diff options
author | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-19 02:56:35 +0000 |
---|---|---|
committer | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-19 02:56:35 +0000 |
commit | eba0cfa6b0bef4f2e73c8630a7efa3944df8b0f8 (patch) | |
tree | 74c37eede1f0634cc5de1c63c934edaa1630c6bc /kexec/arch/arm64 | |
parent | Initial commit. (diff) | |
download | kexec-tools-7bbe6b040c9123991377749057cfc0356c289ceb.tar.xz kexec-tools-7bbe6b040c9123991377749057cfc0356c289ceb.zip |
Adding upstream version 1:2.0.27.upstream/1%2.0.27upstream
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'kexec/arch/arm64')
-rw-r--r-- | kexec/arch/arm64/Makefile | 50 | ||||
-rw-r--r-- | kexec/arch/arm64/crashdump-arm64.c | 248 | ||||
-rw-r--r-- | kexec/arch/arm64/crashdump-arm64.h | 29 | ||||
-rw-r--r-- | kexec/arch/arm64/image-header.h | 147 | ||||
-rw-r--r-- | kexec/arch/arm64/include/arch/options.h | 43 | ||||
-rw-r--r-- | kexec/arch/arm64/iomem.h | 10 | ||||
-rw-r--r-- | kexec/arch/arm64/kexec-arm64.c | 1365 | ||||
-rw-r--r-- | kexec/arch/arm64/kexec-arm64.h | 84 | ||||
-rw-r--r-- | kexec/arch/arm64/kexec-elf-arm64.c | 170 | ||||
-rw-r--r-- | kexec/arch/arm64/kexec-image-arm64.c | 119 | ||||
-rw-r--r-- | kexec/arch/arm64/kexec-uImage-arm64.c | 52 | ||||
-rw-r--r-- | kexec/arch/arm64/kexec-vmlinuz-arm64.c | 110 |
12 files changed, 2427 insertions, 0 deletions
diff --git a/kexec/arch/arm64/Makefile b/kexec/arch/arm64/Makefile new file mode 100644 index 0000000..59212f1 --- /dev/null +++ b/kexec/arch/arm64/Makefile @@ -0,0 +1,50 @@ + +arm64_FS2DT += kexec/fs2dt.c +arm64_FS2DT_INCLUDE += \ + -include $(srcdir)/kexec/arch/arm64/crashdump-arm64.h \ + -include $(srcdir)/kexec/arch/arm64/kexec-arm64.h + +arm64_DT_OPS += kexec/dt-ops.c + +arm64_MEM_REGIONS = kexec/mem_regions.c + +arm64_CPPFLAGS += -I $(srcdir)/kexec/ + +arm64_KEXEC_SRCS += \ + kexec/arch/arm64/crashdump-arm64.c \ + kexec/arch/arm64/kexec-arm64.c \ + kexec/arch/arm64/kexec-elf-arm64.c \ + kexec/arch/arm64/kexec-uImage-arm64.c \ + kexec/arch/arm64/kexec-image-arm64.c \ + kexec/arch/arm64/kexec-vmlinuz-arm64.c + +arm64_UIMAGE = kexec/kexec-uImage.c + +arm64_ARCH_REUSE_INITRD = +arm64_ADD_SEGMENT = +arm64_VIRT_TO_PHYS = +arm64_PHYS_TO_VIRT = + +dist += $(arm64_KEXEC_SRCS) \ + kexec/arch/arm64/include/arch/options.h \ + kexec/arch/arm64/crashdump-arm64.h \ + kexec/arch/arm64/image-header.h \ + kexec/arch/arm64/iomem.h \ + kexec/arch/arm64/kexec-arm64.h \ + kexec/arch/arm64/Makefile + +ifdef HAVE_LIBFDT + +LIBS += -lfdt + +else + +include $(srcdir)/kexec/libfdt/Makefile.libfdt + +libfdt_SRCS += $(LIBFDT_SRCS:%=kexec/libfdt/%) + +arm64_CPPFLAGS += -I$(srcdir)/kexec/libfdt + +arm64_KEXEC_SRCS += $(libfdt_SRCS) + +endif diff --git a/kexec/arch/arm64/crashdump-arm64.c b/kexec/arch/arm64/crashdump-arm64.c new file mode 100644 index 0000000..3098315 --- /dev/null +++ b/kexec/arch/arm64/crashdump-arm64.c @@ -0,0 +1,248 @@ +/* + * ARM64 crashdump. + * partly derived from arm implementation + * + * Copyright (c) 2014-2017 Linaro Limited + * Author: AKASHI Takahiro <takahiro.akashi@linaro.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#define _GNU_SOURCE + +#include <errno.h> +#include <linux/elf.h> + +#include "kexec.h" +#include "crashdump.h" +#include "crashdump-arm64.h" +#include "iomem.h" +#include "kexec-arm64.h" +#include "kexec-elf.h" +#include "mem_regions.h" + +/* memory ranges of crashed kernel */ +static struct memory_ranges system_memory_rgns; + +/* memory range reserved for crashkernel */ +struct memory_range crash_reserved_mem[CRASH_MAX_RESERVED_RANGES]; +struct memory_ranges usablemem_rgns = { + .size = 0, + .max_size = CRASH_MAX_RESERVED_RANGES, + .ranges = crash_reserved_mem, +}; + +struct memory_range elfcorehdr_mem; + +static struct crash_elf_info elf_info = { + .class = ELFCLASS64, +#if (__BYTE_ORDER == __LITTLE_ENDIAN) + .data = ELFDATA2LSB, +#else + .data = ELFDATA2MSB, +#endif + .machine = EM_AARCH64, +}; + +/* + * iomem_range_callback() - callback called for each iomem region + * @data: not used + * @nr: not used + * @str: name of the memory region + * @base: start address of the memory region + * @length: size of the memory region + * + * This function is called once for each memory region found in /proc/iomem. + * It locates system RAM and crashkernel reserved memory and places these to + * variables, respectively, system_memory_rgns and usablemem_rgns. + */ + +static int iomem_range_callback(void *UNUSED(data), int UNUSED(nr), + char *str, unsigned long long base, + unsigned long long length) +{ + if (strncmp(str, CRASH_KERNEL, strlen(CRASH_KERNEL)) == 0) + return mem_regions_alloc_and_add(&usablemem_rgns, + base, length, RANGE_RAM); + else if (strncmp(str, SYSTEM_RAM, strlen(SYSTEM_RAM)) == 0) + return mem_regions_alloc_and_add(&system_memory_rgns, + base, length, RANGE_RAM); + else if (strncmp(str, KERNEL_CODE, strlen(KERNEL_CODE)) == 0) { + + unsigned long long kva_text = get_kernel_sym("_text"); + unsigned long long kva_stext = get_kernel_sym("_stext"); + unsigned long long kva_text_end = get_kernel_sym("__init_begin"); + + /* + * old: kernel_code.start = __pa_symbol(_text); + * new: kernel_code.start = __pa_symbol(_stext); + * + * For compatibility, deduce by comparing the gap "__init_begin - _stext" + * and the res size of "Kernel code" in /proc/iomem + */ + if (kva_text_end - kva_stext == length) + elf_info.kern_paddr_start = base - (kva_stext - kva_text); + else + elf_info.kern_paddr_start = base; + } + else if (strncmp(str, KERNEL_DATA, strlen(KERNEL_DATA)) == 0) + elf_info.kern_size = base + length - elf_info.kern_paddr_start; + + return 0; +} + +int is_crashkernel_mem_reserved(void) +{ + if (!usablemem_rgns.size) + kexec_iomem_for_each_line(NULL, iomem_range_callback, NULL); + + return usablemem_rgns.size; +} + +/* + * crash_get_memory_ranges() - read system physical memory + * + * Function reads through system physical memory and stores found memory + * regions in system_memory_ranges. + * Regions are sorted in ascending order. + * + * Returns 0 in case of success and a negative value otherwise. + */ +static int crash_get_memory_ranges(void) +{ + int i; + + /* + * First read all memory regions that can be considered as + * system memory including the crash area. + */ + if (!usablemem_rgns.size) + kexec_iomem_for_each_line(NULL, iomem_range_callback, NULL); + + /* allow one or two regions for crash dump kernel */ + if (!usablemem_rgns.size) + return -EINVAL; + + dbgprint_mem_range("Reserved memory range", + usablemem_rgns.ranges, usablemem_rgns.size); + + for (i = 0; i < usablemem_rgns.size; i++) { + if (mem_regions_alloc_and_exclude(&system_memory_rgns, + &crash_reserved_mem[i])) { + fprintf(stderr, "Cannot allocate memory for ranges\n"); + return -ENOMEM; + } + } + + /* + * Make sure that the memory regions are sorted. + */ + mem_regions_sort(&system_memory_rgns); + + dbgprint_mem_range("Coredump memory ranges", + system_memory_rgns.ranges, system_memory_rgns.size); + + /* + * For additional kernel code/data segment. + * kern_paddr_start/kern_size are determined in iomem_range_callback + */ + elf_info.kern_vaddr_start = get_kernel_sym("_text"); + if (!elf_info.kern_vaddr_start) + elf_info.kern_vaddr_start = UINT64_MAX; + + return 0; +} + +/* + * load_crashdump_segments() - load the elf core header + * @info: kexec info structure + * + * This function creates and loads an additional segment of elf core header + : which is used to construct /proc/vmcore on crash dump kernel. + * + * Return 0 in case of success and -1 in case of error. + */ + +int load_crashdump_segments(struct kexec_info *info) +{ + unsigned long elfcorehdr; + unsigned long bufsz; + void *buf; + int err; + + /* + * First fetch all the memory (RAM) ranges that we are going to + * pass to the crash dump kernel during panic. + */ + + err = crash_get_memory_ranges(); + + if (err) + return EFAILED; + + get_page_offset((unsigned long *)&elf_info.page_offset); + dbgprintf("%s: page_offset: %016llx\n", __func__, + elf_info.page_offset); + + err = crash_create_elf64_headers(info, &elf_info, + system_memory_rgns.ranges, system_memory_rgns.size, + &buf, &bufsz, ELF_CORE_HEADER_ALIGN); + + if (err) + return EFAILED; + + elfcorehdr = add_buffer_phys_virt(info, buf, bufsz, bufsz, 0, + crash_reserved_mem[usablemem_rgns.size - 1].start, + crash_reserved_mem[usablemem_rgns.size - 1].end, + -1, 0); + + elfcorehdr_mem.start = elfcorehdr; + elfcorehdr_mem.end = elfcorehdr + bufsz - 1; + + dbgprintf("%s: elfcorehdr 0x%llx-0x%llx\n", __func__, + elfcorehdr_mem.start, elfcorehdr_mem.end); + + return 0; +} + +/* + * e_entry and p_paddr are actually in virtual address space. + * Those values will be translated to physcal addresses by using + * virt_to_phys() in add_segment(). + * So let's fix up those values for later use so the memory base + * (arm64_mm.phys_offset) will be correctly replaced with + * crash_reserved_mem[usablemem_rgns.size - 1].start. + */ +void fixup_elf_addrs(struct mem_ehdr *ehdr) +{ + struct mem_phdr *phdr; + int i; + + ehdr->e_entry += -arm64_mem.phys_offset + + crash_reserved_mem[usablemem_rgns.size - 1].start; + + for (i = 0; i < ehdr->e_phnum; i++) { + phdr = &ehdr->e_phdr[i]; + if (phdr->p_type != PT_LOAD) + continue; + phdr->p_paddr += + (-arm64_mem.phys_offset + + crash_reserved_mem[usablemem_rgns.size - 1].start); + } +} + +int get_crash_kernel_load_range(uint64_t *start, uint64_t *end) +{ + if (!usablemem_rgns.size) + kexec_iomem_for_each_line(NULL, iomem_range_callback, NULL); + + if (!usablemem_rgns.size) + return -1; + + *start = crash_reserved_mem[usablemem_rgns.size - 1].start; + *end = crash_reserved_mem[usablemem_rgns.size - 1].end; + + return 0; +} diff --git a/kexec/arch/arm64/crashdump-arm64.h b/kexec/arch/arm64/crashdump-arm64.h new file mode 100644 index 0000000..82fa69b --- /dev/null +++ b/kexec/arch/arm64/crashdump-arm64.h @@ -0,0 +1,29 @@ +/* + * ARM64 crashdump. + * + * Copyright (c) 2014-2017 Linaro Limited + * Author: AKASHI Takahiro <takahiro.akashi@linaro.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef CRASHDUMP_ARM64_H +#define CRASHDUMP_ARM64_H + +#include "kexec.h" + +#define CRASH_MAX_MEMORY_RANGES 32768 + +/* crash dump kernel support at most two regions, low_region and high region. */ +#define CRASH_MAX_RESERVED_RANGES 2 + +extern struct memory_ranges usablemem_rgns; +extern struct memory_range crash_reserved_mem[]; +extern struct memory_range elfcorehdr_mem; + +extern int load_crashdump_segments(struct kexec_info *info); +extern void fixup_elf_addrs(struct mem_ehdr *ehdr); + +#endif /* CRASHDUMP_ARM64_H */ diff --git a/kexec/arch/arm64/image-header.h b/kexec/arch/arm64/image-header.h new file mode 100644 index 0000000..26bb02f --- /dev/null +++ b/kexec/arch/arm64/image-header.h @@ -0,0 +1,147 @@ +/* + * ARM64 binary image header. + */ + +#if !defined(__ARM64_IMAGE_HEADER_H) +#define __ARM64_IMAGE_HEADER_H + +#include <endian.h> +#include <stdint.h> + +/** + * struct arm64_image_header - arm64 kernel image header. + * + * @pe_sig: Optional PE format 'MZ' signature. + * @branch_code: Reserved for instructions to branch to stext. + * @text_offset: The image load offset in LSB byte order. + * @image_size: An estimated size of the memory image size in LSB byte order. + * @flags: Bit flags in LSB byte order: + * Bit 0: Image byte order: 1=MSB. + * Bit 1-2: Kernel page size: 1=4K, 2=16K, 3=64K. + * Bit 3: Image placement: 0=low. + * @reserved_1: Reserved. + * @magic: Magic number, "ARM\x64". + * @pe_header: Optional offset to a PE format header. + **/ + +struct arm64_image_header { + uint8_t pe_sig[2]; + uint16_t branch_code[3]; + uint64_t text_offset; + uint64_t image_size; + uint64_t flags; + uint64_t reserved_1[3]; + uint8_t magic[4]; + uint32_t pe_header; +}; + +static const uint8_t arm64_image_magic[4] = {'A', 'R', 'M', 0x64U}; +static const uint8_t arm64_image_pe_sig[2] = {'M', 'Z'}; +static const uint8_t arm64_pe_machtype[6] = {'P','E', 0x0, 0x0, 0x64, 0xAA}; +static const uint64_t arm64_image_flag_be = (1UL << 0); +static const uint64_t arm64_image_flag_page_size = (3UL << 1); +static const uint64_t arm64_image_flag_placement = (1UL << 3); + +/** + * enum arm64_header_page_size + */ + +enum arm64_header_page_size { + arm64_header_page_size_invalid = 0, + arm64_header_page_size_4k, + arm64_header_page_size_16k, + arm64_header_page_size_64k +}; + +/** + * arm64_header_check_magic - Helper to check the arm64 image header. + * + * Returns non-zero if header is OK. + */ + +static inline int arm64_header_check_magic(const struct arm64_image_header *h) +{ + if (!h) + return 0; + + return (h->magic[0] == arm64_image_magic[0] + && h->magic[1] == arm64_image_magic[1] + && h->magic[2] == arm64_image_magic[2] + && h->magic[3] == arm64_image_magic[3]); +} + +/** + * arm64_header_check_pe_sig - Helper to check the arm64 image header. + * + * Returns non-zero if 'MZ' signature is found. + */ + +static inline int arm64_header_check_pe_sig(const struct arm64_image_header *h) +{ + if (!h) + return 0; + + return (h->pe_sig[0] == arm64_image_pe_sig[0] + && h->pe_sig[1] == arm64_image_pe_sig[1]); +} + +/** + * arm64_header_check_msb - Helper to check the arm64 image header. + * + * Returns non-zero if the image was built as big endian. + */ + +static inline int arm64_header_check_msb(const struct arm64_image_header *h) +{ + if (!h) + return 0; + + return (le64toh(h->flags) & arm64_image_flag_be) >> 0; +} + +/** + * arm64_header_page_size + */ + +static inline enum arm64_header_page_size arm64_header_page_size( + const struct arm64_image_header *h) +{ + if (!h) + return 0; + + return (le64toh(h->flags) & arm64_image_flag_page_size) >> 1; +} + +/** + * arm64_header_placement + * + * Returns non-zero if the image has no physical placement restrictions. + */ + +static inline int arm64_header_placement(const struct arm64_image_header *h) +{ + if (!h) + return 0; + + return (le64toh(h->flags) & arm64_image_flag_placement) >> 3; +} + +static inline uint64_t arm64_header_text_offset( + const struct arm64_image_header *h) +{ + if (!h) + return 0; + + return le64toh(h->text_offset); +} + +static inline uint64_t arm64_header_image_size( + const struct arm64_image_header *h) +{ + if (!h) + return 0; + + return le64toh(h->image_size); +} + +#endif diff --git a/kexec/arch/arm64/include/arch/options.h b/kexec/arch/arm64/include/arch/options.h new file mode 100644 index 0000000..8c695f3 --- /dev/null +++ b/kexec/arch/arm64/include/arch/options.h @@ -0,0 +1,43 @@ +#if !defined(KEXEC_ARCH_ARM64_OPTIONS_H) +#define KEXEC_ARCH_ARM64_OPTIONS_H + +#define OPT_APPEND ((OPT_MAX)+0) +#define OPT_DTB ((OPT_MAX)+1) +#define OPT_INITRD ((OPT_MAX)+2) +#define OPT_REUSE_CMDLINE ((OPT_MAX)+3) +#define OPT_SERIAL ((OPT_MAX)+4) +#define OPT_ARCH_MAX ((OPT_MAX)+5) + +#define KEXEC_ARCH_OPTIONS \ + KEXEC_OPTIONS \ + { "append", 1, NULL, OPT_APPEND }, \ + { "command-line", 1, NULL, OPT_APPEND }, \ + { "dtb", 1, NULL, OPT_DTB }, \ + { "initrd", 1, NULL, OPT_INITRD }, \ + { "serial", 1, NULL, OPT_SERIAL }, \ + { "ramdisk", 1, NULL, OPT_INITRD }, \ + { "reuse-cmdline", 0, NULL, OPT_REUSE_CMDLINE }, \ + +#define KEXEC_ARCH_OPT_STR KEXEC_OPT_STR /* Only accept long arch options. */ +#define KEXEC_ALL_OPTIONS KEXEC_ARCH_OPTIONS +#define KEXEC_ALL_OPT_STR KEXEC_ARCH_OPT_STR + +static const char arm64_opts_usage[] __attribute__ ((unused)) = +" --append=STRING Set the kernel command line to STRING.\n" +" --command-line=STRING Set the kernel command line to STRING.\n" +" --dtb=FILE Use FILE as the device tree blob.\n" +" --initrd=FILE Use FILE as the kernel initial ramdisk.\n" +" --serial=STRING Name of console used for purgatory printing. (e.g. ttyAMA0)\n" +" --ramdisk=FILE Use FILE as the kernel initial ramdisk.\n" +" --reuse-cmdline Use kernel command line from running system.