[linux-2.6-block.git] / arch / x86 / kernel / kexec-bzimage64.c

/*
 * Kexec bzImage loader
 *
 * Copyright (C) 2014 Red Hat Inc.
 * Authors:
 *      Vivek Goyal <vgoyal@redhat.com>
 *
 * This source code is licensed under the GNU General Public License,
 * Version 2.  See the file COPYING for more details.
 */

#define pr_fmt(fmt)	"kexec-bzImage64: " fmt

#include <linux/string.h>
#include <linux/printk.h>
#include <linux/errno.h>
#include <linux/slab.h>
#include <linux/kexec.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/efi.h>
#include <linux/verify_pefile.h>
#include <keys/system_keyring.h>

#include <asm/bootparam.h>
#include <asm/setup.h>
#include <asm/crash.h>
#include <asm/efi.h>
#include <asm/kexec-bzimage64.h>

#define MAX_ELFCOREHDR_STR_LEN	30	/* elfcorehdr=0x<64bit-value> */

/*
 * Defines lowest physical address for various segments. Not sure where
 * exactly these limits came from. Current bzimage64 loader in kexec-tools
 * uses these so I am retaining it. It can be changed over time as we gain
 * more insight.
 */
#define MIN_PURGATORY_ADDR	0x3000
#define MIN_BOOTPARAM_ADDR	0x3000
#define MIN_KERNEL_LOAD_ADDR	0x100000
#define MIN_INITRD_LOAD_ADDR	0x1000000

/*
 * This is a place holder for all boot loader specific data structure which
 * gets allocated in one call but gets freed much later during cleanup
 * time. Right now there is only one field but it can grow as need be.
 */
struct bzimage64_data {
	/*
	 * Temporary buffer to hold bootparams buffer. This should be
	 * freed once the bootparam segment has been loaded.
	 */
	void *bootparams_buf;
};

static int setup_initrd(struct boot_params *params,
		unsigned long initrd_load_addr, unsigned long initrd_len)
{
	params->hdr.ramdisk_image = initrd_load_addr & 0xffffffffUL;
	params->hdr.ramdisk_size = initrd_len & 0xffffffffUL;

	params->ext_ramdisk_image = initrd_load_addr >> 32;
	params->ext_ramdisk_size = initrd_len >> 32;

	return 0;
}

static int setup_cmdline(struct kimage *image, struct boot_params *params,
			 unsigned long bootparams_load_addr,
			 unsigned long cmdline_offset, char *cmdline,
			 unsigned long cmdline_len)
{
	char *cmdline_ptr = ((char *)params) + cmdline_offset;
	unsigned long cmdline_ptr_phys, len = 0;
	uint32_t cmdline_low_32, cmdline_ext_32;

	if (image->type == KEXEC_TYPE_CRASH) {
		len = sprintf(cmdline_ptr,
			"elfcorehdr=0x%lx ", image->arch.elf_load_addr);
	}
	memcpy(cmdline_ptr + len, cmdline, cmdline_len);
	cmdline_len += len;

	cmdline_ptr[cmdline_len - 1] = '\0';

	pr_debug("Final command line is: %s\n", cmdline_ptr);
	cmdline_ptr_phys = bootparams_load_addr + cmdline_offset;
	cmdline_low_32 = cmdline_ptr_phys & 0xffffffffUL;
	cmdline_ext_32 = cmdline_ptr_phys >> 32;

	params->hdr.cmd_line_ptr = cmdline_low_32;
	if (cmdline_ext_32)
		params->ext_cmd_line_ptr = cmdline_ext_32;

	return 0;
}

static int setup_e820_entries(struct boot_params *params)
{
	unsigned int nr_e820_entries;

	nr_e820_entries = e820_saved.nr_map;

	/* TODO: Pass entries more than E820MAX in bootparams setup data */
	if (nr_e820_entries > E820MAX)
		nr_e820_entries = E820MAX;

	params->e820_entries = nr_e820_entries;
	memcpy(&params->e820_map, &e820_saved.map,
	       nr_e820_entries * sizeof(struct e820entry));

	return 0;
}

#ifdef CONFIG_EFI
static int setup_efi_info_memmap(struct boot_params *params,
				  unsigned long params_load_addr,
				  unsigned int efi_map_offset,
				  unsigned int efi_map_sz)
{
	void *efi_map = (void *)params + efi_map_offset;
	unsigned long efi_map_phys_addr = params_load_addr + efi_map_offset;
	struct efi_info *ei = &params->efi_info;

	if (!efi_map_sz)
		return 0;

	efi_runtime_map_copy(efi_map, efi_map_sz);

	ei->efi_memmap = efi_map_phys_addr & 0xffffffff;
	ei->efi_memmap_hi = efi_map_phys_addr >> 32;
	ei->efi_memmap_size = efi_map_sz;

	return 0;
}

static int
prepare_add_efi_setup_data(struct boot_params *params,
		       unsigned long params_load_addr,
		       unsigned int efi_setup_data_offset)
{
	unsigned long setup_data_phys;
	struct setup_data *sd = (void *)params + efi_setup_data_offset;
	struct efi_setup_data *esd = (void *)sd + sizeof(struct setup_data);

	esd->fw_vendor = efi.fw_vendor;
	esd->runtime = efi.runtime;
	esd->tables = efi.config_table;
	esd->smbios = efi.smbios;

	sd->type = SETUP_EFI;
	sd->len = sizeof(struct efi_setup_data);

