[mirror_ubuntu-zesty-kernel.git] / arch / x86 / kernel / crash.c

/*
 * Architecture specific (i386/x86_64) functions for kexec based crash dumps.
 *
 * Created by: Hariprasad Nellitheertha (hari@in.ibm.com)
 *
 * Copyright (C) IBM Corporation, 2004. All rights reserved.
 * Copyright (C) Red Hat Inc., 2014. All rights reserved.
 * Authors:
 *      Vivek Goyal <vgoyal@redhat.com>
 *
 */

#define pr_fmt(fmt)	"kexec: " fmt

#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/smp.h>
#include <linux/reboot.h>
#include <linux/kexec.h>
#include <linux/delay.h>
#include <linux/elf.h>
#include <linux/elfcore.h>
#include <linux/export.h>
#include <linux/slab.h>
#include <linux/vmalloc.h>

#include <asm/processor.h>
#include <asm/hardirq.h>
#include <asm/nmi.h>
#include <asm/hw_irq.h>
#include <asm/apic.h>
#include <asm/io_apic.h>
#include <asm/hpet.h>
#include <linux/kdebug.h>
#include <asm/cpu.h>
#include <asm/reboot.h>
#include <asm/virtext.h>
#include <asm/intel_pt.h>

/* Alignment required for elf header segment */
#define ELF_CORE_HEADER_ALIGN   4096

/* This primarily represents number of split ranges due to exclusion */
#define CRASH_MAX_RANGES	16

struct crash_mem_range {
	u64 start, end;
};

struct crash_mem {
	unsigned int nr_ranges;
	struct crash_mem_range ranges[CRASH_MAX_RANGES];
};

/* Misc data about ram ranges needed to prepare elf headers */
struct crash_elf_data {
	struct kimage *image;
	/*
	 * Total number of ram ranges we have after various adjustments for
	 * crash reserved region, etc.
	 */
	unsigned int max_nr_ranges;

	/* Pointer to elf header */
	void *ehdr;
	/* Pointer to next phdr */
	void *bufp;
	struct crash_mem mem;
};

/* Used while preparing memory map entries for second kernel */
struct crash_memmap_data {
	struct boot_params *params;
	/* Type of memory */
	unsigned int type;
};

/*
 * This is used to VMCLEAR all VMCSs loaded on the
 * processor. And when loading kvm_intel module, the
 * callback function pointer will be assigned.
 *
 * protected by rcu.
 */
crash_vmclear_fn __rcu *crash_vmclear_loaded_vmcss = NULL;
EXPORT_SYMBOL_GPL(crash_vmclear_loaded_vmcss);
unsigned long crash_zero_bytes;

static inline void cpu_crash_vmclear_loaded_vmcss(void)
{
	crash_vmclear_fn *do_vmclear_operation = NULL;

	rcu_read_lock();
	do_vmclear_operation = rcu_dereference(crash_vmclear_loaded_vmcss);
	if (do_vmclear_operation)
		do_vmclear_operation();
	rcu_read_unlock();
}

#if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)

static void kdump_nmi_callback(int cpu, struct pt_regs *regs)
{
#ifdef CONFIG_X86_32
	struct pt_regs fixed_regs;

	if (!user_mode(regs)) {
		crash_fixup_ss_esp(&fixed_regs, regs);
		regs = &fixed_regs;
	}
#endif
	crash_save_cpu(regs, cpu);

	/*
	 * VMCLEAR VMCSs loaded on all cpus if needed.
	 */
	cpu_crash_vmclear_loaded_vmcss();

	/* Disable VMX or SVM if needed.
	 *
	 * We need to disable virtualization on all CPUs.
	 * Having VMX or SVM enabled on any CPU may break rebooting
	 * after the kdump kernel has finished its task.
	 */
	cpu_emergency_vmxoff();
	cpu_emergency_svm_disable();

	/*
	 * Disable Intel PT to stop its logging
	 */
	cpu_emergency_stop_pt();

	disable_local_APIC();
}

void kdump_nmi_shootdown_cpus(void)
{
	nmi_shootdown_cpus(kdump_nmi_callback);

	disable_local_APIC();
}

/* Override the weak function in kernel/panic.c */
void crash_smp_send_stop(void)
{
	static int cpus_stopped;

	if (cpus_stopped)
		return;

	if (smp_ops.crash_stop_other_cpus)
		smp_ops.crash_stop_other_cpus();
	else
		smp_send_stop();

	cpus_stopped = 1;
}

#else
void crash_smp_send_stop(void)
{
	/* There are no cpus to shootdown */
}
#endif

void native_machine_crash_shutdown(struct pt_regs *regs)
{
	/* This function is only called after the system
	 * has panicked or is otherwise in a critical state.
	 * The minimum amount of code to allow a kexec'd kernel
	 * to run successfully needs to happen here.
	 *
	 * In practice this means shooting down the other cpus in
	 * an SMP system.
	 */
	/* The kernel is broken so disable interrupts */
	local_irq_disable();

	crash_smp_send_stop();

	/*
	 * VMCLEAR VMCSs loaded on this cpu if needed.
	 */
	cpu_crash_vmclear_loaded_vmcss();

	/* Booting kdump kernel with VMX or SVM enabled won't work,
	 * because (among other limitations) we can't disable paging
	 * with the virt flags.
	 */
	cpu_emergency_vmxoff();
	cpu_emergency_svm_disable();

	/*
	 * Disable Intel PT to stop its logging
	 */
	cpu_emergency_stop_pt();

#ifdef CONFIG_X86_IO_APIC
	/* Prevent crash_kexec() from deadlocking on ioapic_lock. */
	ioapic_zap_locks();
	disable_IO_APIC();
#endif
	lapic_shutdown();
#ifdef CONFIG_HPET_TIMER
	hpet_disable();
#endif
	crash_save_cpu(regs, safe_smp_processor_id());
}

#ifdef CONFIG_KEXEC_FILE
static int get_nr_ram_ranges_callback(u64 start, u64 end, void *arg)
{
	unsigned int *nr_ranges = arg;

