[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/verification.h>

#include <asm/bootparam.h>
#include <asm/setup.h>
#include <asm/crash.h>
#include <asm/efi.h>
#include <asm/e820/api.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_table_kexec->nr_entries;

	/* TODO: Pass entries more than E820_MAX_ENTRIES_ZEROPAGE in bootparams setup data */
	if (nr_e820_entries > E820_MAX_ENTRIES_ZEROPAGE)
		nr_e820_entries = E820_MAX_ENTRIES_ZEROPAGE;

	params->e820_entries = nr_e820_entries;
	memcpy(&params->e820_table, &e820_table_kexec->entries, nr_e820_entries*sizeof(struct e820_entry));

	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_table[i].type != E820_TYPE_RAM)
			continue;
		start = params->e820_table[i].addr;
		end = params->e820_table[i].addr + params->e820_table[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;
	struct boot_params *params;
	unsigned long bootparam_load_addr, kernel_load_addr, initrd_load_addr;
	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;
	struct kexec_buf kbuf = { .image = image, .buf_max = ULONG_MAX,
				  .top_down = true };
	struct kexec_buf pbuf = { .image = image, .buf_min = MIN_PURGATORY_ADDR,
				  .buf_max = ULONG_MAX, .top_down = true };

	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, &pbuf);
	if (ret) {
		pr_err("Loading purgatory failed\n");
		return ERR_PTR(ret);
	}

	pr_debug("Loaded purgatory at 0x%lx\n", pbuf.mem);


	/*
	 * 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();
	params_cmdline_sz = sizeof(struct boot_params) + cmdline_len +
				MAX_ELFCOREHDR_STR_LEN;
	params_cmdline_sz = ALIGN(params_cmdline_sz, 16);
	kbuf.bufsz = params_cmdline_sz + ALIGN(efi_map_sz, 16) +
				sizeof(struct setup_data) +
				sizeof(struct efi_setup_data);

	params = kzalloc(kbuf.bufsz, GFP_KERNEL);
	if (!params)
		return ERR_PTR(-ENOMEM);
	efi_map_offset = params_cmdline_sz;
	efi_setup_data_offset = efi_map_offset + ALIGN(efi_map_sz, 16);

	/* 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);

	kbuf.buffer = params;
	kbuf.memsz = kbuf.bufsz;
	kbuf.buf_align = 16;
	kbuf.buf_min = MIN_BOOTPARAM_ADDR;
	ret = kexec_add_buffer(&kbuf);
	if (ret)
		goto out_free_params;
	bootparam_load_addr = kbuf.mem;
	pr_debug("Loaded boot_param, command line and misc at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
		 bootparam_load_addr, kbuf.bufsz, kbuf.bufsz);

	/* Load kernel */
	kbuf.buffer = kernel + kern16_size;
	kbuf.bufsz =  kernel_len - kern16_size;
	kbuf.memsz = PAGE_ALIGN(header->init_size);
	kbuf.buf_align = header->kernel_alignment;
	kbuf.buf_min = MIN_KERNEL_LOAD_ADDR;
	ret = kexec_add_buffer(&kbuf);
	if (ret)
		goto out_free_params;
	kernel_load_addr = kbuf.mem;

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

	/* Load initrd high */
	if (initrd) {
		kbuf.buffer = initrd;
		kbuf.bufsz = kbuf.memsz = initrd_len;
		kbuf.buf_align = PAGE_SIZE;
		kbuf.buf_min = MIN_INITRD_LOAD_ADDR;
		ret = kexec_add_buffer(&kbuf);
		if (ret)
			goto out_free_params;
		initrd_load_addr = kbuf.mem;