\n"; + +struct arm64_opts { + const char *command_line; + const char *dtb; + const char *initrd; + const char *console; +}; + +extern struct arm64_opts arm64_opts; + +#endif diff --git a/kexec/arch/arm64/iomem.h b/kexec/arch/arm64/iomem.h new file mode 100644 index 0000000..d4864bb --- /dev/null +++ b/kexec/arch/arm64/iomem.h @@ -0,0 +1,10 @@ +#ifndef IOMEM_H +#define IOMEM_H + +#define SYSTEM_RAM "System RAM\n" +#define KERNEL_CODE "Kernel code\n" +#define KERNEL_DATA "Kernel data\n" +#define CRASH_KERNEL "Crash kernel\n" +#define IOMEM_RESERVED "reserved\n" + +#endif diff --git a/kexec/arch/arm64/kexec-arm64.c b/kexec/arch/arm64/kexec-arm64.c new file mode 100644 index 0000000..4a67b0d --- /dev/null +++ b/kexec/arch/arm64/kexec-arm64.c @@ -0,0 +1,1365 @@ +/* + * ARM64 kexec. + */ + +#define _GNU_SOURCE + +#include <assert.h> +#include <errno.h> +#include <getopt.h> +#include <inttypes.h> +#include <libfdt.h> +#include <limits.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <sys/stat.h> +#include <linux/elf-em.h> +#include <elf.h> +#include <elf_info.h> + +#include <unistd.h> +#include <syscall.h> +#include <errno.h> +#include <linux/random.h> + +#include "kexec.h" +#include "kexec-arm64.h" +#include "crashdump.h" +#include "crashdump-arm64.h" +#include "dt-ops.h" +#include "fs2dt.h" +#include "iomem.h" +#include "kexec-syscall.h" +#include "mem_regions.h" +#include "arch/options.h" + +#define ROOT_NODE_ADDR_CELLS_DEFAULT 1 +#define ROOT_NODE_SIZE_CELLS_DEFAULT 1 + +#define PROP_ADDR_CELLS "#address-cells" +#define PROP_SIZE_CELLS "#size-cells" +#define PROP_ELFCOREHDR "linux,elfcorehdr" +#define PROP_USABLE_MEM_RANGE "linux,usable-memory-range" + +#define PAGE_OFFSET_36 ((0xffffffffffffffffUL) << 36) +#define PAGE_OFFSET_39 ((0xffffffffffffffffUL) << 39) +#define PAGE_OFFSET_42 ((0xffffffffffffffffUL) << 42) +#define PAGE_OFFSET_47 ((0xffffffffffffffffUL) << 47) +#define PAGE_OFFSET_48 ((0xffffffffffffffffUL) << 48) + +/* Global flag which indicates that we have tried reading + * PHYS_OFFSET from 'kcore' already. + */ +static bool try_read_phys_offset_from_kcore = false; + +/* Machine specific details. */ +static int va_bits = -1; +static unsigned long page_offset; + +/* Global varables the core kexec routines expect. */ + +unsigned char reuse_initrd; + +off_t initrd_base; +off_t initrd_size; + +const struct arch_map_entry arches[] = { + { "aarch64", KEXEC_ARCH_ARM64 }, + { "aarch64_be", KEXEC_ARCH_ARM64 }, + { NULL, 0 }, +}; + +struct file_type file_type[] = { + {"vmlinux", elf_arm64_probe, elf_arm64_load, elf_arm64_usage}, + {"Image", image_arm64_probe, image_arm64_load, image_arm64_usage}, + {"uImage", uImage_arm64_probe, uImage_arm64_load, uImage_arm64_usage}, + {"vmlinuz", pez_arm64_probe, pez_arm64_load, pez_arm64_usage}, +}; + +int file_types = sizeof(file_type) / sizeof(file_type[0]); + +/* arm64 global varables. */ + +struct arm64_opts arm64_opts; +struct arm64_mem arm64_mem = { + .phys_offset = arm64_mem_ngv, + .vp_offset = arm64_mem_ngv, +}; + +uint64_t get_phys_offset(void) +{ + assert(arm64_mem.phys_offset != arm64_mem_ngv); + return arm64_mem.phys_offset; +} + +uint64_t get_vp_offset(void) +{ + assert(arm64_mem.vp_offset != arm64_mem_ngv); + return arm64_mem.vp_offset; +} + +/** + * arm64_process_image_header - Process the arm64 image header. + * + * Make a guess that KERNEL_IMAGE_SIZE will be enough for older kernels. + */ + +int arm64_process_image_header(const struct arm64_image_header *h) +{ +#if !defined(KERNEL_IMAGE_SIZE) +# define KERNEL_IMAGE_SIZE MiB(16) +#endif + + if (!arm64_header_check_magic(h)) + return EFAILED; + + if (h->image_size) { + arm64_mem.text_offset = arm64_header_text_offset(h); + arm64_mem.image_size = arm64_header_image_size(h); + } else { + /* For 3.16 and older kernels. */ + arm64_mem.text_offset = 0x80000; + arm64_mem.image_size = KERNEL_IMAGE_SIZE; + fprintf(stderr, + "kexec: %s: Warning: Kernel image size set to %lu MiB.\n" + " Please verify compatability with lodaed kernel.\n", + __func__, KERNEL_IMAGE_SIZE / 1024UL / 1024UL); + } + + return 0; +} + +void arch_usage(void) +{ + printf(arm64_opts_usage); +} + +int arch_process_options(int argc, char **argv) +{ + static const char short_options[] = KEXEC_OPT_STR ""; + static const struct option options[] = { + KEXEC_ARCH_OPTIONS + { 0 } + }; + int opt; + char *cmdline = NULL; + const char *append = NULL; + int do_kexec_file_syscall = 0; + + for (opt = 0; opt != -1; ) { + opt = getopt_long(argc, argv, short_options, options, 0); + + switch (opt) { + case OPT_APPEND: + append = optarg; + break; + case OPT_REUSE_CMDLINE: + cmdline = get_command_line(); + break; + case OPT_DTB: + arm64_opts.dtb = optarg; + break; + case OPT_INITRD: + arm64_opts.initrd = optarg; + break; + case OPT_KEXEC_FILE_SYSCALL: + do_kexec_file_syscall = 1; + case OPT_SERIAL: + arm64_opts.console = optarg; + break; + default: + break; /* Ignore core and unknown options. */ + } + } + + arm64_opts.command_line = concat_cmdline(cmdline, append); + + dbgprintf("%s:%d: command_line: %s\n", __func__, __LINE__, + arm64_opts.command_line); + dbgprintf("%s:%d: initrd: %s\n", __func__, __LINE__, + arm64_opts.initrd); + dbgprintf("%s:%d: dtb: %s\n", __func__, __LINE__, + (do_kexec_file_syscall && arm64_opts.dtb ? "(ignored)" : + arm64_opts.dtb)); + dbgprintf("%s:%d: console: %s\n", __func__, __LINE__, + arm64_opts.console); + + if (do_kexec_file_syscall) + arm64_opts.dtb = NULL; + + return 0; +} + +/** + * find_purgatory_sink - Find a sink for purgatory output. + */ + +static uint64_t find_purgatory_sink(const char *console) +{ + int fd, ret; + char device[255], mem[255]; + struct stat sb; + char buffer[10]; + uint64_t iomem = 0x0; + + if (!console) + return 0; + + ret = snprintf(device, sizeof(device), "/sys/class/tty/%s", console); + if (ret < 0 || ret >= sizeof(device)) { + fprintf(stderr, "snprintf failed: %s\n", strerror(errno)); + return 0; + } + + if (stat(device, &sb) || !S_ISDIR(sb.st_mode)) { + fprintf(stderr, "kexec: %s: No valid console found for %s\n", + __func__, device); + return 0; + } + + ret = snprintf(mem, sizeof(mem), "%s%s", device, "/iomem_base"); + if (ret < 0 || ret >= sizeof(mem)) { + fprintf(stderr, "snprintf failed: %s\n", strerror(errno)); + return 0; + } + + printf("console memory read from %s\n", mem); + + fd = open(mem, O_RDONLY); + if (fd < 0) { + fprintf(stderr, "kexec: %s: No able to open %s\n", + __func__, mem); + return 0; + } + + memset(buffer, '\0', sizeof(buffer)); + ret = read(fd, buffer, sizeof(buffer)); + if (ret < 0) { + fprintf(stderr, "kexec: %s: not able to read fd\n", __func__); + close(fd); + return 0; + } + + sscanf(buffer, "%lx", &iomem); + printf("console memory is at %#lx\n", iomem); + + close(fd); + return iomem; +} + +/** + * struct dtb - Info about a binary device tree. + * + * @buf: Device tree data. + * @size: Device tree data size. + * @name: Shorthand name of this dtb for messages. + * @path: Filesystem path. + */ + +struct dtb { + char *buf; + off_t size; + const char *name; + const char *path; +}; + +/** + * dump_reservemap - Dump the dtb's reservemap. + */ + +static void dump_reservemap(const struct dtb *dtb) +{ + int i; + + for (i = 0; ; i++) { + uint64_t address; + uint64_t size; + + fdt_get_mem_rsv(dtb->buf, i, &address, &size); + + if (!size) + break; + + dbgprintf("%s: %s {%" PRIx64 ", %" PRIx64 "}\n", __func__, + dtb->name, address, size); + } +} + +/** + * set_bootargs - Set the dtb's bootargs. + */ + +static int set_bootargs(struct dtb *dtb, const char *command_line) +{ + int result; + + if (!command_line || !command_line[0]) + return 0; + + result = dtb_set_bootargs(&dtb->buf, &dtb->size, command_line); + + if (result) { + fprintf(stderr, + "kexec: Set device tree bootargs failed.