[PATCH v2 00/14] UEFI support for arm(64)

[PATCH v2 00/14] UEFI support for arm(64)

[Leif Lindholm]

Since all prerequisites are not ready for the (32-bit) arm patches to
go in, I have broken these out of this set and will be resubmitting
them separately.

This set now provides arm64 support only - but includes the code that
is common between the ARM ports.

Prerequisites for this set are:
- Mark Salter's generic early_ioremap()
- Matt Fleming's efi_enabled() rework

Changes from the last upstream posting:
- Some cleanup based on feedback
- Fixed range test in create_id_mapping()
- Fixed name of guard ifdef in arch/arm64/include/asm/efi.h
- Fixed memory corruption bug caused by late use of
  create_id_mapping(). That function uses memblock_alloc()
  to allocate new pmd/pte pages and it was being called
  after slab allocator was initialized. This led to
  possibility that the allocated memory for pmd/pte was
  already in use by something else. Now, create_id_mapping()
  gets called just after paging_init() when it is still okay
  to use memblock_alloc().
- Fixed whitespace error in efi-entry.S
- Now support 64K pagesize in EFI runtime
- Undid the breakage I caused to Ard's dtb= SecureBoot patch

---

Ard Biesheuvel (1):
  efi/arm64: ignore dtb= when UEFI SecureBoot is enabled

H. Peter Anvin (1):
  efi: x86: Improve cmdline conversion

Leif Lindholm (2):
  efi: delete stray ARM ifdef
  doc: arm: add UEFI support documentation

Mark Salter (7):
  efi: create memory map iteration helper
  lib: add fdt_empty_tree.c
  efi: add helper function to get UEFI params from FDT
  arm64: Add function to create identity mappings
  arm64: add EFI stub
  doc: arm64: add description of EFI stub support
  arm64: add EFI runtime services

Roy Franz (4):
  doc: efi-stub.txt updates for ARM
  efi: Add shared printk wrapper for consistent prefixing
  efi: Add get_dram_base() helper function
  efi: Add shared FDT related functions for ARM/ARM64

 Documentation/arm/00-INDEX             |    2 +
 Documentation/arm/uefi.txt             |   64 +++++
 Documentation/arm64/booting.txt        |    4 +
 Documentation/efi-stub.txt             |   33 ++-
 arch/arm64/Kconfig                     |   26 ++
 arch/arm64/include/asm/efi.h           |   14 +
 arch/arm64/include/asm/mmu.h           |    2 +
 arch/arm64/kernel/Makefile             |    4 +
 arch/arm64/kernel/efi-entry.S          |   93 +++++++
 arch/arm64/kernel/efi-stub.c           |   83 ++++++
 arch/arm64/kernel/efi.c                |  465 ++++++++++++++++++++++++++++++++
 arch/arm64/kernel/head.S               |  112 ++++++++
 arch/arm64/kernel/setup.c              |    5 +
 arch/arm64/mm/mmu.c                    |   65 +++--
 arch/x86/boot/compressed/eboot.c       |    3 +-
 drivers/firmware/efi/Kconfig           |    7 +
 drivers/firmware/efi/arm-stub.c        |  149 ++++++++++
 drivers/firmware/efi/efi-stub-helper.c |  182 ++++++++++---
 drivers/firmware/efi/efi.c             |   79 ++++++
 drivers/firmware/efi/fdt.c             |  268 ++++++++++++++++++
 include/linux/efi.h                    |   18 ++
 init/main.c                            |    4 +
 lib/Makefile                           |    3 +-
 lib/fdt_empty_tree.c                   |    2 +
 24 files changed, 1615 insertions(+), 72 deletions(-)
 create mode 100644 Documentation/arm/uefi.txt
 create mode 100644 arch/arm64/include/asm/efi.h
 create mode 100644 arch/arm64/kernel/efi-entry.S
 create mode 100644 arch/arm64/kernel/efi-stub.c
 create mode 100644 arch/arm64/kernel/efi.c
 create mode 100644 drivers/firmware/efi/arm-stub.c
 create mode 100644 drivers/firmware/efi/fdt.c
 create mode 100644 lib/fdt_empty_tree.c

-- 
1.7.10.4

[PATCH v2 01/15] efi: delete stray ARM ifdef

[Leif Lindholm]

An #ifdef CONFIG_ARM clause in efi-stub-helper.c got included with some
of the generic stub rework by Roy Franz. Drop it here to make subsequent
patches less confusing.

Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
---
 drivers/firmware/efi/efi-stub-helper.c |   14 ++------------
 1 file changed, 2 insertions(+), 12 deletions(-)

diff --git a/drivers/firmware/efi/efi-stub-helper.c b/drivers/firmware/efi/efi-stub-helper.c
index b6bffbf..6811af7 100644
--- a/drivers/firmware/efi/efi-stub-helper.c
+++ b/drivers/firmware/efi/efi-stub-helper.c
@@ -608,18 +608,8 @@ static char *efi_convert_cmdline_to_ascii(efi_system_table_t *sys_table_arg,
 	}
 
 	options_size++;  /* NUL termination */
-#ifdef CONFIG_ARM
-	/*
-	 * For ARM, allocate at a high address to avoid reserved
-	 * regions at low addresses that we don't know the specfics of
-	 * at the time we are processing the command line.
-	 */
-	status = efi_high_alloc(sys_table_arg, options_size, 0,
-			    &cmdline_addr, 0xfffff000);
-#else
-	status = efi_low_alloc(sys_table_arg, options_size, 0,
-			    &cmdline_addr);
-#endif
+
+	status = efi_low_alloc(sys_table_arg, options_size, 0, &cmdline_addr);
 	if (status != EFI_SUCCESS)
 		return NULL;
 
-- 
1.7.10.4

[PATCH v2 02/15] efi: x86: Improve cmdline conversion

[Leif Lindholm]

From: "H. Peter Anvin" <hpa@zytor.com>

Improve the conversion of the UTF-16 EFI command line
to UTF-8 for passing to the kernel.

Signed-off-by: Roy Franz <roy.franz@linaro.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
---
 arch/x86/boot/compressed/eboot.c       |    3 +-
 drivers/firmware/efi/efi-stub-helper.c |   91 ++++++++++++++++++++++++--------
 2 files changed, 70 insertions(+), 24 deletions(-)

diff --git a/arch/x86/boot/compressed/eboot.c b/arch/x86/boot/compressed/eboot.c
index a7677ba..feca05f 100644
--- a/arch/x86/boot/compressed/eboot.c
+++ b/arch/x86/boot/compressed/eboot.c
@@ -488,8 +488,7 @@ struct boot_params *make_boot_params(void *handle, efi_system_table_t *_table)
 	hdr->type_of_loader = 0x21;
 
 	/* Convert unicode cmdline to ascii */
-	cmdline_ptr = efi_convert_cmdline_to_ascii(sys_table, image,
-						   &options_size);
+	cmdline_ptr = efi_convert_cmdline(sys_table, image, &options_size);
 	if (!cmdline_ptr)
 		goto fail;
 	hdr->cmd_line_ptr = (unsigned long)cmdline_ptr;
diff --git a/drivers/firmware/efi/efi-stub-helper.c b/drivers/firmware/efi/efi-stub-helper.c
index 6811af7..661f425 100644
--- a/drivers/firmware/efi/efi-stub-helper.c
+++ b/drivers/firmware/efi/efi-stub-helper.c
@@ -575,52 +575,99 @@ static efi_status_t efi_relocate_kernel(efi_system_table_t *sys_table_arg,
 }
 
 /*
- * Convert the unicode UEFI command line to ASCII to pass to kernel.
+ * Get the number of UTF-8 bytes corresponding to an UTF-16 character.
+ * This overestimates for surrogates, but that is okay.
+ */
+static int efi_utf8_bytes(u16 c)
+{
+	return 1 + (c >= 0x80) + (c >= 0x800);
+}
+
+/*
+ * Convert an UTF-16 string, not necessarily null terminated, to UTF-8.
+ */
+static u8 *efi_utf16_to_utf8(u8 *dst, const u16 *src, int n)
+{
+	unsigned int c;
+
+	while (n--) {
+		c = *src++;
+		if (n && c >= 0xd800 && c <= 0xdbff &&
+		    *src >= 0xdc00 && *src <= 0xdfff) {
+			c = 0x10000 + ((c & 0x3ff) << 10) + (*src & 0x3ff);
+			src++;
+			n--;
+		}
+		if (c >= 0xd800 && c <= 0xdfff)
+			c = 0xfffd; /* Unmatched surrogate */
+		if (c < 0x80) {
+			*dst++ = c;
+			continue;
+		}
+		if (c < 0x800) {
+			*dst++ = 0xc0 + (c >> 6);
+			goto t1;
+		}
+		if (c < 0x10000) {
+			*dst++ = 0xe0 + (c >> 12);
+			goto t2;
+		}
+		*dst++ = 0xf0 + (c >> 18);
+		*dst++ = 0x80 + ((c >> 12) & 0x3f);
+t2:
+		*dst++ = 0x80 + ((c >> 6) & 0x3f);
+t1:
+		*dst++ = 0x80 + (c & 0x3f);
+	}
+
+	return dst;
+}
+
+/*
+ * Do proper conversion from UTF-16 to UTF-8
  * Size of memory allocated return in *cmd_line_len.
  * Returns NULL on error.
  */
-static char *efi_convert_cmdline_to_ascii(efi_system_table_t *sys_table_arg,
-				      efi_loaded_image_t *image,
-				      int *cmd_line_len)
+static char *efi_convert_cmdline(efi_system_table_t *sys_table_arg,
+				 efi_loaded_image_t *image,
+				 int *cmd_line_len)
 {
-	u16 *s2;
+	const u16 *s2;
 	u8 *s1 = NULL;
 	unsigned long cmdline_addr = 0;
-	int load_options_size = image->load_options_size / 2; /* ASCII */
-	void *options = image->load_options;
-	int options_size = 0;
+	int load_options_chars = image->load_options_size / 2; /* UTF-16 */
+	const u16 *options = image->load_options;
+	int options_bytes = 0;	/* UTF-8 bytes */
+	int options_chars = 0;	/* UTF-16 chars */
 	efi_status_t status;
-	int i;
 	u16 zero = 0;
 
 	if (options) {
 		s2 = options;
-		while (*s2 && *s2 != '\n' && options_size < load_options_size) {
-			s2++;
-			options_size++;
+		while (options_chars < load_options_chars
+		       && *s2 && *s2 != '\n') {
+			options_bytes += efi_utf8_bytes(*s2++);
+			options_chars++;
 		}
 	}
 
-	if (options_size == 0) {
-		/* No command line options, so return empty string*/
-		options_size = 1;
+	if (!options_chars) {
+		/* No command line options, so return empty string */
 		options = &zero;
 	}
 
-	options_size++;  /* NUL termination */
+	options_bytes++;  /* NUL termination */
 
-	status = efi_low_alloc(sys_table_arg, options_size, 0, &cmdline_addr);
+	status = efi_low_alloc(sys_table_arg, options_bytes, 0, &cmdline_addr);
 	if (status != EFI_SUCCESS)
 		return NULL;
 
 	s1 = (u8 *)cmdline_addr;
-	s2 = (u16 *)options;
-
-	for (i = 0; i < options_size - 1; i++)
-		*s1++ = *s2++;
+	s2 = (const u16 *)options;
 
+	s1 = efi_utf16_to_utf8(s1, s2, options_chars);
 	*s1 = '\0';
 
-	*cmd_line_len = options_size;
+	*cmd_line_len = options_bytes;
 	return (char *)cmdline_addr;
 }
-- 
1.7.10.4

[PATCH v2 03/15] efi: create memory map iteration helper

[Leif Lindholm]

From: Mark Salter <msalter@redhat.com>

There are a lot of places in the kernel which iterate through an
EFI memory map. Most of these places use essentially the same
for-loop code. This patch adds a for_each_efi_memory_desc()
helper to clean up all of the existing duplicate code and avoid
more in the future.

Signed-off-by: Mark Salter <msalter@redhat.com>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
---
 include/linux/efi.h |    6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/include/linux/efi.h b/include/linux/efi.h
index 64d532c..a3276da 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -622,6 +622,12 @@ extern int efi_set_rtc_mmss(const struct timespec *now);
 extern void efi_reserve_boot_services(void);
 extern struct efi_memory_map memmap;
 
+/* Iterate through an efi_memory_map */
+#define for_each_efi_memory_desc(m, md)					   \
+	for ((md) = (m)->map;						   \
+	     (md) <= (efi_memory_desc_t *)((m)->map_end - (m)->desc_size); \
+	     (md) = (void *)(md) + (m)->desc_size)
+
 /**
  * efi_range_is_wc - check the WC bit on an address range
  * @start: starting kvirt address
-- 
1.7.10.4

[PATCH v2 04/15] lib: add fdt_empty_tree.c

[Leif Lindholm]

From: Mark Salter <msalter@redhat.com>

CONFIG_LIBFDT support does not include fdt_empty_tree.c which is
needed by arm64 EFI stub. Add it to libfdt_files.

Signed-off-by: Mark Salter <msalter@redhat.com>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
---
 lib/Makefile         |    3 ++-
 lib/fdt_empty_tree.c |    2 ++
 2 files changed, 4 insertions(+), 1 deletion(-)
 create mode 100644 lib/fdt_empty_tree.c

diff --git a/lib/Makefile b/lib/Makefile
index 48140e3..befe555 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -147,7 +147,8 @@ obj-$(CONFIG_GENERIC_NET_UTILS) += net_utils.o
 
 obj-$(CONFIG_STMP_DEVICE) += stmp_device.o
 
-libfdt_files = fdt.o fdt_ro.o fdt_wip.o fdt_rw.o fdt_sw.o fdt_strerror.o
+libfdt_files = fdt.o fdt_ro.o fdt_wip.o fdt_rw.o fdt_sw.o fdt_strerror.o \
+	       fdt_empty_tree.o
 $(foreach file, $(libfdt_files), \
 	$(eval CFLAGS_$(file) = -I$(src)/../scripts/dtc/libfdt))
 lib-$(CONFIG_LIBFDT) += $(libfdt_files)
diff --git a/lib/fdt_empty_tree.c b/lib/fdt_empty_tree.c
new file mode 100644
index 0000000..5d30c58
--- /dev/null
+++ b/lib/fdt_empty_tree.c
@@ -0,0 +1,2 @@
+#include <linux/libfdt_env.h>
+#include "../scripts/dtc/libfdt/fdt_empty_tree.c"
-- 
1.7.10.4

[PATCH v2 05/15] efi: add helper function to get UEFI params from FDT

[Leif Lindholm]

From: Mark Salter <msalter@redhat.com>

ARM and ARM64 architectures use the device tree to pass UEFI parameters
from stub to kernel. These parameters are things known to the stub but
not discoverable by the kernel after the stub calls ExitBootSerives().
There is a helper function in:

   drivers/firmware/efi/fdt.c

which the stub uses to add the UEFI parameters to the device tree.
This patch adds a complimentary helper function which UEFI runtime
support may use to retrieve the parameters from the device tree.
If an architecture wants to use this helper, it should select
CONFIG_UEFI_PARAMS_FROM_FDT.

Signed-off-by: Mark Salter <msalter@redhat.com>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
---
 drivers/firmware/efi/Kconfig |    7 ++++
 drivers/firmware/efi/efi.c   |   79 ++++++++++++++++++++++++++++++++++++++++++
 include/linux/efi.h          |    9 +++++
 3 files changed, 95 insertions(+)

diff --git a/drivers/firmware/efi/Kconfig b/drivers/firmware/efi/Kconfig
index 1e75f48..d3fe28d 100644
--- a/drivers/firmware/efi/Kconfig
+++ b/drivers/firmware/efi/Kconfig
@@ -47,6 +47,13 @@ config EFI_RUNTIME_MAP
 
 	  See also Documentation/ABI/testing/sysfs-firmware-efi-runtime-map.
 
+config UEFI_PARAMS_FROM_FDT
+	bool
+	help
+	  Select this config option from the architecture Kconfig if
+	  the EFI runtime support gets system table address, memory
+          map address, and other parameters from the device tree.
+
 endmenu
 
 config UEFI_CPER
diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c
index b25b36b..7f6e977 100644
--- a/drivers/firmware/efi/efi.c
+++ b/drivers/firmware/efi/efi.c
@@ -20,6 +20,8 @@
 #include <linux/init.h>
 #include <linux/device.h>
 #include <linux/efi.h>
+#include <linux/of.h>
+#include <linux/of_fdt.h>
 #include <linux/io.h>
 
 struct efi __read_mostly efi = {
@@ -318,3 +320,80 @@ int __init efi_config_init(efi_config_table_type_t *arch_tables)
 
 	return 0;
 }
+
+#ifdef CONFIG_UEFI_PARAMS_FROM_FDT
+
+#define UEFI_PARAM(name, prop, field)			   \
+	{						   \
+		{ name },				   \
+		{ prop },				   \
+		offsetof(struct efi_fdt_params, field),    \
+		FIELD_SIZEOF(struct efi_fdt_params, field) \
+	}
+
+static __initdata struct {
+	const char name[32];
+	const char propname[32];
+	int offset;
+	int size;
+} dt_params[] = {
+	UEFI_PARAM("System Table", "linux,uefi-system-table", system_table),
+	UEFI_PARAM("MemMap Address", "linux,uefi-mmap-start", mmap),
+	UEFI_PARAM("MemMap Size", "linux,uefi-mmap-size", mmap_size),
+	UEFI_PARAM("MemMap Desc. Size", "linux,uefi-mmap-desc-size", desc_size),
+	UEFI_PARAM("MemMap Desc. Version", "linux,uefi-mmap-desc-ver", desc_ver)
+};
+
+struct param_info {
+	int verbose;
+	void *params;
+};
+
+static int __init fdt_find_uefi_params(unsigned long node, const char *uname,
+				       int depth, void *data)
+{
+	struct param_info *info = data;
+	void *prop, *dest;
+	unsigned long len;
+	u64 val;
+	int i;
+
+	if (depth != 1 ||
+	    (strcmp(uname, "chosen") != 0 && strcmp(uname, "chosen at 0") != 0))
+		return 0;
+
+	pr_info("Getting parameters from FDT:\n");
+
+	for (i = 0; i < ARRAY_SIZE(dt_params); i++) {
+		prop = of_get_flat_dt_prop(node, dt_params[i].propname, &len);
+		if (!prop) {
+			pr_err("Can't find %s in device tree!\n",
+			       dt_params[i].name);
+			return 0;
+		}
+		dest = info->params + dt_params[i].offset;
+
+		val = of_read_number(prop, len / sizeof(u32));
+
+		if (dt_params[i].size == sizeof(u32))
+			*(u32 *)dest = val;
+		else
+			*(u64 *)dest = val;
+
+		if (info->verbose)
+			pr_info("  %s: 0x%0*llx\n", dt_params[i].name,
+				dt_params[i].size * 2, val);
+	}
+	return 1;
+}
+
+int __init efi_get_fdt_params(struct efi_fdt_params *params, int verbose)
+{
+	struct param_info info;
+
+	info.verbose = verbose;
+	info.params = params;
+
+	return of_scan_flat_dt(fdt_find_uefi_params, &info);
+}
+#endif /* CONFIG_UEFI_PARAMS_FROM_FDT */
diff --git a/include/linux/efi.h b/include/linux/efi.h
index a3276da..d450673 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -483,6 +483,14 @@ struct efi_memory_map {
 	unsigned long desc_size;
 };
 
+struct efi_fdt_params {
+	u64 system_table;
+	u64 mmap;
+	u32 mmap_size;
+	u32 desc_size;
+	u32 desc_ver;
+};
+
 typedef struct {
 	u32 revision;
 	void *parent_handle;
@@ -620,6 +628,7 @@ extern void efi_initialize_iomem_resources(struct resource *code_resource,
 extern void efi_get_time(struct timespec *now);
 extern int efi_set_rtc_mmss(const struct timespec *now);
 extern void efi_reserve_boot_services(void);
+extern int efi_get_fdt_params(struct efi_fdt_params *params, int verbose);
 extern struct efi_memory_map memmap;
 
 /* Iterate through an efi_memory_map */
-- 
1.7.10.4

[PATCH v2 06/15] doc: efi-stub.txt updates for ARM

[Leif Lindholm]

From: Roy Franz <roy.franz@linaro.org>

Update efi-stub.txt documentation to be more general
and not x86 specific.  Add ARM only "dtb=" command
line option description.