	/* Add setup data */
	setup_data_phys = params_load_addr + efi_setup_data_offset;
	sd->next = params->hdr.setup_data;
	params->hdr.setup_data = setup_data_phys;

	return 0;
}

static int
setup_efi_state(struct boot_params *params, unsigned long params_load_addr,
		unsigned int efi_map_offset, unsigned int efi_map_sz,
		unsigned int efi_setup_data_offset)
{
	struct efi_info *current_ei = &boot_params.efi_info;
	struct efi_info *ei = &params->efi_info;

	if (!current_ei->efi_memmap_size)
		return 0;

	/*
	 * If 1:1 mapping is not enabled, second kernel can not setup EFI
	 * and use EFI run time services. User space will have to pass
	 * acpi_rsdp=<addr> on kernel command line to make second kernel boot
	 * without efi.
	 */
	if (efi_enabled(EFI_OLD_MEMMAP))
		return 0;

	ei->efi_loader_signature = current_ei->efi_loader_signature;
	ei->efi_systab = current_ei->efi_systab;
	ei->efi_systab_hi = current_ei->efi_systab_hi;

	ei->efi_memdesc_version = current_ei->efi_memdesc_version;
	ei->efi_memdesc_size = efi_get_runtime_map_desc_size();

	setup_efi_info_memmap(params, params_load_addr, efi_map_offset,
			      efi_map_sz);
	prepare_add_efi_setup_data(params, params_load_addr,
				   efi_setup_data_offset);
	return 0;
}
#endif /* CONFIG_EFI */

static int
setup_boot_parameters(struct kimage *image, struct boot_params *params,
		      unsigned long params_load_addr,
		      unsigned int efi_map_offset, unsigned int efi_map_sz,
		      unsigned int efi_setup_data_offset)
{
	unsigned int nr_e820_entries;
	unsigned long long mem_k, start, end;
	int i, ret = 0;

	/* Get subarch from existing bootparams */
	params->hdr.hardware_subarch = boot_params.hdr.hardware_subarch;

	/* Copying screen_info will do? */
	memcpy(&params->screen_info, &boot_params.screen_info,
				sizeof(struct screen_info));

	/* Fill in memsize later */
	params->screen_info.ext_mem_k = 0;
	params->alt_mem_k = 0;

	/* Default APM info */
	memset(&params->apm_bios_info, 0, sizeof(params->apm_bios_info));

	/* Default drive info */
	memset(&params->hd0_info, 0, sizeof(params->hd0_info));
	memset(&params->hd1_info, 0, sizeof(params->hd1_info));

	if (image->type == KEXEC_TYPE_CRASH) {
		ret = crash_setup_memmap_entries(image, params);
		if (ret)
			return ret;
	} else
		setup_e820_entries(params);

	nr_e820_entries = params->e820_entries;

	for (i = 0; i < nr_e820_entries; i++) {
		if (params->e820_map[i].type != E820_RAM)
			continue;
		start = params->e820_map[i].addr;
		end = params->e820_map[i].addr + params->e820_map[i].size - 1;

		if ((start <= 0x100000) && end > 0x100000) {
			mem_k = (end >> 10) - (0x100000 >> 10);
			params->screen_info.ext_mem_k = mem_k;
			params->alt_mem_k = mem_k;
			if (mem_k > 0xfc00)
				params->screen_info.ext_mem_k = 0xfc00; /* 64M*/
			if (mem_k > 0xffffffff)
				params->alt_mem_k = 0xffffffff;
		}
	}

#ifdef CONFIG_EFI
	/* Setup EFI state */
	setup_efi_state(params, params_load_addr, efi_map_offset, efi_map_sz,
			efi_setup_data_offset);
#endif

	/* Setup EDD info */
	memcpy(params->eddbuf, boot_params.eddbuf,
				EDDMAXNR * sizeof(struct edd_info));
	params->eddbuf_entries = boot_params.eddbuf_entries;

	memcpy(params->edd_mbr_sig_buffer, boot_params.edd_mbr_sig_buffer,
	       EDD_MBR_SIG_MAX * sizeof(unsigned int));

	return ret;
}

static int bzImage64_probe(const char *buf, unsigned long len)
{
	int ret = -ENOEXEC;
	struct setup_header *header;

	/* kernel should be at least two sectors long */
	if (len < 2 * 512) {
		pr_err("File is too short to be a bzImage\n");
		return ret;
	}

	header = (struct setup_header *)(buf + offsetof(struct boot_params, hdr));
	if (memcmp((char *)&header->header, "HdrS", 4) != 0) {
		pr_err("Not a bzImage\n");
		return ret;
	}

	if (header->boot_flag != 0xAA55) {
		pr_err("No x86 boot sector present\n");
		return ret;
	}

	if (header->version < 0x020C) {
		pr_err("Must be at least protocol version 2.12\n");
		return ret;
	}

	if (!(header->loadflags & LOADED_HIGH)) {
		pr_err("zImage not a bzImage\n");
		return ret;
	}

	if (!(header->xloadflags & XLF_KERNEL_64)) {
		pr_err("Not a bzImage64. XLF_KERNEL_64 is not set.\n");
		return ret;
	}

	if (!(header->xloadflags & XLF_CAN_BE_LOADED_ABOVE_4G)) {
		pr_err("XLF_CAN_BE_LOADED_ABOVE_4G is not set.\n");
		return ret;
	}

