3 files changed, 229 insertions, 97 deletions

diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c
index b97acecf3fd9..45d4f7674678 100644
--- a/arch/x86/platform/efi/efi.c
+++ b/arch/x86/platform/efi/efi.c
@@ -68,9 +68,7 @@ struct efi_memory_map memmap;
 static struct efi efi_phys __initdata;
 static efi_system_table_t efi_systab __initdata;
 
-unsigned long x86_efi_facility;
-
-static __initdata efi_config_table_type_t arch_tables[] = {
+static efi_config_table_type_t arch_tables[] __initdata = {
 #ifdef CONFIG_X86_UV
 	{UV_SYSTEM_TABLE_GUID, "UVsystab", &efi.uv_systab},
 #endif
@@ -79,16 +77,7 @@ static __initdata efi_config_table_type_t arch_tables[] = {
 
 u64 efi_setup;		/* efi setup_data physical address */
 
-/*
- * Returns 1 if 'facility' is enabled, 0 otherwise.
- */
-int efi_enabled(int facility)
-{
-	return test_bit(facility, &x86_efi_facility) != 0;
-}
-EXPORT_SYMBOL(efi_enabled);
-
-static bool __initdata disable_runtime = false;
+static bool disable_runtime __initdata = false;
 static int __init setup_noefi(char *arg)
 {
 	disable_runtime = true;
@@ -275,9 +264,9 @@ static efi_status_t __init phys_efi_get_time(efi_time_t *tm,
 int efi_set_rtc_mmss(const struct timespec *now)
 {
 	unsigned long nowtime = now->tv_sec;
-	efi_status_t 	status;
-	efi_time_t 	eft;
-	efi_time_cap_t 	cap;
+	efi_status_t	status;
+	efi_time_t	eft;
+	efi_time_cap_t	cap;
 	struct rtc_time	tm;
 
 	status = efi.get_time(&eft, &cap);
@@ -295,9 +284,8 @@ int efi_set_rtc_mmss(const struct timespec *now)
 		eft.second = tm.tm_sec;
 		eft.nanosecond = 0;
 	} else {
-		printk(KERN_ERR
-		       "%s: Invalid EFI RTC value: write of %lx to EFI RTC failed\n",
-		       __FUNCTION__, nowtime);
+		pr_err("%s: Invalid EFI RTC value: write of %lx to EFI RTC failed\n",
+		       __func__, nowtime);
 		return -1;
 	}
 
@@ -413,8 +401,7 @@ static void __init print_efi_memmap(void)
 	     p < memmap.map_end;
 	     p += memmap.desc_size, i++) {
 		md = p;
-		pr_info("mem%02u: type=%u, attr=0x%llx, "
-			"range=[0x%016llx-0x%016llx) (%lluMB)\n",
+		pr_info("mem%02u: type=%u, attr=0x%llx, range=[0x%016llx-0x%016llx) (%lluMB)\n",
 			i, md->type, md->attribute, md->phys_addr,
 			md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT),
 			(md->num_pages >> (20 - EFI_PAGE_SHIFT)));
@@ -446,9 +433,8 @@ void __init efi_reserve_boot_services(void)
 			memblock_is_region_reserved(start, size)) {
 			/* Could not reserve, skip it */
 			md->num_pages = 0;
-			memblock_dbg("Could not reserve boot range "
-					"[0x%010llx-0x%010llx]\n",
-						start, start+size-1);
+			memblock_dbg("Could not reserve boot range [0x%010llx-0x%010llx]\n",
+				     start, start+size-1);
 		} else
 			memblock_reserve(start, size);
 	}
@@ -456,7 +442,7 @@ void __init efi_reserve_boot_services(void)
 
 void __init efi_unmap_memmap(void)
 {
-	clear_bit(EFI_MEMMAP, &x86_efi_facility);
+	clear_bit(EFI_MEMMAP, &efi.flags);
 	if (memmap.map) {
 		early_iounmap(memmap.map, memmap.nr_map * memmap.desc_size);
 		memmap.map = NULL;
@@ -584,11 +570,12 @@ static int __init efi_systab_init(void *phys)
 		return -EINVAL;
 	}
 	if ((efi.systab->hdr.revision >> 16) == 0)
-		pr_err("Warning: System table version "
-		       "%d.%02d, expected 1.00 or greater!\n",
+		pr_err("Warning: System table version %d.%02d, expected 1.00 or greater!\n",
 		       efi.systab->hdr.revision >> 16,
 		       efi.systab->hdr.revision & 0xffff);
 