Signed-off-by: Roy Franz <roy.franz@linaro.org>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
Acked-by: Grant Likely <grant.likely@linaro.org>
---
 Documentation/efi-stub.txt |   27 ++++++++++++++++++++-------
 1 file changed, 20 insertions(+), 7 deletions(-)

diff --git a/Documentation/efi-stub.txt b/Documentation/efi-stub.txt
index c628788..26be7b0 100644
--- a/Documentation/efi-stub.txt
+++ b/Documentation/efi-stub.txt
@@ -1,13 +1,16 @@
 			  The EFI Boot Stub
 		     ---------------------------
 
-On the x86 platform, a bzImage can masquerade as a PE/COFF image,
-thereby convincing EFI firmware loaders to load it as an EFI
-executable. The code that modifies the bzImage header, along with the
-EFI-specific entry point that the firmware loader jumps to are
-collectively known as the "EFI boot stub", and live in
+On the x86 and ARM platforms, a kernel zImage/bzImage can masquerade
+as a PE/COFF image, thereby convincing EFI firmware loaders to load
+it as an EFI executable. The code that modifies the bzImage header,
+along with the EFI-specific entry point that the firmware loader
+jumps to are collectively known as the "EFI boot stub", and live in
 arch/x86/boot/header.S and arch/x86/boot/compressed/eboot.c,
-respectively.
+respectively. For ARM the EFI stub is implemented in
+arch/arm/boot/compressed/efi-header.S and
+arch/arm/boot/compressed/efi-stub.c. EFI stub code that is shared
+between architectures is in drivers/firmware/efi/efi-stub-helper.c.
 
 By using the EFI boot stub it's possible to boot a Linux kernel
 without the use of a conventional EFI boot loader, such as grub or
@@ -23,7 +26,9 @@ The bzImage located in arch/x86/boot/bzImage must be copied to the EFI
 System Partition (ESP) and renamed with the extension ".efi". Without
 the extension the EFI firmware loader will refuse to execute it. It's
 not possible to execute bzImage.efi from the usual Linux file systems
-because EFI firmware doesn't have support for them.
+because EFI firmware doesn't have support for them. For ARM the
+arch/arm/boot/zImage should be copied to the system partition, and it
+may not need to be renamed.
 
 
 **** Passing kernel parameters from the EFI shell
@@ -63,3 +68,11 @@ Notice how bzImage.efi can be specified with a relative path. That's
 because the image we're executing is interpreted by the EFI shell,
 which understands relative paths, whereas the rest of the command line
 is passed to bzImage.efi.
+
+
+**** The "dtb=" option
+
+For the ARM architecture, we also need to be able to provide a device
+tree to the kernel. This is done with the "dtb=" command line option,
+and is processed in the same manner as the "initrd=" option that is
+described above.
-- 
1.7.10.4

[PATCH v2 07/15] efi: Add shared printk wrapper for consistent prefixing

[Leif Lindholm]

From: Roy Franz <roy.franz@linaro.org>

Add a wrapper for printk to standardize the prefix for informational and
error messages from the EFI stub.

Signed-off-by: Roy Franz <roy.franz@linaro.org>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
---
 drivers/firmware/efi/efi-stub-helper.c |   25 ++++++++++++++-----------
 1 file changed, 14 insertions(+), 11 deletions(-)

diff --git a/drivers/firmware/efi/efi-stub-helper.c b/drivers/firmware/efi/efi-stub-helper.c
index 661f425..4c8ab86 100644
--- a/drivers/firmware/efi/efi-stub-helper.c
+++ b/drivers/firmware/efi/efi-stub-helper.c
@@ -45,6 +45,9 @@ static void efi_printk(efi_system_table_t *sys_table_arg, char *str)
 	}
 }
 
+#define pr_efi(sys_table, msg)     efi_printk(sys_table, "EFI stub: "msg)
+#define pr_efi_err(sys_table, msg) efi_printk(sys_table, "EFI stub: ERROR: "msg)
+
 
 static efi_status_t efi_get_memory_map(efi_system_table_t *sys_table_arg,
 				       efi_memory_desc_t **map,
@@ -324,7 +327,7 @@ static efi_status_t handle_cmdline_files(efi_system_table_t *sys_table_arg,
 				nr_files * sizeof(*files),
 				(void **)&files);
 	if (status != EFI_SUCCESS) {
-		efi_printk(sys_table_arg, "Failed to alloc mem for file handle list\n");
+		pr_efi_err(sys_table_arg, "Failed to alloc mem for file handle list\n");
 		goto fail;
 	}
 
@@ -376,13 +379,13 @@ static efi_status_t handle_cmdline_files(efi_system_table_t *sys_table_arg,
 					image->device_handle, &fs_proto,
 						(void **)&io);
 			if (status != EFI_SUCCESS) {
-				efi_printk(sys_table_arg, "Failed to handle fs_proto\n");
+				pr_efi_err(sys_table_arg, "Failed to handle fs_proto\n");
 				goto free_files;
 			}
 
 			status = efi_call_phys2(io->open_volume, io, &fh);
 			if (status != EFI_SUCCESS) {
-				efi_printk(sys_table_arg, "Failed to open volume\n");
+				pr_efi_err(sys_table_arg, "Failed to open volume\n");
 				goto free_files;
 			}
 		}
@@ -390,7 +393,7 @@ static efi_status_t handle_cmdline_files(efi_system_table_t *sys_table_arg,
 		status = efi_call_phys5(fh->open, fh, &h, filename_16,
 					EFI_FILE_MODE_READ, (u64)0);
 		if (status != EFI_SUCCESS) {
-			efi_printk(sys_table_arg, "Failed to open file: ");
+			pr_efi_err(sys_table_arg, "Failed to open file: ");
 			efi_char16_printk(sys_table_arg, filename_16);
 			efi_printk(sys_table_arg, "\n");
 			goto close_handles;
@@ -402,7 +405,7 @@ static efi_status_t handle_cmdline_files(efi_system_table_t *sys_table_arg,
 		status = efi_call_phys4(h->get_info, h, &info_guid,
 					&info_sz, NULL);
 		if (status != EFI_BUFFER_TOO_SMALL) {
-			efi_printk(sys_table_arg, "Failed to get file info size\n");
+			pr_efi_err(sys_table_arg, "Failed to get file info size\n");
 			goto close_handles;
 		}
 
@@ -411,7 +414,7 @@ grow:
 					EFI_LOADER_DATA, info_sz,
 					(void **)&info);
 		if (status != EFI_SUCCESS) {
-			efi_printk(sys_table_arg, "Failed to alloc mem for file info\n");
+			pr_efi_err(sys_table_arg, "Failed to alloc mem for file info\n");
 			goto close_handles;
 		}
 
@@ -427,7 +430,7 @@ grow:
 		efi_call_phys1(sys_table_arg->boottime->free_pool, info);
 
 		if (status != EFI_SUCCESS) {
-			efi_printk(sys_table_arg, "Failed to get file info\n");
+			pr_efi_err(sys_table_arg, "Failed to get file info\n");
 			goto close_handles;
 		}
 
@@ -446,13 +449,13 @@ grow:
 		status = efi_high_alloc(sys_table_arg, file_size_total, 0x1000,
 				    &file_addr, max_addr);
 		if (status != EFI_SUCCESS) {
-			efi_printk(sys_table_arg, "Failed to alloc highmem for files\n");
+			pr_efi_err(sys_table_arg, "Failed to alloc highmem for files\n");
 			goto close_handles;
 		}
 
 		/* We've run out of free low memory. */
 		if (file_addr > max_addr) {
-			efi_printk(sys_table_arg, "We've run out of free low memory\n");
+			pr_efi_err(sys_table_arg, "We've run out of free low memory\n");
 			status = EFI_INVALID_PARAMETER;
 			goto free_file_total;
 		}
@@ -473,7 +476,7 @@ grow:
 							&chunksize,
 							(void *)addr);
 				if (status != EFI_SUCCESS) {
-					efi_printk(sys_table_arg, "Failed to read file\n");
+					pr_efi_err(sys_table_arg, "Failed to read file\n");
 					goto free_file_total;
 				}
 				addr += chunksize;
@@ -558,7 +561,7 @@ static efi_status_t efi_relocate_kernel(efi_system_table_t *sys_table_arg,
 				       &new_addr);
 	}
 	if (status != EFI_SUCCESS) {
-		efi_printk(sys_table_arg, "ERROR: Failed to allocate usable memory for kernel.\n");
+		pr_efi_err(sys_table_arg, "ERROR: Failed to allocate usable memory for kernel.\n");
 		return status;
 	}
 
-- 
1.7.10.4

[PATCH v2 08/15] efi: Add get_dram_base() helper function

[Leif Lindholm]

From: Roy Franz <roy.franz@linaro.org>

Add the get_dram_base() function, shared by arm/arm64.

Signed-off-by: Roy Franz <roy.franz@linaro.org>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
---
 drivers/firmware/efi/efi-stub-helper.c |   30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/drivers/firmware/efi/efi-stub-helper.c b/drivers/firmware/efi/efi-stub-helper.c
index 4c8ab86..4a9986b 100644
--- a/drivers/firmware/efi/efi-stub-helper.c
+++ b/drivers/firmware/efi/efi-stub-helper.c
@@ -11,6 +11,10 @@
  */
 #define EFI_READ_CHUNK_SIZE	(1024 * 1024)
 
+/* error code which can't be mistaken for valid address */
+#define EFI_ERROR	(~0UL)
+
+
 struct file_info {
 	efi_file_handle_t *handle;
 	u64 size;
@@ -92,6 +96,32 @@ fail:
 	return status;
 }
 
+
+static unsigned long __init get_dram_base(efi_system_table_t *sys_table)
+{
+	efi_status_t status;
+	unsigned long map_size;
+	unsigned long membase  = EFI_ERROR;
+	struct efi_memory_map map;
+	efi_memory_desc_t *md;
+
+	status = efi_get_memory_map(sys_table, (efi_memory_desc_t **)&map.map,
+				    &map_size, &map.desc_size, NULL, NULL);
+	if (status != EFI_SUCCESS)
+		return membase;
+
+	map.map_end = map.map + map_size;
+
+	for_each_efi_memory_desc(&map, md)
+		if (md->attribute & EFI_MEMORY_WB)
+			if (membase > md->phys_addr)
+				membase = md->phys_addr;
+
+	efi_call_phys1(sys_table->boottime->free_pool, map.map);
+
+	return membase;
+}
+
 /*
  * Allocate at the highest possible address that is not above 'max'.
  */
-- 
1.7.10.4

[PATCH v2 09/15] efi: Add shared FDT related functions for ARM/ARM64

[Leif Lindholm]

From: Roy Franz <roy.franz@linaro.org>

Both ARM and ARM64 stubs will update the device tree that they pass to
the kernel.  In both cases they primarily need to add the same UEFI
related information, so the function can be shared.  Create a new FDT
related file for this to avoid use of architecture #ifdefs in
efi-stub-helper.c.

Signed-off-by: Roy Franz <roy.franz@linaro.org>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
Acked-by: Grant Likely <grant.likely@linaro.org>
---
 drivers/firmware/efi/fdt.c |  268 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/efi.h        |    3 +
 2 files changed, 271 insertions(+)
 create mode 100644 drivers/firmware/efi/fdt.c

diff --git a/drivers/firmware/efi/fdt.c b/drivers/firmware/efi/fdt.c
new file mode 100644
index 0000000..4510f97
--- /dev/null
+++ b/drivers/firmware/efi/fdt.c
@@ -0,0 +1,268 @@
+/*
+ * FDT related Helper functions used by the EFI stub on multiple
+ * architectures. This should be #included by the EFI stub
+ * implementation files.
+ *
+ * Copyright 2013 Linaro Limited; author Roy Franz
+ *
+ * This file is part of the Linux kernel, and is made available
+ * under the terms of the GNU General Public License version 2.
+ *
+ */
+
+static efi_status_t update_fdt(efi_system_table_t *sys_table, void *orig_fdt,
+			       unsigned long orig_fdt_size,
+			       void *fdt, int new_fdt_size, char *cmdline_ptr,
+			       u64 initrd_addr, u64 initrd_size,
+			       efi_memory_desc_t *memory_map,
+			       unsigned long map_size, unsigned long desc_size,
+			       u32 desc_ver)
+{
+	int node;
+	int status;
+	u32 fdt_val32;
+	u64 fdt_val64;
+
+	/*
+	 * Copy definition of linux_banner here.  Since this code is
+	 * built as part of the decompressor for ARM v7, pulling
+	 * in version.c where linux_banner is defined for the
+	 * kernel brings other kernel dependencies with it.
+	 */
+	const char linux_banner[] =
+	    "Linux version " UTS_RELEASE " (" LINUX_COMPILE_BY "@"
+	    LINUX_COMPILE_HOST ") (" LINUX_COMPILER ") " UTS_VERSION "\n";
+
+	/* Do some checks on provided FDT, if it exists*/
+	if (orig_fdt) {
+		if (fdt_check_header(orig_fdt)) {
+			pr_efi_err(sys_table, "Device Tree header not valid!\n");
+			return EFI_LOAD_ERROR;
+		}
+		/*
+		 * We don't get the size of the FDT if we get if from a
+		 * configuration table.
+		 */
+		if (orig_fdt_size && fdt_totalsize(orig_fdt) > orig_fdt_size) {
+			pr_efi_err(sys_table, "Truncated device tree! foo!\n");
+			return EFI_LOAD_ERROR;
+		}
+	}
+
+	if (orig_fdt)
+		status = fdt_open_into(orig_fdt, fdt, new_fdt_size);
+	else
+		status = fdt_create_empty_tree(fdt, new_fdt_size);
+
+	if (status != 0)
+		goto fdt_set_fail;
+
+	/* Delete any memory nodes present */
+	while ((node = fdt_subnode_offset(fdt, 0, "memory")) >= 0)
+		fdt_del_node(fdt, node);
+
+	node = fdt_subnode_offset(fdt, 0, "chosen");
+	if (node < 0) {
+		node = fdt_add_subnode(fdt, 0, "chosen");
+		if (node < 0) {
+			status = node; /* node is error code when negative */
+			goto fdt_set_fail;
+		}
+	}
+
+	if ((cmdline_ptr != NULL) && (strlen(cmdline_ptr) > 0)) {
+		status = fdt_setprop(fdt, node, "bootargs", cmdline_ptr,
+				     strlen(cmdline_ptr) + 1);
+		if (status)
+			goto fdt_set_fail;
+	}
+
+	/* Set initrd address/end in device tree, if present */
+	if (initrd_size != 0) {
+		u64 initrd_image_end;
+		u64 initrd_image_start = cpu_to_fdt64(initrd_addr);
+		status = fdt_setprop(fdt, node, "linux,initrd-start",
+				     &initrd_image_start, sizeof(u64));
+		if (status)
+			goto fdt_set_fail;
+		initrd_image_end = cpu_to_fdt64(initrd_addr + initrd_size);
+		status = fdt_setprop(fdt, node, "linux,initrd-end",
+				     &initrd_image_end, sizeof(u64));
+		if (status)
+			goto fdt_set_fail;
+	}
+
+	/* Add FDT entries for EFI runtime services in chosen node. */
+	node = fdt_subnode_offset(fdt, 0, "chosen");
+	fdt_val64 = cpu_to_fdt64((u64)(unsigned long)sys_table);
+	status = fdt_setprop(fdt, node, "linux,uefi-system-table",
+			     &fdt_val64, sizeof(fdt_val64));
+	if (status)
+		goto fdt_set_fail;
+
+	fdt_val64 = cpu_to_fdt64((u64)(unsigned long)memory_map);
+	status = fdt_setprop(fdt, node, "linux,uefi-mmap-start",
+			     &fdt_val64,  sizeof(fdt_val64));
+	if (status)
+		goto fdt_set_fail;
+
+	fdt_val32 = cpu_to_fdt32(map_size);
+	status = fdt_setprop(fdt, node, "linux,uefi-mmap-size",
+			     &fdt_val32,  sizeof(fdt_val32));
+	if (status)
+		goto fdt_set_fail;
+
+	fdt_val32 = cpu_to_fdt32(desc_size);
+	status = fdt_setprop(fdt, node, "linux,uefi-mmap-desc-size",
+			     &fdt_val32, sizeof(fdt_val32));
+	if (status)
+		goto fdt_set_fail;
+
+	fdt_val32 = cpu_to_fdt32(desc_ver);
+	status = fdt_setprop(fdt, node, "linux,uefi-mmap-desc-ver",
+			     &fdt_val32, sizeof(fdt_val32));
+	if (status)
+		goto fdt_set_fail;
+
+	/*
+	 * Add kernel version banner so stub/kernel match can be
+	 * verified.
+	 */
+	status = fdt_setprop_string(fdt, node, "linux,uefi-stub-kern-ver",
+			     linux_banner);
+	if (status)
+		goto fdt_set_fail;
+
+	return EFI_SUCCESS;
+
+fdt_set_fail:
+	if (status == -FDT_ERR_NOSPACE)
+		return EFI_BUFFER_TOO_SMALL;
+
+	return EFI_LOAD_ERROR;
+}
+
+#ifndef EFI_FDT_ALIGN
+#define EFI_FDT_ALIGN EFI_PAGE_SIZE
+#endif
+
+/*
+ * Allocate memory for a new FDT, then add EFI, commandline, and
+ * initrd related fields to the FDT.  This routine increases the
+ * FDT allocation size until the allocated memory is large
+ * enough.  EFI allocations are in EFI_PAGE_SIZE granules,
+ * which are fixed at 4K bytes, so in most cases the first
+ * allocation should succeed.
+ * EFI boot services are exited at the end of this function.
+ * There must be no allocations between the get_memory_map()
+ * call and the exit_boot_services() call, so the exiting of
+ * boot services is very tightly tied to the creation of the FDT
+ * with the final memory map in it.
+ */
+
+efi_status_t allocate_new_fdt_and_exit_boot(efi_system_table_t *sys_table,
+					    void *handle,
+					    unsigned long *new_fdt_addr,
+					    unsigned long max_addr,
+					    u64 initrd_addr, u64 initrd_size,
+					    char *cmdline_ptr,
+					    unsigned long fdt_addr,
+					    unsigned long fdt_size)
+{
+	unsigned long map_size, desc_size;
+	u32 desc_ver;
+	unsigned long mmap_key;
+	efi_memory_desc_t *memory_map;
+	unsigned long new_fdt_size;
+	efi_status_t status;
+
+	/*
+	 * Estimate size of new FDT, and allocate memory for it. We
+	 * will allocate a bigger buffer if this ends up being too
+	 * small, so a rough guess is OK here.
+	 */
+	new_fdt_size = fdt_size + EFI_PAGE_SIZE;
+	while (1) {
+		status = efi_high_alloc(sys_table, new_fdt_size, EFI_FDT_ALIGN,
+					new_fdt_addr, max_addr);
+		if (status != EFI_SUCCESS) {
+			pr_efi_err(sys_table, "Unable to allocate memory for new device tree.\n");
+			goto fail;
+		}
+
+		/*
+		 * Now that we have done our final memory allocation (and free)
+		 * we can get the memory map key  needed for
+		 * exit_boot_services().
+		 */
+		status = efi_get_memory_map(sys_table, &memory_map, &map_size,
+					    &desc_size, &desc_ver, &mmap_key);
+		if (status != EFI_SUCCESS)
+			goto fail_free_new_fdt;
+
+		status = update_fdt(sys_table,
+				    (void *)fdt_addr, fdt_size,
+				    (void *)*new_fdt_addr, new_fdt_size,
+				    cmdline_ptr, initrd_addr, initrd_size,
+				    memory_map, map_size, desc_size, desc_ver);
+
+		/* Succeeding the first time is the expected case. */
+		if (status == EFI_SUCCESS)
+			break;
+
+		if (status == EFI_BUFFER_TOO_SMALL) {
+			/*
+			 * We need to allocate more space for the new
+			 * device tree, so free existing buffer that is
+			 * too small.  Also free memory map, as we will need
+			 * to get new one that reflects the free/alloc we do
+			 * on the device tree buffer.
+			 */
+			efi_free(sys_table, new_fdt_size, *new_fdt_addr);
+			efi_call_phys1(sys_table->boottime->free_pool,
+				       memory_map);
+			new_fdt_size += EFI_PAGE_SIZE;
+		} else {
+			pr_efi_err(sys_table, "Unable to constuct new device tree.\n");
+			goto fail_free_mmap;
+		}
+	}
+
+	/* Now we are ready to exit_boot_services.*/
+	status = efi_call_phys2(sys_table->boottime->exit_boot_services,
+				handle, mmap_key);
+
+
+	if (status == EFI_SUCCESS)
+		return status;
+
+	pr_efi_err(sys_table, "Exit boot services failed.\n");
+
+fail_free_mmap:
+	efi_call_phys1(sys_table->boottime->free_pool, memory_map);
+
+fail_free_new_fdt:
+	efi_free(sys_table, new_fdt_size, *new_fdt_addr);
+
+fail:
+	return EFI_LOAD_ERROR;
+}
+
+static void *get_fdt(efi_system_table_t *sys_table)
+{
+	efi_guid_t fdt_guid = EFI_DEVICE_TREE_GUID;
+	efi_config_table_t *tables;
+	void *fdt;
+	int i;
+
+	tables = (efi_config_table_t *) sys_table->tables;
+	fdt = NULL;
+
+	for (i = 0; i < sys_table->nr_tables; i++)
+		if (efi_guidcmp(tables[i].guid, fdt_guid) == 0) {
+			fdt = (void *) tables[i].table;
+			break;
+	 }
+
+	return fdt;
+}
diff --git a/include/linux/efi.h b/include/linux/efi.h
index d450673..cd112bb 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -394,6 +394,9 @@ typedef efi_status_t efi_query_variable_store_t(u32 attributes, unsigned long si
 #define EFI_FILE_SYSTEM_GUID \
     EFI_GUID(  0x964e5b22, 0x6459, 0x11d2, 0x8e, 0x39, 0x00, 0xa0, 0xc9, 0x69, 0x72, 0x3b )
 