	/*
	 * Can't handle 32bit EFI as it does not allow loading kernel
	 * above 4G. This should be handled by 32bit bzImage loader
	 */
	if (efi_enabled(EFI_RUNTIME_SERVICES) && !efi_enabled(EFI_64BIT)) {
		pr_debug("EFI is 32 bit. Can't load kernel above 4G.\n");
		return ret;
	}

	/* I've got a bzImage */
	pr_debug("It's a relocatable bzImage64\n");
	ret = 0;

	return ret;
}

static void *bzImage64_load(struct kimage *image, char *kernel,
			    unsigned long kernel_len, char *initrd,
			    unsigned long initrd_len, char *cmdline,
			    unsigned long cmdline_len)
{

	struct setup_header *header;
	int setup_sects, kern16_size, ret = 0;
	unsigned long setup_header_size, params_cmdline_sz, params_misc_sz;
	struct boot_params *params;
	unsigned long bootparam_load_addr, kernel_load_addr, initrd_load_addr;
	unsigned long purgatory_load_addr;
	unsigned long kernel_bufsz, kernel_memsz, kernel_align;
	char *kernel_buf;
	struct bzimage64_data *ldata;
	struct kexec_entry64_regs regs64;
	void *stack;
	unsigned int setup_hdr_offset = offsetof(struct boot_params, hdr);
	unsigned int efi_map_offset, efi_map_sz, efi_setup_data_offset;

	header = (struct setup_header *)(kernel + setup_hdr_offset);
	setup_sects = header->setup_sects;
	if (setup_sects == 0)
		setup_sects = 4;

	kern16_size = (setup_sects + 1) * 512;
	if (kernel_len < kern16_size) {
		pr_err("bzImage truncated\n");
		return ERR_PTR(-ENOEXEC);
	}

	if (cmdline_len > header->cmdline_size) {
		pr_err("Kernel command line too long\n");
		return ERR_PTR(-EINVAL);
	}

	/*
	 * In case of crash dump, we will append elfcorehdr=<addr> to
	 * command line. Make sure it does not overflow
	 */
	if (cmdline_len + MAX_ELFCOREHDR_STR_LEN > header->cmdline_size) {
		pr_debug("Appending elfcorehdr=<addr> to command line exceeds maximum allowed length\n");
		return ERR_PTR(-EINVAL);
	}

	/* Allocate and load backup region */
	if (image->type == KEXEC_TYPE_CRASH) {
		ret = crash_load_segments(image);
		if (ret)
			return ERR_PTR(ret);
	}

	/*
	 * Load purgatory. For 64bit entry point, purgatory  code can be
	 * anywhere.
	 */
	ret = kexec_load_purgatory(image, MIN_PURGATORY_ADDR, ULONG_MAX, 1,
				   &purgatory_load_addr);
	if (ret) {
		pr_err("Loading purgatory failed\n");
		return ERR_PTR(ret);
	}

	pr_debug("Loaded purgatory at 0x%lx\n", purgatory_load_addr);


	/*
	 * Load Bootparams and cmdline and space for efi stuff.
	 *
	 * Allocate memory together for multiple data structures so
	 * that they all can go in single area/segment and we don't
	 * have to create separate segment for each. Keeps things
	 * little bit simple
	 */
	efi_map_sz = efi_get_runtime_map_size();
	efi_map_sz = ALIGN(efi_map_sz, 16);
	params_cmdline_sz = sizeof(struct boot_params) + cmdline_len +
				MAX_ELFCOREHDR_STR_LEN;
	params_cmdline_sz = ALIGN(params_cmdline_sz, 16);
	params_misc_sz = params_cmdline_sz + efi_map_sz +
				sizeof(struct setup_data) +
				sizeof(struct efi_setup_data);

	params = kzalloc(params_misc_sz, GFP_KERNEL);
	if (!params)
		return ERR_PTR(-ENOMEM);
	efi_map_offset = params_cmdline_sz;
	efi_setup_data_offset = efi_map_offset + efi_map_sz;

	/* Copy setup header onto bootparams. Documentation/x86/boot.txt */
	setup_header_size = 0x0202 + kernel[0x0201] - setup_hdr_offset;

	/* Is there a limit on setup header size? */
	memcpy(&params->hdr, (kernel + setup_hdr_offset), setup_header_size);

	ret = kexec_add_buffer(image, (char *)params, params_misc_sz,
			       params_misc_sz, 16, MIN_BOOTPARAM_ADDR,
			       ULONG_MAX, 1, &bootparam_load_addr);
	if (ret)
		goto out_free_params;
	pr_debug("Loaded boot_param, command line and misc at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
		 bootparam_load_addr, params_misc_sz, params_misc_sz);

	/* Load kernel */
	kernel_buf = kernel + kern16_size;
	kernel_bufsz =  kernel_len - kern16_size;
	kernel_memsz = PAGE_ALIGN(header->init_size);
	kernel_align = header->kernel_alignment;

	ret = kexec_add_buffer(image, kernel_buf,
			       kernel_bufsz, kernel_memsz, kernel_align,
			       MIN_KERNEL_LOAD_ADDR, ULONG_MAX, 1,
			       &kernel_load_addr);
	if (ret)
		goto out_free_params;

	pr_debug("Loaded 64bit kernel at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
		 kernel_load_addr, kernel_memsz, kernel_memsz);

	/* Load initrd high */
	if (initrd) {
		ret = kexec_add_buffer(image, initrd, initrd_len, initrd_len,
				       PAGE_SIZE, MIN_INITRD_LOAD_ADDR,
				       ULONG_MAX, 1, &initrd_load_addr);
		if (ret)
			goto out_free_params;

		pr_debug("Loaded initrd at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
				initrd_load_addr, initrd_len, initrd_len);

		setup_initrd(params, initrd_load_addr, initrd_len);
	}

	setup_cmdline(image, params, bootparam_load_addr,
		      sizeof(struct boot_params), cmdline, cmdline_len);