		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)
{
	return verify_pefile_signature(kernel, kernel_len,
				       NULL,
				       VERIFYING_KEXEC_PE_SIGNATURE);
}
#endif

const 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>
e68503bd	22	#include <linux/verification.h>
27f48d3e VG	23
	24	#include <asm/bootparam.h>
	25	#include <asm/setup.h>
dd5f7260	26	#include <asm/crash.h>
6a2c20e7	27	#include <asm/efi.h>
5520b7e7	28	#include <asm/e820/api.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
a09bae0f	103	nr_e820_entries = e820_table_kexec->nr_entries;
27f48d3e	104
08b46d5d IM	105	/* TODO: Pass entries more than E820_MAX_ENTRIES_ZEROPAGE in bootparams setup data */
	106	if (nr_e820_entries > E820_MAX_ENTRIES_ZEROPAGE)
	107	nr_e820_entries = E820_MAX_ENTRIES_ZEROPAGE;
27f48d3e VG	108
27f48d3e VG	109	params->e820_entries = nr_e820_entries;
a09bae0f	110	memcpy(&params->e820_table, &e820_table_kexec->entries, nr_e820_entries*sizeof(struct e820_entry));
27f48d3e VG	111
	112	return 0;
	113	}
	114
6a2c20e7 VG	115	#ifdef CONFIG_EFI
	116	static int setup_efi_info_memmap(struct boot_params *params,
	117	unsigned long params_load_addr,
	118	unsigned int efi_map_offset,
	119	unsigned int efi_map_sz)
	120	{
	121	void efi_map = (void )params + efi_map_offset;
	122	unsigned long efi_map_phys_addr = params_load_addr + efi_map_offset;
	123	struct efi_info *ei = &params->efi_info;
	124
	125	if (!efi_map_sz)
	126	return 0;
	127
	128	efi_runtime_map_copy(efi_map, efi_map_sz);
	129
	130	ei->efi_memmap = efi_map_phys_addr & 0xffffffff;
	131	ei->efi_memmap_hi = efi_map_phys_addr >> 32;
	132	ei->efi_memmap_size = efi_map_sz;
	133
	134	return 0;
	135	}
	136
	137	static int
	138	prepare_add_efi_setup_data(struct boot_params *params,
	139	unsigned long params_load_addr,
	140	unsigned int efi_setup_data_offset)
	141	{
	142	unsigned long setup_data_phys;
	143	struct setup_data sd = (void )params + efi_setup_data_offset;
	144	struct efi_setup_data esd = (void )sd + sizeof(struct setup_data);
	145
	146	esd->fw_vendor = efi.fw_vendor;
	147	esd->runtime = efi.runtime;
	148	esd->tables = efi.config_table;
	149	esd->smbios = efi.smbios;
	150
	151	sd->type = SETUP_EFI;
	152	sd->len = sizeof(struct efi_setup_data);
	153
	154	/* Add setup data */
	155	setup_data_phys = params_load_addr + efi_setup_data_offset;
	156	sd->next = params->hdr.setup_data;
	157	params->hdr.setup_data = setup_data_phys;
	158
	159	return 0;
	160	}
	161
	162	static int
	163	setup_efi_state(struct boot_params *params, unsigned long params_load_addr,
	164	unsigned int efi_map_offset, unsigned int efi_map_sz,
	165	unsigned int efi_setup_data_offset)
	166	{
	167	struct efi_info *current_ei = &boot_params.efi_info;
	168	struct efi_info *ei = &params->efi_info;
	169
	170	if (!current_ei->efi_memmap_size)
	171	return 0;
	172
	173	/*
	174	* If 1:1 mapping is not enabled, second kernel can not setup EFI
