x86/efi: Make efi_set_virtual_address_map IBT safe

Niklāvs reported a boot regression on an Alderlake machine and bisected it
to commit 9df9d2f ("init: Invoke arch_cpu_finalize_init() earlier").

By moving the invocation of arch_cpu_finalize_init() further down he
identified that efi_enter_virtual_mode() is the function which causes the
boot hang.

The main difference of the earlier invocation is that the boot CPU is
already fully initialized and mitigations and alternatives are applied.

But the only really interesting change turned out to be IBT, which is now
enabled before efi_enter_virtual_mode(). "ibt=off" on the kernel command
line cured the problem.

Inspection of the involved calls in efi_enter_virtual_mode() unearthed that
efi_set_virtual_address_map() is the only place in the kernel which invokes
an EFI call without the IBT safe wrapper. This went obviously unnoticed so
far as IBT was enabled later.

Use arch_efi_call_virt() instead of efi_call() to cure that.

Fixes: fe379fa ("x86/ibt: Disable IBT around firmware")
Fixes: 9df9d2f ("init: Invoke arch_cpu_finalize_init() earlier")
Reported-by: Niklāvs Koļesņikovs <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Reviewed-by: Ard Biesheuvel <[email protected]>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=217602
Link: https://lore.kernel.org/r/87jzvm12q0.ffs@tglx

Author: KAGA-KOKO

arch/x86/platform/efi/efi_64.c

@@ -853,9 +853,9 @@ efi_set_virtual_address_map(unsigned long memory_map_size,
 
 	/* Disable interrupts around EFI calls: */
 	local_irq_save(flags);
-	status = efi_call(efi.runtime->set_virtual_address_map,
-			  memory_map_size, descriptor_size,
-			  descriptor_version, virtual_map);
+	status = arch_efi_call_virt(efi.runtime, set_virtual_address_map,
+				    memory_map_size, descriptor_size,
+				    descriptor_version, virtual_map);
 	local_irq_restore(flags);
 
 	efi_fpu_end();