Merge tag 'efi-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mfleming/efi into efi/core

Pull EFI updates from Matt Fleming:

"* Refactor the EFI memory map code into architecture neutral files
   and allow drivers to permanently reserve EFI boot services regions
   on x86, as well as ARM/arm64 - Matt Fleming

 * Add ARM support for the EFI esrt driver - Ard Biesheuvel

 * Make the EFI runtime services and efivar API interruptible by
   swapping spinlocks for semaphores - Sylvain Chouleur

 * Provide the EFI identity mapping for kexec which allows kexec to
   work on SGI/UV platforms with requiring the "noefi" kernel command
   line parameter - Alex Thorlton

 * Add debugfs node to dump EFI page tables on arm64 - Ard Biesheuvel

 * Merge the EFI test driver being carried out of tree until now in
   the FWTS project - Ivan Hu

 * Expand the list of flags for classifying EFI regions as "RAM" on
   arm64 so we align with the UEFI spec - Ard Biesheuvel

 * Optimise out the EFI mixed mode if it's unsupported (CONFIG_X86_32)
   or disabled (CONFIG_EFI_MIXED=n) and switch the early EFI boot
   services function table for direct calls, alleviating us from
   having to maintain the custom function table - Lukas Wunner

 * Miscellaneous cleanups and fixes"

Signed-off-by: Ingo Molnar <mingo@kernel.org>

4 files changed, 204 insertions, 153 deletions

diff --git a/arch/x86/platform/efi/efi-bgrt.c b/arch/x86/platform/efi/efi-bgrt.c
index 6a2f5691b1ab..6aad870e8962 100644
--- a/arch/x86/platform/efi/efi-bgrt.c
+++ b/arch/x86/platform/efi/efi-bgrt.c
@@ -82,21 +82,12 @@ void __init efi_bgrt_init(void)
 	}
 	bgrt_image_size = bmp_header.size;
 
-	bgrt_image = kmalloc(bgrt_image_size, GFP_KERNEL | __GFP_NOWARN);
+	bgrt_image = memremap(bgrt_tab->image_address, bmp_header.size, MEMREMAP_WB);
 	if (!bgrt_image) {
-		pr_notice("Ignoring BGRT: failed to allocate memory for image (wanted %zu bytes)\n",
-		       bgrt_image_size);
-		return;
-	}
-
-	image = memremap(bgrt_tab->image_address, bmp_header.size, MEMREMAP_WB);
-	if (!image) {
 		pr_notice("Ignoring BGRT: failed to map image memory\n");
-		kfree(bgrt_image);
 		bgrt_image = NULL;
 		return;
 	}
 
-	memcpy(bgrt_image, image, bgrt_image_size);
-	memunmap(image);
+	efi_mem_reserve(bgrt_tab->image_address, bgrt_image_size);
 }
diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c
index 1fbb408e2e72..0955c70897ae 100644
--- a/arch/x86/platform/efi/efi.c
+++ b/arch/x86/platform/efi/efi.c
@@ -172,7 +172,9 @@ static void __init do_add_efi_memmap(void)
 int __init efi_memblock_x86_reserve_range(void)
 {
 	struct efi_info *e = &boot_params.efi_info;
+	struct efi_memory_map_data data;
 	phys_addr_t pmap;
+	int rv;
 
 	if (efi_enabled(EFI_PARAVIRT))
 		return 0;
@@ -187,11 +189,17 @@ int __init efi_memblock_x86_reserve_range(void)
 #else
 	pmap = (e->efi_memmap |	((__u64)e->efi_memmap_hi << 32));
 #endif
-	efi.memmap.phys_map	= pmap;
-	efi.memmap.nr_map	= e->efi_memmap_size /
-				  e->efi_memdesc_size;
-	efi.memmap.desc_size	= e->efi_memdesc_size;
-	efi.memmap.desc_version	= e->efi_memdesc_version;
+	data.phys_map		= pmap;
+	data.size 		= e->efi_memmap_size;
+	data.desc_size		= e->efi_memdesc_size;
+	data.desc_version	= e->efi_memdesc_version;
+
+	rv = efi_memmap_init_early(&data);
+	if (rv)
+		return rv;
+
+	if (add_efi_memmap)
+		do_add_efi_memmap();
 
 	WARN(efi.memmap.desc_version != 1,
 	     "Unexpected EFI_MEMORY_DESCRIPTOR version %ld",
@@ -218,19 +226,6 @@ void __init efi_print_memmap(void)
 	}
 }
 