\n"); + return EFAILED; + } + + return 0; +} + +/** + * read_proc_dtb - Read /proc/device-tree. + */ + +static int read_proc_dtb(struct dtb *dtb) +{ + int result; + struct stat s; + static const char path[] = "/proc/device-tree"; + + result = stat(path, &s); + + if (result) { + dbgprintf("%s: %s\n", __func__, strerror(errno)); + return EFAILED; + } + + dtb->path = path; + create_flatten_tree((char **)&dtb->buf, &dtb->size, NULL); + + return 0; +} + +/** + * read_sys_dtb - Read /sys/firmware/fdt. + */ + +static int read_sys_dtb(struct dtb *dtb) +{ + int result; + struct stat s; + static const char path[] = "/sys/firmware/fdt"; + + result = stat(path, &s); + + if (result) { + dbgprintf("%s: %s\n", __func__, strerror(errno)); + return EFAILED; + } + + dtb->path = path; + dtb->buf = slurp_file(path, &dtb->size); + + return 0; +} + +/** + * read_1st_dtb - Read the 1st stage kernel's dtb. + */ + +static int read_1st_dtb(struct dtb *dtb) +{ + int result; + + dtb->name = "dtb_sys"; + result = read_sys_dtb(dtb); + + if (!result) + goto on_success; + + dtb->name = "dtb_proc"; + result = read_proc_dtb(dtb); + + if (!result) + goto on_success; + + dbgprintf("%s: not found\n", __func__); + return EFAILED; + +on_success: + dbgprintf("%s: found %s\n", __func__, dtb->path); + return 0; +} + +static int get_cells_size(void *fdt, uint32_t *address_cells, + uint32_t *size_cells) +{ + int nodeoffset; + const uint32_t *prop = NULL; + int prop_len; + + /* default values */ + *address_cells = ROOT_NODE_ADDR_CELLS_DEFAULT; + *size_cells = ROOT_NODE_SIZE_CELLS_DEFAULT; + + /* under root node */ + nodeoffset = fdt_path_offset(fdt, "/"); + if (nodeoffset < 0) + goto on_error; + + prop = fdt_getprop(fdt, nodeoffset, PROP_ADDR_CELLS, &prop_len); + if (prop) { + if (prop_len == sizeof(*prop)) + *address_cells = fdt32_to_cpu(*prop); + else + goto on_error; + } + + prop = fdt_getprop(fdt, nodeoffset, PROP_SIZE_CELLS, &prop_len); + if (prop) { + if (prop_len == sizeof(*prop)) + *size_cells = fdt32_to_cpu(*prop); + else + goto on_error; + } + + dbgprintf("%s: #address-cells:%d #size-cells:%d\n", __func__, + *address_cells, *size_cells); + return 0; + +on_error: + return EFAILED; +} + +static bool cells_size_fitted(uint32_t address_cells, uint32_t size_cells, + struct memory_range *range) +{ + dbgprintf("%s: %llx-%llx\n", __func__, range->start, range->end); + + /* if *_cells >= 2, cells can hold 64-bit values anyway */ + if ((address_cells == 1) && (range->start >= (1ULL << 32))) + return false; + + if ((size_cells == 1) && + ((range->end - range->start + 1) >= (1ULL << 32))) + return false; + + return true; +} + +static void fill_property(void *buf, uint64_t val, uint32_t cells) +{ + uint32_t val32; + int i; + + if (cells == 1) { + val32 = cpu_to_fdt32((uint32_t)val); + memcpy(buf, &val32, sizeof(uint32_t)); + } else { + for (i = 0; + i < (cells * sizeof(uint32_t) - sizeof(uint64_t)); i++) + *(char *)buf++ = 0; + + val = cpu_to_fdt64(val); + memcpy(buf, &val, sizeof(uint64_t)); + } +} + +static int fdt_setprop_ranges(void *fdt, int nodeoffset, const char *name, + struct memory_range *ranges, int nr_ranges, bool reverse, + uint32_t address_cells, uint32_t size_cells) +{ + void *buf, *prop; + size_t buf_size; + int i, result; + struct memory_range *range; + + buf_size = (address_cells + size_cells) * sizeof(uint32_t) * nr_ranges; + prop = buf = xmalloc(buf_size); + if (!buf) + return -ENOMEM; + + for (i = 0; i < nr_ranges; i++) { + if (reverse) + range = ranges + (nr_ranges - 1 - i); + else + range = ranges + i; + + fill_property(prop, range->start, address_cells); + prop += address_cells * sizeof(uint32_t); + + fill_property(prop, range->end - range->start + 1, size_cells); + prop += size_cells * sizeof(uint32_t); + } + + result = fdt_setprop(fdt, nodeoffset, name, buf, buf_size); + + free(buf); + + return result; +} + +/** + * setup_2nd_dtb - Setup the 2nd stage kernel's dtb. + */ + +static int setup_2nd_dtb(struct dtb *dtb, char *command_line, int on_crash) +{ + uint32_t address_cells, size_cells; + uint64_t fdt_val64; + uint64_t *prop; + char *new_buf = NULL; + int len, range_len; + int nodeoffset; + int new_size; + int i, result, kaslr_seed; + + result = fdt_check_header(dtb->buf); + + if (result) { + fprintf(stderr, "kexec: Invalid 2nd device tree.\n"); + return EFAILED; + } + + result = set_bootargs(dtb, command_line); + if (result) { + fprintf(stderr, "kexec: cannot set bootargs.\n"); + result = -EINVAL; + goto on_error; + } + + /* determine #address-cells and #size-cells */ + result = get_cells_size(dtb->buf, &address_cells, &size_cells); + if (result) { + fprintf(stderr, "kexec: cannot determine cells-size.\n"); + result = -EINVAL; + goto on_error; + } + + if (!cells_size_fitted(address_cells, size_cells, + &elfcorehdr_mem)) { + fprintf(stderr, "kexec: elfcorehdr doesn't fit cells-size.\n"); + result = -EINVAL; + goto on_error; + } + + for (i = 0; i < usablemem_rgns.size; i++) { + if (!cells_size_fitted(address_cells, size_cells, + &crash_reserved_mem[i])) { + fprintf(stderr, "kexec: usable memory range doesn't fit cells-size.\n"); + result = -EINVAL; + goto on_error; + } + } + + /* duplicate dt blob */ + range_len = sizeof(uint32_t) * (address_cells + size_cells); + new_size = fdt_totalsize(dtb->buf) + + fdt_prop_len(PROP_ELFCOREHDR, range_len) + + fdt_prop_len(PROP_USABLE_MEM_RANGE, range_len * usablemem_rgns.size); + + new_buf = xmalloc(new_size); + result = fdt_open_into(dtb->buf, new_buf, new_size); + if (result) { + dbgprintf("%s: fdt_open_into failed: %s\n", __func__, + fdt_strerror(result)); + result = -ENOSPC; + goto on_error; + } + + /* fixup 'kaslr-seed' with a random value, if supported */ + nodeoffset = fdt_path_offset(new_buf, "/chosen"); + prop = fdt_getprop_w(new_buf, nodeoffset, + "kaslr-seed", &len); + if (!prop || len != sizeof(uint64_t)) { + dbgprintf("%s: no kaslr-seed found\n", + __func__); + /* for kexec warm reboot case, we don't need to fixup + * other dtb properties + */ + if (!on_crash) { + dump_reservemap(dtb); + if (new_buf) + free(new_buf); + + return result; + } + } else { + kaslr_seed = fdt64_to_cpu(*prop); + + /* kaslr_seed must be wiped clean by primary + * kernel during boot + */ + if (kaslr_seed != 0) { + dbgprintf("%s: kaslr-seed is not wiped to 0.\n", + __func__); + result = -EINVAL; + goto on_error; + } + + /* + * Invoke the getrandom system call with + * GRND_NONBLOCK, to make sure we + * have a valid random seed to pass to the + * secondary kernel. + */ + result = syscall(SYS_getrandom, &fdt_val64, + sizeof(fdt_val64), + GRND_NONBLOCK); + + if(result == -1) { + fprintf(stderr, "%s: Reading random bytes failed.\n", + __func__); + + /* Currently on some arm64 platforms this + * 'getrandom' system call fails while booting + * the platform. + * + * In case, this happens at best we can set + * the 'kaslr_seed' as 0, indicating that the + * 2nd kernel will be booted with a 'nokaslr' + * like behaviour. + */ + fdt_val64 = 0UL; + dbgprintf("%s: Disabling KASLR in secondary kernel.\n", + __func__); + } + + nodeoffset = fdt_path_offset(new_buf, "/chosen"); + result = fdt_setprop_inplace(new_buf, + nodeoffset, "kaslr-seed", + &fdt_val64, sizeof(fdt_val64)); + if (result) { + dbgprintf("%s: fdt_setprop failed: %s\n", + __func__, fdt_strerror(result)); + result = -EINVAL; + goto on_error; + } + } + + if (on_crash) { + /* add linux,elfcorehdr */ + nodeoffset = fdt_path_offset(new_buf, "/chosen"); + result = fdt_setprop_ranges(new_buf, nodeoffset, + PROP_ELFCOREHDR, &elfcorehdr_mem, 1, false, + address_cells, size_cells); + if (result) { + dbgprintf("%s: fdt_setprop failed: %s\n", __func__, + fdt_strerror(result)); + result = -EINVAL; + goto on_error; + } + + /* + * add linux,usable-memory-range + * + * crash dump kernel support one or two regions, to make + * compatibility with existing user-space and older kdump, the + * low region is always the last one. + */ + nodeoffset = fdt_path_offset(new_buf, "/chosen"); + result = fdt_setprop_ranges(new_buf, nodeoffset, + PROP_USABLE_MEM_RANGE, + usablemem_rgns.ranges, usablemem_rgns.size, true, + address_cells, size_cells); + if (result) { + dbgprintf("%s: fdt_setprop failed: %s\n", __func__, + fdt_strerror(result)); + result = -EINVAL; + goto on_error; + } + } + + fdt_pack(new_buf); + dtb->buf = new_buf; + dtb->size = fdt_totalsize(new_buf); + + dump_reservemap(dtb); + + return result; + +on_error: + fprintf(stderr, "kexec: %s failed.