+#define EFI_DEVICE_TREE_GUID \
+    EFI_GUID(  0xb1b621d5, 0xf19c, 0x41a5, 0x83, 0x0b, 0xd9, 0x15, 0x2c, 0x69, 0xaa, 0xe0 )
+
 typedef struct {
 	efi_guid_t guid;
 	u64 table;
-- 
1.7.10.4

[PATCH v2 10/15] arm64: Add function to create identity mappings

[Leif Lindholm]

From: Mark Salter <msalter@redhat.com>

At boot time, before switching to a virtual UEFI memory map, firmware
expects UEFI memory and IO regions to be identity mapped whenever
kernel makes runtime services calls. The existing early boot code
creates an identity map of kernel text/data but this is not sufficient
for UEFI. This patch adds a create_id_mapping() function which reuses
the core code of the existing create_mapping().

Signed-off-by: Mark Salter <msalter@redhat.com>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
---
 arch/arm64/include/asm/mmu.h |    2 ++
 arch/arm64/mm/mmu.c          |   65 ++++++++++++++++++++++++++++++------------
 2 files changed, 49 insertions(+), 18 deletions(-)

diff --git a/arch/arm64/include/asm/mmu.h b/arch/arm64/include/asm/mmu.h
index f600d40..29ed1d8 100644
--- a/arch/arm64/include/asm/mmu.h
+++ b/arch/arm64/include/asm/mmu.h
@@ -28,5 +28,7 @@ extern void paging_init(void);
 extern void setup_mm_for_reboot(void);
 extern void __iomem *early_io_map(phys_addr_t phys, unsigned long virt);
 extern void init_mem_pgprot(void);
+/* create an identity mapping for memory (or io if map_io is true) */
+extern void create_id_mapping(phys_addr_t addr, phys_addr_t size, int map_io);
 
 #endif
diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c
index 6b7e895..357956a 100644
--- a/arch/arm64/mm/mmu.c
+++ b/arch/arm64/mm/mmu.c
@@ -168,7 +168,8 @@ static void __init *early_alloc(unsigned long sz)
 }
 
 static void __init alloc_init_pte(pmd_t *pmd, unsigned long addr,
-				  unsigned long end, unsigned long pfn)
+				  unsigned long end, unsigned long pfn,
+				  pgprot_t prot)
 {
 	pte_t *pte;
 
@@ -180,16 +181,28 @@ static void __init alloc_init_pte(pmd_t *pmd, unsigned long addr,
 
 	pte = pte_offset_kernel(pmd, addr);
 	do {
-		set_pte(pte, pfn_pte(pfn, PAGE_KERNEL_EXEC));
+		set_pte(pte, pfn_pte(pfn, prot));
 		pfn++;
 	} while (pte++, addr += PAGE_SIZE, addr != end);
 }
 
 static void __init alloc_init_pmd(pud_t *pud, unsigned long addr,
-				  unsigned long end, phys_addr_t phys)
+				  unsigned long end, phys_addr_t phys,
+				  int map_io)
 {
 	pmd_t *pmd;
 	unsigned long next;
+	pmdval_t prot_sect;
+	pgprot_t prot_pte;
+
+	if (map_io) {
+		prot_sect = PMD_TYPE_SECT | PMD_SECT_AF |
+			    PMD_ATTRINDX(MT_DEVICE_nGnRE);
+		prot_pte = __pgprot(PROT_DEVICE_nGnRE);
+	} else {
+		prot_sect = prot_sect_kernel;
+		prot_pte = PAGE_KERNEL_EXEC;
+	}
 
 	/*
 	 * Check for initial section mappings in the pgd/pud and remove them.
@@ -205,7 +218,7 @@ static void __init alloc_init_pmd(pud_t *pud, unsigned long addr,
 		/* try section mapping first */
 		if (((addr | next | phys) & ~SECTION_MASK) == 0) {
 			pmd_t old_pmd =*pmd;
-			set_pmd(pmd, __pmd(phys | prot_sect_kernel));
+			set_pmd(pmd, __pmd(phys | prot_sect));
 			/*
 			 * Check for previous table entries created during
 			 * boot (__create_page_tables) and flush them.
@@ -213,21 +226,23 @@ static void __init alloc_init_pmd(pud_t *pud, unsigned long addr,
 			if (!pmd_none(old_pmd))
 				flush_tlb_all();
 		} else {
-			alloc_init_pte(pmd, addr, next, __phys_to_pfn(phys));
+			alloc_init_pte(pmd, addr, next, __phys_to_pfn(phys),
+				       prot_pte);
 		}
 		phys += next - addr;
 	} while (pmd++, addr = next, addr != end);
 }
 
 static void __init alloc_init_pud(pgd_t *pgd, unsigned long addr,
-				  unsigned long end, unsigned long phys)
+				  unsigned long end, unsigned long phys,
+				  int map_io)
 {
 	pud_t *pud = pud_offset(pgd, addr);
 	unsigned long next;
 
 	do {
 		next = pud_addr_end(addr, end);
-		alloc_init_pmd(pud, addr, next, phys);
+		alloc_init_pmd(pud, addr, next, phys, map_io);
 		phys += next - addr;
 	} while (pud++, addr = next, addr != end);
 }
@@ -236,30 +251,44 @@ static void __init alloc_init_pud(pgd_t *pgd, unsigned long addr,
  * Create the page directory entries and any necessary page tables for the
  * mapping specified by 'md'.
  */
-static void __init create_mapping(phys_addr_t phys, unsigned long virt,
-				  phys_addr_t size)
+static void __init __create_mapping(pgd_t *pgd, phys_addr_t phys,
+				    unsigned long virt, phys_addr_t size,
+				    int map_io)
 {
 	unsigned long addr, length, end, next;
-	pgd_t *pgd;
-
-	if (virt < VMALLOC_START) {
-		pr_warning("BUG: not creating mapping for 0x%016llx at 0x%016lx - outside kernel range\n",
-			   phys, virt);
-		return;
-	}
 
 	addr = virt & PAGE_MASK;
 	length = PAGE_ALIGN(size + (virt & ~PAGE_MASK));
 
-	pgd = pgd_offset_k(addr);
 	end = addr + length;
 	do {
 		next = pgd_addr_end(addr, end);
-		alloc_init_pud(pgd, addr, next, phys);
+		alloc_init_pud(pgd, addr, next, phys, map_io);
 		phys += next - addr;
 	} while (pgd++, addr = next, addr != end);
 }
 
+static void __init create_mapping(phys_addr_t phys, unsigned long virt,
+				  phys_addr_t size)
+{
+	if (virt < VMALLOC_START) {
+		pr_warn("BUG: not creating mapping for 0x%016llx@0x%016lx - outside kernel range\n",
+			phys, virt);
+		return;
+	}
+	__create_mapping(pgd_offset_k(virt & PAGE_MASK), phys, virt, size, 0);
+}
+
+void __init create_id_mapping(phys_addr_t addr, phys_addr_t size, int map_io)
+{
+	if ((addr >> PGDIR_SHIFT) >= ARRAY_SIZE(idmap_pg_dir)) {
+		pr_warn("BUG: not creating id mapping for 0x%016llx\n", addr);
+		return;
+	}
+	__create_mapping(&idmap_pg_dir[pgd_index(addr)],
+			 addr, addr, size, map_io);
+}
+
 static void __init map_mem(void)
 {
 	struct memblock_region *reg;
-- 
1.7.10.4

[PATCH v2 11/15] arm64: add EFI stub

[Leif Lindholm]

From: Mark Salter <msalter@redhat.com>

This patch adds PE/COFF header fields to the start of the Image
so that it appears as an EFI application to EFI firmware. An EFI
stub is included to allow direct booting of the kernel Image.
Support in the COFF header for signed images was provided by
Ard Biesheuvel.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Mark Salter <msalter@redhat.com>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
---
 arch/arm64/Kconfig              |   10 +++
 arch/arm64/kernel/Makefile      |    3 +
 arch/arm64/kernel/efi-entry.S   |   93 +++++++++++++++++++++++++
 arch/arm64/kernel/efi-stub.c    |   83 ++++++++++++++++++++++
 arch/arm64/kernel/head.S        |  112 ++++++++++++++++++++++++++++++
 drivers/firmware/efi/arm-stub.c |  144 +++++++++++++++++++++++++++++++++++++++
 6 files changed, 445 insertions(+)
 create mode 100644 arch/arm64/kernel/efi-entry.S
 create mode 100644 arch/arm64/kernel/efi-stub.c
 create mode 100644 drivers/firmware/efi/arm-stub.c

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index da4304a..6333d49 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -259,6 +259,16 @@ config CMDLINE_FORCE
 	  This is useful if you cannot or don't want to change the
 	  command-line options your boot loader passes to the kernel.
 
+config EFI_STUB
+	bool "EFI stub support"
+	depends on OF
+	select LIBFDT
+	default y
+	help
+	  This kernel feature allows an Image to be loaded directly
+	  by EFI firmware without the use of a bootloader.
+	  See Documentation/efi-stub.txt for more information.
+
 endmenu
 
 menu "Userspace binary formats"
diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile
index 2d4554b..0f60b45 100644
--- a/arch/arm64/kernel/Makefile
+++ b/arch/arm64/kernel/Makefile
@@ -4,6 +4,8 @@
 
 CPPFLAGS_vmlinux.lds	:= -DTEXT_OFFSET=$(TEXT_OFFSET)
 AFLAGS_head.o		:= -DTEXT_OFFSET=$(TEXT_OFFSET)
+CFLAGS_efi-stub.o 	:= -DTEXT_OFFSET=$(TEXT_OFFSET) \
+			   -I$(src)/../../../scripts/dtc/libfdt
 
 # Object file lists.
 arm64-obj-y		:= cputable.o debug-monitors.o entry.o irq.o fpsimd.o	\
@@ -20,6 +22,7 @@ arm64-obj-$(CONFIG_HAVE_HW_BREAKPOINT)+= hw_breakpoint.o
 arm64-obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 arm64-obj-$(CONFIG_ARM64_CPU_SUSPEND)	+= sleep.o suspend.o
 arm64-obj-$(CONFIG_JUMP_LABEL)		+= jump_label.o
+arm64-obj-$(CONFIG_EFI_STUB)		+= efi-stub.o efi-entry.o
 
