as an urgent fix because we have the SEV fix in place now and I guess we'll leave it in in case something else is in boot services and is needed for SEV guests to boot.

So I guess just a normal, unexpedited fix please, so that it sees more testing before it goes up.

Thx Ard!

Signature · 4 lines
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
maintainer
#3 lore raw
Quoted 6 lines · Borislav Petkov
Quoted 1 line · Ard Biesheuvel

Thanks for checking me here.

Quoted 9 lines
Quoted 1 line

Would you like me to queue this as a fix?

Well, I wanna say it is unnecessary to send it as an urgent fix because we have the SEV fix in place now and I guess we'll leave it in in case something else is in boot services and is needed for SEV guests to boot.

So I guess just a normal, unexpedited fix please, so that it sees more testing before it goes up.

I have accumulated a handful of EFI fixes which are not urgent, but fixes nonetheless. I will add it to the set.