\n", __func__); + if (new_buf) + free(new_buf); + + return result; +} + +unsigned long arm64_locate_kernel_segment(struct kexec_info *info) +{ + unsigned long hole; + + if (info->kexec_flags & KEXEC_ON_CRASH) { + unsigned long hole_end; + + hole = (crash_reserved_mem[usablemem_rgns.size - 1].start < mem_min ? + mem_min : crash_reserved_mem[usablemem_rgns.size - 1].start); + hole = _ALIGN_UP(hole, MiB(2)); + hole_end = hole + arm64_mem.text_offset + arm64_mem.image_size; + + if ((hole_end > mem_max) || + (hole_end > crash_reserved_mem[usablemem_rgns.size - 1].end)) { + dbgprintf("%s: Crash kernel out of range\n", __func__); + hole = ULONG_MAX; + } + } else { + hole = locate_hole(info, + arm64_mem.text_offset + arm64_mem.image_size, + MiB(2), 0, ULONG_MAX, 1); + + if (hole == ULONG_MAX) + dbgprintf("%s: locate_hole failed\n", __func__); + } + + return hole; +} + +/** + * arm64_load_other_segments - Prepare the dtb, initrd and purgatory segments. + */ + +int arm64_load_other_segments(struct kexec_info *info, + unsigned long image_base) +{ + int result; + unsigned long dtb_base; + unsigned long hole_min; + unsigned long hole_max; + unsigned long initrd_end; + uint64_t purgatory_sink; + char *initrd_buf = NULL; + struct dtb dtb; + char command_line[COMMAND_LINE_SIZE] = ""; + + if (arm64_opts.command_line) { + if (strlen(arm64_opts.command_line) > + sizeof(command_line) - 1) { + fprintf(stderr, + "Kernel command line too long for kernel!\n"); + return EFAILED; + } + + strncpy(command_line, arm64_opts.command_line, + sizeof(command_line) - 1); + command_line[sizeof(command_line) - 1] = 0; + } + + purgatory_sink = find_purgatory_sink(arm64_opts.console); + + dbgprintf("%s:%d: purgatory sink: 0x%" PRIx64 "\n", __func__, __LINE__, + purgatory_sink); + + if (arm64_opts.dtb) { + dtb.name = "dtb_user"; + dtb.buf = slurp_file(arm64_opts.dtb, &dtb.size); + } else { + result = read_1st_dtb(&dtb); + + if (result) { + fprintf(stderr, + "kexec: Error: No device tree available.\n"); + return EFAILED; + } + } + + result = setup_2nd_dtb(&dtb, command_line, + info->kexec_flags & KEXEC_ON_CRASH); + + if (result) + return EFAILED; + + /* Put the other segments after the image. */ + + hole_min = image_base + arm64_mem.image_size; + if (info->kexec_flags & KEXEC_ON_CRASH) + hole_max = crash_reserved_mem[usablemem_rgns.size - 1].end; + else + hole_max = ULONG_MAX; + + if (arm64_opts.initrd) { + initrd_buf = slurp_file(arm64_opts.initrd, &initrd_size); + + if (!initrd_buf) + fprintf(stderr, "kexec: Empty ramdisk file.\n"); + else { + /* Put the initrd after the kernel. */ + + initrd_base = add_buffer_phys_virt(info, initrd_buf, + initrd_size, initrd_size, 0, + hole_min, hole_max, 1, 0); + + initrd_end = initrd_base + initrd_size; + + /* Check limits as specified in booting.txt. + * The kernel may have as little as 32 GB of address space to map + * system memory and both kernel and initrd must be 1GB aligend. + */ + + if (_ALIGN_UP(initrd_end, GiB(1)) - _ALIGN_DOWN(image_base, GiB(1)) > GiB(32)) { + fprintf(stderr, "kexec: Error: image + initrd too big.\n"); + return EFAILED; + } + + dbgprintf("initrd: base %lx, size %lxh (%ld)\n", + initrd_base, initrd_size, initrd_size); + + result = dtb_set_initrd((char **)&dtb.buf, + &dtb.size, initrd_base, + initrd_base + initrd_size); + + if (result) + return EFAILED; + } + } + + if (!initrd_buf) { + /* Don't reuse the initrd addresses from 1st DTB */ + dtb_clear_initrd((char **)&dtb.buf, &dtb.size); + } + + /* Check size limit as specified in booting.txt. */ + + if (dtb.size > MiB(2)) { + fprintf(stderr, "kexec: Error: dtb too big.\n"); + return EFAILED; + } + + dtb_base = add_buffer_phys_virt(info, dtb.buf, dtb.size, dtb.size, + 0, hole_min, hole_max, 1, 0); + + /* dtb_base is valid if we got here. */ + + dbgprintf("dtb: base %lx, size %lxh (%ld)\n", dtb_base, dtb.size, + dtb.size); + + elf_rel_build_load(info, &info->rhdr, purgatory, purgatory_size, + hole_min, hole_max, 1, 0); + + info->entry = (void *)elf_rel_get_addr(&info->rhdr, "purgatory_start"); + + elf_rel_set_symbol(&info->rhdr, "arm64_sink", &purgatory_sink, + sizeof(purgatory_sink)); + + elf_rel_set_symbol(&info->rhdr, "arm64_kernel_entry", &image_base, + sizeof(image_base)); + + elf_rel_set_symbol(&info->rhdr, "arm64_dtb_addr", &dtb_base, + sizeof(dtb_base)); + + return 0; +} + +/** + * virt_to_phys - For processing elf file values. + */ + +unsigned long virt_to_phys(unsigned long v) +{ + unsigned long p; + + p = v - get_vp_offset() + get_phys_offset(); + + return p; +} + +/** + * phys_to_virt - For crashdump setup. + */ + +unsigned long phys_to_virt(struct crash_elf_info *elf_info, + unsigned long long p) +{ + unsigned long v; + + v = p - get_phys_offset() + elf_info->page_offset; + + return v; +} + +/** + * add_segment - Use virt_to_phys when loading elf files. + */ + +void add_segment(struct kexec_info *info, const void *buf, size_t bufsz, + unsigned long base, size_t memsz) +{ + add_segment_phys_virt(info, buf, bufsz, base, memsz, 1); +} + +static inline void set_phys_offset(int64_t v, char *set_method) +{ + if (arm64_mem.phys_offset == arm64_mem_ngv + || v < arm64_mem.phys_offset) { + arm64_mem.phys_offset = v; + dbgprintf("%s: phys_offset : %016lx (method : %s)\n", + __func__, arm64_mem.phys_offset, + set_method); + } +} + +/** + * get_va_bits - Helper for getting VA_BITS + */ + +static int get_va_bits(void) +{ + unsigned long long stext_sym_addr; + + /* + * if already got from kcore + */ + if (va_bits != -1) + goto out; + + + /* For kernel older than v4.19 */ + fprintf(stderr, "Warning, can't get the VA_BITS from kcore\n"); + stext_sym_addr = get_kernel_sym("_stext"); + + if (stext_sym_addr == 0) { + fprintf(stderr, "Can't get the symbol of _stext.\n"); + return -1; + } + + /* Derive va_bits as per arch/arm64/Kconfig */ + if ((stext_sym_addr & PAGE_OFFSET_36) == PAGE_OFFSET_36) { + va_bits = 36; + } else if ((stext_sym_addr & PAGE_OFFSET_39) == PAGE_OFFSET_39) { + va_bits = 39; + } else if ((stext_sym_addr & PAGE_OFFSET_42) == PAGE_OFFSET_42) { + va_bits = 42; + } else if ((stext_sym_addr & PAGE_OFFSET_47) == PAGE_OFFSET_47) { + va_bits = 47; + } else if ((stext_sym_addr & PAGE_OFFSET_48) == PAGE_OFFSET_48) { + va_bits = 48; + } else { + fprintf(stderr, + "Cannot find a proper _stext for calculating VA_BITS\n"); + return -1; + } + +out: + dbgprintf("va_bits : %d\n", va_bits); + + return 0; +} + +/** + * get_page_offset - Helper for getting PAGE_OFFSET + */ + +int get_page_offset(unsigned long *page_offset) +{ + unsigned long long text_sym_addr, kernel_va_mid; + int ret; + + text_sym_addr = get_kernel_sym("_text"); + if (text_sym_addr == 0) { + fprintf(stderr, "Can't get the symbol of _text to calculate page_offset.\n"); + return -1; + } + + ret = get_va_bits(); + if (ret < 0) + return ret; + + /* Since kernel 5.4, kernel image is put above + * UINT64_MAX << (va_bits - 1) + */ + kernel_va_mid = UINT64_MAX << (va_bits - 1); + /* older kernel */ + if (text_sym_addr < kernel_va_mid) + *page_offset = UINT64_MAX << (va_bits - 1); + else + *page_offset = UINT64_MAX << va_bits; + + dbgprintf("page_offset : %lx\n", *page_offset); + + return 0; +} + +static void arm64_scan_vmcoreinfo(char *pos) +{ + const char *str; + + str = "NUMBER(VA_BITS)="; + if (memcmp(str, pos, strlen(str)) == 0) + va_bits = strtoul(pos + strlen(str), NULL, 10); +} + +/** + * get_phys_offset_from_vmcoreinfo_pt_note - Helper for getting PHYS_OFFSET (and va_bits) + * from VMCOREINFO note inside 'kcore'. + */ + +static int get_phys_offset_from_vmcoreinfo_pt_note(long *phys_offset) +{ + int fd, ret = 0; + + if ((fd = open("/proc/kcore", O_RDONLY)) < 0) { + fprintf(stderr, "Can't open (%s).