 obj-y					+= $(arm64-obj-y) vdso/
 obj-m					+= $(arm64-obj-m)
diff --git a/arch/arm64/kernel/efi-entry.S b/arch/arm64/kernel/efi-entry.S
new file mode 100644
index 0000000..bfe8c9b
--- /dev/null
+++ b/arch/arm64/kernel/efi-entry.S
@@ -0,0 +1,93 @@
+/*
+ * EFI entry point.
+ *
+ * Copyright (C) 2013 Red Hat, Inc.
+ * Author: Mark Salter <msalter@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/linkage.h>
+#include <linux/init.h>
+
+#include <asm/assembler.h>
+
+#define EFI_LOAD_ERROR 0x8000000000000001
+
+	__INIT
+
+	/*
+	 * We arrive here from the EFI boot manager with:
+	 *
+	 *    * MMU on with identity-mapped RAM.
+	 *    * Icache and Dcache on
+	 *
+	 * We will most likely be running from some place other than where
+	 * we want to be. The kernel image wants to be placed at TEXT_OFFSET
+	 * from start of RAM.
+	 */
+ENTRY(efi_stub_entry)
+	stp	x29, x30, [sp, #-32]!
+
+	/*
+	 * Call efi_entry to do the real work.
+	 * x0 and x1 are already set up by firmware. Current runtime
+	 * address of image is calculated and passed via *image_addr.
+	 *
+	 * unsigned long efi_entry(void *handle,
+	 *                         efi_system_table_t *sys_table,
+	 *                         unsigned long *image_addr) ;
+	 */
+	adrp	x8, _text
+	add	x8, x8, #:lo12:_text
+	add	x2, sp, 16
+	str	x8, [x2]
+	bl	efi_entry
+	cmn	x0, #1
+	b.eq	efi_load_fail
+
+	/*
+	 * efi_entry() will have relocated the kernel image if necessary
+	 * and we return here with device tree address in x0 and the kernel
+	 * entry point stored at *image_addr. Save those values in registers
+	 * which are preserved by __flush_dcache_all.
+	 */
+	ldr	x1, [sp, #16]
+	mov	x20, x0
+	mov	x21, x1
+
+	/* Turn off Dcache and MMU */
+	mrs	x0, CurrentEL
+	cmp	x0, #PSR_MODE_EL2t
+	ccmp	x0, #PSR_MODE_EL2h, #0x4, ne
+	b.ne	1f
+	mrs	x0, sctlr_el2
+	bic	x0, x0, #1 << 0	// clear SCTLR.M
+	bic	x0, x0, #1 << 2	// clear SCTLR.C
+	msr	sctlr_el2, x0
+	isb
+	b	2f
+1:
+	mrs	x0, sctlr_el1
+	bic	x0, x0, #1 << 0	// clear SCTLR.M
+	bic	x0, x0, #1 << 2	// clear SCTLR.C
+	msr	sctlr_el1, x0
+	isb
+2:
+	bl	__flush_dcache_all
+
+	/* Jump to real entry point */
+	mov	x0, x20
+	mov	x1, xzr
+	mov	x2, xzr
+	mov	x3, xzr
+	br	x21
+
+efi_load_fail:
+	mov	x0, #EFI_LOAD_ERROR
+	ldp	x29, x30, [sp], #32
+	ret
+
+ENDPROC(efi_stub_entry)
diff --git a/arch/arm64/kernel/efi-stub.c b/arch/arm64/kernel/efi-stub.c
new file mode 100644
index 0000000..bf30913
--- /dev/null
+++ b/arch/arm64/kernel/efi-stub.c
@@ -0,0 +1,83 @@
+/*
+ * linux/arch/arm/boot/compressed/efi-stub.c
+ *
+ * Copyright (C) 2013, 2014 Linaro Ltd;  <roy.franz@linaro.org>
+ *
+ * This file implements the EFI boot stub for the arm64 kernel.
+ * Adapted from ARM version by Mark Salter <msalter@redhat.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/efi.h>
+#include <linux/libfdt.h>
+#include <asm/sections.h>
+#include <generated/compile.h>
+#include <generated/utsrelease.h>
+
+/*
+ * EFI function call wrappers. These are not required for arm/arm64, but
+ * wrappers are required for X86 to convert between ABIs. These wrappers are
+ * provided to allow code sharing between X86 and other architectures. Since
+ * these wrappers directly invoke the EFI function pointer, the function
+ * pointer type must be properly defined, which is not the case for X86. One
+ * advantage of this is it allows for type checking of arguments, which is not
+ * possible with the X86 wrappers.
+ */
+#define efi_call_phys0(f)			f()
+#define efi_call_phys1(f, a1)			f(a1)
+#define efi_call_phys2(f, a1, a2)		f(a1, a2)
+#define efi_call_phys3(f, a1, a2, a3)		f(a1, a2, a3)
+#define efi_call_phys4(f, a1, a2, a3, a4)	f(a1, a2, a3, a4)
+#define efi_call_phys5(f, a1, a2, a3, a4, a5)	f(a1, a2, a3, a4, a5)
+
+/*
+ * AArch64 requires the DTB to be 8-byte aligned in the first 512MiB from
+ * start of kernel and may not cross a 2MiB boundary. We set alignment to
+ * 2MiB so we know it won't cross a 2MiB boundary.
+ */
+#define EFI_FDT_ALIGN	SZ_2M   /* used by allocate_new_fdt_and_exit_boot() */
+#define MAX_FDT_OFFSET	SZ_512M
+
+/* Include shared EFI stub code */
+#include "../../../drivers/firmware/efi/efi-stub-helper.c"
+#include "../../../drivers/firmware/efi/fdt.c"
+#include "../../../drivers/firmware/efi/arm-stub.c"
+
+
+static efi_status_t handle_kernel_image(efi_system_table_t *sys_table,
+					unsigned long *image_addr,
+					unsigned long *image_size,
+					unsigned long *reserve_addr,
+					unsigned long *reserve_size,
+					unsigned long dram_base,
+					efi_loaded_image_t *image)
+{
+	efi_status_t status;
+	unsigned long kernel_size, kernel_memsize = 0;
+
+	/* Relocate the image, if required. */
+	kernel_size = _edata - _text;
+	if (*image_addr != (dram_base + TEXT_OFFSET)) {
+		kernel_memsize = kernel_size + (_end - _edata);
+		status = efi_relocate_kernel(sys_table, image_addr,
+					     kernel_size, kernel_memsize,
+					     dram_base + TEXT_OFFSET,
+					     PAGE_SIZE);
+		if (status != EFI_SUCCESS) {
+			pr_efi_err(sys_table, "Failed to relocate kernel\n");
+			return status;
+		}
+		if (*image_addr != (dram_base + TEXT_OFFSET)) {
+			pr_efi_err(sys_table, "Failed to alloc kernel memory\n");
+			efi_free(sys_table, kernel_memsize, *image_addr);
+			return EFI_ERROR;
+		}
+		*image_size = kernel_memsize;
+	}
+
+
+	return EFI_SUCCESS;
+}
diff --git a/arch/arm64/kernel/head.S b/arch/arm64/kernel/head.S
index c86bfdf..2f07979 100644
--- a/arch/arm64/kernel/head.S
+++ b/arch/arm64/kernel/head.S
@@ -107,8 +107,18 @@
 	/*
 	 * DO NOT MODIFY. Image header expected by Linux boot-loaders.
 	 */
+#ifdef CONFIG_EFI_STUB
+	/*
+	 * Magic "MZ" signature for PE/COFF
+	 * Little Endian:  add x13, x18, #0x16
+	 */
+efi_head:
+	.long   0x91005a4d
+	b	stext
+#else
 	b	stext				// branch to kernel start, magic
 	.long	0				// reserved
+#endif
 	.quad	TEXT_OFFSET			// Image load offset from start of RAM
 	.quad	0				// reserved
 	.quad	0				// reserved
@@ -119,7 +129,109 @@
 	.byte	0x52
 	.byte	0x4d
 	.byte	0x64
+#ifdef CONFIG_EFI_STUB
+	.long	pe_header - efi_head		// Offset to the PE header.
+#else
 	.word	0				// reserved
+#endif
+
+#ifdef CONFIG_EFI_STUB
+	.align 3
+pe_header:
+	.ascii	"PE"
+	.short 	0
+coff_header:
+	.short	0xaa64				// AArch64
+	.short	2				// nr_sections
+	.long	0 				// TimeDateStamp
+	.long	0				// PointerToSymbolTable
+	.long	1				// NumberOfSymbols
+	.short	section_table - optional_header	// SizeOfOptionalHeader
+	.short	0x206				// Characteristics.
+						// IMAGE_FILE_DEBUG_STRIPPED |
+						// IMAGE_FILE_EXECUTABLE_IMAGE |
+						// IMAGE_FILE_LINE_NUMS_STRIPPED
+optional_header:
+	.short	0x20b				// PE32+ format
+	.byte	0x02				// MajorLinkerVersion
+	.byte	0x14				// MinorLinkerVersion
+	.long	_edata - stext			// SizeOfCode
+	.long	0				// SizeOfInitializedData
+	.long	0				// SizeOfUninitializedData
+	.long	efi_stub_entry - efi_head	// AddressOfEntryPoint
+	.long	stext - efi_head		// BaseOfCode
+
+extra_header_fields:
+	.quad	0				// ImageBase
+	.long	0x20				// SectionAlignment
+	.long	0x8				// FileAlignment
+	.short	0				// MajorOperatingSystemVersion
+	.short	0				// MinorOperatingSystemVersion
+	.short	0				// MajorImageVersion
+	.short	0				// MinorImageVersion
+	.short	0				// MajorSubsystemVersion
+	.short	0				// MinorSubsystemVersion
+	.long	0				// Win32VersionValue
+
+	.long	_edata - efi_head		// SizeOfImage
+
+	// Everything before the kernel image is considered part of the header
+	.long	stext - efi_head			// SizeOfHeaders
+	.long	0				// CheckSum
+	.short	0xa				// Subsystem (EFI application)
+	.short	0				// DllCharacteristics
+	.quad	0				// SizeOfStackReserve
+	.quad	0				// SizeOfStackCommit
+	.quad	0				// SizeOfHeapReserve
+	.quad	0				// SizeOfHeapCommit
+	.long	0				// LoaderFlags
+	.long	0x6				// NumberOfRvaAndSizes
+
+	.quad	0				// ExportTable
+	.quad	0				// ImportTable
+	.quad	0				// ResourceTable
+	.quad	0				// ExceptionTable
+	.quad	0				// CertificationTable
+	.quad	0				// BaseRelocationTable
+
+	// Section table
+section_table:
+
+	/*
+	 * The EFI application loader requires a relocation section
+	 * because EFI applications must be relocatable.  This is a
+	 * dummy section as far as we are concerned.
+	 */
+	.ascii	".reloc"
+	.byte	0
+	.byte	0			// end of 0 padding of section name
+	.long	0
+	.long	0
+	.long	0			// SizeOfRawData
+	.long	0			// PointerToRawData
+	.long	0			// PointerToRelocations
+	.long	0			// PointerToLineNumbers
+	.short	0			// NumberOfRelocations
+	.short	0			// NumberOfLineNumbers
+	.long	0x42100040		// Characteristics (section flags)
+
+
+	.ascii	".text"
+	.byte	0
+	.byte	0
+	.byte	0        		// end of 0 padding of section name
+	.long	_edata - stext		// VirtualSize
+	.long	stext - efi_head	// VirtualAddress
+	.long	_edata - stext		// SizeOfRawData
+	.long	stext - efi_head	// PointerToRawData
+
+	.long	0		// PointerToRelocations (0 for executables)
+	.long	0		// PointerToLineNumbers (0 for executables)
+	.short	0		// NumberOfRelocations  (0 for executables)
+	.short	0		// NumberOfLineNumbers  (0 for executables)
+	.long	0xe0500020	// Characteristics (section flags)
+	.align 5
+#endif
 
 ENTRY(stext)
 	mov	x21, x0				// x21=FDT
diff --git a/drivers/firmware/efi/arm-stub.c b/drivers/firmware/efi/arm-stub.c
new file mode 100644
index 0000000..b505fde
--- /dev/null
+++ b/drivers/firmware/efi/arm-stub.c
@@ -0,0 +1,144 @@
+/*
+ * EFI stub implementation that is shared by arm and arm64 architectures.
+ * This should be #included by the EFI stub implementation files.
+ *
+ * Copyright (C) 2013,2014 Linaro Limited
+ *     Roy Franz <roy.franz at linaro.org
+ * Copyright (C) 2013 Red Hat, Inc.
+ *     Mark Salter <msalter@redhat.com>
+ *
+ * This file is part of the Linux kernel, and is made available under the
+ * terms of the GNU General Public License version 2.
+ *
+ */
+
+/*
+ * This function handles the architcture specific differences between arm and
+ * arm64 regarding where the kernel image must be loaded and any memory that
+ * must be reserved. On failure it is required to free all
+ * all allocations it has made.
+ */
+static efi_status_t handle_kernel_image(efi_system_table_t *sys_table,
+					unsigned long *image_addr,
+					unsigned long *image_size,
+					unsigned long *reserve_addr,
+					unsigned long *reserve_size,
+					unsigned long dram_base,
+					efi_loaded_image_t *image);
+/*
+ * EFI entry point for the arm/arm64 EFI stubs.  This is the entrypoint
+ * that is described in the PE/COFF header.  Most of the code is the same
+ * for both archictectures, with the arch-specific code provided in the
+ * handle_kernel_image() function.
+ */
+unsigned long efi_entry(void *handle, efi_system_table_t *sys_table,
+			       unsigned long *image_addr)
+{
+	efi_loaded_image_t *image;
+	efi_status_t status;
+	unsigned long image_size = 0;
+	unsigned long dram_base;
+	/* addr/point and size pairs for memory management*/
+	unsigned long initrd_addr;
+	u64 initrd_size = 0;
+	unsigned long fdt_addr;  /* Original DTB */
+	u64 fdt_size = 0;  /* We don't get size from configuration table */
+	char *cmdline_ptr = NULL;
+	int cmdline_size = 0;
+	unsigned long new_fdt_addr;
+	efi_guid_t loaded_image_proto = LOADED_IMAGE_PROTOCOL_GUID;
+	unsigned long reserve_addr = 0;
+	unsigned long reserve_size = 0;
+
+	/* Check if we were booted by the EFI firmware */
+	if (sys_table->hdr.signature != EFI_SYSTEM_TABLE_SIGNATURE)
+		goto fail;
+
+	pr_efi(sys_table, "Booting Linux Kernel...\n");
+
+	/*
+	 * Get a handle to the loaded image protocol.  This is used to get
+	 * information about the running image, such as size and the command
+	 * line.
+	 */
+	status = efi_call_phys3(sys_table->boottime->handle_protocol,
+				handle, &loaded_image_proto, (void *)&image);
+	if (status != EFI_SUCCESS) {
+		pr_efi_err(sys_table, "Failed to get loaded image protocol\n");
+		goto fail;
+	}
+
+	dram_base = get_dram_base(sys_table);
+	if (dram_base == EFI_ERROR) {
+		pr_efi_err(sys_table, "Failed to find DRAM base\n");
+		goto fail;
+	}
+	status = handle_kernel_image(sys_table, image_addr, &image_size,
+				     &reserve_addr,
+				     &reserve_size,
+				     dram_base, image);
+	if (status != EFI_SUCCESS) {
+		pr_efi_err(sys_table, "Failed to relocate kernel\n");
+		goto fail;
+	}
+
+	/*
+	 * Get the command line from EFI, using the LOADED_IMAGE
+	 * protocol. We are going to copy the command line into the
+	 * device tree, so this can be allocated anywhere.
+	 */
+	cmdline_ptr = efi_convert_cmdline(sys_table, image, &cmdline_size);
+	if (!cmdline_ptr) {
+		pr_efi_err(sys_table, "getting command line via LOADED_IMAGE_PROTOCOL\n");
+		goto fail_free_image;
+	}
+
+	/* Load a device tree from the configuration table, if present. */
+	fdt_addr = (uintptr_t)get_fdt(sys_table);
+	if (!fdt_addr) {
+		status = handle_cmdline_files(sys_table, image, cmdline_ptr,
+					      "dtb=",
+					      ~0UL, (unsigned long *)&fdt_addr,
+					      (unsigned long *)&fdt_size);
+
+		if (status != EFI_SUCCESS) {
+			pr_efi_err(sys_table, "Failed to load device tree!\n");
+			goto fail_free_cmdline;
+		}
+	}
+
+	status = handle_cmdline_files(sys_table, image, cmdline_ptr,
+				      "initrd=", dram_base + SZ_512M,
+				      (unsigned long *)&initrd_addr,
+				      (unsigned long *)&initrd_size);
+	if (status != EFI_SUCCESS)
+		pr_efi_err(sys_table, "Failed initrd from command line!\n");
+
+	new_fdt_addr = fdt_addr;
+	status = allocate_new_fdt_and_exit_boot(sys_table, handle,
+				&new_fdt_addr, dram_base + MAX_FDT_OFFSET,
+				initrd_addr, initrd_size, cmdline_ptr,
+				fdt_addr, fdt_size);
+
+	/*
+	 * If all went well, we need to return the FDT address to the
+	 * calling function so it can be passed to kernel as part of
+	 * the kernel boot protocol.
+	 */
+	if (status == EFI_SUCCESS)
+		return new_fdt_addr;
+
+	pr_efi_err(sys_table, "Failed to update FDT and exit boot services\n");
+
+	efi_free(sys_table, initrd_size, initrd_addr);
+	efi_free(sys_table, fdt_size, fdt_addr);
+
+fail_free_cmdline:
+	efi_free(sys_table, cmdline_size, (unsigned long)cmdline_ptr);
+
+fail_free_image:
+	efi_free(sys_table, image_size, *image_addr);
+	efi_free(sys_table, reserve_size, reserve_addr);
+fail:
+	return EFI_ERROR;
+}
-- 
1.7.10.4

[PATCH v2 12/15] doc: arm64: add description of EFI stub support

[Leif Lindholm]

From: Mark Salter <msalter@redhat.com>

Add explanation of arm64 EFI stub and kernel image header changes
needed to masquerade as a PE/COFF application.

Signed-off-by: Mark Salter <msalter@redhat.com>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
Acked-by: Grant Likely <grant.likely@linaro.org>
CC: linux-doc at vger.kernel.org
CC: Rob Landley <rob@landley.net>
---
 Documentation/arm64/booting.txt |    4 ++++
 Documentation/efi-stub.txt      |   12 +++++++++---
 2 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/Documentation/arm64/booting.txt b/Documentation/arm64/booting.txt
index a9691cc..aa95d38c 100644
--- a/Documentation/arm64/booting.txt
+++ b/Documentation/arm64/booting.txt
@@ -85,6 +85,10 @@ The decompressed kernel image contains a 64-byte header as follows:
 Header notes:
 
 - code0/code1 are responsible for branching to stext.
+- when booting through EFI, code0/code1 are initially skipped.
+  res5 is an offset to the PE header and the PE header has the EFI
+  entry point (efi_stub_entry). When the stub has done its work, it
+  jumps to code0 to resume the normal boot process.
 
 The image must be placed at the specified offset (currently 0x80000)
 from the start of the system RAM and called there. The start of the
diff --git a/Documentation/efi-stub.txt b/Documentation/efi-stub.txt
index 26be7b0..7747024 100644
--- a/Documentation/efi-stub.txt
+++ b/Documentation/efi-stub.txt
@@ -12,6 +12,11 @@ arch/arm/boot/compressed/efi-header.S and
 arch/arm/boot/compressed/efi-stub.c. EFI stub code that is shared
 between architectures is in drivers/firmware/efi/efi-stub-helper.c.
 
+For arm64, there is no compressed kernel support, so the Image itself
+masquerades as a PE/COFF image and the EFI stub is linked into the
+kernel. The arm64 EFI stub lives in arch/arm64/kernel/efi-entry.S
+and arch/arm64/kernel/efi-stub.c.
+
 By using the EFI boot stub it's possible to boot a Linux kernel
 without the use of a conventional EFI boot loader, such as grub or
 elilo. Since the EFI boot stub performs the jobs of a boot loader, in
@@ -28,7 +33,8 @@ the extension the EFI firmware loader will refuse to execute it. It's
 not possible to execute bzImage.efi from the usual Linux file systems
 because EFI firmware doesn't have support for them. For ARM the
 arch/arm/boot/zImage should be copied to the system partition, and it
-may not need to be renamed.
+may not need to be renamed. Similarly for arm64, arch/arm64/boot/Image
+should be copied but not necessarily renamed.
 
 
 **** Passing kernel parameters from the EFI shell
@@ -72,7 +78,7 @@ is passed to bzImage.efi.
 
 **** The "dtb=" option
 
-For the ARM architecture, we also need to be able to provide a device
-tree to the kernel. This is done with the "dtb=" command line option,
+For the ARM and arm64 architectures, we also need to be able to provide a
+device tree to the kernel. This is done with the "dtb=" command line option,
 and is processed in the same manner as the "initrd=" option that is
 described above.
-- 
1.7.10.4

[PATCH v2 13/15] arm64: add EFI runtime services

[Leif Lindholm]

From: Mark Salter <msalter@redhat.com>

This patch adds EFI runtime support for arm64. The runtime support allows
the kernel to access various EFI runtime services provided by EFI firmware.
Things like reboot, real time clock, EFI boot variables, and others.

Signed-off-by: Mark Salter <msalter@redhat.com>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
---
 arch/arm64/Kconfig           |   16 ++
 arch/arm64/include/asm/efi.h |   14 ++
 arch/arm64/kernel/Makefile   |    1 +
 arch/arm64/kernel/efi.c      |  465 ++++++++++++++++++++++++++++++++++++++++++
 arch/arm64/kernel/setup.c    |    5 +
 init/main.c                  |    4 +
 6 files changed, 505 insertions(+)
 create mode 100644 arch/arm64/include/asm/efi.h
 create mode 100644 arch/arm64/kernel/efi.c

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 6333d49..52d846e 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -259,6 +259,20 @@ config CMDLINE_FORCE
 	  This is useful if you cannot or don't want to change the
 	  command-line options your boot loader passes to the kernel.
 