+	set_bit(EFI_SYSTEM_TABLES, &efi.flags);
+
 	return 0;
 }
 
@@ -624,6 +611,8 @@ static int __init efi_runtime_init(void)
 	efi.get_time = phys_efi_get_time;
 	early_iounmap(runtime, sizeof(efi_runtime_services_t));
 
+	set_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+
 	return 0;
 }
 
@@ -641,6 +630,8 @@ static int __init efi_memmap_init(void)
 	if (add_efi_memmap)
 		do_add_efi_memmap();
 
+	set_bit(EFI_MEMMAP, &efi.flags);
+
 	return 0;
 }
 
@@ -723,7 +714,7 @@ void __init efi_init(void)
 	if (efi_systab_init(efi_phys.systab))
 		return;
 
-	set_bit(EFI_SYSTEM_TABLES, &x86_efi_facility);
+	set_bit(EFI_SYSTEM_TABLES, &efi.flags);
 
 	efi.config_table = (unsigned long)efi.systab->tables;
 	efi.fw_vendor	 = (unsigned long)efi.systab->fw_vendor;
@@ -751,8 +742,6 @@ void __init efi_init(void)
 	if (efi_config_init(arch_tables))
 		return;
 
-	set_bit(EFI_CONFIG_TABLES, &x86_efi_facility);
-
 	/*
 	 * Note: We currently don't support runtime services on an EFI
 	 * that doesn't match the kernel 32/64-bit mode.
@@ -763,12 +752,11 @@ void __init efi_init(void)
 	else {
 		if (disable_runtime || efi_runtime_init())
 			return;
-		set_bit(EFI_RUNTIME_SERVICES, &x86_efi_facility);
 	}
 	if (efi_memmap_init())
 		return;
 
-	set_bit(EFI_MEMMAP, &x86_efi_facility);
+	set_bit(EFI_MEMMAP, &efi.flags);
 
 	print_efi_memmap();
 }
@@ -892,8 +880,9 @@ static void __init get_systab_virt_addr(efi_memory_desc_t *md)
 	}
 }
 
-static int __init save_runtime_map(void)
+static void __init save_runtime_map(void)
 {
+#ifdef CONFIG_KEXEC
 	efi_memory_desc_t *md;
 	void *tmp, *p, *q = NULL;
 	int count = 0;
@@ -915,38 +904,44 @@ static int __init save_runtime_map(void)
 	}
 
 	efi_runtime_map_setup(q, count, memmap.desc_size);
+	return;
 
-	return 0;
 out:
 	kfree(q);
-	return -ENOMEM;
+	pr_err("Error saving runtime map, efi runtime on kexec non-functional!!\n");
+#endif
 }
 
-/*
- * Map efi regions which were passed via setup_data. The virt_addr is a fixed
- * addr which was used in first kernel of a kexec boot.
- */
-static void __init efi_map_regions_fixed(void)
+static void *realloc_pages(void *old_memmap, int old_shift)
 {
-	void *p;
-	efi_memory_desc_t *md;
+	void *ret;
 
-	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
-		md = p;
-		efi_map_region_fixed(md); /* FIXME: add error handling */
-		get_systab_virt_addr(md);
-	}
+	ret = (void *)__get_free_pages(GFP_KERNEL, old_shift + 1);
+	if (!ret)
+		goto out;
 
+	/*
+	 * A first-time allocation doesn't have anything to copy.
+	 */
+	if (!old_memmap)
+		return ret;
+
+	memcpy(ret, old_memmap, PAGE_SIZE << old_shift);
+
+out:
+	free_pages((unsigned long)old_memmap, old_shift);
+	return ret;
 }
 
 /*
- * Map efi memory ranges for runtime serivce and update new_memmap with virtual
- * addresses.
+ * Map the efi memory ranges of the runtime services and update new_mmap with
+ * virtual addresses.
  */
-static void * __init efi_map_regions(int *count)
+static void * __init efi_map_regions(int *count, int *pg_shift)
 {
+	void *p, *new_memmap = NULL;
+	unsigned long left = 0;
 	efi_memory_desc_t *md;
-	void *p, *tmp, *new_memmap = NULL;
 