	(*nr_ranges)++;
	return 0;
}


/* Gather all the required information to prepare elf headers for ram regions */
static void fill_up_crash_elf_data(struct crash_elf_data *ced,
				   struct kimage *image)
{
	unsigned int nr_ranges = 0;

	ced->image = image;

	walk_system_ram_res(0, -1, &nr_ranges,
				get_nr_ram_ranges_callback);

	ced->max_nr_ranges = nr_ranges;

	/* Exclusion of crash region could split memory ranges */
	ced->max_nr_ranges++;

	/* If crashk_low_res is not 0, another range split possible */
	if (crashk_low_res.end)
		ced->max_nr_ranges++;
}

static int exclude_mem_range(struct crash_mem *mem,
		unsigned long long mstart, unsigned long long mend)
{
	int i, j;
	unsigned long long start, end;
	struct crash_mem_range temp_range = {0, 0};

	for (i = 0; i < mem->nr_ranges; i++) {
		start = mem->ranges[i].start;
		end = mem->ranges[i].end;

		if (mstart > end || mend < start)
			continue;

		/* Truncate any area outside of range */
		if (mstart < start)
			mstart = start;
		if (mend > end)
			mend = end;

		/* Found completely overlapping range */
		if (mstart == start && mend == end) {
			mem->ranges[i].start = 0;
			mem->ranges[i].end = 0;
			if (i < mem->nr_ranges - 1) {
				/* Shift rest of the ranges to left */
				for (j = i; j < mem->nr_ranges - 1; j++) {
					mem->ranges[j].start =
						mem->ranges[j+1].start;
					mem->ranges[j].end =
							mem->ranges[j+1].end;
				}
			}
			mem->nr_ranges--;
			return 0;
		}

		if (mstart > start && mend < end) {
			/* Split original range */
			mem->ranges[i].end = mstart - 1;
			temp_range.start = mend + 1;
			temp_range.end = end;
		} else if (mstart != start)
			mem->ranges[i].end = mstart - 1;
		else
			mem->ranges[i].start = mend + 1;
		break;
	}

	/* If a split happend, add the split to array */
	if (!temp_range.end)
		return 0;

	/* Split happened */
	if (i == CRASH_MAX_RANGES - 1) {
		pr_err("Too many crash ranges after split\n");
		return -ENOMEM;
	}

	/* Location where new range should go */
	j = i + 1;
	if (j < mem->nr_ranges) {
		/* Move over all ranges one slot towards the end */
		for (i = mem->nr_ranges - 1; i >= j; i--)
			mem->ranges[i + 1] = mem->ranges[i];
	}

	mem->ranges[j].start = temp_range.start;
	mem->ranges[j].end = temp_range.end;
	mem->nr_ranges++;
	return 0;
}

/*
 * Look for any unwanted ranges between mstart, mend and remove them. This
 * might lead to split and split ranges are put in ced->mem.ranges[] array
 */
static int elf_header_exclude_ranges(struct crash_elf_data *ced,
		unsigned long long mstart, unsigned long long mend)
{
	struct crash_mem *cmem = &ced->mem;
	int ret = 0;

	memset(cmem->ranges, 0, sizeof(cmem->ranges));

	cmem->ranges[0].start = mstart;
	cmem->ranges[0].end = mend;
	cmem->nr_ranges = 1;

	/* Exclude crashkernel region */
	ret = exclude_mem_range(cmem, crashk_res.start, crashk_res.end);
	if (ret)
		return ret;

	if (crashk_low_res.end) {
		ret = exclude_mem_range(cmem, crashk_low_res.start, crashk_low_res.end);
		if (ret)
			return ret;
	}

	return ret;
}

static int prepare_elf64_ram_headers_callback(u64 start, u64 end, void *arg)
{
	struct crash_elf_data *ced = arg;
	Elf64_Ehdr *ehdr;
	Elf64_Phdr *phdr;
	unsigned long mstart, mend;
	struct kimage *image = ced->image;
	struct crash_mem *cmem;
	int ret, i;

	ehdr = ced->ehdr;

	/* Exclude unwanted mem ranges */
	ret = elf_header_exclude_ranges(ced, start, end);
	if (ret)
		return ret;

	/* Go through all the ranges in ced->mem.ranges[] and prepare phdr */
	cmem = &ced->mem;

	for (i = 0; i < cmem->nr_ranges; i++) {
		mstart = cmem->ranges[i].start;
		mend = cmem->ranges[i].end;

		phdr = ced->bufp;
		ced->bufp += sizeof(Elf64_Phdr);

		phdr->p_type = PT_LOAD;
		phdr->p_flags = PF_R|PF_W|PF_X;
		phdr->p_offset  = mstart;

		/*
		 * If a range matches backup region, adjust offset to backup
		 * segment.
		 */
		if (mstart == image->arch.backup_src_start &&
		    (mend - mstart + 1) == image->arch.backup_src_sz)
			phdr->p_offset = image->arch.backup_load_addr;

		phdr->p_paddr = mstart;
		phdr->p_vaddr = (unsigned long long) __va(mstart);
		phdr->p_filesz = phdr->p_memsz = mend - mstart + 1;
		phdr->p_align = 0;
		ehdr->e_phnum++;
		pr_debug("Crash PT_LOAD elf header. phdr=%p vaddr=0x%llx, paddr=0x%llx, sz=0x%llx e_phnum=%d p_offset=0x%llx\n",
			phdr, phdr->p_vaddr, phdr->p_paddr, phdr->p_filesz,
			ehdr->e_phnum, phdr->p_offset);
	}

	return ret;
}

static int prepare_elf64_headers(struct crash_elf_data *ced,
		void **addr, unsigned long *sz)
{
	Elf64_Ehdr *ehdr;
	Elf64_Phdr *phdr;
	unsigned long nr_cpus = num_possible_cpus(), nr_phdr, elf_sz;
	unsigned char *buf, *bufp;
	unsigned int cpu;
	unsigned long long notes_addr;
	int ret;

	/* extra phdr for vmcoreinfo elf note */
	nr_phdr = nr_cpus + 1;
	nr_phdr += ced->max_nr_ranges;

