x86/efi: Use efi_switch_mm() rather than manually twiddling with %cr3

Use helper function efi_switch_mm() to switch to/from efi_mm when
invoking any UEFI runtime services.

Likewise, we need to switch back to previous mm (mm context stolen
by efi_mm) after the above calls return successfully. We can use
efi_switch_mm() helper function only with x86_64 kernel and
"efi=old_map" disabled because, x86_32 and efi=old_map do not use
efi_pgd, rather they use swapper_pg_dir.

Tested-by: Bhupesh Sharma <bhsharma@redhat.com>
[ardb: add #include of sched/task.h for task_lock/_unlock]
Signed-off-by: Sai Praneeth Prakhya <sai.praneeth.prakhya@intel.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Reviewed-by: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Lee, Chun-Yi <jlee@suse.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Shankar <ravi.v.shankar@intel.com>
Cc: Ricardo Neri <ricardo.neri@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Cc: linux-efi@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>

arch/x86/include/asm/efi.h

@@ -70,14 +70,13 @@ extern asmlinkage u64 efi_call(void *fp, ...);
 #define efi_call_phys(f, args...)		efi_call((f), args)
 
 /*
- * Scratch space used for switching the pagetable in the EFI stub
+ * struct efi_scratch - Scratch space used while switching to/from efi_mm
+ * @phys_stack: stack used during EFI Mixed Mode
+ * @prev_mm:    store/restore stolen mm_struct while switching to/from efi_mm
  */
 struct efi_scratch {
-	u64	r15;
+	u64			phys_stack;
-	u64	prev_cr3;
+	struct mm_struct	*prev_mm;
-	pgd_t	*efi_pgt;
-	bool	use_pgd;
-	u64	phys_stack;
 } __packed;
 
 #define arch_efi_call_virt_setup()					\
@@ -87,11 +86,8 @@ struct efi_scratch {
 	__kernel_fpu_begin();						\
 	firmware_restrict_branch_speculation_start();			\
 									\
-	if (efi_scratch.use_pgd) {					\
+	if (!efi_enabled(EFI_OLD_MEMMAP))				\
-		efi_scratch.prev_cr3 = __read_cr3();			\
+		efi_switch_mm(&efi_mm);					\
-		write_cr3((unsigned long)efi_scratch.efi_pgt);		\
-		__flush_tlb_all();					\
-	}								\
 })
 
 #define arch_efi_call_virt(p, f, args...)				\
@@ -99,10 +95,8 @@ struct efi_scratch {
 
 #define arch_efi_call_virt_teardown()					\
 ({									\
-	if (efi_scratch.use_pgd) {					\
+	if (!efi_enabled(EFI_OLD_MEMMAP))				\
-		write_cr3(efi_scratch.prev_cr3);			\
+		efi_switch_mm(efi_scratch.prev_mm);			\
-		__flush_tlb_all();					\
-	}								\
 									\
 	firmware_restrict_branch_speculation_end();			\
 	__kernel_fpu_end();						\
@@ -145,6 +139,7 @@ extern void __init efi_dump_pagetable(void);
 extern void __init efi_apply_memmap_quirks(void);
 extern int __init efi_reuse_config(u64 tables, int nr_tables);
 extern void efi_delete_dummy_variable(void);
+extern void efi_switch_mm(struct mm_struct *mm);
 
 struct efi_setup_data {
 	u64 fw_vendor;

arch/x86/platform/efi/efi_64.c

@@ -34,6 +34,7 @@
 #include <linux/slab.h>
 #include <linux/ucs2_string.h>
 #include <linux/mem_encrypt.h>
+#include <linux/sched/task.h>
 
 #include <asm/setup.h>
 #include <asm/page.h>
@@ -82,9 +83,8 @@ pgd_t * __init efi_call_phys_prolog(void)
 	int n_pgds, i, j;
 
 	if (!efi_enabled(EFI_OLD_MEMMAP)) {
-		save_pgd = (pgd_t *)__read_cr3();
+		efi_switch_mm(&efi_mm);
-		write_cr3((unsigned long)efi_scratch.efi_pgt);
+		return NULL;
-		goto out;
 	}
 
 	early_code_mapping_set_exec(1);
@@ -156,8 +156,7 @@ void __init efi_call_phys_epilog(pgd_t *save_pgd)
 	pud_t *pud;
 
 	if (!efi_enabled(EFI_OLD_MEMMAP)) {
-		write_cr3((unsigned long)save_pgd);
+		efi_switch_mm(efi_scratch.prev_mm);
-		__flush_tlb_all();
 		return;
 	}
 
@@ -347,13 +346,6 @@ int __init efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages)
 	if (efi_enabled(EFI_OLD_MEMMAP))
 		return 0;
 
-	/*
-	 * Since the PGD is encrypted, set the encryption mask so that when
-	 * this value is loaded into cr3 the PGD will be decrypted during
-	 * the pagetable walk.
-	 */
-	efi_scratch.efi_pgt = (pgd_t *)__sme_pa(pgd);
-
 	/*
 	 * It can happen that the physical address of new_memmap lands in memory
 	 * which is not mapped in the EFI page table. Therefore we need to go
@@ -367,8 +359,6 @@ int __init efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages)
 		return 1;
 	}
 
-	efi_scratch.use_pgd = true;
-
 	/*
 	 * Certain firmware versions are way too sentimential and still believe
 	 * they are exclusive and unquestionable owners of the first physical page,
@@ -627,6 +617,22 @@ void __init efi_dump_pagetable(void)
 #endif
 }
 
+/*
+ * Makes the calling thread switch to/from efi_mm context. Can be used
+ * for SetVirtualAddressMap() i.e. current->active_mm == init_mm as well
+ * as during efi runtime calls i.e current->active_mm == current_mm.
+ * We are not mm_dropping()/mm_grabbing() any mm, because we are not
+ * losing/creating any references.
+ */
+void efi_switch_mm(struct mm_struct *mm)
+{
+	task_lock(current);
+	efi_scratch.prev_mm = current->active_mm;
+	current->active_mm = mm;
+	switch_mm(efi_scratch.prev_mm, mm, NULL);
+	task_unlock(current);
+}
+
 #ifdef CONFIG_EFI_MIXED
 extern efi_status_t efi64_thunk(u32, ...);
 
@@ -680,16 +686,13 @@ efi_status_t efi_thunk_set_virtual_address_map(
 	efi_sync_low_kernel_mappings();
 	local_irq_save(flags);
 
-	efi_scratch.prev_cr3 = __read_cr3();
+	efi_switch_mm(&efi_mm);
-	write_cr3((unsigned long)efi_scratch.efi_pgt);
-	__flush_tlb_all();
 
 	func = (u32)(unsigned long)phys_set_virtual_address_map;
 	status = efi64_thunk(func, memory_map_size, descriptor_size,
 			     descriptor_version, virtual_map);
 
-	write_cr3(efi_scratch.prev_cr3);
+	efi_switch_mm(efi_scratch.prev_mm);
-	__flush_tlb_all();
 	local_irq_restore(flags);
 
 	return status;

arch/x86/platform/efi/efi_thunk_64.S

@@ -33,7 +33,7 @@ ENTRY(efi64_thunk)
 	 * Switch to 1:1 mapped 32-bit stack pointer.
 	 */
 	movq	%rsp, efi_saved_sp(%rip)
-	movq	efi_scratch+25(%rip), %rsp
+	movq	efi_scratch(%rip), %rsp
 
 	/*
 	 * Calculate the physical address of the kernel text.