	/* bootloader info. Do we need a separate ID for kexec kernel loader? */
	params->hdr.type_of_loader = 0x0D << 4;
	params->hdr.loadflags = 0;

	/* Setup purgatory regs for entry */
	ret = kexec_purgatory_get_set_symbol(image, "entry64_regs", &regs64,
					     sizeof(regs64), 1);
	if (ret)
		goto out_free_params;

	regs64.rbx = 0; /* Bootstrap Processor */
	regs64.rsi = bootparam_load_addr;
	regs64.rip = kernel_load_addr + 0x200;
	stack = kexec_purgatory_get_symbol_addr(image, "stack_end");
	if (IS_ERR(stack)) {
		pr_err("Could not find address of symbol stack_end\n");
		ret = -EINVAL;
		goto out_free_params;
	}

	regs64.rsp = (unsigned long)stack;
	ret = kexec_purgatory_get_set_symbol(image, "entry64_regs", &regs64,
					     sizeof(regs64), 0);
	if (ret)
		goto out_free_params;

	ret = setup_boot_parameters(image, params, bootparam_load_addr,
				    efi_map_offset, efi_map_sz,
				    efi_setup_data_offset);
	if (ret)
		goto out_free_params;

	/* Allocate loader specific data */
	ldata = kzalloc(sizeof(struct bzimage64_data), GFP_KERNEL);
	if (!ldata) {
		ret = -ENOMEM;
		goto out_free_params;
	}

	/*
	 * Store pointer to params so that it could be freed after loading
	 * params segment has been loaded and contents have been copied
	 * somewhere else.
	 */
	ldata->bootparams_buf = params;
	return ldata;

out_free_params:
	kfree(params);
	return ERR_PTR(ret);
}

/* This cleanup function is called after various segments have been loaded */
static int bzImage64_cleanup(void *loader_data)
{
	struct bzimage64_data *ldata = loader_data;

	if (!ldata)
		return 0;

	kfree(ldata->bootparams_buf);
	ldata->bootparams_buf = NULL;

	return 0;
}

#ifdef CONFIG_KEXEC_BZIMAGE_VERIFY_SIG
static int bzImage64_verify_sig(const char *kernel, unsigned long kernel_len)
{
	bool trusted;
	int ret;

	ret = verify_pefile_signature(kernel, kernel_len,
				      system_trusted_keyring,
				      VERIFYING_KEXEC_PE_SIGNATURE,
				      &trusted);
	if (ret < 0)
		return ret;
	if (!trusted)
		return -EKEYREJECTED;
	return 0;
}
#endif