	175	* and use EFI run time services. User space will have to pass
	176	* acpi_rsdp=<addr> on kernel command line to make second kernel boot
	177	* without efi.
	178	*/
179	if (efi_enabled(EFI_OLD_MEMMAP))
180	return 0;
181
182	ei->efi_loader_signature = current_ei->efi_loader_signature;
183	ei->efi_systab = current_ei->efi_systab;
184	ei->efi_systab_hi = current_ei->efi_systab_hi;
185
186	ei->efi_memdesc_version = current_ei->efi_memdesc_version;
187	ei->efi_memdesc_size = efi_get_runtime_map_desc_size();
188
189	setup_efi_info_memmap(params, params_load_addr, efi_map_offset,
190	efi_map_sz);
191	prepare_add_efi_setup_data(params, params_load_addr,
192	efi_setup_data_offset);
193	return 0;
194	}
195	#endif /* CONFIG_EFI */
196
197	static int
198	setup_boot_parameters(struct kimage image, struct boot_params params,
199	unsigned long params_load_addr,
200	unsigned int efi_map_offset, unsigned int efi_map_sz,
201	unsigned int efi_setup_data_offset)
27f48d3e VG	202	{
	203	unsigned int nr_e820_entries;
	204	unsigned long long mem_k, start, end;
dd5f7260	205	int i, ret = 0;
27f48d3e VG	206
	207	/* Get subarch from existing bootparams */
	208	params->hdr.hardware_subarch = boot_params.hdr.hardware_subarch;
	209
	210	/* Copying screen_info will do? */
	211	memcpy(&params->screen_info, &boot_params.screen_info,
	212	sizeof(struct screen_info));
	213
	214	/* Fill in memsize later */
	215	params->screen_info.ext_mem_k = 0;
	216	params->alt_mem_k = 0;
	217
	218	/* Default APM info */
	219	memset(&params->apm_bios_info, 0, sizeof(params->apm_bios_info));
	220
	221	/* Default drive info */
	222	memset(&params->hd0_info, 0, sizeof(params->hd0_info));
	223	memset(&params->hd1_info, 0, sizeof(params->hd1_info));
	224
dd5f7260 VG	225	if (image->type == KEXEC_TYPE_CRASH) {
	226	ret = crash_setup_memmap_entries(image, params);
	227	if (ret)
	228	return ret;
	229	} else
6a2c20e7	230	setup_e820_entries(params);
dd5f7260	231
27f48d3e VG	232	nr_e820_entries = params->e820_entries;
	233
	234	for (i = 0; i < nr_e820_entries; i++) {
09821ff1	235	if (params->e820_table[i].type != E820_TYPE_RAM)
27f48d3e	236	continue;
61a50101 IM	237	start = params->e820_table[i].addr;
61a50101 IM	238	end = params->e820_table[i].addr + params->e820_table[i].size - 1;
27f48d3e VG	239
	240	if ((start <= 0x100000) && end > 0x100000) {
	241	mem_k = (end >> 10) - (0x100000 >> 10);
	242	params->screen_info.ext_mem_k = mem_k;
	243	params->alt_mem_k = mem_k;
	244	if (mem_k > 0xfc00)
	245	params->screen_info.ext_mem_k = 0xfc00; /* 64M*/
	246	if (mem_k > 0xffffffff)
	247	params->alt_mem_k = 0xffffffff;
	248	}
	249	}
	250
6a2c20e7 VG	251	#ifdef CONFIG_EFI
	252	/* Setup EFI state */
	253	setup_efi_state(params, params_load_addr, efi_map_offset, efi_map_sz,
	254	efi_setup_data_offset);
	255	#endif
	256
27f48d3e VG	257	/* Setup EDD info */
	258	memcpy(params->eddbuf, boot_params.eddbuf,
	259	EDDMAXNR * sizeof(struct edd_info));