-void __init efi_unmap_memmap(void)
-{
-	unsigned long size;
-
-	clear_bit(EFI_MEMMAP, &efi.flags);
-
-	size = efi.memmap.nr_map * efi.memmap.desc_size;
-	if (efi.memmap.map) {
-		early_memunmap(efi.memmap.map, size);
-		efi.memmap.map = NULL;
-	}
-}
-
 static int __init efi_systab_init(void *phys)
 {
 	if (efi_enabled(EFI_64BIT)) {
@@ -414,33 +409,6 @@ static int __init efi_runtime_init(void)
 	return 0;
 }
 
-static int __init efi_memmap_init(void)
-{
-	unsigned long addr, size;
-
-	if (efi_enabled(EFI_PARAVIRT))
-		return 0;
-
-	/* Map the EFI memory map */
-	size = efi.memmap.nr_map * efi.memmap.desc_size;
-	addr = (unsigned long)efi.memmap.phys_map;
-
-	efi.memmap.map = early_memremap(addr, size);
-	if (efi.memmap.map == NULL) {
-		pr_err("Could not map the memory map!\n");
-		return -ENOMEM;
-	}
-
-	efi.memmap.map_end = efi.memmap.map + size;
-
-	if (add_efi_memmap)
-		do_add_efi_memmap();
-
-	set_bit(EFI_MEMMAP, &efi.flags);
-
-	return 0;
-}
-
 void __init efi_init(void)
 {
 	efi_char16_t *c16;
@@ -498,16 +466,14 @@ void __init efi_init(void)
 	if (!efi_runtime_supported())
 		pr_info("No EFI runtime due to 32/64-bit mismatch with kernel\n");
 	else {
-		if (efi_runtime_disabled() || efi_runtime_init())
+		if (efi_runtime_disabled() || efi_runtime_init()) {
+			efi_memmap_unmap();
 			return;
+		}
 	}
-	if (efi_memmap_init())
-		return;
 
 	if (efi_enabled(EFI_DBG))
 		efi_print_memmap();
-
-	efi_esrt_init();
 }
 
 void __init efi_late_init(void)
@@ -624,42 +590,6 @@ static void __init get_systab_virt_addr(efi_memory_desc_t *md)
 	}
 }
 
-static void __init save_runtime_map(void)
-{
-#ifdef CONFIG_KEXEC_CORE
-	unsigned long desc_size;
-	efi_memory_desc_t *md;
-	void *tmp, *q = NULL;
-	int count = 0;
-
-	if (efi_enabled(EFI_OLD_MEMMAP))
-		return;
-
-	desc_size = efi.memmap.desc_size;
-
-	for_each_efi_memory_desc(md) {
-		if (!(md->attribute & EFI_MEMORY_RUNTIME) ||
-		    (md->type == EFI_BOOT_SERVICES_CODE) ||
-		    (md->type == EFI_BOOT_SERVICES_DATA))
-			continue;
-		tmp = krealloc(q, (count + 1) * desc_size, GFP_KERNEL);
-		if (!tmp)
-			goto out;
-		q = tmp;
-
-		memcpy(q + count * desc_size, md, desc_size);
-		count++;
-	}
-
-	efi_runtime_map_setup(q, count, desc_size);
-	return;
-
-out:
-	kfree(q);
-	pr_err("Error saving runtime map, efi runtime on kexec non-functional!!\n");
-#endif
-}
-
 static void *realloc_pages(void *old_memmap, int old_shift)
 {
 	void *ret;
@@ -745,6 +675,46 @@ static void *efi_map_next_entry(void *entry)
 	return entry;
 }
 