struct kexec_file_ops kexec_bzImage64_ops = {
	.probe = bzImage64_probe,
	.load = bzImage64_load,
	.cleanup = bzImage64_cleanup,
#ifdef CONFIG_KEXEC_BZIMAGE_VERIFY_SIG
	.verify_sig = bzImage64_verify_sig,
#endif
};
Commit	Line	Data
27f48d3e VG	1	/*
	2	* Kexec bzImage loader
	3	*
	4	* Copyright (C) 2014 Red Hat Inc.
	5	* Authors:
	6	* Vivek Goyal <vgoyal@redhat.com>
	7	*
	8	* This source code is licensed under the GNU General Public License,
	9	* Version 2. See the file COPYING for more details.
	10	*/
	11
	12	#define pr_fmt(fmt) "kexec-bzImage64: " fmt
	13
	14	#include <linux/string.h>
	15	#include <linux/printk.h>
	16	#include <linux/errno.h>
	17	#include <linux/slab.h>
	18	#include <linux/kexec.h>
	19	#include <linux/kernel.h>
	20	#include <linux/mm.h>
6a2c20e7	21	#include <linux/efi.h>
8e7d8381 VG	22	#include <linux/verify_pefile.h>
8e7d8381 VG	23	#include <keys/system_keyring.h>
27f48d3e VG	24
	25	#include <asm/bootparam.h>
	26	#include <asm/setup.h>
dd5f7260	27	#include <asm/crash.h>
6a2c20e7	28	#include <asm/efi.h>
f8da964d	29	#include <asm/kexec-bzimage64.h>
dd5f7260 VG	30
dd5f7260 VG	31	#define MAX_ELFCOREHDR_STR_LEN 30 /* elfcorehdr=0x<64bit-value> */
27f48d3e VG	32
	33	/*
	34	* Defines lowest physical address for various segments. Not sure where
	35	* exactly these limits came from. Current bzimage64 loader in kexec-tools
	36	* uses these so I am retaining it. It can be changed over time as we gain
	37	* more insight.
	38	*/
	39	#define MIN_PURGATORY_ADDR 0x3000
	40	#define MIN_BOOTPARAM_ADDR 0x3000
	41	#define MIN_KERNEL_LOAD_ADDR 0x100000
	42	#define MIN_INITRD_LOAD_ADDR 0x1000000
	43
	44	/*
	45	* This is a place holder for all boot loader specific data structure which
	46	* gets allocated in one call but gets freed much later during cleanup
	47	* time. Right now there is only one field but it can grow as need be.
	48	*/
	49	struct bzimage64_data {
	50	/*
	51	* Temporary buffer to hold bootparams buffer. This should be
	52	* freed once the bootparam segment has been loaded.
	53	*/
	54	void *bootparams_buf;
	55	};
	56
	57	static int setup_initrd(struct boot_params *params,
	58	unsigned long initrd_load_addr, unsigned long initrd_len)
	59	{
	60	params->hdr.ramdisk_image = initrd_load_addr & 0xffffffffUL;
	61	params->hdr.ramdisk_size = initrd_len & 0xffffffffUL;
	62
	63	params->ext_ramdisk_image = initrd_load_addr >> 32;
	64	params->ext_ramdisk_size = initrd_len >> 32;
	65
	66	return 0;
	67	}
	68
dd5f7260	69	static int setup_cmdline(struct kimage image, struct boot_params params,
27f48d3e VG	70	unsigned long bootparams_load_addr,
	71	unsigned long cmdline_offset, char *cmdline,
	72	unsigned long cmdline_len)
	73	{
	74	char cmdline_ptr = ((char )params) + cmdline_offset;
a846f479	75	unsigned long cmdline_ptr_phys, len = 0;
27f48d3e VG	76	uint32_t cmdline_low_32, cmdline_ext_32;
27f48d3e VG	77
dd5f7260	78	if (image->type == KEXEC_TYPE_CRASH) {
a846f479 KA	79	len = sprintf(cmdline_ptr,
a846f479 KA	80	"elfcorehdr=0x%lx ", image->arch.elf_load_addr);
dd5f7260	81	}
a846f479 KA	82	memcpy(cmdline_ptr + len, cmdline, cmdline_len);
	83	cmdline_len += len;
	84
27f48d3e VG	85	cmdline_ptr[cmdline_len - 1] = '\0';
27f48d3e VG	86
dd5f7260	87	pr_debug("Final command line is: %s\n", cmdline_ptr);
27f48d3e VG	88	cmdline_ptr_phys = bootparams_load_addr + cmdline_offset;
	89	cmdline_low_32 = cmdline_ptr_phys & 0xffffffffUL;
	90	cmdline_ext_32 = cmdline_ptr_phys >> 32;
	91
	92	params->hdr.cmd_line_ptr = cmdline_low_32;
	93	if (cmdline_ext_32)
	94	params->ext_cmd_line_ptr = cmdline_ext_32;
	95
	96	return 0;
	97	}
	98
6a2c20e7	99	static int setup_e820_entries(struct boot_params *params)
27f48d3e VG	100	{
	101	unsigned int nr_e820_entries;
	102
	103	nr_e820_entries = e820_saved.nr_map;
	104
	105	/* TODO: Pass entries more than E820MAX in bootparams setup data */
	106	if (nr_e820_entries > E820MAX)
	107	nr_e820_entries = E820MAX;
	108
	109	params->e820_entries = nr_e820_entries;
	110	memcpy(&params->e820_map, &e820_saved.map,
	111	nr_e820_entries * sizeof(struct e820entry));
	112
	113	return 0;
	114	}
	115
6a2c20e7 VG	116	#ifdef CONFIG_EFI
	117	static int setup_efi_info_memmap(struct boot_params *params,
	118	unsigned long params_load_addr,
	119	unsigned int efi_map_offset,
	120	unsigned int efi_map_sz)
	121	{
	122	void efi_map = (void )params + efi_map_offset;
	123	unsigned long efi_map_phys_addr = params_load_addr + efi_map_offset;
	124	struct efi_info *ei = &params->efi_info;
	125
	126	if (!efi_map_sz)
	127	return 0;
	128
	129	efi_runtime_map_copy(efi_map, efi_map_sz);
	130
	131	ei->efi_memmap = efi_map_phys_addr & 0xffffffff;