	260	params->eddbuf_entries = boot_params.eddbuf_entries;
	261
	262	memcpy(params->edd_mbr_sig_buffer, boot_params.edd_mbr_sig_buffer,
	263	EDD_MBR_SIG_MAX * sizeof(unsigned int));
	264
dd5f7260	265	return ret;
27f48d3e VG	266	}
27f48d3e VG	267
f8da964d	268	static int bzImage64_probe(const char *buf, unsigned long len)
27f48d3e VG	269	{
	270	int ret = -ENOEXEC;
	271	struct setup_header *header;
	272
6a6256f9	273	/* kernel should be at least two sectors long */
27f48d3e VG	274	if (len < 2 * 512) {
	275	pr_err("File is too short to be a bzImage\n");
	276	return ret;
	277	}
	278
	279	header = (struct setup_header *)(buf + offsetof(struct boot_params, hdr));
	280	if (memcmp((char *)&header->header, "HdrS", 4) != 0) {
	281	pr_err("Not a bzImage\n");
	282	return ret;
	283	}
	284
	285	if (header->boot_flag != 0xAA55) {
	286	pr_err("No x86 boot sector present\n");
	287	return ret;
	288	}
	289
	290	if (header->version < 0x020C) {
	291	pr_err("Must be at least protocol version 2.12\n");
	292	return ret;
	293	}
	294
	295	if (!(header->loadflags & LOADED_HIGH)) {
	296	pr_err("zImage not a bzImage\n");
	297	return ret;
	298	}
	299
	300	if (!(header->xloadflags & XLF_KERNEL_64)) {
	301	pr_err("Not a bzImage64. XLF_KERNEL_64 is not set.\n");
	302	return ret;
	303	}
	304
	305	if (!(header->xloadflags & XLF_CAN_BE_LOADED_ABOVE_4G)) {
	306	pr_err("XLF_CAN_BE_LOADED_ABOVE_4G is not set.\n");
	307	return ret;
	308	}
	309
6a2c20e7 VG	310	/*
	311	* Can't handle 32bit EFI as it does not allow loading kernel
	312	* above 4G. This should be handled by 32bit bzImage loader
	313	*/
	314	if (efi_enabled(EFI_RUNTIME_SERVICES) && !efi_enabled(EFI_64BIT)) {
	315	pr_debug("EFI is 32 bit. Can't load kernel above 4G.\n");
	316	return ret;
	317	}
	318
27f48d3e VG	319	/* I've got a bzImage */
	320	pr_debug("It's a relocatable bzImage64\n");
	321	ret = 0;
	322
	323	return ret;
	324	}
	325
f8da964d VG	326	static void bzImage64_load(struct kimage image, char *kernel,
	327	unsigned long kernel_len, char *initrd,
	328	unsigned long initrd_len, char *cmdline,
	329	unsigned long cmdline_len)
27f48d3e VG	330	{
	331
	332	struct setup_header *header;
	333	int setup_sects, kern16_size, ret = 0;
ec2b9bfa	334	unsigned long setup_header_size, params_cmdline_sz;
27f48d3e VG	335	struct boot_params *params;
27f48d3e VG	336	unsigned long bootparam_load_addr, kernel_load_addr, initrd_load_addr;
27f48d3e VG	337	struct bzimage64_data *ldata;
	338	struct kexec_entry64_regs regs64;
	339	void *stack;
	340	unsigned int setup_hdr_offset = offsetof(struct boot_params, hdr);
6a2c20e7	341	unsigned int efi_map_offset, efi_map_sz, efi_setup_data_offset;
ec2b9bfa TJB	342	struct kexec_buf kbuf = { .image = image, .buf_max = ULONG_MAX,
ec2b9bfa TJB	343	.top_down = true };
3be3f61d PR	344	struct kexec_buf pbuf = { .image = image, .buf_min = MIN_PURGATORY_ADDR,
3be3f61d PR	345	.buf_max = ULONG_MAX, .top_down = true };
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	*/