	/*
	 * kexec-tools creates an extra PT_LOAD phdr for kernel text mapping
	 * area on x86_64 (ffffffff80000000 - ffffffffa0000000).
	 * I think this is required by tools like gdb. So same physical
	 * memory will be mapped in two elf headers. One will contain kernel
	 * text virtual addresses and other will have __va(physical) addresses.
	 */

	nr_phdr++;
	elf_sz = sizeof(Elf64_Ehdr) + nr_phdr * sizeof(Elf64_Phdr);
	elf_sz = ALIGN(elf_sz, ELF_CORE_HEADER_ALIGN);

	buf = vzalloc(elf_sz);
	if (!buf)
		return -ENOMEM;

	bufp = buf;
	ehdr = (Elf64_Ehdr *)bufp;
	bufp += sizeof(Elf64_Ehdr);
	memcpy(ehdr->e_ident, ELFMAG, SELFMAG);
	ehdr->e_ident[EI_CLASS] = ELFCLASS64;
	ehdr->e_ident[EI_DATA] = ELFDATA2LSB;
	ehdr->e_ident[EI_VERSION] = EV_CURRENT;
	ehdr->e_ident[EI_OSABI] = ELF_OSABI;
	memset(ehdr->e_ident + EI_PAD, 0, EI_NIDENT - EI_PAD);
	ehdr->e_type = ET_CORE;
	ehdr->e_machine = ELF_ARCH;
	ehdr->e_version = EV_CURRENT;
	ehdr->e_phoff = sizeof(Elf64_Ehdr);
	ehdr->e_ehsize = sizeof(Elf64_Ehdr);
	ehdr->e_phentsize = sizeof(Elf64_Phdr);

	/* Prepare one phdr of type PT_NOTE for each present cpu */
	for_each_present_cpu(cpu) {
		phdr = (Elf64_Phdr *)bufp;
		bufp += sizeof(Elf64_Phdr);
		phdr->p_type = PT_NOTE;
		notes_addr = per_cpu_ptr_to_phys(per_cpu_ptr(crash_notes, cpu));
		phdr->p_offset = phdr->p_paddr = notes_addr;
		phdr->p_filesz = phdr->p_memsz = sizeof(note_buf_t);
		(ehdr->e_phnum)++;
	}

	/* Prepare one PT_NOTE header for vmcoreinfo */
	phdr = (Elf64_Phdr *)bufp;
	bufp += sizeof(Elf64_Phdr);
	phdr->p_type = PT_NOTE;
	phdr->p_offset = phdr->p_paddr = paddr_vmcoreinfo_note();
	phdr->p_filesz = phdr->p_memsz = sizeof(vmcoreinfo_note);
	(ehdr->e_phnum)++;

#ifdef CONFIG_X86_64
	/* Prepare PT_LOAD type program header for kernel text region */
	phdr = (Elf64_Phdr *)bufp;
	bufp += sizeof(Elf64_Phdr);
	phdr->p_type = PT_LOAD;
	phdr->p_flags = PF_R|PF_W|PF_X;
	phdr->p_vaddr = (Elf64_Addr)_text;
	phdr->p_filesz = phdr->p_memsz = _end - _text;
	phdr->p_offset = phdr->p_paddr = __pa_symbol(_text);
	(ehdr->e_phnum)++;
#endif

	/* Prepare PT_LOAD headers for system ram chunks. */
	ced->ehdr = ehdr;
	ced->bufp = bufp;
	ret = walk_system_ram_res(0, -1, ced,
			prepare_elf64_ram_headers_callback);
	if (ret < 0)
		return ret;

	*addr = buf;
	*sz = elf_sz;
	return 0;
}

/* Prepare elf headers. Return addr and size */
static int prepare_elf_headers(struct kimage *image, void **addr,
					unsigned long *sz)
{
	struct crash_elf_data *ced;
	int ret;

	ced = kzalloc(sizeof(*ced), GFP_KERNEL);
	if (!ced)
		return -ENOMEM;

	fill_up_crash_elf_data(ced, image);

	/* By default prepare 64bit headers */
	ret =  prepare_elf64_headers(ced, addr, sz);
	kfree(ced);
	return ret;
}

static int add_e820_entry(struct boot_params *params, struct e820entry *entry)
{
	unsigned int nr_e820_entries;

	nr_e820_entries = params->e820_entries;
	if (nr_e820_entries >= E820MAX)
		return 1;

	memcpy(&params->e820_map[nr_e820_entries], entry,
			sizeof(struct e820entry));
	params->e820_entries++;
	return 0;
}

static int memmap_entry_callback(u64 start, u64 end, void *arg)
{
	struct crash_memmap_data *cmd = arg;
	struct boot_params *params = cmd->params;
	struct e820entry ei;

	ei.addr = start;
	ei.size = end - start + 1;
	ei.type = cmd->type;
	add_e820_entry(params, &ei);

	return 0;
}

static int memmap_exclude_ranges(struct kimage *image, struct crash_mem *cmem,
				 unsigned long long mstart,
				 unsigned long long mend)
{
	unsigned long start, end;
	int ret = 0;

	cmem->ranges[0].start = mstart;
	cmem->ranges[0].end = mend;
	cmem->nr_ranges = 1;

	/* Exclude Backup region */
	start = image->arch.backup_load_addr;
	end = start + image->arch.backup_src_sz - 1;
	ret = exclude_mem_range(cmem, start, end);
	if (ret)
		return ret;

	/* Exclude elf header region */
	start = image->arch.elf_load_addr;
	end = start + image->arch.elf_headers_sz - 1;
	return exclude_mem_range(cmem, start, end);
}

/* Prepare memory map for crash dump kernel */
int crash_setup_memmap_entries(struct kimage *image, struct boot_params *params)
{
	int i, ret = 0;
	unsigned long flags;
	struct e820entry ei;
	struct crash_memmap_data cmd;
	struct crash_mem *cmem;

	cmem = vzalloc(sizeof(struct crash_mem));
	if (!cmem)
		return -ENOMEM;

	memset(&cmd, 0, sizeof(struct crash_memmap_data));
	cmd.params = params;

	/* Add first 640K segment */
	ei.addr = image->arch.backup_src_start;
	ei.size = image->arch.backup_src_sz;
	ei.type = E820_RAM;
	add_e820_entry(params, &ei);