\n", "/proc/kcore"); + return EFAILED; + } + + arch_scan_vmcoreinfo = arm64_scan_vmcoreinfo; + ret = read_phys_offset_elf_kcore(fd, phys_offset); + + close(fd); + return ret; +} + +/** + * get_phys_base_from_pt_load - Helper for getting PHYS_OFFSET + * from PT_LOADs inside 'kcore'. + */ + +int get_phys_base_from_pt_load(long *phys_offset) +{ + int i, fd, ret; + unsigned long long phys_start; + unsigned long long virt_start; + + ret = get_page_offset(&page_offset); + if (ret < 0) + return ret; + + if ((fd = open("/proc/kcore", O_RDONLY)) < 0) { + fprintf(stderr, "Can't open (%s).\n", "/proc/kcore"); + return EFAILED; + } + + read_elf(fd); + + for (i = 0; get_pt_load(i, + &phys_start, NULL, &virt_start, NULL); + i++) { + if (virt_start != NOT_KV_ADDR + && virt_start >= page_offset + && phys_start != NOT_PADDR) + *phys_offset = phys_start - + (virt_start & ~page_offset); + } + + close(fd); + return 0; +} + +static bool to_be_excluded(char *str, unsigned long long start, unsigned long long end) +{ + if (!strncmp(str, CRASH_KERNEL, strlen(CRASH_KERNEL))) { + uint64_t load_start, load_end; + + if (!get_crash_kernel_load_range(&load_start, &load_end) && + (load_start == start) && (load_end == end)) + return false; + + return true; + } + + if (!strncmp(str, SYSTEM_RAM, strlen(SYSTEM_RAM)) || + !strncmp(str, KERNEL_CODE, strlen(KERNEL_CODE)) || + !strncmp(str, KERNEL_DATA, strlen(KERNEL_DATA))) + return false; + else + return true; +} + +/** + * get_memory_ranges - Try to get the memory ranges from + * /proc/iomem. + */ +int get_memory_ranges(struct memory_range **range, int *ranges, + unsigned long kexec_flags) +{ + long phys_offset = -1; + FILE *fp; + const char *iomem = proc_iomem(); + char line[MAX_LINE], *str; + unsigned long long start, end; + int n, consumed; + struct memory_ranges memranges; + struct memory_range *last, excl_range; + int ret; + + if (!try_read_phys_offset_from_kcore) { + /* Since kernel version 4.19, 'kcore' contains + * a new PT_NOTE which carries the VMCOREINFO + * information. + * If the same is available, one should prefer the + * same to retrieve 'PHYS_OFFSET' value exported by + * the kernel as this is now the standard interface + * exposed by kernel for sharing machine specific + * details with the userland. + */ + ret = get_phys_offset_from_vmcoreinfo_pt_note(&phys_offset); + if (!ret) { + if (phys_offset != -1) + set_phys_offset(phys_offset, + "vmcoreinfo pt_note"); + } else { + /* If we are running on a older kernel, + * try to retrieve the 'PHYS_OFFSET' value + * exported by the kernel in the 'kcore' + * file by reading the PT_LOADs and determining + * the correct combination. + */ + ret = get_phys_base_from_pt_load(&phys_offset); + if (!ret) + if (phys_offset != -1) + set_phys_offset(phys_offset, + "pt_load"); + } + + try_read_phys_offset_from_kcore = true; + } + + fp = fopen(iomem, "r"); + if (!fp) + die("Cannot open %s\n", iomem); + + memranges.ranges = NULL; + memranges.size = memranges.max_size = 0; + + while (fgets(line, sizeof(line), fp) != 0) { + n = sscanf(line, "%llx-%llx : %n", &start, &end, &consumed); + if (n != 2) + continue; + str = line + consumed; + + if (!strncmp(str, SYSTEM_RAM, strlen(SYSTEM_RAM))) { + ret = mem_regions_alloc_and_add(&memranges, + start, end - start + 1, RANGE_RAM); + if (ret) { + fprintf(stderr, + "Cannot allocate memory for ranges\n"); + fclose(fp); + return -ENOMEM; + } + + dbgprintf("%s:+[%d] %016llx - %016llx\n", __func__, + memranges.size - 1, + memranges.ranges[memranges.size - 1].start, + memranges.ranges[memranges.size - 1].end); + } else if (to_be_excluded(str, start, end)) { + if (!memranges.size) + continue; + + /* + * Note: mem_regions_exclude() doesn't guarantee + * that the ranges are sorted out, but as long as + * we cope with /proc/iomem, we only operate on + * the last entry and so it is safe. + */ + + /* The last System RAM range */ + last = &memranges.ranges[memranges.size - 1]; + + if (last->end < start) + /* New resource outside of System RAM */ + continue; + if (end < last->start) + /* Already excluded by parent resource */ + continue; + + excl_range.start = start; + excl_range.end = end; + ret = mem_regions_alloc_and_exclude(&memranges, &excl_range); + if (ret) { + fprintf(stderr, + "Cannot allocate memory for ranges (exclude)\n"); + fclose(fp); + return -ENOMEM; + } + dbgprintf("%s:- %016llx - %016llx\n", + __func__, start, end); + } + } + + fclose(fp); + + *range = memranges.ranges; + *ranges = memranges.size; + + /* As a fallback option, we can try determining the PHYS_OFFSET + * value from the '/proc/iomem' entries as well. + * + * But note that this can be flaky, as on certain arm64 + * platforms, it has been noticed that due to a hole at the + * start of physical ram exposed to kernel + * (i.e. it doesn't start from address 0), the kernel still + * calculates the 'memstart_addr' kernel variable as 0. + * + * Whereas the SYSTEM_RAM or IOMEM_RESERVED range in + * '/proc/iomem' would carry a first entry whose start address + * is non-zero (as the physical ram exposed to the kernel + * starts from a non-zero address). + * + * In such cases, if we rely on '/proc/iomem' entries to + * calculate the phys_offset, then we will have mismatch + * between the user-space and kernel space 'PHYS_OFFSET' + * value. + */ + if (memranges.size) + set_phys_offset(memranges.ranges[0].start, "iomem"); + + dbgprint_mem_range("System RAM ranges;", + memranges.ranges, memranges.size); + + return 0; +} + +int arch_compat_trampoline(struct kexec_info *info) +{ + return 0; +} + +int machine_verify_elf_rel(struct mem_ehdr *ehdr) +{ + return (ehdr->e_machine == EM_AARCH64); +} + +enum aarch64_rel_type { + R_AARCH64_NONE = 0, + R_AARCH64_ABS64 = 257, + R_AARCH64_PREL32 = 261, + R_AARCH64_MOVW_UABS_G0_NC = 264, + R_AARCH64_MOVW_UABS_G1_NC = 266, + R_AARCH64_MOVW_UABS_G2_NC = 268, + R_AARCH64_MOVW_UABS_G3 =269, + R_AARCH64_LD_PREL_LO19 = 273, + R_AARCH64_ADR_PREL_LO21 = 274, + R_AARCH64_ADR_PREL_PG_HI21 = 275, + R_AARCH64_ADD_ABS_LO12_NC = 277, + R_AARCH64_JUMP26 = 282, + R_AARCH64_CALL26 = 283, + R_AARCH64_LDST64_ABS_LO12_NC = 286, + R_AARCH64_LDST128_ABS_LO12_NC = 299 +}; + +static uint32_t get_bits(uint32_t value, int start, int end) +{ + uint32_t mask = ((uint32_t)1 << (end + 1 - start)) - 1; + return (value >> start) & mask; +} + +void machine_apply_elf_rel(struct mem_ehdr *ehdr, struct mem_sym *UNUSED(sym), + unsigned long r_type, void *ptr, unsigned long address, + unsigned long value) +{ + uint64_t *loc64; + uint32_t *loc32; + uint64_t *location = (uint64_t *)ptr; + uint64_t data = *location; + uint64_t imm; + const char *type = NULL; + + switch((enum aarch64_rel_type)r_type) { + case R_AARCH64_ABS64: + type = "ABS64"; + loc64 = ptr; + *loc64 = cpu_to_elf64(ehdr, value); + break; + case R_AARCH64_PREL32: + type = "PREL32"; + loc32 = ptr; + *loc32 = cpu_to_elf32(ehdr, value - address); + break; + + /* Set a MOV[KZ] immediate field to bits [15:0] of X. No overflow check */ + case R_AARCH64_MOVW_UABS_G0_NC: + type = "MOVW_UABS_G0_NC"; + loc32 = ptr; + imm = get_bits(value, 0, 15); + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + (imm << 5)); + break; + /* Set a MOV[KZ] immediate field to bits [31:16] of X. No overflow check */ + case R_AARCH64_MOVW_UABS_G1_NC: + type = "MOVW_UABS_G1_NC"; + loc32 = ptr; + imm = get_bits(value, 16, 31); + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + (imm << 5)); + break; + /* Set a MOV[KZ] immediate field to bits [47:32] of X. No overflow check */ + case R_AARCH64_MOVW_UABS_G2_NC: + type = "MOVW_UABS_G2_NC"; + loc32 = ptr; + imm = get_bits(value, 32, 47); + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + (imm << 5)); + break; + /* Set a MOV[KZ] immediate field to bits [63:48] of X */ + case R_AARCH64_MOVW_UABS_G3: + type = "MOVW_UABS_G3"; + loc32 = ptr; + imm = get_bits(value, 48, 63); + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + (imm << 5)); + break; + + case R_AARCH64_LD_PREL_LO19: + type = "LD_PREL_LO19"; + loc32 = ptr; + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + + (((value - address) << 3) & 0xffffe0)); + break; + case R_AARCH64_ADR_PREL_LO21: + if (value & 3) + die("%s: ERROR Unaligned value: %lx\n", __func__, + value); + type = "ADR_PREL_LO21"; + loc32 = ptr; + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + + (((value - address) << 3) & 0xffffe0)); + break; + case R_AARCH64_ADR_PREL_PG_HI21: + type = "ADR_PREL_PG_HI21"; + imm = ((value & ~0xfff) - (address & ~0xfff)) >> 12; + loc32 = ptr; + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + + ((imm & 3) << 29) + ((imm & 0x1ffffc) << (5 - 2))); + break; + case R_AARCH64_ADD_ABS_LO12_NC: + type = "ADD_ABS_LO12_NC"; + loc32 = ptr; + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + + ((value & 0xfff) << 10)); + break; + case R_AARCH64_JUMP26: + type = "JUMP26"; + loc32 = ptr; + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + + (((value - address) >> 2) & 0x3ffffff)); + break; + case R_AARCH64_CALL26: + type = "CALL26"; + loc32 = ptr; + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + + (((value - address) >> 2) & 0x3ffffff)); + break; + /* encode imm field with bits [11:3] of value */ + case R_AARCH64_LDST64_ABS_LO12_NC: + if (value & 7) + die("%s: ERROR Unaligned value: %lx\n", __func__, + value); + type = "LDST64_ABS_LO12_NC"; + loc32 = ptr; + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + + ((value & 0xff8) << (10 - 3))); + break; + + /* encode imm field with bits [11:4] of value */ + case R_AARCH64_LDST128_ABS_LO12_NC: + if (value & 15) + die("%s: ERROR Unaligned value: %lx\n", __func__, + value); + type = "LDST128_ABS_LO12_NC"; + loc32 = ptr; + imm = value & 0xff0; + *loc32 = cpu_to_le32(le32_to_cpu(*loc32) + (imm << (10 - 4))); + break; + default: + die("%s: ERROR Unknown type: %lu\n", __func__, r_type); + break; + } + + dbgprintf("%s: %s %016lx->%016lx\n", __func__, type, data, *location); +} + +void arch_reuse_initrd(void) +{ + reuse_initrd = 1; +} + +void arch_update_purgatory(struct kexec_info *UNUSED(info)) +{ +} diff --git a/kexec/arch/arm64/kexec-arm64.h b/kexec/arch/arm64/kexec-arm64.h new file mode 100644 index 0000000..95fb5c2 --- /dev/null +++ b/kexec/arch/arm64/kexec-arm64.h @@ -0,0 +1,84 @@ +/* + * ARM64 kexec. + */ + +#if !defined(KEXEC_ARM64_H) +#define KEXEC_ARM64_H + +#include <stdbool.h> +#include <sys/types.h> + +#include "image-header.h" +#include "kexec.h" + +#define KEXEC_SEGMENT_MAX 64 + +#define BOOT_BLOCK_VERSION 17 +#define BOOT_BLOCK_LAST_COMP_VERSION 16 +#define COMMAND_LINE_SIZE 2048 /* from kernel */ + +#define KiB(x) ((x) * 1024UL) +#define MiB(x) (KiB(x) * 1024UL) +#define GiB(x) (MiB(x) * 1024UL) + +#define ULONGLONG_MAX (~0ULL) + +/* + * Incorrect address + */ +#define NOT_KV_ADDR (0x0) +#define NOT_PADDR (ULONGLONG_MAX) + +int elf_arm64_probe(const char *kernel_buf, off_t kernel_size); +int elf_arm64_load(int argc, char **argv, const char *kernel_buf, + off_t kernel_size, struct kexec_info *info); +void elf_arm64_usage(void); + +int image_arm64_probe(const char *kernel_buf, off_t kernel_size); +int image_arm64_load(int argc, char **argv, const char *kernel_buf, + off_t kernel_size, struct kexec_info *info); +void image_arm64_usage(void); + +int uImage_arm64_probe(const char *buf, off_t len); +int uImage_arm64_load(int argc, char **argv, const char *buf, off_t len, + struct kexec_info *info); +void uImage_arm64_usage(void); + +int pez_arm64_probe(const char *kernel_buf, off_t kernel_size); +int pez_arm64_load(int argc, char **argv, const char *buf, off_t len, + struct kexec_info *info); +void pez_arm64_usage(void); + + +extern off_t initrd_base; +extern off_t initrd_size; + +/** + * struct arm64_mem - Memory layout info. + */ + +struct arm64_mem { + int64_t phys_offset; + uint64_t text_offset; + uint64_t image_size; + uint64_t vp_offset; +}; + +#define arm64_mem_ngv UINT64_MAX +extern struct arm64_mem arm64_mem; + +uint64_t get_phys_offset(void); +uint64_t get_vp_offset(void); +int get_page_offset(unsigned long *offset); + +static inline void reset_vp_offset(void) +{ + arm64_mem.vp_offset = arm64_mem_ngv; +} + +int arm64_process_image_header(const struct arm64_image_header *h); +unsigned long arm64_locate_kernel_segment(struct kexec_info *info); +int arm64_load_other_segments(struct kexec_info *info, + unsigned long image_base); + +#endif diff --git a/kexec/arch/arm64/kexec-elf-arm64.c b/kexec/arch/arm64/kexec-elf-arm64.c new file mode 100644 index 0000000..e14f8e9 --- /dev/null +++ b/kexec/arch/arm64/kexec-elf-arm64.c @@ -0,0 +1,170 @@ +/* + * ARM64 kexec elf support. + */ + +#define _GNU_SOURCE + +#include <errno.h> +#include <fcntl.h> +#include <limits.h> +#include <stdlib.h> +#include <linux/elf.h> + +#include "arch/options.h" +#include "crashdump-arm64.h" +#include "kexec-arm64.h" +#include "kexec-elf.h" +#include "kexec-syscall.h" + +int elf_arm64_probe(const char *kernel_buf, off_t kernel_size) +{ + struct mem_ehdr ehdr; + int result; + + result = build_elf_exec_info(kernel_buf, kernel_size, &ehdr, 0); + + if (result < 0) { + dbgprintf("%s: Not an ELF executable.\n", __func__); + goto on_exit; + } + + if (ehdr.e_machine != EM_AARCH64) { + dbgprintf("%s: Not an AARCH64 ELF executable.\n", __func__); + result = -1; + goto on_exit; + } + + result = 0; +on_exit: + free_elf_info(&ehdr); + return result; +} + +int elf_arm64_load(int argc, char **argv, const char *kernel_buf, + off_t kernel_size, struct kexec_info *info) +{ + const struct arm64_image_header *header = NULL; + unsigned long kernel_segment; + struct mem_ehdr ehdr; + int result; + int i; + + if (info->file_mode) { + fprintf(stderr, + "ELF executable is not supported in kexec_file\n"); + + return EFAILED; + } + + result = build_elf_exec_info(kernel_buf, kernel_size, &ehdr, 0); + + if (result < 0) { + dbgprintf("%s: build_elf_exec_info failed\n", __func__); + goto exit; + } + + /* Find and process the arm64 image header. */ + + for (i = 0; i < ehdr.e_phnum; i++) { + struct mem_phdr *phdr = &ehdr.e_phdr[i]; + unsigned long header_offset; + + if (phdr->p_type != PT_LOAD) + continue; + + /* + * When CONFIG_ARM64_RANDOMIZE_TEXT_OFFSET=y the image header + * could be offset in the elf segment. The linker script sets + * ehdr.e_entry to the start of text. + */ + + header_offset = ehdr.e_entry - phdr->p_vaddr; + + header = (const struct arm64_image_header *)( + kernel_buf + phdr->p_offset + header_offset); + + if (!arm64_process_image_header(header)) { + dbgprintf("%s: e_entry: %016llx\n", __func__, + ehdr.e_entry); + dbgprintf("%s: p_vaddr: %016llx\n", __func__, + phdr->p_vaddr); + dbgprintf("%s: header_offset: %016lx\n", __func__, + header_offset); + + break; + } + } + + if (i == ehdr.e_phnum) { + dbgprintf("%s: Valid arm64 header not found\n", __func__); + result = EFAILED; + goto exit; + } + + kernel_segment = arm64_locate_kernel_segment(info); + + if (kernel_segment == ULONG_MAX) { + dbgprintf("%s: Kernel segment is not allocated\n", __func__); + result = EFAILED; + goto exit; + } + + arm64_mem.vp_offset = _ALIGN_DOWN(ehdr.e_entry, MiB(2)); + if (!(info->kexec_flags & KEXEC_ON_CRASH)) + arm64_mem.vp_offset -= kernel_segment - get_phys_offset(); + + dbgprintf("%s: kernel_segment: %016lx\n", __func__, kernel_segment); + dbgprintf("%s: text_offset: %016lx\n", __func__, + arm64_mem.text_offset); + dbgprintf("%s: image_size: %016lx\n", __func__, + arm64_mem.image_size); + dbgprintf("%s: phys_offset: %016lx\n", __func__, + arm64_mem.phys_offset); + dbgprintf("%s: vp_offset: %016lx\n", __func__, + arm64_mem.vp_offset); + dbgprintf("%s: PE format: %s\n", __func__, + (arm64_header_check_pe_sig(header) ? "yes" : "no")); + + /* create and initialize elf core header segment */ + if (info->kexec_flags & KEXEC_ON_CRASH) { + result = load_crashdump_segments(info); + if (result) { + dbgprintf("%s: Creating eflcorehdr failed.\n", + __func__); + goto exit; + } + } + + /* load the kernel */ + if (info->kexec_flags & KEXEC_ON_CRASH) + /* + * offset addresses in elf header in order to load + * vmlinux (elf_exec) into crash kernel's memory + */ + fixup_elf_addrs(&ehdr); + + result = elf_exec_load(&ehdr, info); + + if (result) { + dbgprintf("%s: elf_exec_load failed\n", __func__); + goto exit; + } + + /* load additional data */ + result = arm64_load_other_segments(info, kernel_segment + + arm64_mem.text_offset); + +exit: + reset_vp_offset(); + free_elf_info(&ehdr); + if (result) + fprintf(stderr, "kexec: Bad elf image file, load failed.