 	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
 		md = p;
@@ -961,20 +956,89 @@ static void * __init efi_map_regions(int *count)
 		efi_map_region(md);
 		get_systab_virt_addr(md);
 
-		tmp = krealloc(new_memmap, (*count + 1) * memmap.desc_size,
-			       GFP_KERNEL);
-		if (!tmp)
-			goto out;
-		new_memmap = tmp;
+		if (left < memmap.desc_size) {
+			new_memmap = realloc_pages(new_memmap, *pg_shift);
+			if (!new_memmap)
+				return NULL;
+
+			left += PAGE_SIZE << *pg_shift;
+			(*pg_shift)++;
+		}
+
 		memcpy(new_memmap + (*count * memmap.desc_size), md,
 		       memmap.desc_size);
+
+		left -= memmap.desc_size;
 		(*count)++;
 	}
 
 	return new_memmap;
-out:
-	kfree(new_memmap);
-	return NULL;
+}
+
+static void __init kexec_enter_virtual_mode(void)
+{
+#ifdef CONFIG_KEXEC
+	efi_memory_desc_t *md;
+	void *p;
+
+	efi.systab = NULL;
+
+	/*
+	 * We don't do virtual mode, since we don't do runtime services, on
+	 * non-native EFI
+	 */
+	if (!efi_is_native()) {
+		efi_unmap_memmap();
+		return;
+	}
+
+	/*
+	* Map efi regions which were passed via setup_data. The virt_addr is a
+	* fixed addr which was used in first kernel of a kexec boot.
+	*/
+	for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
+		md = p;
+		efi_map_region_fixed(md); /* FIXME: add error handling */
+		get_systab_virt_addr(md);
+	}
+
+	save_runtime_map();
+
+	BUG_ON(!efi.systab);
+
+	efi_sync_low_kernel_mappings();
+
+	/*
+	 * Now that EFI is in virtual mode, update the function
+	 * pointers in the runtime service table to the new virtual addresses.
+	 *
+	 * Call EFI services through wrapper functions.
+	 */
+	efi.runtime_version = efi_systab.hdr.revision;
+	efi.get_time = virt_efi_get_time;
+	efi.set_time = virt_efi_set_time;
+	efi.get_wakeup_time = virt_efi_get_wakeup_time;
+	efi.set_wakeup_time = virt_efi_set_wakeup_time;
+	efi.get_variable = virt_efi_get_variable;
+	efi.get_next_variable = virt_efi_get_next_variable;
+	efi.set_variable = virt_efi_set_variable;
+	efi.get_next_high_mono_count = virt_efi_get_next_high_mono_count;
+	efi.reset_system = virt_efi_reset_system;
+	efi.set_virtual_address_map = NULL;
+	efi.query_variable_info = virt_efi_query_variable_info;
+	efi.update_capsule = virt_efi_update_capsule;
+	efi.query_capsule_caps = virt_efi_query_capsule_caps;
+
+	if (efi_enabled(EFI_OLD_MEMMAP) && (__supported_pte_mask & _PAGE_NX))
+		runtime_code_page_mkexec();
+
+	/* clean DUMMY object */
+	efi.set_variable(efi_dummy_name, &EFI_DUMMY_GUID,
+			 EFI_VARIABLE_NON_VOLATILE |
+			 EFI_VARIABLE_BOOTSERVICE_ACCESS |
+			 EFI_VARIABLE_RUNTIME_ACCESS,
+			 0, NULL);
+#endif
 }
 
 /*
@@ -996,13 +1060,14 @@ out:
  *
  * Specially for kexec boot, efi runtime maps in previous kernel should
  * be passed in via setup_data. In that case runtime ranges will be mapped
- * to the same virtual addresses as the first kernel.
+ * to the same virtual addresses as the first kernel, see
+ * kexec_enter_virtual_mode().
  */
-void __init efi_enter_virtual_mode(void)
+static void __init __efi_enter_virtual_mode(void)
 {
-	efi_status_t status;
+	int count = 0, pg_shift = 0;
 	void *new_memmap = NULL;
-	int err, count = 0;
+	efi_status_t status;
 
 	efi.systab = NULL;
 