	/* Add ACPI tables */
	cmd.type = E820_ACPI;
	flags = IORESOURCE_MEM | IORESOURCE_BUSY;
	walk_iomem_res_desc(IORES_DESC_ACPI_TABLES, flags, 0, -1, &cmd,
		       memmap_entry_callback);

	/* Add ACPI Non-volatile Storage */
	cmd.type = E820_NVS;
	walk_iomem_res_desc(IORES_DESC_ACPI_NV_STORAGE, flags, 0, -1, &cmd,
			memmap_entry_callback);

	/* Add crashk_low_res region */
	if (crashk_low_res.end) {
		ei.addr = crashk_low_res.start;
		ei.size = crashk_low_res.end - crashk_low_res.start + 1;
		ei.type = E820_RAM;
		add_e820_entry(params, &ei);
	}

	/* Exclude some ranges from crashk_res and add rest to memmap */
	ret = memmap_exclude_ranges(image, cmem, crashk_res.start,
						crashk_res.end);
	if (ret)
		goto out;

	for (i = 0; i < cmem->nr_ranges; i++) {
		ei.size = cmem->ranges[i].end - cmem->ranges[i].start + 1;

		/* If entry is less than a page, skip it */
		if (ei.size < PAGE_SIZE)
			continue;
		ei.addr = cmem->ranges[i].start;
		ei.type = E820_RAM;
		add_e820_entry(params, &ei);
	}

out:
	vfree(cmem);
	return ret;
}

static int determine_backup_region(u64 start, u64 end, void *arg)
{
	struct kimage *image = arg;

	image->arch.backup_src_start = start;
	image->arch.backup_src_sz = end - start + 1;

	/* Expecting only one range for backup region */
	return 1;
}

int crash_load_segments(struct kimage *image)
{
	unsigned long src_start, src_sz, elf_sz;
	void *elf_addr;
	int ret;

	/*
	 * Determine and load a segment for backup area. First 640K RAM
	 * region is backup source
	 */

	ret = walk_system_ram_res(KEXEC_BACKUP_SRC_START, KEXEC_BACKUP_SRC_END,
				image, determine_backup_region);

	/* Zero or postive return values are ok */
	if (ret < 0)
		return ret;

	src_start = image->arch.backup_src_start;
	src_sz = image->arch.backup_src_sz;

	/* Add backup segment. */
	if (src_sz) {
		/*
		 * Ideally there is no source for backup segment. This is
		 * copied in purgatory after crash. Just add a zero filled
		 * segment for now to make sure checksum logic works fine.
		 */
		ret = kexec_add_buffer(image, (char *)&crash_zero_bytes,
				       sizeof(crash_zero_bytes), src_sz,
				       PAGE_SIZE, 0, -1, 0,
				       &image->arch.backup_load_addr);
		if (ret)
			return ret;
		pr_debug("Loaded backup region at 0x%lx backup_start=0x%lx memsz=0x%lx\n",
			 image->arch.backup_load_addr, src_start, src_sz);
	}

	/* Prepare elf headers and add a segment */
	ret = prepare_elf_headers(image, &elf_addr, &elf_sz);
	if (ret)
		return ret;

	image->arch.elf_headers = elf_addr;
	image->arch.elf_headers_sz = elf_sz;

	ret = kexec_add_buffer(image, (char *)elf_addr, elf_sz, elf_sz,
			ELF_CORE_HEADER_ALIGN, 0, -1, 0,
			&image->arch.elf_load_addr);
	if (ret) {
		vfree((void *)image->arch.elf_headers);
		return ret;
	}
	pr_debug("Loaded ELF headers at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
		 image->arch.elf_load_addr, elf_sz, elf_sz);