+config EFI
+        bool "EFI runtime service support"
+	depends on OF
+	select UCS2_STRING
+	select LIBFDT
+	select UEFI_PARAMS_FROM_FDT
+	help
+	  This enables the kernel to use UEFI runtime services that are
+	  available (such as the UEFI variable services).
+
+	  This option is only useful on systems that have UEFI firmware.
+	  However, even with this option, the resultant kernel should
+	  continue to boot on existing non-UEFI platforms.
+
 config EFI_STUB
 	bool "EFI stub support"
 	depends on OF
@@ -318,6 +332,8 @@ source "net/Kconfig"
 
 source "drivers/Kconfig"
 
+source "drivers/firmware/Kconfig"
+
 source "fs/Kconfig"
 
 source "arch/arm64/kvm/Kconfig"
diff --git a/arch/arm64/include/asm/efi.h b/arch/arm64/include/asm/efi.h
new file mode 100644
index 0000000..5a46c4e
--- /dev/null
+++ b/arch/arm64/include/asm/efi.h
@@ -0,0 +1,14 @@
+#ifndef _ASM_EFI_H
+#define _ASM_EFI_H
+
+#include <asm/io.h>
+
+#ifdef CONFIG_EFI
+extern void efi_init(void);
+extern void efi_idmap_init(void);
+#else
+#define efi_init()
+#define efi_idmap_init()
+#endif
+
+#endif /* _ASM_EFI_H */
diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile
index 0f60b45..4c5a797 100644
--- a/arch/arm64/kernel/Makefile
+++ b/arch/arm64/kernel/Makefile
@@ -23,6 +23,7 @@ arm64-obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 arm64-obj-$(CONFIG_ARM64_CPU_SUSPEND)	+= sleep.o suspend.o
 arm64-obj-$(CONFIG_JUMP_LABEL)		+= jump_label.o
 arm64-obj-$(CONFIG_EFI_STUB)		+= efi-stub.o efi-entry.o
+arm64-obj-$(CONFIG_EFI)			+= efi.o
 
 obj-y					+= $(arm64-obj-y) vdso/
 obj-m					+= $(arm64-obj-m)
diff --git a/arch/arm64/kernel/efi.c b/arch/arm64/kernel/efi.c
new file mode 100644
index 0000000..87c4910
--- /dev/null
+++ b/arch/arm64/kernel/efi.c
@@ -0,0 +1,465 @@
+/*
+ * Extensible Firmware Interface
+ *
+ * Based on Extensible Firmware Interface Specification version 2.4
+ *
+ * Copyright (C) 2013 Linaro Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/efi.h>
+#include <linux/export.h>
+#include <linux/memblock.h>
+#include <linux/bootmem.h>
+#include <linux/of.h>
+#include <linux/of_fdt.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+
+#include <asm/cacheflush.h>
+#include <asm/efi.h>
+#include <asm/tlbflush.h>
+#include <asm/mmu_context.h>
+
+struct efi_memory_map memmap;
+
+static efi_runtime_services_t *runtime;
+
+static u64 efi_system_table;
+
+static int uefi_debug __initdata;
+static int __init uefi_debug_setup(char *str)
+{
+	uefi_debug = 1;
+
+	return 0;
+}
+early_param("uefi_debug", uefi_debug_setup);
+
+static int __init is_normal_ram(efi_memory_desc_t *md)
+{
+	if (md->attribute & EFI_MEMORY_WB)
+		return 1;
+	return 0;
+}
+
+static void __init efi_setup_idmap(void)
+{
+	struct memblock_region *r;
+	efi_memory_desc_t *md;
+	u64 paddr, npages, size;
+
+	for_each_memblock(memory, r)
+		create_id_mapping(r->base, r->size, 0);
+
+	/* map runtime io spaces */
+	for_each_efi_memory_desc(&memmap, md) {
+		if (!(md->attribute & EFI_MEMORY_RUNTIME) || is_normal_ram(md))
+			continue;
+		paddr = md->phys_addr;
+		npages = md->num_pages;
+		memrange_efi_to_native(&paddr, &npages);
+		size = npages << PAGE_SHIFT;
+		create_id_mapping(paddr, size, 1);
+	}
+}
+
+static int __init uefi_init(void)
+{
+	efi_char16_t *c16;
+	char vendor[100] = "unknown";
+	int i, retval;
+
+	efi.systab = early_memremap(efi_system_table,
+				    sizeof(efi_system_table_t));
+	if (efi.systab == NULL) {
+		pr_warn("Unable to map EFI system table.\n");
+		return -ENOMEM;
+	}
+
+	set_bit(EFI_BOOT, &efi.flags);
+	set_bit(EFI_64BIT, &efi.flags);
+
+	/*
+	 * Verify the EFI Table
+	 */
+	if (efi.systab->hdr.signature != EFI_SYSTEM_TABLE_SIGNATURE) {
+		pr_err("System table signature incorrect\n");
+		return -EINVAL;
+	}
+	if ((efi.systab->hdr.revision >> 16) < 2)
+		pr_warn("Warning: EFI system table version %d.%02d, expected 2.00 or greater\n",
+			efi.systab->hdr.revision >> 16,
+			efi.systab->hdr.revision & 0xffff);
+
+	/* Show what we know for posterity */
+	c16 = early_memremap(efi.systab->fw_vendor,
+			     sizeof(vendor));
+	if (c16) {
+		for (i = 0; i < (int) sizeof(vendor) - 1 && *c16; ++i)
+			vendor[i] = c16[i];
+		vendor[i] = '\0';
+	}
+
+	pr_info("EFI v%u.%.02u by %s\n",
+		efi.systab->hdr.revision >> 16,
+		efi.systab->hdr.revision & 0xffff, vendor);
+
+	retval = efi_config_init(NULL);
+	if (retval == 0)
+		set_bit(EFI_CONFIG_TABLES, &efi.flags);
+
+	early_memunmap(c16, sizeof(vendor));
+	early_memunmap(efi.systab,  sizeof(efi_system_table_t));
+
+	return retval;
+}
+
+static __initdata char memory_type_name[][32] = {
+	{"Reserved"},
+	{"Loader Code"},
+	{"Loader Data"},
+	{"Boot Code"},
+	{"Boot Data"},
+	{"Runtime Code"},
+	{"Runtime Data"},
+	{"Conventional Memory"},
+	{"Unusable Memory"},
+	{"ACPI Reclaim Memory"},
+	{"ACPI Memory NVS"},
+	{"Memory Mapped I/O"},
+	{"MMIO Port Space"},
+	{"PAL Code"},
+};
+
+/*
+ * Return true for RAM regions we want to permanently reserve.
+ */
+static __init int is_reserve_region(efi_memory_desc_t *md)
+{
+	if (!is_normal_ram(md))
+		return 0;
+
+	if (md->attribute & EFI_MEMORY_RUNTIME)
+		return 1;
+
+	if (md->type == EFI_ACPI_RECLAIM_MEMORY)
+		return 1;
+
+	return 0;
+}
+
+static __init void reserve_regions(void)
+{
+	efi_memory_desc_t *md;
+	u64 paddr, npages, size;
+
+	if (uefi_debug)
+		pr_info("Processing EFI memory map:\n");
+
+	for_each_efi_memory_desc(&memmap, md) {
+		paddr = md->phys_addr;
+		npages = md->num_pages;
+
+		if (uefi_debug)
+			pr_info("  0x%012llx-0x%012llx [%s]",
+				paddr, paddr + (npages << EFI_PAGE_SHIFT) - 1,
+				memory_type_name[md->type]);
+
+		memrange_efi_to_native(&paddr, &npages);
+		size = npages << PAGE_SHIFT;
+
+		if (is_normal_ram(md))
+			early_init_dt_add_memory_arch(paddr, size);
+
+		if (is_reserve_region(md) ||
+		    md->type == EFI_BOOT_SERVICES_CODE ||
+		    md->type == EFI_BOOT_SERVICES_DATA) {
+			memblock_reserve(paddr, size);
+			if (uefi_debug)
+				pr_cont("*");
+		}
+
+		if (uefi_debug)
+			pr_cont("\n");
+	}
+}
+
+
+static u64 __init free_one_region(u64 start, u64 end)
+{
+	u64 size = end - start;
+
+	if (uefi_debug)
+		pr_info("  EFI freeing: 0x%012llx-0x%012llx\n",	start, end - 1);
+
+	free_bootmem_late(start, size);
+	return size;
+}
+
+static u64 __init free_region(u64 start, u64 end)
+{
+	u64 map_start, map_end, total = 0;
+
+	if (end <= start)
+		return total;
+
+	map_start = (u64)memmap.phys_map;
+	map_end = PAGE_ALIGN(map_start + (memmap.map_end - memmap.map));
+	map_start &= PAGE_MASK;
+
+	if (start < map_end && end > map_start) {
+		/* region overlaps UEFI memmap */
+		if (start < map_start)
+			total += free_one_region(start, map_start);
+
+		if (map_end < end)
+			total += free_one_region(map_end, end);
+	} else
+		total += free_one_region(start, end);
+
+	return total;
+}
+
+static void __init free_boot_services(void)
+{
+	u64 total_freed = 0;
+	u64 keep_end, free_start, free_end;
+	efi_memory_desc_t *md;
+
+	/*
+	 * If kernel uses larger pages than UEFI, we have to be careful
+	 * not to inadvertantly free memory we want to keep if there is
+	 * overlap@the kernel page size alignment. We do not want to
+	 * free is_reserve_region() memory nor the UEFI memmap itself.
+	 *
+	 * The memory map is sorted, so we keep track of the end of
+	 * any previous region we want to keep, remember any region
+	 * we want to free and defer freeing it until we encounter
+	 * the next region we want to keep. This way, before freeing
+	 * it, we can clip it as needed to avoid freeing memory we
+	 * want to keep for UEFI.
+	 */
+
+	keep_end = 0;
+	free_start = 0;
+
+	for_each_efi_memory_desc(&memmap, md) {
+		u64 paddr, npages, size;
+
+		if (is_reserve_region(md)) {
+			/*
+			 * We don't want to free any memory from this region.
+			 */
+			if (free_start) {
+				/* adjust free_end then free region */
+				if (free_end > md->phys_addr)
+					free_end -= PAGE_SIZE;
+				total_freed += free_region(free_start, free_end);
+				free_start = 0;
+			}
+			keep_end = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT);
+			continue;
+		}
+
+		if (md->type != EFI_BOOT_SERVICES_CODE &&
+		    md->type != EFI_BOOT_SERVICES_DATA) {
+			/* no need to free this region */
+			continue;
+		}
+
+		/*
+		 * We want to free memory from this region.
+		 */
+		paddr = md->phys_addr;
+		npages = md->num_pages;
+		memrange_efi_to_native(&paddr, &npages);
+		size = npages << PAGE_SHIFT;
+
+		if (free_start) {
+			if (paddr <= free_end)
+				free_end = paddr + size;
+			else {
+				total_freed += free_region(free_start, free_end);
+				free_start = paddr;
+				free_end = paddr + size;
+			}
+		} else {
+			free_start = paddr;
+			free_end = paddr + size;
+		}
+		if (free_start < keep_end) {
+			free_start += PAGE_SIZE;
+			if (free_start >= free_end)
+				free_start = 0;
+		}
+	}
+	if (free_start)
+		total_freed += free_region(free_start, free_end);
+
+	if (total_freed)
+		pr_info("Freed 0x%llx bytes of EFI boot services memory",
+			total_freed);
+}
+
+void __init efi_init(void)
+{
+	struct efi_fdt_params params;
+
+	/* Grab UEFI information placed in FDT by stub */
+	if (!efi_get_fdt_params(&params, uefi_debug))
+		return;
+
+	efi_system_table = params.system_table;
+
+	memblock_reserve(params.mmap & PAGE_MASK,
+			 PAGE_ALIGN(params.mmap_size + (params.mmap & ~PAGE_MASK)));
+	memmap.phys_map = (void *)params.mmap;
+	memmap.map = early_memremap(params.mmap, params.mmap_size);
+	memmap.map_end = memmap.map + params.mmap_size;
+	memmap.desc_size = params.desc_size;
+	memmap.desc_version = params.desc_ver;
+
+	if (uefi_init() < 0)
+		return;
+
+	reserve_regions();
+}
+
+void __init efi_idmap_init(void)
+{
+	/* boot time idmap_pg_dir is incomplete, so fill in missing parts */
+	efi_setup_idmap();
+}
+
+static int __init remap_region(efi_memory_desc_t *md, void **new)
+{
+	u64 paddr, vaddr, npages, size;
+
+	paddr = md->phys_addr;
+	npages = md->num_pages;
+	memrange_efi_to_native(&paddr, &npages);
+	size = npages << PAGE_SHIFT;
+
+	if (is_normal_ram(md))
+		vaddr = (__force u64)ioremap_cache(paddr, size);
+	else
+		vaddr = (__force u64)ioremap(paddr, size);
+
+	if (!vaddr) {
+		pr_err("Unable to remap 0x%llx pages @ %p\n",
+		       npages, (void *)paddr);
+		return 0;
+	}
+
+	/* adjust for any rounding when EFI and system pagesize differs */
+	md->virt_addr = vaddr + (md->phys_addr - paddr);
+
+	if (uefi_debug)
+		pr_info("  EFI remap 0x%012llx => %p\n",
+			md->phys_addr, (void *)md->virt_addr);
+
+	memcpy(*new, md, memmap.desc_size);
+	*new += memmap.desc_size;
+
+	return 1;
+}
+
+/*
+ * Called from setup_arch with interrupts disabled.
+ */
+void __init efi_enter_virtual_mode(void)
+{
+	efi_memory_desc_t *md;
+	phys_addr_t virtmap_phys;
+	void *virtmap, *virt_md;
+	efi_status_t status;
+	u64 mapsize;
+	int count = 0;
+	unsigned long flags;
+
+	if (!efi_enabled(EFI_BOOT)) {
+		pr_info("EFI services will not be available.\n");
+		return;
+	}
+
+	pr_info("Remapping and enabling EFI services.\n");
+
+	/* replace early memmap mapping with permanent mapping */
+	mapsize = memmap.map_end - memmap.map;
+	early_memunmap(memmap.map, mapsize);
+	memmap.map = (__force void *)ioremap_cache((phys_addr_t)memmap.phys_map,
+						   mapsize);
+	memmap.map_end = memmap.map + mapsize;
+
+	efi.memmap = &memmap;
+
+	/* Map the runtime regions */
+	virtmap = kmalloc(mapsize, GFP_KERNEL);
+	if (!virtmap) {
+		pr_err("Failed to allocate EFI virtual memmap\n");
+		return;
+	}
+	virtmap_phys = virt_to_phys(virtmap);
+	virt_md = virtmap;
+
+	for_each_efi_memory_desc(&memmap, md) {
+		if (!(md->attribute & EFI_MEMORY_RUNTIME))
+			continue;
+		if (remap_region(md, &virt_md))
+			++count;
+	}
+
+	efi.systab = (__force void *)efi_lookup_mapped_addr(efi_system_table);
+	if (efi.systab)
+		set_bit(EFI_SYSTEM_TABLES, &efi.flags);
+
+	local_irq_save(flags);
+	cpu_switch_mm(idmap_pg_dir, &init_mm);
+	flush_tlb_all();
+	flush_cache_all();
+
+	/* Call SetVirtualAddressMap with the physical address of the map */
+	runtime = efi.systab->runtime;
+	efi.set_virtual_address_map = runtime->set_virtual_address_map;
+
+	status = efi.set_virtual_address_map(count * memmap.desc_size,
+					     memmap.desc_size,
+					     memmap.desc_version,
+					     (efi_memory_desc_t *)virtmap_phys);
+	cpu_set_reserved_ttbr0();
+	flush_tlb_all();
+	flush_cache_all();
+	local_irq_restore(flags);
+
+	kfree(virtmap);
+
+	free_boot_services();
+
+	if (status != EFI_SUCCESS) {
+		pr_err("Failed to set EFI virtual address map! [%lx]\n",
+			status);
+		return;
+	}
+
+	/* Set up runtime services function pointers */
+	runtime = efi.systab->runtime;
+	efi.get_time = runtime->get_time;
+	efi.set_time = runtime->set_time;
+	efi.get_wakeup_time = runtime->get_wakeup_time;
+	efi.set_wakeup_time = runtime->set_wakeup_time;
+	efi.get_variable = runtime->get_variable;
+	efi.get_next_variable = runtime->get_next_variable;
+	efi.set_variable = runtime->set_variable;
+	efi.query_variable_info = runtime->query_variable_info;
+	efi.update_capsule = runtime->update_capsule;
+	efi.query_capsule_caps = runtime->query_capsule_caps;
+	efi.get_next_high_mono_count = runtime->get_next_high_mono_count;
+	efi.reset_system = runtime->reset_system;
+
+	set_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+}
diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c
index 4d2ac74..17e7e30 100644
--- a/arch/arm64/kernel/setup.c
+++ b/arch/arm64/kernel/setup.c
@@ -41,6 +41,7 @@
 #include <linux/memblock.h>
 #include <linux/of_fdt.h>
 #include <linux/of_platform.h>
+#include <linux/efi.h>
 
 #include <asm/fixmap.h>
 #include <asm/cputype.h>
@@ -55,6 +56,7 @@
 #include <asm/traps.h>
 #include <asm/memblock.h>
 #include <asm/psci.h>
+#include <asm/efi.h>
 
 unsigned int processor_id;
 EXPORT_SYMBOL(processor_id);
@@ -333,11 +335,14 @@ void __init setup_arch(char **cmdline_p)
 
 	parse_early_param();
 
+	efi_init();
 	arm64_memblock_init();
 
 	paging_init();
 	request_standard_resources();
 
+	efi_idmap_init();
+
 	unflatten_device_tree();
 
 	psci_init();
diff --git a/init/main.c b/init/main.c
index eb03090..7cd57b2 100644
--- a/init/main.c
+++ b/init/main.c
@@ -902,6 +902,10 @@ static noinline void __init kernel_init_freeable(void)
 	smp_prepare_cpus(setup_max_cpus);
 
 	do_pre_smp_initcalls();
+
+	if (IS_ENABLED(CONFIG_ARM64) && efi_enabled(EFI_BOOT))
+		efi_enter_virtual_mode();
+
 	lockup_detector_init();
 
 	smp_init();
-- 
1.7.10.4

[PATCH v2 14/15] doc: arm: add UEFI support documentation

[Leif Lindholm]

This patch provides documentation of the [U]EFI runtime service and
configuration features for the arm architecture.