@@ -1015,38 +1080,33 @@ void __init efi_enter_virtual_mode(void)
 		return;
 	}
 
-	if (efi_setup) {
-		efi_map_regions_fixed();
-	} else {
-		efi_merge_regions();
-		new_memmap = efi_map_regions(&count);
-		if (!new_memmap) {
-			pr_err("Error reallocating memory, EFI runtime non-functional!\n");
-			return;
-		}
+	efi_merge_regions();
+	new_memmap = efi_map_regions(&count, &pg_shift);
+	if (!new_memmap) {
+		pr_err("Error reallocating memory, EFI runtime non-functional!\n");
+		return;
 	}
 
-	err = save_runtime_map();
-	if (err)
-		pr_err("Error saving runtime map, efi runtime on kexec non-functional!!\n");
+	save_runtime_map();
 
 	BUG_ON(!efi.systab);
 
-	efi_setup_page_tables();
+	if (efi_setup_page_tables(__pa(new_memmap), 1 << pg_shift))
+		return;
+
 	efi_sync_low_kernel_mappings();
+	efi_dump_pagetable();
 
-	if (!efi_setup) {
-		status = phys_efi_set_virtual_address_map(
+	status = phys_efi_set_virtual_address_map(
 			memmap.desc_size * count,
 			memmap.desc_size,
 			memmap.desc_version,
 			(efi_memory_desc_t *)__pa(new_memmap));
 
-		if (status != EFI_SUCCESS) {
-			pr_alert("Unable to switch EFI into virtual mode (status=%lx)!\n",
-				 status);
-			panic("EFI call to SetVirtualAddressMap() failed!");
-		}
+	if (status != EFI_SUCCESS) {
+		pr_alert("Unable to switch EFI into virtual mode (status=%lx)!\n",
+			 status);
+		panic("EFI call to SetVirtualAddressMap() failed!");
 	}
 
 	/*
@@ -1072,7 +1132,33 @@ void __init efi_enter_virtual_mode(void)
 
 	efi_runtime_mkexec();
 
-	kfree(new_memmap);
+	/*
+	 * We mapped the descriptor array into the EFI pagetable above but we're
+	 * not unmapping it here. Here's why:
+	 *
+	 * We're copying select PGDs from the kernel page table to the EFI page
+	 * table and when we do so and make changes to those PGDs like unmapping
+	 * stuff from them, those changes appear in the kernel page table and we
+	 * go boom.
+	 *
+	 * From setup_real_mode():
+	 *
+	 * ...
+	 * trampoline_pgd[0] = init_level4_pgt[pgd_index(__PAGE_OFFSET)].pgd;
+	 *
+	 * In this particular case, our allocation is in PGD 0 of the EFI page
+	 * table but we've copied that PGD from PGD[272] of the EFI page table:
+	 *
+	 *	pgd_index(__PAGE_OFFSET = 0xffff880000000000) = 272
+	 *
+	 * where the direct memory mapping in kernel space is.
+	 *
+	 * new_memmap's VA comes from that direct mapping and thus clearing it,
+	 * it would get cleared in the kernel page table too.
+	 *
+	 * efi_cleanup_page_tables(__pa(new_memmap), 1 << pg_shift);
+	 */
+	free_pages((unsigned long)new_memmap, pg_shift);
 
 	/* clean DUMMY object */
 	efi.set_variable(efi_dummy_name, &EFI_DUMMY_GUID,
@@ -1082,6 +1168,14 @@ void __init efi_enter_virtual_mode(void)
 			 0, NULL);
 }
 
+void __init efi_enter_virtual_mode(void)
+{
+	if (efi_setup)
+		kexec_enter_virtual_mode();
+	else
+		__efi_enter_virtual_mode();
+}
+
 /*
  * Convenience functions to obtain memory types and attributes
  */
@@ -1119,9 +1213,8 @@ u64 efi_mem_attributes(unsigned long phys_addr)
 }
 
 /*
- * Some firmware has serious problems when using more than 50% of the EFI
- * variable store, i.e. it triggers bugs that can brick machines. Ensure that
- * we never use more than this safe limit.
+ * Some firmware implementations refuse to boot if there's insufficient space
+ * in the variable store. Ensure that we never use more than a safe limit.
  *
  * Return EFI_SUCCESS if it is safe to write 'size' bytes to the variable
  * store.
@@ -1140,10 +1233,9 @@ efi_status_t efi_query_variable_store(u32 attributes, unsigned long size)
 		return status;
 