	132	ei->efi_memmap_hi = efi_map_phys_addr >> 32;
	133	ei->efi_memmap_size = efi_map_sz;
	134
	135	return 0;
	136	}
	137
	138	static int
	139	prepare_add_efi_setup_data(struct boot_params *params,
	140	unsigned long params_load_addr,
	141	unsigned int efi_setup_data_offset)
	142	{
	143	unsigned long setup_data_phys;
	144	struct setup_data sd = (void )params + efi_setup_data_offset;
	145	struct efi_setup_data esd = (void )sd + sizeof(struct setup_data);
	146
	147	esd->fw_vendor = efi.fw_vendor;
	148	esd->runtime = efi.runtime;
	149	esd->tables = efi.config_table;
	150	esd->smbios = efi.smbios;
	151
	152	sd->type = SETUP_EFI;
	153	sd->len = sizeof(struct efi_setup_data);
	154
	155	/* Add setup data */
	156	setup_data_phys = params_load_addr + efi_setup_data_offset;
	157	sd->next = params->hdr.setup_data;
	158	params->hdr.setup_data = setup_data_phys;
	159
	160	return 0;
	161	}
	162
	163	static int
	164	setup_efi_state(struct boot_params *params, unsigned long params_load_addr,
	165	unsigned int efi_map_offset, unsigned int efi_map_sz,
	166	unsigned int efi_setup_data_offset)
	167	{
	168	struct efi_info *current_ei = &boot_params.efi_info;
	169	struct efi_info *ei = &params->efi_info;
	170
	171	if (!current_ei->efi_memmap_size)
	172	return 0;
	173
	174	/*
	175	* If 1:1 mapping is not enabled, second kernel can not setup EFI
	176	* and use EFI run time services. User space will have to pass
	177	* acpi_rsdp=<addr> on kernel command line to make second kernel boot
	178	* without efi.
	179	*/
180	if (efi_enabled(EFI_OLD_MEMMAP))
181	return 0;
182
183	ei->efi_loader_signature = current_ei->efi_loader_signature;
184	ei->efi_systab = current_ei->efi_systab;
185	ei->efi_systab_hi = current_ei->efi_systab_hi;
186
187	ei->efi_memdesc_version = current_ei->efi_memdesc_version;
188	ei->efi_memdesc_size = efi_get_runtime_map_desc_size();
189
190	setup_efi_info_memmap(params, params_load_addr, efi_map_offset,
191	efi_map_sz);
192	prepare_add_efi_setup_data(params, params_load_addr,
193	efi_setup_data_offset);
194	return 0;
195	}
196	#endif /* CONFIG_EFI */
197
198	static int
199	setup_boot_parameters(struct kimage image, struct boot_params params,
200	unsigned long params_load_addr,
201	unsigned int efi_map_offset, unsigned int efi_map_sz,
202	unsigned int efi_setup_data_offset)
27f48d3e VG	203	{
	204	unsigned int nr_e820_entries;
	205	unsigned long long mem_k, start, end;
dd5f7260	206	int i, ret = 0;
27f48d3e VG	207
	208	/* Get subarch from existing bootparams */
	209	params->hdr.hardware_subarch = boot_params.hdr.hardware_subarch;
	210
	211	/* Copying screen_info will do? */
	212	memcpy(&params->screen_info, &boot_params.screen_info,
	213	sizeof(struct screen_info));
	214
	215	/* Fill in memsize later */
	216	params->screen_info.ext_mem_k = 0;
	217	params->alt_mem_k = 0;
	218
	219	/* Default APM info */
	220	memset(&params->apm_bios_info, 0, sizeof(params->apm_bios_info));
	221
	222	/* Default drive info */
	223	memset(&params->hd0_info, 0, sizeof(params->hd0_info));
	224	memset(&params->hd1_info, 0, sizeof(params->hd1_info));
	225
dd5f7260 VG	226	if (image->type == KEXEC_TYPE_CRASH) {
	227	ret = crash_setup_memmap_entries(image, params);
	228	if (ret)
	229	return ret;
	230	} else
6a2c20e7	231	setup_e820_entries(params);
dd5f7260	232
27f48d3e VG	233	nr_e820_entries = params->e820_entries;
	234
	235	for (i = 0; i < nr_e820_entries; i++) {
	236	if (params->e820_map[i].type != E820_RAM)
	237	continue;
	238	start = params->e820_map[i].addr;
	239	end = params->e820_map[i].addr + params->e820_map[i].size - 1;
	240
	241	if ((start <= 0x100000) && end > 0x100000) {
	242	mem_k = (end >> 10) - (0x100000 >> 10);
	243	params->screen_info.ext_mem_k = mem_k;
	244	params->alt_mem_k = mem_k;
	245	if (mem_k > 0xfc00)
	246	params->screen_info.ext_mem_k = 0xfc00; /* 64M*/
	247	if (mem_k > 0xffffffff)
	248	params->alt_mem_k = 0xffffffff;
	249	}
	250	}
	251
6a2c20e7 VG	252	#ifdef CONFIG_EFI
	253	/* Setup EFI state */
	254	setup_efi_state(params, params_load_addr, efi_map_offset, efi_map_sz,
	255	efi_setup_data_offset);
	256	#endif
	257
27f48d3e VG	258	/* Setup EDD info */
	259	memcpy(params->eddbuf, boot_params.eddbuf,
	260	EDDMAXNR * sizeof(struct edd_info));
	261	params->eddbuf_entries = boot_params.eddbuf_entries;
	262
	263	memcpy(params->edd_mbr_sig_buffer, boot_params.edd_mbr_sig_buffer,
	264	EDD_MBR_SIG_MAX * sizeof(unsigned int));
	265
dd5f7260	266	return ret;
27f48d3e VG	267	}
27f48d3e VG	268
f8da964d	269	static int bzImage64_probe(const char *buf, unsigned long len)
27f48d3e VG	270	{
	271	int ret = -ENOEXEC;
	272	struct setup_header *header;
	273
6a6256f9	274	/* kernel should be at least two sectors long */