	return ret;
}
#endif /* CONFIG_KEXEC_FILE */
Commit	Line	Data
5033cba0	1	/*
62a31a03	2	* Architecture specific (i386/x86_64) functions for kexec based crash dumps.
5033cba0 EB	3	*
	4	* Created by: Hariprasad Nellitheertha (hari@in.ibm.com)
	5	*
	6	* Copyright (C) IBM Corporation, 2004. All rights reserved.
dd5f7260 VG	7	* Copyright (C) Red Hat Inc., 2014. All rights reserved.
	8	* Authors:
	9	* Vivek Goyal <vgoyal@redhat.com>
5033cba0 EB	10	*
	11	*/
	12
dd5f7260 VG	13	#define pr_fmt(fmt) "kexec: " fmt
dd5f7260 VG	14
5033cba0 EB	15	#include <linux/types.h>
	16	#include <linux/kernel.h>
	17	#include <linux/smp.h>
5033cba0 EB	18	#include <linux/reboot.h>
5033cba0 EB	19	#include <linux/kexec.h>
5033cba0 EB	20	#include <linux/delay.h>
	21	#include <linux/elf.h>
	22	#include <linux/elfcore.h>
186f4360	23	#include <linux/export.h>
dd5f7260	24	#include <linux/slab.h>
d6472302	25	#include <linux/vmalloc.h>
5033cba0 EB	26
	27	#include <asm/processor.h>
	28	#include <asm/hardirq.h>
	29	#include <asm/nmi.h>
	30	#include <asm/hw_irq.h>
19842d67	31	#include <asm/apic.h>
8643e28d	32	#include <asm/io_apic.h>
0c1b2724	33	#include <asm/hpet.h>
1eeb66a1	34	#include <linux/kdebug.h>
96b89dc6	35	#include <asm/cpu.h>
ed23dc6f	36	#include <asm/reboot.h>
2340b62f	37	#include <asm/virtext.h>
da06a43d	38	#include <asm/intel_pt.h>
8e294786	39
dd5f7260 VG	40	/* Alignment required for elf header segment */
	41	#define ELF_CORE_HEADER_ALIGN 4096
	42
	43	/* This primarily represents number of split ranges due to exclusion */
	44	#define CRASH_MAX_RANGES 16
	45
	46	struct crash_mem_range {
	47	u64 start, end;
	48	};
	49
	50	struct crash_mem {
	51	unsigned int nr_ranges;
	52	struct crash_mem_range ranges[CRASH_MAX_RANGES];
	53	};
	54
	55	/* Misc data about ram ranges needed to prepare elf headers */
	56	struct crash_elf_data {
	57	struct kimage *image;
	58	/*
	59	* Total number of ram ranges we have after various adjustments for
f296f263	60	* crash reserved region, etc.
dd5f7260 VG	61	*/
dd5f7260 VG	62	unsigned int max_nr_ranges;
dd5f7260 VG	63
	64	/* Pointer to elf header */
	65	void *ehdr;
	66	/* Pointer to next phdr */
	67	void *bufp;
	68	struct crash_mem mem;
	69	};
	70
	71	/* Used while preparing memory map entries for second kernel */
	72	struct crash_memmap_data {
	73	struct boot_params *params;
	74	/* Type of memory */
	75	unsigned int type;
	76	};
	77
f23d1f4a ZY	78	/*
	79	* This is used to VMCLEAR all VMCSs loaded on the
	80	* processor. And when loading kvm_intel module, the
	81	* callback function pointer will be assigned.
	82	*
	83	* protected by rcu.
	84	*/
0ca0d818	85	crash_vmclear_fn __rcu *crash_vmclear_loaded_vmcss = NULL;
f23d1f4a	86	EXPORT_SYMBOL_GPL(crash_vmclear_loaded_vmcss);
dd5f7260	87	unsigned long crash_zero_bytes;
f23d1f4a ZY	88
	89	static inline void cpu_crash_vmclear_loaded_vmcss(void)
	90	{
0ca0d818	91	crash_vmclear_fn *do_vmclear_operation = NULL;
f23d1f4a ZY	92
	93	rcu_read_lock();
	94	do_vmclear_operation = rcu_dereference(crash_vmclear_loaded_vmcss);
	95	if (do_vmclear_operation)
	96	do_vmclear_operation();
	97	rcu_read_unlock();
	98	}
	99
b2bbe71b EH	100	#if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
b2bbe71b EH	101
9c48f1c6	102	static void kdump_nmi_callback(int cpu, struct pt_regs *regs)
c4ac4263	103	{
1fb473d8	104	#ifdef CONFIG_X86_32
4d55476c	105	struct pt_regs fixed_regs;
a7d41820	106
f39b6f0e	107	if (!user_mode(regs)) {
a7d41820 EH	108	crash_fixup_ss_esp(&fixed_regs, regs);
	109	regs = &fixed_regs;
	110	}
	111	#endif
	112	crash_save_cpu(regs, cpu);
	113
f23d1f4a ZY	114	/*
	115	* VMCLEAR VMCSs loaded on all cpus if needed.
	116	*/
	117	cpu_crash_vmclear_loaded_vmcss();
	118
2340b62f EH	119	/* Disable VMX or SVM if needed.
	120	*
	121	* We need to disable virtualization on all CPUs.
	122	* Having VMX or SVM enabled on any CPU may break rebooting
	123	* after the kdump kernel has finished its task.
	124	*/
	125	cpu_emergency_vmxoff();
	126	cpu_emergency_svm_disable();
	127
da06a43d TI	128	/*
	129	* Disable Intel PT to stop its logging
	130	*/
	131	cpu_emergency_stop_pt();
	132
a7d41820 EH	133	disable_local_APIC();
	134	}
	135
0ee59413	136	void kdump_nmi_shootdown_cpus(void)
d1e7b91c	137	{
8e294786	138	nmi_shootdown_cpus(kdump_nmi_callback);
d1e7b91c	139
19842d67	140	disable_local_APIC();
c4ac4263	141	}
d1e7b91c	142
0ee59413 HK	143	/* Override the weak function in kernel/panic.c */
	144	void crash_smp_send_stop(void)
	145	{
	146	static int cpus_stopped;
	147
	148	if (cpus_stopped)
	149	return;
	150
	151	if (smp_ops.crash_stop_other_cpus)
	152	smp_ops.crash_stop_other_cpus();
	153	else
	154	smp_send_stop();
	155
	156	cpus_stopped = 1;
	157	}
	158
c4ac4263	159	#else
0ee59413	160	void crash_smp_send_stop(void)
c4ac4263 EB	161	{
	162	/* There are no cpus to shootdown */
	163	}
	164	#endif
	165
ed23dc6f	166	void native_machine_crash_shutdown(struct pt_regs *regs)
5033cba0 EB	167	{
5033cba0 EB	168	/* This function is only called after the system
f18190bd	169	* has panicked or is otherwise in a critical state.
5033cba0 EB	170	* The minimum amount of code to allow a kexec'd kernel