 	/*
-	 * Some firmware implementations refuse to boot if there's insufficient
-	 * space in the variable store. We account for that by refusing the
-	 * write if permitting it would reduce the available space to under
-	 * 5KB. This figure was provided by Samsung, so should be safe.
+	 * We account for that by refusing the write if permitting it would
+	 * reduce the available space to under 5KB. This figure was provided by
+	 * Samsung, so should be safe.
 	 */
 	if ((remaining_size - size < EFI_MIN_RESERVE) &&
 		!efi_no_storage_paranoia) {
@@ -1206,7 +1298,7 @@ static int __init parse_efi_cmdline(char *str)
 		str++;
 
 	if (!strncmp(str, "old_map", 7))
-		set_bit(EFI_OLD_MEMMAP, &x86_efi_facility);
+		set_bit(EFI_OLD_MEMMAP, &efi.flags);
 
 	return 0;
 }
@@ -1228,5 +1320,5 @@ void __init efi_apply_memmap_quirks(void)
 	 * UV doesn't support the new EFI pagetable mapping yet.
 	 */
 	if (is_uv_system())
-		set_bit(EFI_OLD_MEMMAP, &x86_efi_facility);
+		set_bit(EFI_OLD_MEMMAP, &efi.flags);
 }
diff --git a/arch/x86/platform/efi/efi_32.c b/arch/x86/platform/efi/efi_32.c
index 0b74cdf7f816..9ee3491e31fb 100644
--- a/arch/x86/platform/efi/efi_32.c
+++ b/arch/x86/platform/efi/efi_32.c
@@ -40,7 +40,12 @@
 static unsigned long efi_rt_eflags;
 
 void efi_sync_low_kernel_mappings(void) {}
-void efi_setup_page_tables(void) {}
+void __init efi_dump_pagetable(void) {}
+int efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages)
+{
+	return 0;
+}
+void efi_cleanup_page_tables(unsigned long pa_memmap, unsigned num_pages) {}
 
 void __init efi_map_region(efi_memory_desc_t *md)
 {
diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c
index 0c2a234fef1e..19280900ec25 100644
--- a/arch/x86/platform/efi/efi_64.c
+++ b/arch/x86/platform/efi/efi_64.c
@@ -137,12 +137,38 @@ void efi_sync_low_kernel_mappings(void)
 		sizeof(pgd_t) * num_pgds);
 }
 
-void efi_setup_page_tables(void)
+int efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages)
 {
+	pgd_t *pgd;
+
+	if (efi_enabled(EFI_OLD_MEMMAP))
+		return 0;
+
 	efi_scratch.efi_pgt = (pgd_t *)(unsigned long)real_mode_header->trampoline_pgd;
+	pgd = __va(efi_scratch.efi_pgt);
 
-	if (!efi_enabled(EFI_OLD_MEMMAP))
-		efi_scratch.use_pgd = true;
+	/*
+	 * 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
+	 * and ident-map those pages containing the map before calling
+	 * phys_efi_set_virtual_address_map().
+	 */
+	if (kernel_map_pages_in_pgd(pgd, pa_memmap, pa_memmap, num_pages, _PAGE_NX)) {
+		pr_err("Error ident-mapping new memmap (0x%lx)!\n", pa_memmap);
+		return 1;
+	}
+
+	efi_scratch.use_pgd = true;
+
+
+	return 0;
+}
+
+void efi_cleanup_page_tables(unsigned long pa_memmap, unsigned num_pages)
+{
+	pgd_t *pgd = (pgd_t *)__va(real_mode_header->trampoline_pgd);
+
+	kernel_unmap_pages_in_pgd(pgd, pa_memmap, num_pages);
 }
 
 static void __init __map_region(efi_memory_desc_t *md, u64 va)
@@ -242,3 +268,12 @@ void __init efi_runtime_mkexec(void)
 	if (__supported_pte_mask & _PAGE_NX)
 		runtime_code_page_mkexec();
 }
+
+void __init efi_dump_pagetable(void)
+{
+#ifdef CONFIG_EFI_PGT_DUMP
+	pgd_t *pgd = (pgd_t *)__va(real_mode_header->trampoline_pgd);
+
+	ptdump_walk_pgd_level(NULL, pgd);
+#endif
+}