27f48d3e VG	275	if (len < 2 * 512) {
	276	pr_err("File is too short to be a bzImage\n");
	277	return ret;
	278	}
	279
	280	header = (struct setup_header *)(buf + offsetof(struct boot_params, hdr));
	281	if (memcmp((char *)&header->header, "HdrS", 4) != 0) {
	282	pr_err("Not a bzImage\n");
	283	return ret;
	284	}
	285
	286	if (header->boot_flag != 0xAA55) {
	287	pr_err("No x86 boot sector present\n");
	288	return ret;
	289	}
	290
	291	if (header->version < 0x020C) {
	292	pr_err("Must be at least protocol version 2.12\n");
	293	return ret;
	294	}
	295
	296	if (!(header->loadflags & LOADED_HIGH)) {
	297	pr_err("zImage not a bzImage\n");
	298	return ret;
	299	}
	300
	301	if (!(header->xloadflags & XLF_KERNEL_64)) {
	302	pr_err("Not a bzImage64. XLF_KERNEL_64 is not set.\n");
	303	return ret;
	304	}
	305
	306	if (!(header->xloadflags & XLF_CAN_BE_LOADED_ABOVE_4G)) {
	307	pr_err("XLF_CAN_BE_LOADED_ABOVE_4G is not set.\n");
	308	return ret;
	309	}
	310
6a2c20e7 VG	311	/*
	312	* Can't handle 32bit EFI as it does not allow loading kernel
	313	* above 4G. This should be handled by 32bit bzImage loader
	314	*/
	315	if (efi_enabled(EFI_RUNTIME_SERVICES) && !efi_enabled(EFI_64BIT)) {
	316	pr_debug("EFI is 32 bit. Can't load kernel above 4G.\n");
	317	return ret;
	318	}
	319
27f48d3e VG	320	/* I've got a bzImage */
	321	pr_debug("It's a relocatable bzImage64\n");
	322	ret = 0;
	323
	324	return ret;
	325	}
	326
f8da964d VG	327	static void bzImage64_load(struct kimage image, char *kernel,
	328	unsigned long kernel_len, char *initrd,
	329	unsigned long initrd_len, char *cmdline,
	330	unsigned long cmdline_len)
27f48d3e VG	331	{
	332
	333	struct setup_header *header;
	334	int setup_sects, kern16_size, ret = 0;
6a2c20e7	335	unsigned long setup_header_size, params_cmdline_sz, params_misc_sz;
27f48d3e VG	336	struct boot_params *params;
	337	unsigned long bootparam_load_addr, kernel_load_addr, initrd_load_addr;
	338	unsigned long purgatory_load_addr;
	339	unsigned long kernel_bufsz, kernel_memsz, kernel_align;
	340	char *kernel_buf;
	341	struct bzimage64_data *ldata;
	342	struct kexec_entry64_regs regs64;
	343	void *stack;
	344	unsigned int setup_hdr_offset = offsetof(struct boot_params, hdr);
6a2c20e7	345	unsigned int efi_map_offset, efi_map_sz, efi_setup_data_offset;
27f48d3e VG	346
	347	header = (struct setup_header *)(kernel + setup_hdr_offset);
	348	setup_sects = header->setup_sects;
	349	if (setup_sects == 0)
	350	setup_sects = 4;
	351
	352	kern16_size = (setup_sects + 1) * 512;
	353	if (kernel_len < kern16_size) {
	354	pr_err("bzImage truncated\n");
	355	return ERR_PTR(-ENOEXEC);
	356	}
	357
	358	if (cmdline_len > header->cmdline_size) {
	359	pr_err("Kernel command line too long\n");
	360	return ERR_PTR(-EINVAL);
	361	}
	362
dd5f7260 VG	363	/*
	364	* In case of crash dump, we will append elfcorehdr=<addr> to
	365	* command line. Make sure it does not overflow
	366	*/
	367	if (cmdline_len + MAX_ELFCOREHDR_STR_LEN > header->cmdline_size) {
	368	pr_debug("Appending elfcorehdr=<addr> to command line exceeds maximum allowed length\n");
	369	return ERR_PTR(-EINVAL);
	370	}
	371
	372	/* Allocate and load backup region */
	373	if (image->type == KEXEC_TYPE_CRASH) {
	374	ret = crash_load_segments(image);
	375	if (ret)
	376	return ERR_PTR(ret);
	377	}
	378
27f48d3e VG	379	/*
	380	* Load purgatory. For 64bit entry point, purgatory code can be
	381	* anywhere.
	382	*/
	383	ret = kexec_load_purgatory(image, MIN_PURGATORY_ADDR, ULONG_MAX, 1,
	384	&purgatory_load_addr);
	385	if (ret) {
	386	pr_err("Loading purgatory failed\n");
	387	return ERR_PTR(ret);
	388	}
	389
	390	pr_debug("Loaded purgatory at 0x%lx\n", purgatory_load_addr);
	391
6a2c20e7 VG	392
	393	/*
	394	* Load Bootparams and cmdline and space for efi stuff.
	395	*
	396	* Allocate memory together for multiple data structures so
	397	* that they all can go in single area/segment and we don't
	398	* have to create separate segment for each. Keeps things
	399	* little bit simple
	400	*/
	401	efi_map_sz = efi_get_runtime_map_size();
	402	efi_map_sz = ALIGN(efi_map_sz, 16);
dd5f7260 VG	403	params_cmdline_sz = sizeof(struct boot_params) + cmdline_len +
dd5f7260 VG	404	MAX_ELFCOREHDR_STR_LEN;
6a2c20e7 VG	405	params_cmdline_sz = ALIGN(params_cmdline_sz, 16);
	406	params_misc_sz = params_cmdline_sz + efi_map_sz +
	407	sizeof(struct setup_data) +
	408	sizeof(struct efi_setup_data);
	409
	410	params = kzalloc(params_misc_sz, GFP_KERNEL);
27f48d3e VG	411	if (!params)
27f48d3e VG	412	return ERR_PTR(-ENOMEM);
6a2c20e7 VG	413	efi_map_offset = params_cmdline_sz;
6a2c20e7 VG	414	efi_setup_data_offset = efi_map_offset + efi_map_sz;