+static bool should_map_region(efi_memory_desc_t *md)
+{
+	/*
+	 * Runtime regions always require runtime mappings (obviously).
+	 */
+	if (md->attribute & EFI_MEMORY_RUNTIME)
+		return true;
+
+	/*
+	 * 32-bit EFI doesn't suffer from the bug that requires us to
+	 * reserve boot services regions, and mixed mode support
+	 * doesn't exist for 32-bit kernels.
+	 */
+	if (IS_ENABLED(CONFIG_X86_32))
+		return false;
+
+	/*
+	 * Map all of RAM so that we can access arguments in the 1:1
+	 * mapping when making EFI runtime calls.
+	 */
+	if (IS_ENABLED(CONFIG_EFI_MIXED) && !efi_is_native()) {
+		if (md->type == EFI_CONVENTIONAL_MEMORY ||
+		    md->type == EFI_LOADER_DATA ||
+		    md->type == EFI_LOADER_CODE)
+			return true;
+	}
+
+	/*
+	 * Map boot services regions as a workaround for buggy
+	 * firmware that accesses them even when they shouldn't.
+	 *
+	 * See efi_{reserve,free}_boot_services().
+	 */
+	if (md->type == EFI_BOOT_SERVICES_CODE ||
+	    md->type == EFI_BOOT_SERVICES_DATA)
+		return true;
+
+	return false;
+}
+
 /*
  * Map the efi memory ranges of the runtime services and update new_mmap with
  * virtual addresses.
@@ -761,13 +731,9 @@ static void * __init efi_map_regions(int *count, int *pg_shift)
 	p = NULL;
 	while ((p = efi_map_next_entry(p))) {
 		md = p;
-		if (!(md->attribute & EFI_MEMORY_RUNTIME)) {
-#ifdef CONFIG_X86_64
-			if (md->type != EFI_BOOT_SERVICES_CODE &&
-			    md->type != EFI_BOOT_SERVICES_DATA)
-#endif
-				continue;
-		}
+
+		if (!should_map_region(md))
+			continue;
 
 		efi_map_region(md);
 		get_systab_virt_addr(md);
@@ -803,7 +769,7 @@ static void __init kexec_enter_virtual_mode(void)
 	 * non-native EFI
 	 */
 	if (!efi_is_native()) {
-		efi_unmap_memmap();
+		efi_memmap_unmap();
 		clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
 		return;
 	}
@@ -823,7 +789,18 @@ static void __init kexec_enter_virtual_mode(void)
 		get_systab_virt_addr(md);
 	}
 
-	save_runtime_map();
+	/*
+	 * Unregister the early EFI memmap from efi_init() and install
+	 * the new EFI memory map.
+	 */
+	efi_memmap_unmap();
+
+	if (efi_memmap_init_late(efi.memmap.phys_map,
+				 efi.memmap.desc_size * efi.memmap.nr_map)) {
+		pr_err("Failed to remap late EFI memory map\n");
+		clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+		return;
+	}
 
 	BUG_ON(!efi.systab);
 
@@ -884,6 +861,7 @@ static void __init __efi_enter_virtual_mode(void)
 	int count = 0, pg_shift = 0;
 	void *new_memmap = NULL;
 	efi_status_t status;
+	phys_addr_t pa;
 
 	efi.systab = NULL;
 
@@ -901,11 +879,24 @@ static void __init __efi_enter_virtual_mode(void)
 		return;
 	}
 
-	save_runtime_map();
+	pa = __pa(new_memmap);
+
+	/*
+	 * Unregister the early EFI memmap from efi_init() and install
+	 * the new EFI memory map that we are about to pass to the
+	 * firmware via SetVirtualAddressMap().
+	 */
+	efi_memmap_unmap();
+
+	if (efi_memmap_init_late(pa, efi.memmap.desc_size * count)) {
+		pr_err("Failed to remap late EFI memory map\n");
+		clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+		return;
+	}
 
 	BUG_ON(!efi.systab);
 
-	if (efi_setup_page_tables(__pa(new_memmap), 1 << pg_shift)) {
+	if (efi_setup_page_tables(pa, 1 << pg_shift)) {
 		clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
 		return;
 	}
@@ -917,14 +908,14 @@ static void __init __efi_enter_virtual_mode(void)
 				efi.memmap.desc_size * count,
 				efi.memmap.desc_size,
 				efi.memmap.desc_version,
-				(efi_memory_desc_t *)__pa(new_memmap));
+				(efi_memory_desc_t *)pa);
 	} else {
 		status = efi_thunk_set_virtual_address_map(
 				efi_phys.set_virtual_address_map,
 				efi.memmap.desc_size * count,
 				efi.memmap.desc_size,
 				efi.memmap.desc_version,
-				(efi_memory_desc_t *)__pa(new_memmap));
+				(efi_memory_desc_t *)pa);
 	}
 
 	if (status != EFI_SUCCESS) {
@@ -956,15 +947,6 @@ static void __init __efi_enter_virtual_mode(void)
 	efi_runtime_update_mappings();
 	efi_dump_pagetable();
 