	171	* to run successfully needs to happen here.
	172	*
	173	* In practice this means shooting down the other cpus in
	174	* an SMP system.
	175	*/
c4ac4263 EB	176	/* The kernel is broken so disable interrupts */
c4ac4263 EB	177	local_irq_disable();
a3ea8ac8	178
0ee59413	179	crash_smp_send_stop();
2340b62f	180
f23d1f4a ZY	181	/*
	182	* VMCLEAR VMCSs loaded on this cpu if needed.
	183	*/
	184	cpu_crash_vmclear_loaded_vmcss();
	185
2340b62f EH	186	/* Booting kdump kernel with VMX or SVM enabled won't work,
	187	* because (among other limitations) we can't disable paging
	188	* with the virt flags.
	189	*/
	190	cpu_emergency_vmxoff();
	191	cpu_emergency_svm_disable();
	192
da06a43d TI	193	/*
	194	* Disable Intel PT to stop its logging
	195	*/
	196	cpu_emergency_stop_pt();
	197
17405453 YY	198	#ifdef CONFIG_X86_IO_APIC
	199	/* Prevent crash_kexec() from deadlocking on ioapic_lock. */
	200	ioapic_zap_locks();
19842d67	201	disable_IO_APIC();
0c1b2724	202	#endif
522e6646	203	lapic_shutdown();
0c1b2724 OH	204	#ifdef CONFIG_HPET_TIMER
0c1b2724 OH	205	hpet_disable();
19842d67	206	#endif
85916f81	207	crash_save_cpu(regs, safe_smp_processor_id());
5033cba0	208	}
dd5f7260	209
74ca317c	210	#ifdef CONFIG_KEXEC_FILE
e3c41e37	211	static int get_nr_ram_ranges_callback(u64 start, u64 end, void *arg)
dd5f7260	212	{
e3c41e37	213	unsigned int *nr_ranges = arg;
dd5f7260 VG	214
	215	(*nr_ranges)++;
	216	return 0;
	217	}
	218
dd5f7260 VG	219
	220	/* Gather all the required information to prepare elf headers for ram regions */
	221	static void fill_up_crash_elf_data(struct crash_elf_data *ced,
	222	struct kimage *image)
	223	{
	224	unsigned int nr_ranges = 0;
	225
	226	ced->image = image;
	227
e3c41e37	228	walk_system_ram_res(0, -1, &nr_ranges,
dd5f7260 VG	229	get_nr_ram_ranges_callback);
	230
	231	ced->max_nr_ranges = nr_ranges;
	232
dd5f7260 VG	233	/* Exclusion of crash region could split memory ranges */
	234	ced->max_nr_ranges++;
	235
	236	/* If crashk_low_res is not 0, another range split possible */
a2d6aa8f	237	if (crashk_low_res.end)
dd5f7260 VG	238	ced->max_nr_ranges++;
	239	}
	240
	241	static int exclude_mem_range(struct crash_mem *mem,
	242	unsigned long long mstart, unsigned long long mend)
	243	{
	244	int i, j;
	245	unsigned long long start, end;
	246	struct crash_mem_range temp_range = {0, 0};
	247
	248	for (i = 0; i < mem->nr_ranges; i++) {
	249	start = mem->ranges[i].start;
	250	end = mem->ranges[i].end;
	251
	252	if (mstart > end \|\| mend < start)
	253	continue;
	254
	255	/* Truncate any area outside of range */
	256	if (mstart < start)
	257	mstart = start;
	258	if (mend > end)
	259	mend = end;
	260
	261	/* Found completely overlapping range */
	262	if (mstart == start && mend == end) {
	263	mem->ranges[i].start = 0;
	264	mem->ranges[i].end = 0;
	265	if (i < mem->nr_ranges - 1) {
	266	/* Shift rest of the ranges to left */
	267	for (j = i; j < mem->nr_ranges - 1; j++) {
	268	mem->ranges[j].start =
	269	mem->ranges[j+1].start;
	270	mem->ranges[j].end =
	271	mem->ranges[j+1].end;
	272	}
	273	}
	274	mem->nr_ranges--;
	275	return 0;
	276	}
	277
	278	if (mstart > start && mend < end) {
	279	/* Split original range */
	280	mem->ranges[i].end = mstart - 1;
	281	temp_range.start = mend + 1;
	282	temp_range.end = end;
	283	} else if (mstart != start)
	284	mem->ranges[i].end = mstart - 1;
	285	else
	286	mem->ranges[i].start = mend + 1;
	287	break;
	288	}
	289
	290	/* If a split happend, add the split to array */
	291	if (!temp_range.end)
	292	return 0;
	293
	294	/* Split happened */
	295	if (i == CRASH_MAX_RANGES - 1) {
	296	pr_err("Too many crash ranges after split\n");
	297	return -ENOMEM;
	298	}
	299
	300	/* Location where new range should go */
	301	j = i + 1;
302	if (j < mem->nr_ranges) {
303	/* Move over all ranges one slot towards the end */
304	for (i = mem->nr_ranges - 1; i >= j; i--)
305	mem->ranges[i + 1] = mem->ranges[i];
306	}
307
308	mem->ranges[j].start = temp_range.start;
309	mem->ranges[j].end = temp_range.end;
310	mem->nr_ranges++;
311	return 0;
312	}
313
314	/*
315	* Look for any unwanted ranges between mstart, mend and remove them. This
316	* might lead to split and split ranges are put in ced->mem.ranges[] array
317	*/
318	static int elf_header_exclude_ranges(struct crash_elf_data *ced,
319	unsigned long long mstart, unsigned long long mend)
320	{
321	struct crash_mem *cmem = &ced->mem;
322	int ret = 0;
323
324	memset(cmem->ranges, 0, sizeof(cmem->ranges));
325
326	cmem->ranges[0].start = mstart;
327	cmem->ranges[0].end = mend;
328	cmem->nr_ranges = 1;
329
330	/* Exclude crashkernel region */
331	ret = exclude_mem_range(cmem, crashk_res.start, crashk_res.end);
332	if (ret)
333	return ret;
334
a2d6aa8f BH	335	if (crashk_low_res.end) {
	336	ret = exclude_mem_range(cmem, crashk_low_res.start, crashk_low_res.end);
	337	if (ret)
	338	return ret;
	339	}
dd5f7260	340
dd5f7260 VG	341	return ret;
	342	}
	343
	344	static int prepare_elf64_ram_headers_callback(u64 start, u64 end, void *arg)
	345	{
	346	struct crash_elf_data *ced = arg;