27f48d3e VG	415
	416	/* Copy setup header onto bootparams. Documentation/x86/boot.txt */
	417	setup_header_size = 0x0202 + kernel[0x0201] - setup_hdr_offset;
	418
	419	/* Is there a limit on setup header size? */
	420	memcpy(&params->hdr, (kernel + setup_hdr_offset), setup_header_size);
	421
6a2c20e7 VG	422	ret = kexec_add_buffer(image, (char *)params, params_misc_sz,
6a2c20e7 VG	423	params_misc_sz, 16, MIN_BOOTPARAM_ADDR,
27f48d3e VG	424	ULONG_MAX, 1, &bootparam_load_addr);
	425	if (ret)
	426	goto out_free_params;
6a2c20e7 VG	427	pr_debug("Loaded boot_param, command line and misc at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
6a2c20e7 VG	428	bootparam_load_addr, params_misc_sz, params_misc_sz);
27f48d3e VG	429
	430	/* Load kernel */
	431	kernel_buf = kernel + kern16_size;
	432	kernel_bufsz = kernel_len - kern16_size;
	433	kernel_memsz = PAGE_ALIGN(header->init_size);
	434	kernel_align = header->kernel_alignment;
	435
	436	ret = kexec_add_buffer(image, kernel_buf,
	437	kernel_bufsz, kernel_memsz, kernel_align,
	438	MIN_KERNEL_LOAD_ADDR, ULONG_MAX, 1,
	439	&kernel_load_addr);
	440	if (ret)
	441	goto out_free_params;
	442
	443	pr_debug("Loaded 64bit kernel at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
	444	kernel_load_addr, kernel_memsz, kernel_memsz);
	445
	446	/* Load initrd high */
	447	if (initrd) {
	448	ret = kexec_add_buffer(image, initrd, initrd_len, initrd_len,
	449	PAGE_SIZE, MIN_INITRD_LOAD_ADDR,
	450	ULONG_MAX, 1, &initrd_load_addr);
	451	if (ret)
	452	goto out_free_params;
	453
	454	pr_debug("Loaded initrd at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
	455	initrd_load_addr, initrd_len, initrd_len);
	456
	457	setup_initrd(params, initrd_load_addr, initrd_len);
	458	}
	459
dd5f7260 VG	460	setup_cmdline(image, params, bootparam_load_addr,
dd5f7260 VG	461	sizeof(struct boot_params), cmdline, cmdline_len);
27f48d3e VG	462
	463	/* bootloader info. Do we need a separate ID for kexec kernel loader? */
	464	params->hdr.type_of_loader = 0x0D << 4;
	465	params->hdr.loadflags = 0;
	466
	467	/* Setup purgatory regs for entry */
	468	ret = kexec_purgatory_get_set_symbol(image, "entry64_regs", &regs64,
	469	sizeof(regs64), 1);
	470	if (ret)
	471	goto out_free_params;
	472
	473	regs64.rbx = 0; /* Bootstrap Processor */
	474	regs64.rsi = bootparam_load_addr;
	475	regs64.rip = kernel_load_addr + 0x200;
	476	stack = kexec_purgatory_get_symbol_addr(image, "stack_end");
	477	if (IS_ERR(stack)) {
	478	pr_err("Could not find address of symbol stack_end\n");
	479	ret = -EINVAL;
	480	goto out_free_params;
	481	}
	482
	483	regs64.rsp = (unsigned long)stack;
	484	ret = kexec_purgatory_get_set_symbol(image, "entry64_regs", &regs64,
	485	sizeof(regs64), 0);
	486	if (ret)
	487	goto out_free_params;
	488
6a2c20e7 VG	489	ret = setup_boot_parameters(image, params, bootparam_load_addr,
	490	efi_map_offset, efi_map_sz,
	491	efi_setup_data_offset);
dd5f7260 VG	492	if (ret)
dd5f7260 VG	493	goto out_free_params;
27f48d3e VG	494
	495	/* Allocate loader specific data */
	496	ldata = kzalloc(sizeof(struct bzimage64_data), GFP_KERNEL);
	497	if (!ldata) {
	498	ret = -ENOMEM;
	499	goto out_free_params;
	500	}
	501
	502	/*
	503	* Store pointer to params so that it could be freed after loading
	504	* params segment has been loaded and contents have been copied
	505	* somewhere else.
	506	*/
	507	ldata->bootparams_buf = params;
	508	return ldata;
	509
	510	out_free_params:
	511	kfree(params);
	512	return ERR_PTR(ret);
	513	}
	514
	515	/* This cleanup function is called after various segments have been loaded */
f8da964d	516	static int bzImage64_cleanup(void *loader_data)
27f48d3e VG	517	{
	518	struct bzimage64_data *ldata = loader_data;
	519
	520	if (!ldata)
	521	return 0;
	522
	523	kfree(ldata->bootparams_buf);
	524	ldata->bootparams_buf = NULL;
	525
	526	return 0;
	527	}
	528
8e7d8381	529	#ifdef CONFIG_KEXEC_BZIMAGE_VERIFY_SIG
f8da964d	530	static int bzImage64_verify_sig(const char *kernel, unsigned long kernel_len)
8e7d8381 VG	531	{
	532	bool trusted;
	533	int ret;
	534
	535	ret = verify_pefile_signature(kernel, kernel_len,
99db4435 DH	536	system_trusted_keyring,
	537	VERIFYING_KEXEC_PE_SIGNATURE,
	538	&trusted);
8e7d8381 VG	539	if (ret < 0)
	540	return ret;
	541	if (!trusted)
	542	return -EKEYREJECTED;
	543	return 0;
	544	}
	545	#endif
	546
27f48d3e VG	547	struct kexec_file_ops kexec_bzImage64_ops = {
	548	.probe = bzImage64_probe,
	549	.load = bzImage64_load,
	550	.cleanup = bzImage64_cleanup,
8e7d8381 VG	551	#ifdef CONFIG_KEXEC_BZIMAGE_VERIFY_SIG
	552	.verify_sig = bzImage64_verify_sig,
	553	#endif
27f48d3e	554	};