Cc: Rob Landley <rob@landley.net>
Cc: linux-doc at vger.kernel.org

Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
Acked-by: Grant Likely <grant.likely@linaro.org>
---
 Documentation/arm/00-INDEX |    2 ++
 Documentation/arm/uefi.txt |   64 ++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 66 insertions(+)
 create mode 100644 Documentation/arm/uefi.txt

diff --git a/Documentation/arm/00-INDEX b/Documentation/arm/00-INDEX
index a94090c..3b08bc2 100644
--- a/Documentation/arm/00-INDEX
+++ b/Documentation/arm/00-INDEX
@@ -46,5 +46,7 @@ swp_emulation
 	- SWP/SWPB emulation handler/logging description
 tcm.txt
 	- ARM Tightly Coupled Memory
+uefi.txt
+	- [U]EFI configuration and runtime services documentation
 vlocks.txt
 	- Voting locks, low-level mechanism relying on memory system atomic writes.
diff --git a/Documentation/arm/uefi.txt b/Documentation/arm/uefi.txt
new file mode 100644
index 0000000..d60030a
--- /dev/null
+++ b/Documentation/arm/uefi.txt
@@ -0,0 +1,64 @@
+UEFI, the Unified Extensible Firmware Interface, is a specification
+governing the behaviours of compatible firmware interfaces. It is
+maintained by the UEFI Forum - http://www.uefi.org/.
+
+UEFI is an evolution of its predecessor 'EFI', so the terms EFI and
+UEFI are used somewhat interchangeably in this document and associated
+source code. As a rule, anything new uses 'UEFI', whereas 'EFI' refers
+to legacy code or specifications.
+
+UEFI support in Linux
+=====================
+Booting on a platform with firmware compliant with the UEFI specification
+makes it possible for the kernel to support additional features:
+- UEFI Runtime Services
+- Retrieving various configuration information through the standardised
+  interface of UEFI configuration tables. (ACPI, SMBIOS, ...)
+
+For actually enabling [U]EFI support, enable:
+- CONFIG_EFI=y
+- CONFIG_EFI_VARS=y or m
+
+The implementation depends on receiving information about the UEFI environment
+in a Flattened Device Tree (FDT) - so is only available with CONFIG_OF.
+
+UEFI stub
+=========
+The "stub" is a feature that extends the Image/zImage into a valid UEFI
+PE/COFF executable, including a loader application that makes it possible to
+load the kernel directly from the UEFI shell, boot menu, or one of the
+lightweight bootloaders like Gummiboot or rEFInd.
+
+The kernel image built with stub support remains a valid kernel image for
+booting in non-UEFI environments.
+
+UEFI kernel support on ARM
+==========================
+UEFI kernel support on the ARM architectures (arm and arm64) is only available
+when boot is performed through the stub.
+
+When booting in UEFI mode, the stub deletes any memory nodes from a provided DT.
+Instead, the kernel reads the UEFI memory map.
+
+The stub populates the FDT /chosen node with (and the kernel scans for) the
+following parameters:
+________________________________________________________________________________
+Name                      | Size   | Description
+================================================================================
+linux,uefi-system-table   | 64-bit | Physical address of the UEFI System Table.
+--------------------------------------------------------------------------------
+linux,uefi-mmap-start     | 64-bit | Physical address of the UEFI memory map,
+                          |        | populated by the UEFI GetMemoryMap() call.
+--------------------------------------------------------------------------------
+linux,uefi-mmap-size      | 32-bit | Size in bytes of the UEFI memory map
+                          |        | pointed to in previous entry.
+--------------------------------------------------------------------------------
+linux,uefi-mmap-desc-size | 32-bit | Size in bytes of each entry in the UEFI
+                          |        | memory map.
+--------------------------------------------------------------------------------
+linux,uefi-mmap-desc-ver  | 32-bit | Version of the mmap descriptor format.
+--------------------------------------------------------------------------------
+linux,uefi-stub-kern-ver  | string | Copy of linux_banner from build.
+--------------------------------------------------------------------------------
+
+For verbose debug messages, specify 'uefi_debug' on the kernel command line.
-- 
1.7.10.4

[PATCH v2 15/15] efi/arm64: ignore dtb= when UEFI SecureBoot is enabled

[Leif Lindholm]

From: Ard Biesheuvel <ard.biesheuvel@linaro.org>

Loading unauthenticated FDT blobs directly from storage is a security hazard,
so this should only be allowed when running with UEFI Secure Boot disabled.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Leif Lindholm <leif.lindholm@linaro.org>
---
 drivers/firmware/efi/arm-stub.c        |   23 ++++++++++++++---------
 drivers/firmware/efi/efi-stub-helper.c |   24 ++++++++++++++++++++++++
 2 files changed, 38 insertions(+), 9 deletions(-)

diff --git a/drivers/firmware/efi/arm-stub.c b/drivers/firmware/efi/arm-stub.c
index b505fde..733723c 100644
--- a/drivers/firmware/efi/arm-stub.c
+++ b/drivers/firmware/efi/arm-stub.c
@@ -41,7 +41,7 @@ unsigned long efi_entry(void *handle, efi_system_table_t *sys_table,
 	/* addr/point and size pairs for memory management*/
 	unsigned long initrd_addr;
 	u64 initrd_size = 0;
-	unsigned long fdt_addr;  /* Original DTB */
+	unsigned long fdt_addr = 0;  /* Original DTB */
 	u64 fdt_size = 0;  /* We don't get size from configuration table */
 	char *cmdline_ptr = NULL;
 	int cmdline_size = 0;
@@ -93,15 +93,20 @@ unsigned long efi_entry(void *handle, efi_system_table_t *sys_table,
 		goto fail_free_image;
 	}
 
-	/* Load a device tree from the configuration table, if present. */
-	fdt_addr = (uintptr_t)get_fdt(sys_table);
+	/*
+	 * Unauthenticated device tree data is a security hazard, so
+	 * ignore 'dtb=' unless UEFI Secure Boot is disabled.
+	 */
+	if (efi_secureboot_enabled(sys_table))
+		pr_efi(sys_table, "UEFI Secure Boot is enabled.\n");
+	else
+		handle_cmdline_files(sys_table, image, cmdline_ptr, "dtb=",
+				     ~0UL, (unsigned long *)&fdt_addr,
+				     (unsigned long *)&fdt_size);
 	if (!fdt_addr) {
-		status = handle_cmdline_files(sys_table, image, cmdline_ptr,
-					      "dtb=",
-					      ~0UL, (unsigned long *)&fdt_addr,
-					      (unsigned long *)&fdt_size);
-
-		if (status != EFI_SUCCESS) {
+		/* Look for a device tree configuration table entry. */
+		fdt_addr = (uintptr_t)get_fdt(sys_table);
+		if (!fdt_addr) {
 			pr_efi_err(sys_table, "Failed to load device tree!\n");
 			goto fail_free_cmdline;
 		}
diff --git a/drivers/firmware/efi/efi-stub-helper.c b/drivers/firmware/efi/efi-stub-helper.c
index 4a9986b..f2014f2 100644
--- a/drivers/firmware/efi/efi-stub-helper.c
+++ b/drivers/firmware/efi/efi-stub-helper.c
@@ -704,3 +704,27 @@ static char *efi_convert_cmdline(efi_system_table_t *sys_table_arg,
 	*cmd_line_len = options_bytes;
 	return (char *)cmdline_addr;
 }
+
+static int __init efi_secureboot_enabled(efi_system_table_t *sys_table_arg)
+{
+	static efi_guid_t const var_guid __initconst = EFI_GLOBAL_VARIABLE_GUID;
+	static efi_char16_t const var_name[] __initconst = {
+		'S', 'e', 'c', 'u', 'r', 'e', 'B', 'o', 'o', 't', 0 };
+
+	efi_get_variable_t *f_getvar = sys_table_arg->runtime->get_variable;
+	unsigned long size = sizeof(u8);
+	efi_status_t status;
+	u8 val;
+
+	status = efi_call_phys5(f_getvar, (efi_char16_t *)var_name,
+				(efi_guid_t *)&var_guid, NULL, &size, &val);
+
+	switch (status) {
+	case EFI_SUCCESS:
+		return val;
+	case EFI_NOT_FOUND:
+		return 0;
+	default:
+		return 1;
+	}
+}
-- 
1.7.10.4

[PATCH v2 00/14] UEFI support for arm(64)

[Matt Fleming]

On Thu, 13 Mar, at 10:46:53PM, Leif Lindholm wrote:
> Since all prerequisites are not ready for the (32-bit) arm patches to
> go in, I have broken these out of this set and will be resubmitting
> them separately.
> 
> This set now provides arm64 support only - but includes the code that
> is common between the ARM ports.

This series looks OK to me.

-- 
Matt Fleming, Intel Open Source Technology Center

[PATCH v2 11/15] arm64: add EFI stub

[Catalin Marinas]

On Thu, Mar 13, 2014 at 10:47:04PM +0000, Leif Lindholm wrote:
> --- /dev/null
> +++ b/arch/arm64/kernel/efi-entry.S
> @@ -0,0 +1,93 @@
> +/*
> + * EFI entry point.
> + *
> + * Copyright (C) 2013 Red Hat, Inc.
> + * Author: Mark Salter <msalter@redhat.com>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + *
> + */
> +#include <linux/linkage.h>
> +#include <linux/init.h>
> +
> +#include <asm/assembler.h>
> +
> +#define EFI_LOAD_ERROR 0x8000000000000001
> +
> +       __INIT
> +
> +       /*
> +        * We arrive here from the EFI boot manager with:
> +        *
> +        *    * MMU on with identity-mapped RAM.
> +        *    * Icache and Dcache on
> +        *
> +        * We will most likely be running from some place other than where
> +        * we want to be. The kernel image wants to be placed at TEXT_OFFSET
> +        * from start of RAM.
> +        */
> +ENTRY(efi_stub_entry)
> +       stp     x29, x30, [sp, #-32]!
> +
> +       /*
> +        * Call efi_entry to do the real work.
> +        * x0 and x1 are already set up by firmware. Current runtime
> +        * address of image is calculated and passed via *image_addr.
> +        *
> +        * unsigned long efi_entry(void *handle,
> +        *                         efi_system_table_t *sys_table,
> +        *                         unsigned long *image_addr) ;
> +        */
> +       adrp    x8, _text
> +       add     x8, x8, #:lo12:_text
> +       add     x2, sp, 16
> +       str     x8, [x2]
> +       bl      efi_entry
> +       cmn     x0, #1
> +       b.eq    efi_load_fail
> +
> +       /*
> +        * efi_entry() will have relocated the kernel image if necessary
> +        * and we return here with device tree address in x0 and the kernel
> +        * entry point stored at *image_addr. Save those values in registers
> +        * which are preserved by __flush_dcache_all.
> +        */
> +       ldr     x1, [sp, #16]
> +       mov     x20, x0
> +       mov     x21, x1
> +
> +       /* Turn off Dcache and MMU */
> +       mrs     x0, CurrentEL
> +       cmp     x0, #PSR_MODE_EL2t
> +       ccmp    x0, #PSR_MODE_EL2h, #0x4, ne
> +       b.ne    1f
> +       mrs     x0, sctlr_el2
> +       bic     x0, x0, #1 << 0 // clear SCTLR.M
> +       bic     x0, x0, #1 << 2 // clear SCTLR.C
> +       msr     sctlr_el2, x0
> +       isb
> +       b       2f
> +1:
> +       mrs     x0, sctlr_el1
> +       bic     x0, x0, #1 << 0 // clear SCTLR.M
> +       bic     x0, x0, #1 << 2 // clear SCTLR.C
> +       msr     sctlr_el1, x0
> +       isb
> +2:
> +       bl      __flush_dcache_all

In linux-next I'm pushing a patch which no longer exports the
__flush_dcache_all function. The reason is that it doesn't really work
if you have a (not fully transparent) external cache like on the Applied
Micro boards. There other issues when running as a guest as well.

If you know exactly what needs to be flushed here, can you use a range
(MVA) operation?

> diff --git a/arch/arm64/kernel/efi-stub.c b/arch/arm64/kernel/efi-stub.c
> new file mode 100644
> index 0000000..bf30913
> --- /dev/null
> +++ b/arch/arm64/kernel/efi-stub.c
> @@ -0,0 +1,83 @@
> +/*
> + * linux/arch/arm/boot/compressed/efi-stub.c

Nitpick: arch/arm64/... But we don't really need to write the file name
here, I use a smart editor that tells me which file I'm viewing ;).
Better write a one-line summary of what this file is about.

> + *
> + * Copyright (C) 2013, 2014 Linaro Ltd;  <roy.franz@linaro.org>
> + *
> + * This file implements the EFI boot stub for the arm64 kernel.
> + * Adapted from ARM version by Mark Salter <msalter@redhat.com>
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + *
> + */
> +#include <linux/efi.h>
> +#include <linux/libfdt.h>
> +#include <asm/sections.h>
> +#include <generated/compile.h>
> +#include <generated/utsrelease.h>
> +
> +/*
> + * EFI function call wrappers. These are not required for arm/arm64, but
> + * wrappers are required for X86 to convert between ABIs. These wrappers are
> + * provided to allow code sharing between X86 and other architectures. Since
> + * these wrappers directly invoke the EFI function pointer, the function
> + * pointer type must be properly defined, which is not the case for X86. One
> + * advantage of this is it allows for type checking of arguments, which is not
> + * possible with the X86 wrappers.
> + */
> +#define efi_call_phys0(f)                      f()
> +#define efi_call_phys1(f, a1)                  f(a1)
> +#define efi_call_phys2(f, a1, a2)              f(a1, a2)
> +#define efi_call_phys3(f, a1, a2, a3)          f(a1, a2, a3)
> +#define efi_call_phys4(f, a1, a2, a3, a4)      f(a1, a2, a3, a4)
> +#define efi_call_phys5(f, a1, a2, a3, a4, a5)  f(a1, a2, a3, a4, a5)
> +
> +/*
> + * AArch64 requires the DTB to be 8-byte aligned in the first 512MiB from
> + * start of kernel and may not cross a 2MiB boundary. We set alignment to
> + * 2MiB so we know it won't cross a 2MiB boundary.
> + */
> +#define EFI_FDT_ALIGN  SZ_2M   /* used by allocate_new_fdt_and_exit_boot() */
> +#define MAX_FDT_OFFSET SZ_512M
> +
> +/* Include shared EFI stub code */
> +#include "../../../drivers/firmware/efi/efi-stub-helper.c"

It looks like this is done by x86 as well.

> +#include "../../../drivers/firmware/efi/fdt.c"
> +#include "../../../drivers/firmware/efi/arm-stub.c"

But why do we need to create more stuff like this? Is it because on arm
we need this as part of the decompressor (which would be a good enough
argument)?

-- 
Catalin

[PATCH v2 13/15] arm64: add EFI runtime services

[Catalin Marinas]

On Thu, Mar 13, 2014 at 10:47:06PM +0000, Leif Lindholm wrote:
> --- /dev/null
> +++ b/arch/arm64/kernel/efi.c
[...]
> +/*
> + * Called from setup_arch with interrupts disabled.
> + */
> +void __init efi_enter_virtual_mode(void)
[...]
> --- a/init/main.c
> +++ b/init/main.c
> @@ -902,6 +902,10 @@ static noinline void __init kernel_init_freeable(void)
>         smp_prepare_cpus(setup_max_cpus);
> 
>         do_pre_smp_initcalls();
> +
> +       if (IS_ENABLED(CONFIG_ARM64) && efi_enabled(EFI_BOOT))
> +               efi_enter_virtual_mode();

The comment for the efi_enter_virtual_mode() function says "called from
setup_arch with interrupts disabled". None of these are true for the
call above (and I would really prefer an arch call than this arm64
conditional call in init/main.c.

-- 
Catalin

[PATCH v2 13/15] arm64: add EFI runtime services

[Mark Salter]

On Tue, 2014-03-18 at 12:34 +0000, Catalin Marinas wrote:
> On Thu, Mar 13, 2014 at 10:47:06PM +0000, Leif Lindholm wrote:
> > --- /dev/null
> > +++ b/arch/arm64/kernel/efi.c
> [...]
> > +/*
> > + * Called from setup_arch with interrupts disabled.
> > + */
> > +void __init efi_enter_virtual_mode(void)
> [...]
> > --- a/init/main.c
> > +++ b/init/main.c
> > @@ -902,6 +902,10 @@ static noinline void __init kernel_init_freeable(void)
> >         smp_prepare_cpus(setup_max_cpus);
> > 
> >         do_pre_smp_initcalls();
> > +
> > +       if (IS_ENABLED(CONFIG_ARM64) && efi_enabled(EFI_BOOT))
> > +               efi_enter_virtual_mode();
> 
> The comment for the efi_enter_virtual_mode() function says "called from
> setup_arch with interrupts disabled". None of these are true for the
> call above (and I would really prefer an arch call than this arm64
> conditional call in init/main.c.
> 

Right, the call changed to later in boot but the comment didn't. Calling
from setup_arch is too early. But an early_initcall would work and would
get rid of the ugly CONFIG_ARM64 test.

[PATCH v2 11/15] arm64: add EFI stub

[Mark Salter]

On Tue, 2014-03-18 at 12:09 +0000, Catalin Marinas wrote:
> On Thu, Mar 13, 2014 at 10:47:04PM +0000, Leif Lindholm wrote:
> > --- /dev/null
> > +++ b/arch/arm64/kernel/efi-entry.S
> > @@ -0,0 +1,93 @@
> > +/*
> > + * EFI entry point.
> > + *
> > + * Copyright (C) 2013 Red Hat, Inc.
> > + * Author: Mark Salter <msalter@redhat.com>
> > + *
> > + * This program is free software; you can redistribute it and/or modify
> > + * it under the terms of the GNU General Public License version 2 as
> > + * published by the Free Software Foundation.
> > + *
> > + */
> > +#include <linux/linkage.h>
> > +#include <linux/init.h>
> > +
> > +#include <asm/assembler.h>
> > +
> > +#define EFI_LOAD_ERROR 0x8000000000000001
> > +
> > +       __INIT
> > +
> > +       /*
> > +        * We arrive here from the EFI boot manager with:
> > +        *
> > +        *    * MMU on with identity-mapped RAM.
> > +        *    * Icache and Dcache on
> > +        *
> > +        * We will most likely be running from some place other than where
> > +        * we want to be. The kernel image wants to be placed at TEXT_OFFSET
> > +        * from start of RAM.
> > +        */
> > +ENTRY(efi_stub_entry)
> > +       stp     x29, x30, [sp, #-32]!
> > +
> > +       /*
> > +        * Call efi_entry to do the real work.
> > +        * x0 and x1 are already set up by firmware. Current runtime
> > +        * address of image is calculated and passed via *image_addr.
> > +        *
> > +        * unsigned long efi_entry(void *handle,
> > +        *                         efi_system_table_t *sys_table,
> > +        *                         unsigned long *image_addr) ;
> > +        */
> > +       adrp    x8, _text
> > +       add     x8, x8, #:lo12:_text
> > +       add     x2, sp, 16
> > +       str     x8, [x2]
> > +       bl      efi_entry
> > +       cmn     x0, #1
> > +       b.eq    efi_load_fail
> > +
> > +       /*
> > +        * efi_entry() will have relocated the kernel image if necessary
> > +        * and we return here with device tree address in x0 and the kernel
> > +        * entry point stored at *image_addr. Save those values in registers
> > +        * which are preserved by __flush_dcache_all.
> > +        */
> > +       ldr     x1, [sp, #16]
> > +       mov     x20, x0
> > +       mov     x21, x1
> > +
> > +       /* Turn off Dcache and MMU */
> > +       mrs     x0, CurrentEL
> > +       cmp     x0, #PSR_MODE_EL2t
> > +       ccmp    x0, #PSR_MODE_EL2h, #0x4, ne
> > +       b.ne    1f
> > +       mrs     x0, sctlr_el2
> > +       bic     x0, x0, #1 << 0 // clear SCTLR.M
> > +       bic     x0, x0, #1 << 2 // clear SCTLR.C
> > +       msr     sctlr_el2, x0
> > +       isb
> > +       b       2f
> > +1:
> > +       mrs     x0, sctlr_el1
> > +       bic     x0, x0, #1 << 0 // clear SCTLR.M
> > +       bic     x0, x0, #1 << 2 // clear SCTLR.C
> > +       msr     sctlr_el1, x0
> > +       isb
> > +2:
> > +       bl      __flush_dcache_all
> 
> In linux-next I'm pushing a patch which no longer exports the
> __flush_dcache_all function. The reason is that it doesn't really work
> if you have a (not fully transparent) external cache like on the Applied
> Micro boards. There other issues when running as a guest as well.
> 
> If you know exactly what needs to be flushed here, can you use a range
> (MVA) operation?

This is just before the EFI stub jumps to kernel proper. The only things
in the dcache would be from identity mapped references to RAM used by
UEFI. The booting.txt doc says dcache should be off and invalidated. I
am just wanting to comply with that. The code here doesn't really know
the extent of DRAM to flush by address.