-	/*
-	 * We mapped the descriptor array into the EFI pagetable above
-	 * but we're not unmapping it here because if we're running in
-	 * EFI mixed mode we need all of memory to be accessible when
-	 * we pass parameters to the EFI runtime services in the
-	 * thunking code.
-	 */
-	free_pages((unsigned long)new_memmap, pg_shift);
-
 	/* clean DUMMY object */
 	efi_delete_dummy_variable();
 }
diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c
index 677e29e29473..d65cdadaa6b6 100644
--- a/arch/x86/platform/efi/efi_64.c
+++ b/arch/x86/platform/efi/efi_64.c
@@ -85,7 +85,7 @@ pgd_t * __init efi_call_phys_prolog(void)
 	early_code_mapping_set_exec(1);
 
 	n_pgds = DIV_ROUND_UP((max_pfn << PAGE_SHIFT), PGDIR_SIZE);
-	save_pgd = kmalloc(n_pgds * sizeof(pgd_t), GFP_KERNEL);
+	save_pgd = kmalloc_array(n_pgds, sizeof(*save_pgd), GFP_KERNEL);
 
 	for (pgd = 0; pgd < n_pgds; pgd++) {
 		save_pgd[pgd] = *pgd_offset_k(pgd * PGDIR_SIZE);
@@ -214,7 +214,6 @@ void efi_sync_low_kernel_mappings(void)
 int __init efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages)
 {
 	unsigned long pfn, text;
-	efi_memory_desc_t *md;
 	struct page *page;
 	unsigned npages;
 	pgd_t *pgd;
@@ -248,25 +247,6 @@ int __init efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages)
 	if (!IS_ENABLED(CONFIG_EFI_MIXED))
 		return 0;
 
-	/*
-	 * Map all of RAM so that we can access arguments in the 1:1
-	 * mapping when making EFI runtime calls.
-	 */
-	for_each_efi_memory_desc(md) {
-		if (md->type != EFI_CONVENTIONAL_MEMORY &&
-		    md->type != EFI_LOADER_DATA &&
-		    md->type != EFI_LOADER_CODE)
-			continue;
-
-		pfn = md->phys_addr >> PAGE_SHIFT;
-		npages = md->num_pages;
-
-		if (kernel_map_pages_in_pgd(pgd, pfn, md->phys_addr, npages, _PAGE_RW)) {
-			pr_err("Failed to map 1:1 memory\n");
-			return 1;
-		}
-	}
-
 	page = alloc_page(GFP_KERNEL|__GFP_DMA32);
 	if (!page)
 		panic("Unable to allocate EFI runtime stack < 4GB\n");
@@ -359,6 +339,7 @@ void __init efi_map_region(efi_memory_desc_t *md)
  */
 void __init efi_map_region_fixed(efi_memory_desc_t *md)
 {
+	__map_region(md, md->phys_addr);
 	__map_region(md, md->virt_addr);
 }
 
diff --git a/arch/x86/platform/efi/quirks.c b/arch/x86/platform/efi/quirks.c
index 89d1146f5a6f..f14b7a9da24b 100644
--- a/arch/x86/platform/efi/quirks.c
+++ b/arch/x86/platform/efi/quirks.c
@@ -164,6 +164,71 @@ efi_status_t efi_query_variable_store(u32 attributes, unsigned long size,
 EXPORT_SYMBOL_GPL(efi_query_variable_store);
 
 /*
+ * The UEFI specification makes it clear that the operating system is
+ * free to do whatever it wants with boot services code after
+ * ExitBootServices() has been called. Ignoring this recommendation a
+ * significant bunch of EFI implementations continue calling into boot
+ * services code (SetVirtualAddressMap). In order to work around such
+ * buggy implementations we reserve boot services region during EFI
+ * init and make sure it stays executable. Then, after
+ * SetVirtualAddressMap(), it is discarded.
+ *
+ * However, some boot services regions contain data that is required
+ * by drivers, so we need to track which memory ranges can never be
+ * freed. This is done by tagging those regions with the
+ * EFI_MEMORY_RUNTIME attribute.
+ *
+ * Any driver that wants to mark a region as reserved must use
+ * efi_mem_reserve() which will insert a new EFI memory descriptor
+ * into efi.memmap (splitting existing regions if necessary) and tag
+ * it with EFI_MEMORY_RUNTIME.
+ */
+void __init efi_arch_mem_reserve(phys_addr_t addr, u64 size)
+{
+	phys_addr_t new_phys, new_size;
+	struct efi_mem_range mr;
+	efi_memory_desc_t md;
+	int num_entries;
+	void *new;
+
+	if (efi_mem_desc_lookup(addr, &md)) {
+		pr_err("Failed to lookup EFI memory descriptor for %pa\n", &addr);
+		return;
+	}
+
+	if (addr + size > md.phys_addr + (md.num_pages << EFI_PAGE_SHIFT)) {
+		pr_err("Region spans EFI memory descriptors, %pa\n", &addr);
+		return;
+	}
+
+	mr.range.start = addr;
+	mr.range.end = addr + size;
+	mr.attribute = md.attribute | EFI_MEMORY_RUNTIME;
+
+	num_entries = efi_memmap_split_count(&md, &mr.range);
+	num_entries += efi.memmap.nr_map;
+
+	new_size = efi.memmap.desc_size * num_entries;
+
+	new_phys = memblock_alloc(new_size, 0);
+	if (!new_phys) {
+		pr_err("Could not allocate boot services memmap\n");
+		return;
+	}
+
+	new = early_memremap(new_phys, new_size);
+	if (!new) {
+		pr_err("Failed to map new boot services memmap\n");
+		return;
+	}
+
+	efi_memmap_insert(&efi.memmap, new, &mr);
+	early_memunmap(new, new_size);
+
+	efi_memmap_install(new_phys, num_entries);
+}
+
+/*
  * Helper function for efi_reserve_boot_services() to figure out if we
  * can free regions in efi_free_boot_services().
  *
@@ -184,15 +249,6 @@ static bool can_free_region(u64 start, u64 size)
 	return true;
 }
 