3be3f61d	383	ret = kexec_load_purgatory(image, &pbuf);
27f48d3e VG	384	if (ret) {
	385	pr_err("Loading purgatory failed\n");
	386	return ERR_PTR(ret);
	387	}
	388
3be3f61d	389	pr_debug("Loaded purgatory at 0x%lx\n", pbuf.mem);
27f48d3e	390
6a2c20e7 VG	391
	392	/*
	393	* Load Bootparams and cmdline and space for efi stuff.
	394	*
	395	* Allocate memory together for multiple data structures so
	396	* that they all can go in single area/segment and we don't
	397	* have to create separate segment for each. Keeps things
	398	* little bit simple
	399	*/
	400	efi_map_sz = efi_get_runtime_map_size();
dd5f7260 VG	401	params_cmdline_sz = sizeof(struct boot_params) + cmdline_len +
dd5f7260 VG	402	MAX_ELFCOREHDR_STR_LEN;
6a2c20e7	403	params_cmdline_sz = ALIGN(params_cmdline_sz, 16);
a841aa83	404	kbuf.bufsz = params_cmdline_sz + ALIGN(efi_map_sz, 16) +
6a2c20e7 VG	405	sizeof(struct setup_data) +
	406	sizeof(struct efi_setup_data);
	407
ec2b9bfa	408	params = kzalloc(kbuf.bufsz, GFP_KERNEL);
27f48d3e VG	409	if (!params)
27f48d3e VG	410	return ERR_PTR(-ENOMEM);
6a2c20e7	411	efi_map_offset = params_cmdline_sz;
a841aa83	412	efi_setup_data_offset = efi_map_offset + ALIGN(efi_map_sz, 16);
27f48d3e VG	413
	414	/* Copy setup header onto bootparams. Documentation/x86/boot.txt */
	415	setup_header_size = 0x0202 + kernel[0x0201] - setup_hdr_offset;
	416
	417	/* Is there a limit on setup header size? */
	418	memcpy(&params->hdr, (kernel + setup_hdr_offset), setup_header_size);
	419
ec2b9bfa TJB	420	kbuf.buffer = params;
	421	kbuf.memsz = kbuf.bufsz;
	422	kbuf.buf_align = 16;
	423	kbuf.buf_min = MIN_BOOTPARAM_ADDR;
	424	ret = kexec_add_buffer(&kbuf);
27f48d3e VG	425	if (ret)
27f48d3e VG	426	goto out_free_params;
ec2b9bfa	427	bootparam_load_addr = kbuf.mem;
6a2c20e7	428	pr_debug("Loaded boot_param, command line and misc at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
ec2b9bfa	429	bootparam_load_addr, kbuf.bufsz, kbuf.bufsz);
27f48d3e VG	430
27f48d3e VG	431	/* Load kernel */
ec2b9bfa TJB	432	kbuf.buffer = kernel + kern16_size;
	433	kbuf.bufsz = kernel_len - kern16_size;
	434	kbuf.memsz = PAGE_ALIGN(header->init_size);
	435	kbuf.buf_align = header->kernel_alignment;
	436	kbuf.buf_min = MIN_KERNEL_LOAD_ADDR;
	437	ret = kexec_add_buffer(&kbuf);
27f48d3e VG	438	if (ret)
27f48d3e VG	439	goto out_free_params;
ec2b9bfa	440	kernel_load_addr = kbuf.mem;
27f48d3e VG	441
27f48d3e VG	442	pr_debug("Loaded 64bit kernel at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
ec2b9bfa	443	kernel_load_addr, kbuf.bufsz, kbuf.memsz);
27f48d3e VG	444
	445	/* Load initrd high */
	446	if (initrd) {
ec2b9bfa TJB	447	kbuf.buffer = initrd;
	448	kbuf.bufsz = kbuf.memsz = initrd_len;
	449	kbuf.buf_align = PAGE_SIZE;
	450	kbuf.buf_min = MIN_INITRD_LOAD_ADDR;
	451	ret = kexec_add_buffer(&kbuf);
27f48d3e VG	452	if (ret)
27f48d3e VG	453	goto out_free_params;
ec2b9bfa	454	initrd_load_addr = kbuf.mem;