	347	Elf64_Ehdr *ehdr;
	348	Elf64_Phdr *phdr;
	349	unsigned long mstart, mend;
	350	struct kimage *image = ced->image;
	351	struct crash_mem *cmem;
	352	int ret, i;
	353
	354	ehdr = ced->ehdr;
	355
	356	/* Exclude unwanted mem ranges */
	357	ret = elf_header_exclude_ranges(ced, start, end);
	358	if (ret)
	359	return ret;
	360
	361	/* Go through all the ranges in ced->mem.ranges[] and prepare phdr */
	362	cmem = &ced->mem;
	363
	364	for (i = 0; i < cmem->nr_ranges; i++) {
	365	mstart = cmem->ranges[i].start;
	366	mend = cmem->ranges[i].end;
	367
	368	phdr = ced->bufp;
	369	ced->bufp += sizeof(Elf64_Phdr);
	370
	371	phdr->p_type = PT_LOAD;
	372	phdr->p_flags = PF_R\|PF_W\|PF_X;
	373	phdr->p_offset = mstart;
	374
	375	/*
	376	* If a range matches backup region, adjust offset to backup
	377	* segment.
	378	*/
	379	if (mstart == image->arch.backup_src_start &&
	380	(mend - mstart + 1) == image->arch.backup_src_sz)
	381	phdr->p_offset = image->arch.backup_load_addr;
	382
	383	phdr->p_paddr = mstart;
	384	phdr->p_vaddr = (unsigned long long) __va(mstart);
	385	phdr->p_filesz = phdr->p_memsz = mend - mstart + 1;
	386	phdr->p_align = 0;
	387	ehdr->e_phnum++;
	388	pr_debug("Crash PT_LOAD elf header. phdr=%p vaddr=0x%llx, paddr=0x%llx, sz=0x%llx e_phnum=%d p_offset=0x%llx\n",
	389	phdr, phdr->p_vaddr, phdr->p_paddr, phdr->p_filesz,
	390	ehdr->e_phnum, phdr->p_offset);
	391	}
	392
	393	return ret;
	394	}
	395
	396	static int prepare_elf64_headers(struct crash_elf_data *ced,
	397	void *addr, unsigned long sz)
	398	{
	399	Elf64_Ehdr *ehdr;
	400	Elf64_Phdr *phdr;
	401	unsigned long nr_cpus = num_possible_cpus(), nr_phdr, elf_sz;
	402	unsigned char buf, bufp;
	403	unsigned int cpu;
	404	unsigned long long notes_addr;
405	int ret;
406
407	/* extra phdr for vmcoreinfo elf note */
408	nr_phdr = nr_cpus + 1;
409	nr_phdr += ced->max_nr_ranges;
410
411	/*
412	* kexec-tools creates an extra PT_LOAD phdr for kernel text mapping
413	* area on x86_64 (ffffffff80000000 - ffffffffa0000000).
414	* I think this is required by tools like gdb. So same physical
415	* memory will be mapped in two elf headers. One will contain kernel
416	* text virtual addresses and other will have __va(physical) addresses.
417	*/
418
419	nr_phdr++;
420	elf_sz = sizeof(Elf64_Ehdr) + nr_phdr * sizeof(Elf64_Phdr);
421	elf_sz = ALIGN(elf_sz, ELF_CORE_HEADER_ALIGN);
422
423	buf = vzalloc(elf_sz);
424	if (!buf)
425	return -ENOMEM;
426
427	bufp = buf;
428	ehdr = (Elf64_Ehdr *)bufp;
429	bufp += sizeof(Elf64_Ehdr);
430	memcpy(ehdr->e_ident, ELFMAG, SELFMAG);
431	ehdr->e_ident[EI_CLASS] = ELFCLASS64;
432	ehdr->e_ident[EI_DATA] = ELFDATA2LSB;
433	ehdr->e_ident[EI_VERSION] = EV_CURRENT;
434	ehdr->e_ident[EI_OSABI] = ELF_OSABI;
435	memset(ehdr->e_ident + EI_PAD, 0, EI_NIDENT - EI_PAD);
436	ehdr->e_type = ET_CORE;
437	ehdr->e_machine = ELF_ARCH;
438	ehdr->e_version = EV_CURRENT;
439	ehdr->e_phoff = sizeof(Elf64_Ehdr);
440	ehdr->e_ehsize = sizeof(Elf64_Ehdr);
441	ehdr->e_phentsize = sizeof(Elf64_Phdr);
442
443	/* Prepare one phdr of type PT_NOTE for each present cpu */
444	for_each_present_cpu(cpu) {
445	phdr = (Elf64_Phdr *)bufp;
446	bufp += sizeof(Elf64_Phdr);
447	phdr->p_type = PT_NOTE;
448	notes_addr = per_cpu_ptr_to_phys(per_cpu_ptr(crash_notes, cpu));
449	phdr->p_offset = phdr->p_paddr = notes_addr;
450	phdr->p_filesz = phdr->p_memsz = sizeof(note_buf_t);
451	(ehdr->e_phnum)++;
452	}
453
454	/* Prepare one PT_NOTE header for vmcoreinfo */
455	phdr = (Elf64_Phdr *)bufp;
456	bufp += sizeof(Elf64_Phdr);
457	phdr->p_type = PT_NOTE;
458	phdr->p_offset = phdr->p_paddr = paddr_vmcoreinfo_note();
459	phdr->p_filesz = phdr->p_memsz = sizeof(vmcoreinfo_note);
460	(ehdr->e_phnum)++;
461
462	#ifdef CONFIG_X86_64
463	/* Prepare PT_LOAD type program header for kernel text region */
464	phdr = (Elf64_Phdr *)bufp;
465	bufp += sizeof(Elf64_Phdr);
466	phdr->p_type = PT_LOAD;
467	phdr->p_flags = PF_R\|PF_W\|PF_X;
468	phdr->p_vaddr = (Elf64_Addr)_text;
469	phdr->p_filesz = phdr->p_memsz = _end - _text;
470	phdr->p_offset = phdr->p_paddr = __pa_symbol(_text);
471	(ehdr->e_phnum)++;
472	#endif
473
474	/* Prepare PT_LOAD headers for system ram chunks. */
475	ced->ehdr = ehdr;
476	ced->bufp = bufp;
477	ret = walk_system_ram_res(0, -1, ced,
478	prepare_elf64_ram_headers_callback);
479	if (ret < 0)
480	return ret;
481
482	*addr = buf;
483	*sz = elf_sz;
484	return 0;
485	}
486
487	/* Prepare elf headers. Return addr and size */
488	static int prepare_elf_headers(struct kimage image, void *addr,
489	unsigned long *sz)
490	{
491	struct crash_elf_data *ced;
492	int ret;
493
494	ced = kzalloc(sizeof(*ced), GFP_KERNEL);
495	if (!ced)
496	return -ENOMEM;
497
498	fill_up_crash_elf_data(ced, image);
499
500	/* By default prepare 64bit headers */
501	ret = prepare_elf64_headers(ced, addr, sz);
502	kfree(ced);
503	return ret;
504	}
505
506	static int add_e820_entry(struct boot_params params, struct e820entry entry)
507	{
508	unsigned int nr_e820_entries;
509
510	nr_e820_entries = params->e820_entries;
511	if (nr_e820_entries >= E820MAX)
512	return 1;
513