-/*
- * The UEFI specification makes it clear that the operating system is free to do
- * whatever it wants with boot services code after ExitBootServices() has been
- * called. Ignoring this recommendation a significant bunch of EFI implementations 
- * continue calling into boot services code (SetVirtualAddressMap). In order to 
- * work around such buggy implementations we reserve boot services region during 
- * EFI init and make sure it stays executable. Then, after SetVirtualAddressMap(), it
-* is discarded.
-*/
 void __init efi_reserve_boot_services(void)
 {
 	efi_memory_desc_t *md;
@@ -249,7 +305,10 @@ void __init efi_reserve_boot_services(void)
 
 void __init efi_free_boot_services(void)
 {
+	phys_addr_t new_phys, new_size;
 	efi_memory_desc_t *md;
+	int num_entries = 0;
+	void *new, *new_md;
 
 	for_each_efi_memory_desc(md) {
 		unsigned long long start = md->phys_addr;
@@ -257,12 +316,16 @@ void __init efi_free_boot_services(void)
 		size_t rm_size;
 
 		if (md->type != EFI_BOOT_SERVICES_CODE &&
-		    md->type != EFI_BOOT_SERVICES_DATA)
+		    md->type != EFI_BOOT_SERVICES_DATA) {
+			num_entries++;
 			continue;
+		}
 
 		/* Do not free, someone else owns it: */
-		if (md->attribute & EFI_MEMORY_RUNTIME)
+		if (md->attribute & EFI_MEMORY_RUNTIME) {
+			num_entries++;
 			continue;
+		}
 
 		/*
 		 * Nasty quirk: if all sub-1MB memory is used for boot
@@ -287,7 +350,41 @@ void __init efi_free_boot_services(void)
 		free_bootmem_late(start, size);
 	}
 
-	efi_unmap_memmap();
+	new_size = efi.memmap.desc_size * num_entries;
+	new_phys = memblock_alloc(new_size, 0);
+	if (!new_phys) {
+		pr_err("Failed to allocate new EFI memmap\n");
+		return;
+	}
+
+	new = memremap(new_phys, new_size, MEMREMAP_WB);
+	if (!new) {
+		pr_err("Failed to map new EFI memmap\n");
+		return;
+	}
+
+	/*
+	 * Build a new EFI memmap that excludes any boot services
+	 * regions that are not tagged EFI_MEMORY_RUNTIME, since those
+	 * regions have now been freed.
+	 */
+	new_md = new;
+	for_each_efi_memory_desc(md) {
+		if (!(md->attribute & EFI_MEMORY_RUNTIME) &&
+		    (md->type == EFI_BOOT_SERVICES_CODE ||
+		     md->type == EFI_BOOT_SERVICES_DATA))
+			continue;
+
+		memcpy(new_md, md, efi.memmap.desc_size);
+		new_md += efi.memmap.desc_size;
+	}
+
+	memunmap(new);
+
+	if (efi_memmap_install(new_phys, num_entries)) {
+		pr_err("Could not install new EFI memmap\n");
+		return;
+	}
 }
 
 /*
@@ -365,7 +462,7 @@ void __init efi_apply_memmap_quirks(void)
 	 */
 	if (!efi_runtime_supported()) {
 		pr_info("Setup done, disabling due to 32/64-bit mismatch\n");
-		efi_unmap_memmap();
+		efi_memmap_unmap();
 	}
 
 	/* UV2+ BIOS has a fix for this issue.  UV1 still needs the quirk. */