> 
> > diff --git a/arch/arm64/kernel/efi-stub.c b/arch/arm64/kernel/efi-stub.c
> > new file mode 100644
> > index 0000000..bf30913
> > --- /dev/null
> > +++ b/arch/arm64/kernel/efi-stub.c
> > @@ -0,0 +1,83 @@
> > +/*
> > + * linux/arch/arm/boot/compressed/efi-stub.c
> 
> Nitpick: arch/arm64/... But we don't really need to write the file name
> here, I use a smart editor that tells me which file I'm viewing ;).
> Better write a one-line summary of what this file is about.

Okay. Leftover cruft from copy of arm stub.

> 
> > + *
> > + * Copyright (C) 2013, 2014 Linaro Ltd;  <roy.franz@linaro.org>
> > + *
> > + * This file implements the EFI boot stub for the arm64 kernel.
> > + * Adapted from ARM version by Mark Salter <msalter@redhat.com>
> > + *
> > + * This program is free software; you can redistribute it and/or modify
> > + * it under the terms of the GNU General Public License version 2 as
> > + * published by the Free Software Foundation.
> > + *
> > + */
> > +#include <linux/efi.h>
> > +#include <linux/libfdt.h>
> > +#include <asm/sections.h>
> > +#include <generated/compile.h>
> > +#include <generated/utsrelease.h>
> > +
> > +/*
> > + * EFI function call wrappers. These are not required for arm/arm64, but
> > + * wrappers are required for X86 to convert between ABIs. These wrappers are
> > + * provided to allow code sharing between X86 and other architectures. Since
> > + * these wrappers directly invoke the EFI function pointer, the function
> > + * pointer type must be properly defined, which is not the case for X86. One
> > + * advantage of this is it allows for type checking of arguments, which is not
> > + * possible with the X86 wrappers.
> > + */
> > +#define efi_call_phys0(f)                      f()
> > +#define efi_call_phys1(f, a1)                  f(a1)
> > +#define efi_call_phys2(f, a1, a2)              f(a1, a2)
> > +#define efi_call_phys3(f, a1, a2, a3)          f(a1, a2, a3)
> > +#define efi_call_phys4(f, a1, a2, a3, a4)      f(a1, a2, a3, a4)
> > +#define efi_call_phys5(f, a1, a2, a3, a4, a5)  f(a1, a2, a3, a4, a5)
> > +
> > +/*
> > + * AArch64 requires the DTB to be 8-byte aligned in the first 512MiB from
> > + * start of kernel and may not cross a 2MiB boundary. We set alignment to
> > + * 2MiB so we know it won't cross a 2MiB boundary.
> > + */
> > +#define EFI_FDT_ALIGN  SZ_2M   /* used by allocate_new_fdt_and_exit_boot() */
> > +#define MAX_FDT_OFFSET SZ_512M
> > +
> > +/* Include shared EFI stub code */
> > +#include "../../../drivers/firmware/efi/efi-stub-helper.c"
> 
> It looks like this is done by x86 as well.
> 
> > +#include "../../../drivers/firmware/efi/fdt.c"
> > +#include "../../../drivers/firmware/efi/arm-stub.c"
> 
> But why do we need to create more stuff like this? Is it because on arm
> we need this as part of the decompressor (which would be a good enough
> argument)?

That is the reason exactly. I wish there were a better way.

[PATCH v2 13/15] arm64: add EFI runtime services

[Catalin Marinas]

On Tue, Mar 18, 2014 at 02:16:49PM +0000, Mark Salter wrote:
> On Tue, 2014-03-18 at 12:34 +0000, Catalin Marinas wrote:
> > On Thu, Mar 13, 2014 at 10:47:06PM +0000, Leif Lindholm wrote:
> > > --- /dev/null
> > > +++ b/arch/arm64/kernel/efi.c
> > [...]
> > > +/*
> > > + * Called from setup_arch with interrupts disabled.
> > > + */
> > > +void __init efi_enter_virtual_mode(void)
> > [...]
> > > --- a/init/main.c
> > > +++ b/init/main.c
> > > @@ -902,6 +902,10 @@ static noinline void __init kernel_init_freeable(void)
> > >         smp_prepare_cpus(setup_max_cpus);
> > > 
> > >         do_pre_smp_initcalls();
> > > +
> > > +       if (IS_ENABLED(CONFIG_ARM64) && efi_enabled(EFI_BOOT))
> > > +               efi_enter_virtual_mode();
> > 
> > The comment for the efi_enter_virtual_mode() function says "called from
> > setup_arch with interrupts disabled". None of these are true for the
> > call above (and I would really prefer an arch call than this arm64
> > conditional call in init/main.c.
> > 
> 
> Right, the call changed to later in boot but the comment didn't. Calling
> from setup_arch is too early. But an early_initcall would work and would
> get rid of the ugly CONFIG_ARM64 test.

I agree, early_initcall() is better.

-- 
Catalin

[PATCH v2 11/15] arm64: add EFI stub

[Catalin Marinas]

On Tue, Mar 18, 2014 at 02:40:29PM +0000, Mark Salter wrote:
> On Tue, 2014-03-18 at 12:09 +0000, Catalin Marinas wrote:
> > On Thu, Mar 13, 2014 at 10:47:04PM +0000, Leif Lindholm wrote:
> > > --- /dev/null
> > > +++ b/arch/arm64/kernel/efi-entry.S
> > > @@ -0,0 +1,93 @@
> > > +/*
> > > + * EFI entry point.
> > > + *
> > > + * Copyright (C) 2013 Red Hat, Inc.
> > > + * Author: Mark Salter <msalter@redhat.com>
> > > + *
> > > + * This program is free software; you can redistribute it and/or modify
> > > + * it under the terms of the GNU General Public License version 2 as
> > > + * published by the Free Software Foundation.
> > > + *
> > > + */
> > > +#include <linux/linkage.h>
> > > +#include <linux/init.h>
> > > +
> > > +#include <asm/assembler.h>
> > > +
> > > +#define EFI_LOAD_ERROR 0x8000000000000001
> > > +
> > > +       __INIT
> > > +
> > > +       /*
> > > +        * We arrive here from the EFI boot manager with:
> > > +        *
> > > +        *    * MMU on with identity-mapped RAM.
> > > +        *    * Icache and Dcache on
> > > +        *
> > > +        * We will most likely be running from some place other than where
> > > +        * we want to be. The kernel image wants to be placed at TEXT_OFFSET
> > > +        * from start of RAM.
> > > +        */
> > > +ENTRY(efi_stub_entry)
> > > +       stp     x29, x30, [sp, #-32]!
> > > +
> > > +       /*
> > > +        * Call efi_entry to do the real work.
> > > +        * x0 and x1 are already set up by firmware. Current runtime
> > > +        * address of image is calculated and passed via *image_addr.
> > > +        *
> > > +        * unsigned long efi_entry(void *handle,
> > > +        *                         efi_system_table_t *sys_table,
> > > +        *                         unsigned long *image_addr) ;
> > > +        */
> > > +       adrp    x8, _text
> > > +       add     x8, x8, #:lo12:_text
> > > +       add     x2, sp, 16
> > > +       str     x8, [x2]
> > > +       bl      efi_entry
> > > +       cmn     x0, #1
> > > +       b.eq    efi_load_fail
> > > +
> > > +       /*
> > > +        * efi_entry() will have relocated the kernel image if necessary
> > > +        * and we return here with device tree address in x0 and the kernel
> > > +        * entry point stored at *image_addr. Save those values in registers
> > > +        * which are preserved by __flush_dcache_all.
> > > +        */
> > > +       ldr     x1, [sp, #16]
> > > +       mov     x20, x0
> > > +       mov     x21, x1
> > > +
> > > +       /* Turn off Dcache and MMU */
> > > +       mrs     x0, CurrentEL
> > > +       cmp     x0, #PSR_MODE_EL2t
> > > +       ccmp    x0, #PSR_MODE_EL2h, #0x4, ne
> > > +       b.ne    1f
> > > +       mrs     x0, sctlr_el2
> > > +       bic     x0, x0, #1 << 0 // clear SCTLR.M
> > > +       bic     x0, x0, #1 << 2 // clear SCTLR.C
> > > +       msr     sctlr_el2, x0
> > > +       isb
> > > +       b       2f
> > > +1:
> > > +       mrs     x0, sctlr_el1
> > > +       bic     x0, x0, #1 << 0 // clear SCTLR.M
> > > +       bic     x0, x0, #1 << 2 // clear SCTLR.C
> > > +       msr     sctlr_el1, x0
> > > +       isb
> > > +2:
> > > +       bl      __flush_dcache_all
> > 
> > In linux-next I'm pushing a patch which no longer exports the
> > __flush_dcache_all function. The reason is that it doesn't really work
> > if you have a (not fully transparent) external cache like on the Applied
> > Micro boards. There other issues when running as a guest as well.
> > 
> > If you know exactly what needs to be flushed here, can you use a range
> > (MVA) operation?
> 
> This is just before the EFI stub jumps to kernel proper. The only things
> in the dcache would be from identity mapped references to RAM used by
> UEFI. The booting.txt doc says dcache should be off and invalidated. I
> am just wanting to comply with that. The code here doesn't really know
> the extent of DRAM to flush by address.

Does UEFI do anything with the caches before invoking the EFI_STUB code?
I guess it doesn't since that's just another application for it. Can
UEFI flush the caches via exit boot? When is this called?

As I said, we have a real problem here since the EFI_STUB call does not
have information about the SoC to be able to flush all the caches. But
UEFI should know more about the hardware.

If UEFI doesn't handle the caches, the only thing left to EFI_STUB is to
flush by MVA. We don't need to flush the whole DRAM (and I would even
recommend it) but at least the relevant kernel code/data touched with
the MMU disabled.

-- 
Catalin

[PATCH v2 11/15] arm64: add EFI stub

[Mark Salter]

On Tue, 2014-03-18 at 18:28 +0000, Catalin Marinas wrote:
> On Tue, Mar 18, 2014 at 02:40:29PM +0000, Mark Salter wrote:
> > On Tue, 2014-03-18 at 12:09 +0000, Catalin Marinas wrote:
> > > On Thu, Mar 13, 2014 at 10:47:04PM +0000, Leif Lindholm wrote:
> > > > --- /dev/null
> > > > +++ b/arch/arm64/kernel/efi-entry.S
> > > > @@ -0,0 +1,93 @@
> > > > +/*
> > > > + * EFI entry point.
> > > > + *
> > > > + * Copyright (C) 2013 Red Hat, Inc.
> > > > + * Author: Mark Salter <msalter@redhat.com>
> > > > + *
> > > > + * This program is free software; you can redistribute it and/or modify
> > > > + * it under the terms of the GNU General Public License version 2 as
> > > > + * published by the Free Software Foundation.
> > > > + *
> > > > + */
> > > > +#include <linux/linkage.h>
> > > > +#include <linux/init.h>
> > > > +
> > > > +#include <asm/assembler.h>
> > > > +
> > > > +#define EFI_LOAD_ERROR 0x8000000000000001
> > > > +
> > > > +       __INIT
> > > > +
> > > > +       /*
> > > > +        * We arrive here from the EFI boot manager with:
> > > > +        *
> > > > +        *    * MMU on with identity-mapped RAM.
> > > > +        *    * Icache and Dcache on
> > > > +        *
> > > > +        * We will most likely be running from some place other than where
> > > > +        * we want to be. The kernel image wants to be placed at TEXT_OFFSET
> > > > +        * from start of RAM.
> > > > +        */
> > > > +ENTRY(efi_stub_entry)
> > > > +       stp     x29, x30, [sp, #-32]!
> > > > +
> > > > +       /*
> > > > +        * Call efi_entry to do the real work.
> > > > +        * x0 and x1 are already set up by firmware. Current runtime
> > > > +        * address of image is calculated and passed via *image_addr.
> > > > +        *
> > > > +        * unsigned long efi_entry(void *handle,
> > > > +        *                         efi_system_table_t *sys_table,
> > > > +        *                         unsigned long *image_addr) ;
> > > > +        */
> > > > +       adrp    x8, _text
> > > > +       add     x8, x8, #:lo12:_text
> > > > +       add     x2, sp, 16
> > > > +       str     x8, [x2]
> > > > +       bl      efi_entry
> > > > +       cmn     x0, #1
> > > > +       b.eq    efi_load_fail
> > > > +
> > > > +       /*
> > > > +        * efi_entry() will have relocated the kernel image if necessary
> > > > +        * and we return here with device tree address in x0 and the kernel
> > > > +        * entry point stored at *image_addr. Save those values in registers
> > > > +        * which are preserved by __flush_dcache_all.
> > > > +        */
> > > > +       ldr     x1, [sp, #16]
> > > > +       mov     x20, x0
> > > > +       mov     x21, x1
> > > > +
> > > > +       /* Turn off Dcache and MMU */
> > > > +       mrs     x0, CurrentEL
> > > > +       cmp     x0, #PSR_MODE_EL2t
> > > > +       ccmp    x0, #PSR_MODE_EL2h, #0x4, ne
> > > > +       b.ne    1f
> > > > +       mrs     x0, sctlr_el2
> > > > +       bic     x0, x0, #1 << 0 // clear SCTLR.M
> > > > +       bic     x0, x0, #1 << 2 // clear SCTLR.C
> > > > +       msr     sctlr_el2, x0
> > > > +       isb
> > > > +       b       2f
> > > > +1:
> > > > +       mrs     x0, sctlr_el1
> > > > +       bic     x0, x0, #1 << 0 // clear SCTLR.M
> > > > +       bic     x0, x0, #1 << 2 // clear SCTLR.C
> > > > +       msr     sctlr_el1, x0
> > > > +       isb
> > > > +2:
> > > > +       bl      __flush_dcache_all
> > > 
> > > In linux-next I'm pushing a patch which no longer exports the
> > > __flush_dcache_all function. The reason is that it doesn't really work
> > > if you have a (not fully transparent) external cache like on the Applied
> > > Micro boards. There other issues when running as a guest as well.
> > > 
> > > If you know exactly what needs to be flushed here, can you use a range
> > > (MVA) operation?
> > 
> > This is just before the EFI stub jumps to kernel proper. The only things
> > in the dcache would be from identity mapped references to RAM used by
> > UEFI. The booting.txt doc says dcache should be off and invalidated. I
> > am just wanting to comply with that. The code here doesn't really know
> > the extent of DRAM to flush by address.
> 
> Does UEFI do anything with the caches before invoking the EFI_STUB code?
> I guess it doesn't since that's just another application for it. Can
> UEFI flush the caches via exit boot? When is this called?
> 
> As I said, we have a real problem here since the EFI_STUB call does not
> have information about the SoC to be able to flush all the caches. But
> UEFI should know more about the hardware.
> 
> If UEFI doesn't handle the caches, the only thing left to EFI_STUB is to
> flush by MVA. We don't need to flush the whole DRAM (and I would even
> recommend it) but at least the relevant kernel code/data touched with
> the MMU disabled.
> 

So, it goes like this:

  1) UEFI calls stub with MMU/Caches on. Stub/kernel can be anywhere.
  2) Stub runs and relocates kernel to the desired runtime location
     but continues to execute from wherever UEFI loaded it until just
     after ExitBootServices().
  3) After ExitBootServices, efi_entry() returns relocated entry point
     for kernel to efi_stub_entry() in efi-entry.S where the Dcache and
     MMU are turned off, the __flush_dcache_all is called, then the
     code jumps to the kernel proper entry point.

It isn't clear to me if UEFI does cache flushing at ExitBootServices
time, but even so, at least stack use will get cached between then and
the kernel entry point. The stub could conceivably get its hands on the
EFI memmap and invalidate dcache using address ranges from UEFI memory
descriptors so maybe that is the way we should do it.

[PATCH v2 11/15] arm64: add EFI stub

[Roy Franz]

On Tue, Mar 18, 2014 at 2:40 PM, Mark Salter <msalter@redhat.com> wrote:
> On Tue, 2014-03-18 at 18:28 +0000, Catalin Marinas wrote:
>> On Tue, Mar 18, 2014 at 02:40:29PM +0000, Mark Salter wrote:
>> > On Tue, 2014-03-18 at 12:09 +0000, Catalin Marinas wrote:
>> > > On Thu, Mar 13, 2014 at 10:47:04PM +0000, Leif Lindholm wrote:
>> > > > --- /dev/null
>> > > > +++ b/arch/arm64/kernel/efi-entry.S
>> > > > @@ -0,0 +1,93 @@
>> > > > +/*
>> > > > + * EFI entry point.
>> > > > + *
>> > > > + * Copyright (C) 2013 Red Hat, Inc.
>> > > > + * Author: Mark Salter <msalter@redhat.com>
>> > > > + *
>> > > > + * This program is free software; you can redistribute it and/or modify
>> > > > + * it under the terms of the GNU General Public License version 2 as
>> > > > + * published by the Free Software Foundation.
>> > > > + *
>> > > > + */
>> > > > +#include <linux/linkage.h>
>> > > > +#include <linux/init.h>
>> > > > +
>> > > > +#include <asm/assembler.h>
>> > > > +
>> > > > +#define EFI_LOAD_ERROR 0x8000000000000001
>> > > > +
>> > > > +       __INIT
>> > > > +
>> > > > +       /*
>> > > > +        * We arrive here from the EFI boot manager with:
>> > > > +        *
>> > > > +        *    * MMU on with identity-mapped RAM.
>> > > > +        *    * Icache and Dcache on
>> > > > +        *
>> > > > +        * We will most likely be running from some place other than where
>> > > > +        * we want to be. The kernel image wants to be placed at TEXT_OFFSET
>> > > > +        * from start of RAM.
>> > > > +        */
>> > > > +ENTRY(efi_stub_entry)
>> > > > +       stp     x29, x30, [sp, #-32]!
>> > > > +
>> > > > +       /*
>> > > > +        * Call efi_entry to do the real work.
>> > > > +        * x0 and x1 are already set up by firmware. Current runtime
>> > > > +        * address of image is calculated and passed via *image_addr.
>> > > > +        *
>> > > > +        * unsigned long efi_entry(void *handle,
>> > > > +        *                         efi_system_table_t *sys_table,
>> > > > +        *                         unsigned long *image_addr) ;
>> > > > +        */
>> > > > +       adrp    x8, _text
>> > > > +       add     x8, x8, #:lo12:_text
>> > > > +       add     x2, sp, 16
>> > > > +       str     x8, [x2]
>> > > > +       bl      efi_entry
>> > > > +       cmn     x0, #1
>> > > > +       b.eq    efi_load_fail
>> > > > +
>> > > > +       /*
>> > > > +        * efi_entry() will have relocated the kernel image if necessary
>> > > > +        * and we return here with device tree address in x0 and the kernel
>> > > > +        * entry point stored at *image_addr. Save those values in registers
>> > > > +        * which are preserved by __flush_dcache_all.
>> > > > +        */
>> > > > +       ldr     x1, [sp, #16]
>> > > > +       mov     x20, x0
>> > > > +       mov     x21, x1
>> > > > +
>> > > > +       /* Turn off Dcache and MMU */
>> > > > +       mrs     x0, CurrentEL
>> > > > +       cmp     x0, #PSR_MODE_EL2t
>> > > > +       ccmp    x0, #PSR_MODE_EL2h, #0x4, ne
>> > > > +       b.ne    1f
>> > > > +       mrs     x0, sctlr_el2
>> > > > +       bic     x0, x0, #1 << 0 // clear SCTLR.M
>> > > > +       bic     x0, x0, #1 << 2 // clear SCTLR.C
>> > > > +       msr     sctlr_el2, x0
>> > > > +       isb
>> > > > +       b       2f
>> > > > +1:
>> > > > +       mrs     x0, sctlr_el1
>> > > > +       bic     x0, x0, #1 << 0 // clear SCTLR.M
>> > > > +       bic     x0, x0, #1 << 2 // clear SCTLR.C
>> > > > +       msr     sctlr_el1, x0
>> > > > +       isb
>> > > > +2:
>> > > > +       bl      __flush_dcache_all
>> > >
>> > > In linux-next I'm pushing a patch which no longer exports the
>> > > __flush_dcache_all function. The reason is that it doesn't really work
>> > > if you have a (not fully transparent) external cache like on the Applied
>> > > Micro boards. There other issues when running as a guest as well.
>> > >
>> > > If you know exactly what needs to be flushed here, can you use a range
>> > > (MVA) operation?
>> >
>> > This is just before the EFI stub jumps to kernel proper. The only things
>> > in the dcache would be from identity mapped references to RAM used by
>> > UEFI. The booting.txt doc says dcache should be off and invalidated. I
>> > am just wanting to comply with that. The code here doesn't really know
>> > the extent of DRAM to flush by address.
>>
>> Does UEFI do anything with the caches before invoking the EFI_STUB code?
>> I guess it doesn't since that's just another application for it. Can
>> UEFI flush the caches via exit boot? When is this called?
>>
>> As I said, we have a real problem here since the EFI_STUB call does not
>> have information about the SoC to be able to flush all the caches. But
>> UEFI should know more about the hardware.
>>
>> If UEFI doesn't handle the caches, the only thing left to EFI_STUB is to
>> flush by MVA. We don't need to flush the whole DRAM (and I would even
>> recommend it) but at least the relevant kernel code/data touched with
>> the MMU disabled.
>>
>
> So, it goes like this:
>
>   1) UEFI calls stub with MMU/Caches on. Stub/kernel can be anywhere.
>   2) Stub runs and relocates kernel to the desired runtime location
>      but continues to execute from wherever UEFI loaded it until just
>      after ExitBootServices().
>   3) After ExitBootServices, efi_entry() returns relocated entry point
>      for kernel to efi_stub_entry() in efi-entry.S where the Dcache and
>      MMU are turned off, the __flush_dcache_all is called, then the
>      code jumps to the kernel proper entry point.
>
> It isn't clear to me if UEFI does cache flushing at ExitBootServices
> time, but even so, at least stack use will get cached between then and
> the kernel entry point. The stub could conceivably get its hands on the
> EFI memmap and invalidate dcache using address ranges from UEFI memory
> descriptors so maybe that is the way we should do it.
>