514	memcpy(&params->e820_map[nr_e820_entries], entry,
515	sizeof(struct e820entry));
516	params->e820_entries++;
517	return 0;
518	}
519
520	static int memmap_entry_callback(u64 start, u64 end, void *arg)
521	{
522	struct crash_memmap_data *cmd = arg;
523	struct boot_params *params = cmd->params;
524	struct e820entry ei;
525
526	ei.addr = start;
527	ei.size = end - start + 1;
528	ei.type = cmd->type;
529	add_e820_entry(params, &ei);
530
531	return 0;
532	}
533
534	static int memmap_exclude_ranges(struct kimage image, struct crash_mem cmem,
535	unsigned long long mstart,
536	unsigned long long mend)
537	{
538	unsigned long start, end;
539	int ret = 0;
540
541	cmem->ranges[0].start = mstart;
542	cmem->ranges[0].end = mend;
543	cmem->nr_ranges = 1;
544
545	/* Exclude Backup region */
546	start = image->arch.backup_load_addr;
547	end = start + image->arch.backup_src_sz - 1;
548	ret = exclude_mem_range(cmem, start, end);
549	if (ret)
550	return ret;
551
552	/* Exclude elf header region */
553	start = image->arch.elf_load_addr;
554	end = start + image->arch.elf_headers_sz - 1;
555	return exclude_mem_range(cmem, start, end);
556	}
557
558	/* Prepare memory map for crash dump kernel */
559	int crash_setup_memmap_entries(struct kimage image, struct boot_params params)
560	{
561	int i, ret = 0;
562	unsigned long flags;
563	struct e820entry ei;
564	struct crash_memmap_data cmd;
565	struct crash_mem *cmem;
566
567	cmem = vzalloc(sizeof(struct crash_mem));
568	if (!cmem)
569	return -ENOMEM;
570
571	memset(&cmd, 0, sizeof(struct crash_memmap_data));
572	cmd.params = params;
573
574	/* Add first 640K segment */
575	ei.addr = image->arch.backup_src_start;
576	ei.size = image->arch.backup_src_sz;
577	ei.type = E820_RAM;
578	add_e820_entry(params, &ei);
579
580	/* Add ACPI tables */
581	cmd.type = E820_ACPI;
582	flags = IORESOURCE_MEM \| IORESOURCE_BUSY;
f0f4711a	583	walk_iomem_res_desc(IORES_DESC_ACPI_TABLES, flags, 0, -1, &cmd,
dd5f7260 VG	584	memmap_entry_callback);
	585
	586	/* Add ACPI Non-volatile Storage */
	587	cmd.type = E820_NVS;
f0f4711a	588	walk_iomem_res_desc(IORES_DESC_ACPI_NV_STORAGE, flags, 0, -1, &cmd,
dd5f7260 VG	589	memmap_entry_callback);
	590
	591	/* Add crashk_low_res region */
	592	if (crashk_low_res.end) {
	593	ei.addr = crashk_low_res.start;
	594	ei.size = crashk_low_res.end - crashk_low_res.start + 1;
	595	ei.type = E820_RAM;
	596	add_e820_entry(params, &ei);
	597	}
	598
	599	/* Exclude some ranges from crashk_res and add rest to memmap */
	600	ret = memmap_exclude_ranges(image, cmem, crashk_res.start,
	601	crashk_res.end);
	602	if (ret)
	603	goto out;
	604
	605	for (i = 0; i < cmem->nr_ranges; i++) {
	606	ei.size = cmem->ranges[i].end - cmem->ranges[i].start + 1;
	607
	608	/* If entry is less than a page, skip it */
	609	if (ei.size < PAGE_SIZE)
	610	continue;
	611	ei.addr = cmem->ranges[i].start;
	612	ei.type = E820_RAM;
	613	add_e820_entry(params, &ei);
	614	}
	615
	616	out:
	617	vfree(cmem);
	618	return ret;
	619	}
	620
	621	static int determine_backup_region(u64 start, u64 end, void *arg)
	622	{
	623	struct kimage *image = arg;
	624
	625	image->arch.backup_src_start = start;
	626	image->arch.backup_src_sz = end - start + 1;
	627
	628	/* Expecting only one range for backup region */
	629	return 1;
	630	}
	631
	632	int crash_load_segments(struct kimage *image)
	633	{
	634	unsigned long src_start, src_sz, elf_sz;
	635	void *elf_addr;
	636	int ret;
	637
	638	/*
	639	* Determine and load a segment for backup area. First 640K RAM
	640	* region is backup source
	641	*/
	642
	643	ret = walk_system_ram_res(KEXEC_BACKUP_SRC_START, KEXEC_BACKUP_SRC_END,
	644	image, determine_backup_region);
	645
	646	/* Zero or postive return values are ok */
	647	if (ret < 0)
	648	return ret;
	649
	650	src_start = image->arch.backup_src_start;
	651	src_sz = image->arch.backup_src_sz;
	652
653	/* Add backup segment. */
654	if (src_sz) {
655	/*
656	* Ideally there is no source for backup segment. This is
657	* copied in purgatory after crash. Just add a zero filled
658	* segment for now to make sure checksum logic works fine.
659	*/
660	ret = kexec_add_buffer(image, (char *)&crash_zero_bytes,
661	sizeof(crash_zero_bytes), src_sz,
662	PAGE_SIZE, 0, -1, 0,
663	&image->arch.backup_load_addr);
664	if (ret)
665	return ret;
666	pr_debug("Loaded backup region at 0x%lx backup_start=0x%lx memsz=0x%lx\n",
667	image->arch.backup_load_addr, src_start, src_sz);
668	}
669
670	/* Prepare elf headers and add a segment */
671	ret = prepare_elf_headers(image, &elf_addr, &elf_sz);
672	if (ret)
673	return ret;
674
675	image->arch.elf_headers = elf_addr;
676	image->arch.elf_headers_sz = elf_sz;
677
678	ret = kexec_add_buffer(image, (char *)elf_addr, elf_sz, elf_sz,
679	ELF_CORE_HEADER_ALIGN, 0, -1, 0,
680	&image->arch.elf_load_addr);
681	if (ret) {
682	vfree((void *)image->arch.elf_headers);
683	return ret;
684	}
685	pr_debug("Loaded ELF headers at 0x%lx bufsz=0x%lx memsz=0x%lx\n",
686	image->arch.elf_load_addr, elf_sz, elf_sz);
687
688	return ret;
689	}
74ca317c	690	#endif /* CONFIG_KEXEC_FILE */