\n"); + return result; +} + +void elf_arm64_usage(void) +{ + printf( +" An ARM64 ELF image, big or little endian.\n" +" Typically vmlinux or a stripped version of vmlinux.\n\n"); +} diff --git a/kexec/arch/arm64/kexec-image-arm64.c b/kexec/arch/arm64/kexec-image-arm64.c new file mode 100644 index 0000000..a196747 --- /dev/null +++ b/kexec/arch/arm64/kexec-image-arm64.c @@ -0,0 +1,119 @@ +/* + * ARM64 kexec binary image support. + */ + +#define _GNU_SOURCE + +#include <errno.h> +#include <fcntl.h> +#include <limits.h> +#include "crashdump-arm64.h" +#include "image-header.h" +#include "kexec.h" +#include "kexec-arm64.h" +#include "kexec-syscall.h" +#include "arch/options.h" + +int image_arm64_probe(const char *kernel_buf, off_t kernel_size) +{ + const struct arm64_image_header *h; + + if (kernel_size < sizeof(struct arm64_image_header)) { + dbgprintf("%s: No arm64 image header.\n", __func__); + return -1; + } + + h = (const struct arm64_image_header *)(kernel_buf); + + if (!arm64_header_check_magic(h)) { + dbgprintf("%s: Bad arm64 image header.\n", __func__); + return -1; + } + + return 0; +} + +int image_arm64_load(int argc, char **argv, const char *kernel_buf, + off_t kernel_size, struct kexec_info *info) +{ + const struct arm64_image_header *header; + unsigned long kernel_segment; + int result; + + if (info->file_mode) { + if (arm64_opts.initrd) { + info->initrd_fd = open(arm64_opts.initrd, O_RDONLY); + if (info->initrd_fd == -1) { + fprintf(stderr, + "Could not open initrd file %s:%s\n", + arm64_opts.initrd, strerror(errno)); + result = EFAILED; + goto exit; + } + } + + if (arm64_opts.command_line) { + info->command_line = (char *)arm64_opts.command_line; + info->command_line_len = + strlen(arm64_opts.command_line) + 1; + } + + return 0; + } + + header = (const struct arm64_image_header *)(kernel_buf); + + if (arm64_process_image_header(header)) + return EFAILED; + + kernel_segment = arm64_locate_kernel_segment(info); + + if (kernel_segment == ULONG_MAX) { + dbgprintf("%s: Kernel segment is not allocated\n", __func__); + result = EFAILED; + goto exit; + } + + dbgprintf("%s: kernel_segment: %016lx\n", __func__, kernel_segment); + dbgprintf("%s: text_offset: %016lx\n", __func__, + arm64_mem.text_offset); + dbgprintf("%s: image_size: %016lx\n", __func__, + arm64_mem.image_size); + dbgprintf("%s: phys_offset: %016lx\n", __func__, + arm64_mem.phys_offset); + dbgprintf("%s: vp_offset: %016lx\n", __func__, + arm64_mem.vp_offset); + dbgprintf("%s: PE format: %s\n", __func__, + (arm64_header_check_pe_sig(header) ? "yes" : "no")); + + /* create and initialize elf core header segment */ + if (info->kexec_flags & KEXEC_ON_CRASH) { + result = load_crashdump_segments(info); + if (result) { + dbgprintf("%s: Creating eflcorehdr failed.\n", + __func__); + goto exit; + } + } + + /* load the kernel */ + add_segment_phys_virt(info, kernel_buf, kernel_size, + kernel_segment + arm64_mem.text_offset, + arm64_mem.image_size, 0); + + /* load additional data */ + result = arm64_load_other_segments(info, kernel_segment + + arm64_mem.text_offset); + +exit: + if (result) + fprintf(stderr, "kexec: load failed.\n"); + return result; +} + +void image_arm64_usage(void) +{ + printf( +" An ARM64 binary image, compressed or not, big or little endian.\n" +" Typically an Image file.\n\n"); +} diff --git a/kexec/arch/arm64/kexec-uImage-arm64.c b/kexec/arch/arm64/kexec-uImage-arm64.c new file mode 100644 index 0000000..c466913 --- /dev/null +++ b/kexec/arch/arm64/kexec-uImage-arm64.c @@ -0,0 +1,52 @@ +/* + * uImage support added by David Woodhouse <dwmw2@infradead.org> + */ +#include <stdint.h> +#include <string.h> +#include <sys/types.h> +#include <image.h> +#include <kexec-uImage.h> +#include "../../kexec.h" +#include "kexec-arm64.h" + +int uImage_arm64_probe(const char *buf, off_t len) +{ + int ret; + + ret = uImage_probe_kernel(buf, len, IH_ARCH_ARM64); + + /* 0 - valid uImage. + * -1 - uImage is corrupted. + * 1 - image is not a uImage. + */ + if (!ret) + return 0; + else + return -1; +} + +int uImage_arm64_load(int argc, char **argv, const char *buf, off_t len, + struct kexec_info *info) +{ + struct Image_info img; + int ret; + + if (info->file_mode) { + fprintf(stderr, + "uImage is not supported in kexec_file\n"); + + return EFAILED; + } + + ret = uImage_load(buf, len, &img); + if (ret) + return ret; + + return image_arm64_load(argc, argv, img.buf, img.len, info); +} + +void uImage_arm64_usage(void) +{ + printf( +" An ARM64 U-boot uImage file, compressed or not, big or little endian.\n\n"); +} diff --git a/kexec/arch/arm64/kexec-vmlinuz-arm64.c b/kexec/arch/arm64/kexec-vmlinuz-arm64.c new file mode 100644 index 0000000..c0ee47c --- /dev/null +++ b/kexec/arch/arm64/kexec-vmlinuz-arm64.c @@ -0,0 +1,110 @@ +/* + * ARM64 PE compressed Image (vmlinuz, ZBOOT) support. + * + * Several distros use 'make zinstall' rule inside + * 'arch/arm64/boot/Makefile' to install the arm64 + * ZBOOT compressed file inside the boot destination + * directory (for e.g. /boot). + * + * Currently we cannot use kexec_file_load() to load vmlinuz + * PE images that self decompress. + * + * To support ZBOOT, we should: + * a). Copy the compressed contents of vmlinuz to a temporary file. + * b). Decompress (gunzip-decompress) the contents inside the + * temporary file. + * c). Validate the resulting image and write it back to the + * temporary file. + * d). Pass the 'fd' of the temporary file to the kernel space. + * + * Note this, module doesn't provide a _load() function instead + * relying on image_arm64_load() to load the resulting decompressed + * image. + * + * So basically the kernel space still gets a decompressed + * kernel image to load via kexec-tools. + */ + +#define _GNU_SOURCE +#include <fcntl.h> +#include <unistd.h> +#include <stdlib.h> +#include "kexec-arm64.h" +#include <kexec-pe-zboot.h> +#include "arch/options.h" + +static int kernel_fd = -1; + +/* Returns: + * -1 : in case of error/invalid format (not a valid PE+compressed ZBOOT format. + */ +int pez_arm64_probe(const char *kernel_buf, off_t kernel_size) +{ + int ret = -1; + const struct arm64_image_header *h; + char *buf; + off_t buf_sz; + + buf = (char *)kernel_buf; + buf_sz = kernel_size; + if (!buf) + return -1; + h = (const struct arm64_image_header *)buf; + + dbgprintf("%s: PROBE.\n", __func__); + if (buf_sz < sizeof(struct arm64_image_header)) { + dbgprintf("%s: Not large enough to be a PE image.\n", __func__); + return -1; + } + if (!arm64_header_check_pe_sig(h)) { + dbgprintf("%s: Not an PE image.\n", __func__); + return -1; + } + + if (buf_sz < sizeof(struct arm64_image_header) + h->pe_header) { + dbgprintf("%s: PE image offset larger than image.\n", __func__); + return -1; + } + + if (memcmp(&buf[h->pe_header], + arm64_pe_machtype, sizeof(arm64_pe_machtype))) { + dbgprintf("%s: PE header doesn't match machine type.\n", __func__); + return -1; + } + + ret = pez_prepare(buf, buf_sz, &kernel_fd); + + if (!ret) { + /* validate the arm64 specific header */ + struct arm64_image_header hdr_check; + if (read(kernel_fd, &hdr_check, sizeof(hdr_check)) != sizeof(hdr_check)) + goto bad_header; + + lseek(kernel_fd, 0, SEEK_SET); + + if (!arm64_header_check_magic(&hdr_check)) { + dbgprintf("%s: Bad arm64 image header.\n", __func__); + goto bad_header; + } + } + + return ret; +bad_header: + close(kernel_fd); + free(buf); + return -1; +} + +int pez_arm64_load(int argc, char **argv, const char *buf, off_t len, + struct kexec_info *info) +{ + info->kernel_fd = kernel_fd; + return image_arm64_load(argc, argv, buf, len, info); +} + +void pez_arm64_usage(void) +{ + printf( +" An ARM64 vmlinuz, PE image of a compressed, little endian.\n" +" kernel, built with ZBOOT enabled.\n\n"); +} |