27f48d3e VG	455
	456	pr_debug("Loaded initrd at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
	457	initrd_load_addr, initrd_len, initrd_len);
	458
	459	setup_initrd(params, initrd_load_addr, initrd_len);
	460	}
	461
dd5f7260 VG	462	setup_cmdline(image, params, bootparam_load_addr,
dd5f7260 VG	463	sizeof(struct boot_params), cmdline, cmdline_len);
27f48d3e VG	464
	465	/* bootloader info. Do we need a separate ID for kexec kernel loader? */
	466	params->hdr.type_of_loader = 0x0D << 4;
	467	params->hdr.loadflags = 0;
	468
	469	/* Setup purgatory regs for entry */
	470	ret = kexec_purgatory_get_set_symbol(image, "entry64_regs", &regs64,
	471	sizeof(regs64), 1);
	472	if (ret)
	473	goto out_free_params;
	474
	475	regs64.rbx = 0; /* Bootstrap Processor */
	476	regs64.rsi = bootparam_load_addr;
	477	regs64.rip = kernel_load_addr + 0x200;
	478	stack = kexec_purgatory_get_symbol_addr(image, "stack_end");
	479	if (IS_ERR(stack)) {
	480	pr_err("Could not find address of symbol stack_end\n");
	481	ret = -EINVAL;
	482	goto out_free_params;
	483	}
	484
	485	regs64.rsp = (unsigned long)stack;
	486	ret = kexec_purgatory_get_set_symbol(image, "entry64_regs", &regs64,
	487	sizeof(regs64), 0);
	488	if (ret)
	489	goto out_free_params;
	490
6a2c20e7 VG	491	ret = setup_boot_parameters(image, params, bootparam_load_addr,
	492	efi_map_offset, efi_map_sz,
	493	efi_setup_data_offset);
dd5f7260 VG	494	if (ret)
dd5f7260 VG	495	goto out_free_params;
27f48d3e VG	496
	497	/* Allocate loader specific data */
	498	ldata = kzalloc(sizeof(struct bzimage64_data), GFP_KERNEL);
	499	if (!ldata) {
	500	ret = -ENOMEM;
	501	goto out_free_params;
	502	}
	503
	504	/*
	505	* Store pointer to params so that it could be freed after loading
	506	* params segment has been loaded and contents have been copied
	507	* somewhere else.
	508	*/
	509	ldata->bootparams_buf = params;
	510	return ldata;
	511
	512	out_free_params:
	513	kfree(params);
	514	return ERR_PTR(ret);
	515	}
	516
	517	/* This cleanup function is called after various segments have been loaded */
f8da964d	518	static int bzImage64_cleanup(void *loader_data)
27f48d3e VG	519	{
	520	struct bzimage64_data *ldata = loader_data;
	521
	522	if (!ldata)
	523	return 0;
	524
	525	kfree(ldata->bootparams_buf);
	526	ldata->bootparams_buf = NULL;
	527
	528	return 0;
	529	}
	530
8e7d8381	531	#ifdef CONFIG_KEXEC_BZIMAGE_VERIFY_SIG
f8da964d	532	static int bzImage64_verify_sig(const char *kernel, unsigned long kernel_len)
8e7d8381	533	{
e68503bd DH	534	return verify_pefile_signature(kernel, kernel_len,
	535	NULL,
	536	VERIFYING_KEXEC_PE_SIGNATURE);
8e7d8381 VG	537	}
	538	#endif
	539
9ec4ecef	540	const struct kexec_file_ops kexec_bzImage64_ops = {
27f48d3e VG	541	.probe = bzImage64_probe,
	542	.load = bzImage64_load,
	543	.cleanup = bzImage64_cleanup,
8e7d8381 VG	544	#ifdef CONFIG_KEXEC_BZIMAGE_VERIFY_SIG
	545	.verify_sig = bzImage64_verify_sig,
	546	#endif
27f48d3e	547	};