I looked at the UEFI spec and there is no mention of cache flushing in
ExitBootServices(),
so it seems it is up to the OS to do any cache management.

Roy

[PATCH v2 11/15] arm64: add EFI stub

[Jason Gunthorpe]

On Tue, Mar 18, 2014 at 02:48:30PM -0700, Roy Franz wrote:

> > It isn't clear to me if UEFI does cache flushing at ExitBootServices
> > time, but even so, at least stack use will get cached between then and
> > the kernel entry point. The stub could conceivably get its hands on the
> > EFI memmap and invalidate dcache using address ranges from UEFI memory
> > descriptors so maybe that is the way we should do it.

> I looked at the UEFI spec and there is no mention of cache flushing
> in ExitBootServices(), so it seems it is up to the OS to do any
> cache management.

Something to think about: On mvebu we recently confirmed a situation
where turning off the L1 cache is not sufficient for booting. The L2
cache must also be completely off and disabled prior to jumping in to
the kernel.

The issue is the decompressor turns the L1 cache back on, and if the
L2 is also enabled at this point then it gets filled with
decompression data. Things go wrong from here because after
decompression the L1 dcache is switched off, but the L2 isn't
flush-invalidated.

So now the L2 holds writeback data and uncached reads return garbage,
and/or the L2 misses the uncached writes (eg relocation fixup) and
becomes inconsistent with memory. Either case gives a black screen
crash at boot.

Fundementally if the L2 doesn't monitor uncached read/writes then it
will cause a big problem.

Thus, if the UEFI calls the sub with the caches on, and the stub
doesn't know enough to turn off the L2 then you might not be able to
turn the dcache off at all. :(

On ARM64 at least the L1 cache ops are standardized so maybe ARM64
could keep the mmu+caches enabled during boot and do the L1
d-flush/i-inval required for instruction coherency?

Jason

[PATCH v2 11/15] arm64: add EFI stub

[Catalin Marinas]

On Tue, Mar 18, 2014 at 10:21:05PM +0000, Jason Gunthorpe wrote:
> On Tue, Mar 18, 2014 at 02:48:30PM -0700, Roy Franz wrote:
> 
> > > It isn't clear to me if UEFI does cache flushing at ExitBootServices
> > > time, but even so, at least stack use will get cached between then and
> > > the kernel entry point. The stub could conceivably get its hands on the
> > > EFI memmap and invalidate dcache using address ranges from UEFI memory
> > > descriptors so maybe that is the way we should do it.
> 
> > I looked at the UEFI spec and there is no mention of cache flushing
> > in ExitBootServices(), so it seems it is up to the OS to do any
> > cache management.
> 
> Something to think about: On mvebu we recently confirmed a situation
> where turning off the L1 cache is not sufficient for booting. The L2
> cache must also be completely off and disabled prior to jumping in to
> the kernel.
> 
> The issue is the decompressor turns the L1 cache back on, and if the
> L2 is also enabled at this point then it gets filled with
> decompression data. Things go wrong from here because after
> decompression the L1 dcache is switched off, but the L2 isn't
> flush-invalidated.
> 
> So now the L2 holds writeback data and uncached reads return garbage,
> and/or the L2 misses the uncached writes (eg relocation fixup) and
> becomes inconsistent with memory. Either case gives a black screen
> crash at boot.
> 
> Fundementally if the L2 doesn't monitor uncached read/writes then it
> will cause a big problem.

Yes, that's a problem on (ARMv7) SoCs booting in non-secure mode with
external L2 already enabled. For ARMv8 at least, the SoCs I've seen with
external caches detect the data cache by MVA ops.

> Thus, if the UEFI calls the sub with the caches on, and the stub
> doesn't know enough to turn off the L2 then you might not be able to
> turn the dcache off at all. :(
> 
> On ARM64 at least the L1 cache ops are standardized so maybe ARM64
> could keep the mmu+caches enabled during boot and do the L1
> d-flush/i-inval required for instruction coherency?

We thought about keeping the MMU on from EFI_STUB into the kernel but it
gets messier since UEFI has different MMU settings. So with some sane
external cache, we could get away with flushing a range.

-- 
Catalin

[PATCH v2 11/15] arm64: add EFI stub

[Catalin Marinas]

On Tue, Mar 18, 2014 at 09:40:31PM +0000, Mark Salter wrote:
> On Tue, 2014-03-18 at 18:28 +0000, Catalin Marinas wrote:
> > If UEFI doesn't handle the caches, the only thing left to EFI_STUB is to
> > flush by MVA. We don't need to flush the whole DRAM (and I would even
> > recommend it) but at least the relevant kernel code/data touched with
> > the MMU disabled.
> 
> So, it goes like this:
> 
>   1) UEFI calls stub with MMU/Caches on. Stub/kernel can be anywhere.
>   2) Stub runs and relocates kernel to the desired runtime location
>      but continues to execute from wherever UEFI loaded it until just
>      after ExitBootServices().
>   3) After ExitBootServices, efi_entry() returns relocated entry point
>      for kernel to efi_stub_entry() in efi-entry.S where the Dcache and
>      MMU are turned off, the __flush_dcache_all is called, then the
>      code jumps to the kernel proper entry point.
> 
> It isn't clear to me if UEFI does cache flushing at ExitBootServices
> time, but even so, at least stack use will get cached between then and
> the kernel entry point. The stub could conceivably get its hands on the
> EFI memmap and invalidate dcache using address ranges from UEFI memory
> descriptors so maybe that is the way we should do it.

I think the stub just needs to flush the relocated kernel image, ensure
it is sync with the memory. Additional flushing can be done by the
kernel for bits it writes (like page tables, code patching etc). We can
enter the kernel with the SCTLR.I bit set, so it can allocate in an
unified cache already and D-cache maintenance would be needed anyway.

Another aspect in the booting document is that we require any external
cache to be disabled. I think on ARMv8 with a more transparent external
cache we can relax this (Applied's SoC already does this). Otherwise,
since external cache enabling is usually a secure operation, we would
need some SoC specific code just for this early during boot and I'd like
to avoid it.

-- 
Catalin

[PATCH v2 11/15] arm64: add EFI stub

[Mark Salter]

On Wed, 2014-03-19 at 10:57 +0000, Catalin Marinas wrote:
> On Tue, Mar 18, 2014 at 09:40:31PM +0000, Mark Salter wrote:
> > On Tue, 2014-03-18 at 18:28 +0000, Catalin Marinas wrote:
> > > If UEFI doesn't handle the caches, the only thing left to EFI_STUB is to
> > > flush by MVA. We don't need to flush the whole DRAM (and I would even
> > > recommend it) but at least the relevant kernel code/data touched with
> > > the MMU disabled.
> > 
> > So, it goes like this:
> > 
> >   1) UEFI calls stub with MMU/Caches on. Stub/kernel can be anywhere.
> >   2) Stub runs and relocates kernel to the desired runtime location
> >      but continues to execute from wherever UEFI loaded it until just
> >      after ExitBootServices().
> >   3) After ExitBootServices, efi_entry() returns relocated entry point
> >      for kernel to efi_stub_entry() in efi-entry.S where the Dcache and
> >      MMU are turned off, the __flush_dcache_all is called, then the
> >      code jumps to the kernel proper entry point.
> > 
> > It isn't clear to me if UEFI does cache flushing at ExitBootServices
> > time, but even so, at least stack use will get cached between then and
> > the kernel entry point. The stub could conceivably get its hands on the
> > EFI memmap and invalidate dcache using address ranges from UEFI memory
> > descriptors so maybe that is the way we should do it.
> 
> I think the stub just needs to flush the relocated kernel image, ensure
> it is sync with the memory. Additional flushing can be done by the
> kernel for bits it writes (like page tables, code patching etc). We can
> enter the kernel with the SCTLR.I bit set, so it can allocate in an
> unified cache already and D-cache maintenance would be needed anyway.

How about this?

diff --git a/arch/arm64/kernel/efi-entry.S b/arch/arm64/kernel/efi-entry.S
index 83bfb72..ed480b2 100644
--- a/arch/arm64/kernel/efi-entry.S
+++ b/arch/arm64/kernel/efi-entry.S
@@ -52,11 +52,19 @@ ENTRY(efi_stub_entry)
 	 * efi_entry() will have relocated the kernel image if necessary
 	 * and we return here with device tree address in x0 and the kernel
 	 * entry point stored at *image_addr. Save those values in registers
-	 * which are preserved by __flush_dcache_all.
+	 * which are callee preserved.
 	 */
-	ldr	x1, [sp, #16]
 	mov	x20, x0
-	mov	x21, x1
+	ldr	x0, [sp, #16]
+	mov	x21, x0
+
+	adrp	x1, _text
+	add	x1, x1, #:lo12:_text
+	adrp	x2, _edata
+	add	x2, x2, #:lo12:_edata
+	sub	x1, x2, x1
+
+	bl	__flush_dcache_area
 
 	/* Turn off Dcache and MMU */
 	mrs	x0, CurrentEL
@@ -76,8 +84,6 @@ ENTRY(efi_stub_entry)
 	msr	sctlr_el1, x0
 	isb
 2:
-	bl	__flush_dcache_all
-
 	/* Jump to real entry point */
 	mov	x0, x20
 	mov	x1, xzr

[PATCH v2 11/15] arm64: add EFI stub

[Catalin Marinas]

On Wed, Mar 19, 2014 at 03:13:18PM +0000, Mark Salter wrote:
> On Wed, 2014-03-19 at 10:57 +0000, Catalin Marinas wrote:
> > On Tue, Mar 18, 2014 at 09:40:31PM +0000, Mark Salter wrote:
> > > On Tue, 2014-03-18 at 18:28 +0000, Catalin Marinas wrote:
> > > > If UEFI doesn't handle the caches, the only thing left to EFI_STUB is to
> > > > flush by MVA. We don't need to flush the whole DRAM (and I would even
> > > > recommend it) but at least the relevant kernel code/data touched with
> > > > the MMU disabled.
> > > 
> > > So, it goes like this:
> > > 
> > >   1) UEFI calls stub with MMU/Caches on. Stub/kernel can be anywhere.
> > >   2) Stub runs and relocates kernel to the desired runtime location
> > >      but continues to execute from wherever UEFI loaded it until just
> > >      after ExitBootServices().
> > >   3) After ExitBootServices, efi_entry() returns relocated entry point
> > >      for kernel to efi_stub_entry() in efi-entry.S where the Dcache and
> > >      MMU are turned off, the __flush_dcache_all is called, then the
> > >      code jumps to the kernel proper entry point.
> > > 
> > > It isn't clear to me if UEFI does cache flushing at ExitBootServices
> > > time, but even so, at least stack use will get cached between then and
> > > the kernel entry point. The stub could conceivably get its hands on the
> > > EFI memmap and invalidate dcache using address ranges from UEFI memory
> > > descriptors so maybe that is the way we should do it.
> > 
> > I think the stub just needs to flush the relocated kernel image, ensure
> > it is sync with the memory. Additional flushing can be done by the
> > kernel for bits it writes (like page tables, code patching etc). We can
> > enter the kernel with the SCTLR.I bit set, so it can allocate in an
> > unified cache already and D-cache maintenance would be needed anyway.
> 
> How about this?
> 
> diff --git a/arch/arm64/kernel/efi-entry.S b/arch/arm64/kernel/efi-entry.S
> index 83bfb72..ed480b2 100644
> --- a/arch/arm64/kernel/efi-entry.S
> +++ b/arch/arm64/kernel/efi-entry.S
> @@ -52,11 +52,19 @@ ENTRY(efi_stub_entry)
>  	 * efi_entry() will have relocated the kernel image if necessary
>  	 * and we return here with device tree address in x0 and the kernel
>  	 * entry point stored at *image_addr. Save those values in registers
> -	 * which are preserved by __flush_dcache_all.
> +	 * which are callee preserved.
>  	 */
> -	ldr	x1, [sp, #16]
>  	mov	x20, x0
> -	mov	x21, x1
> +	ldr	x0, [sp, #16]
> +	mov	x21, x0

BTW, some comments to make it easier to review the code later about
what's at [sp, #16].

> +
> +	adrp	x1, _text
> +	add	x1, x1, #:lo12:_text
> +	adrp	x2, _edata
> +	add	x2, x2, #:lo12:_edata
> +	sub	x1, x2, x1
> +
> +	bl	__flush_dcache_area

Looks fine. We also need a "ic ialluis", is any other part of EFI_STUB
invalidating the I-cache?

We need a patch to invalidate the D-cache for areas of memory that the
kernel touches but I'll do this as a separate patch.

-- 
Catalin

[PATCH v2 11/15] arm64: add EFI stub

[Mark Salter]

On Wed, 2014-03-19 at 16:01 +0000, Catalin Marinas wrote:
> On Wed, Mar 19, 2014 at 03:13:18PM +0000, Mark Salter wrote:
> > On Wed, 2014-03-19 at 10:57 +0000, Catalin Marinas wrote:
> > > On Tue, Mar 18, 2014 at 09:40:31PM +0000, Mark Salter wrote:
> > > > On Tue, 2014-03-18 at 18:28 +0000, Catalin Marinas wrote:
> > > > > If UEFI doesn't handle the caches, the only thing left to EFI_STUB is to
> > > > > flush by MVA. We don't need to flush the whole DRAM (and I would even
> > > > > recommend it) but at least the relevant kernel code/data touched with
> > > > > the MMU disabled.
> > > > 
> > > > So, it goes like this:
> > > > 
> > > >   1) UEFI calls stub with MMU/Caches on. Stub/kernel can be anywhere.
> > > >   2) Stub runs and relocates kernel to the desired runtime location
> > > >      but continues to execute from wherever UEFI loaded it until just
> > > >      after ExitBootServices().
> > > >   3) After ExitBootServices, efi_entry() returns relocated entry point
> > > >      for kernel to efi_stub_entry() in efi-entry.S where the Dcache and
> > > >      MMU are turned off, the __flush_dcache_all is called, then the
> > > >      code jumps to the kernel proper entry point.
> > > > 
> > > > It isn't clear to me if UEFI does cache flushing at ExitBootServices
> > > > time, but even so, at least stack use will get cached between then and
> > > > the kernel entry point. The stub could conceivably get its hands on the
> > > > EFI memmap and invalidate dcache using address ranges from UEFI memory
> > > > descriptors so maybe that is the way we should do it.
> > > 
> > > I think the stub just needs to flush the relocated kernel image, ensure
> > > it is sync with the memory. Additional flushing can be done by the
> > > kernel for bits it writes (like page tables, code patching etc). We can
> > > enter the kernel with the SCTLR.I bit set, so it can allocate in an
> > > unified cache already and D-cache maintenance would be needed anyway.
> > 
> > How about this?
> > 
> > diff --git a/arch/arm64/kernel/efi-entry.S b/arch/arm64/kernel/efi-entry.S
> > index 83bfb72..ed480b2 100644
> > --- a/arch/arm64/kernel/efi-entry.S
> > +++ b/arch/arm64/kernel/efi-entry.S
> > @@ -52,11 +52,19 @@ ENTRY(efi_stub_entry)
> >  	 * efi_entry() will have relocated the kernel image if necessary
> >  	 * and we return here with device tree address in x0 and the kernel
> >  	 * entry point stored at *image_addr. Save those values in registers
> > -	 * which are preserved by __flush_dcache_all.
> > +	 * which are callee preserved.
> >  	 */
> > -	ldr	x1, [sp, #16]
> >  	mov	x20, x0
> > -	mov	x21, x1
> > +	ldr	x0, [sp, #16]
> > +	mov	x21, x0
> 
> BTW, some comments to make it easier to review the code later about
> what's at [sp, #16].

Okay.

> 
> > +
> > +	adrp	x1, _text
> > +	add	x1, x1, #:lo12:_text
> > +	adrp	x2, _edata
> > +	add	x2, x2, #:lo12:_edata
> > +	sub	x1, x2, x1
> > +
> > +	bl	__flush_dcache_area
> 
> Looks fine. We also need a "ic ialluis", is any other part of EFI_STUB
> invalidating the I-cache?

No. I will add it.