path: root/include/elf/elf.h

#pragma once

// ============================================================================
// elf.h - ELF format definitions
// https://refspecs.linuxbase.org/elf/elf.pdf
// https://gist.github.com/x0nu11byt3/bcb35c3de461e5fb66173071a2379779
// https://docs.oracle.com/cd/E19683-01/816-1386/6m7qcoblh/index.html
// ============================================================================

#include <stdint.h>

/* ELF Types */
#pragma message("TODO: describe ELF types from specification")

// ============================================================================
// =================================== ELF Header =============================
// ============================================================================
#define EI_NIDENT (16)

/* e_ident offset (https://en.wikipedia.org/wiki/Executable_and_Linkable_Format#ELF_header) */
#define EI_MAG0         0
#define EI_MAG1         1
#define EI_MAG2         2
#define EI_MAG3         3
#define EI_CLASS        4
#define EI_DATA         5
#define EI_VERSION      6
#define EI_OSABI        7
#define EI_ABIVERSION   8
#define EI_PAD          9

/* ELF Magic value */
#define ELF_MAGIC 0x464C457F  // "\x7FELF" little-endian

/* ELF Class */
#define ELFCLASS32   0x1
#define ELFCLASS64   0x2

/* Data format: or LSB(Least Significant Bit, little-endian) or MSB(Most Significant Bit, big-endian) */
#define ELFDATA2LSB  0x1
#define ELFDATA2MSB  0x2

/* ELF version */
#define ELFVERSION   0x1

/* Target ABI */
#define ELFABISYSTEMV   0x00
#define ELFABIHPUX      0x01
#define ELFABINETBSD    0x02
#define ELFABILINUX     0x03
#define ELFABIGNUHURD   0x04
#define ELFABISOLARIS   0x06
#define ELFABIAIX       0x07
#define ELFABIIRIX      0x08
#define ELFABIFREEBSD   0x09
#define ELFABITRU64     0x0A
#define ELFABINM        0x0B
#define ELFABIOPENBSD   0x0C
#define ELFABIOPENVMS   0x0D
#define ELFABINSKERNEL  0x0E
#define ELFABIAROS      0x0F
#define ELFABIFENIXOS   0x10
#define ELFABICLOUDABI  0x11
#define ELFABISTOPENVOS 0x12

/* ABI version */
#pragma message("TODO: do something with ABI version")
/* Currently i don't use ABI version */

/* Object File Types, e_type field */
#define ET_NONE     0x0    /* Unknown */
#define ET_REL      0x1    /* Relocatable. (output of gcc -c). Contains sections + symbol tables + relocations. No program header */
#define ET_EXEC     0x2    /* Executable. Statically-linked executable with fixed virtual addresses. Has program headers. */
#define ET_DYN      0x3    /* Shared object. Shared library or PIE executable. Postitoin-independent, relocated at load time. */
#define ET_CORE     0x4    /* Core dump. Process memory image for post-mortem debugging */
#define ET_LOOS     0xFE00 /* Reserved inclusive range. Operating system specific */
#define ET_HIOS     0xFEFF /* Reserved inclusive range. Operating system specific */
#define ET_LOPROC   0xFF00 /* Reserved inclusive range. Processor specific */
#define ET_HIPROC   0xFFFF /* Reserved inclusive range. Processor specific */

/* Arch types, e_machine field */
#pragma message("TODO: add other architecture defines, copy from github")
#define EM_X86      0x03
#define EM_X86_64   0x3E
#define EM_ARM      0x28 /* up to Armv7/AArch32 */
#define EM_AARCH64  0xB7 /* Armv8/AArch64 */

/* Version value, e_version field */
#define EV_NONE     0x00
#define EV_CURRENT  0x01
#define EV_NUM      0x02

/* Other header fields: 
    e_entry: Memory address of the entry point.
    e_phoff: Pointer to the start of the program header table.
    e_shoff: Pointer to the start of section header table.
    e_flags: Interpretation of this field depends on the target arch.
    e_ehsize: Contains the size of header (normally 64 bytes for 64-bit and 52 Bytes for 32-bit)
    e_phentsize: Contains the size of program header table entry. Or 0x20 (for 32-bit) or 0x38 (for 64-bit).
    e_phnum: Contains the number of entries in the program header table.
    e_shentsize: Contains the size of a section header table entry. Or 0x28 (for 32-bit) or 0x40 (for 64-bit)
    e_shnum: Contains the number of entries in the section header table.
    e_shstrndx: Contains index of the section header table entry, that contains the section names.
*/

/* ELF header (atm only for 64 bit) */
#pragma message("TODO: add elf header for 32 bit")
struct Elf64_Ehdr {
    uint8_t  e_ident[EI_NIDENT];
    uint16_t e_type;
    uint16_t e_machine;
    uint32_t e_version;
    uint64_t e_entry;
    uint64_t e_phoff;
    uint64_t e_shoff;
    uint32_t e_flags;
    uint16_t e_ehsize;
    uint16_t e_phentsize;
    uint16_t e_phnum;
    uint16_t e_shentsize;
    uint16_t e_shnum;
    uint16_t e_shstrndx;
};

// ============================================================================
// =================================== Program Header =========================
// ============================================================================

/* Type, p_type field */
#define PT_NULL         0x00000000 /* Program header table entry unused */
#define PT_LOAD         0x00000001 /* Loadable segment */
#define PT_DYNAMIC      0x00000002 /* Dynamic linking information */
#define PT_INTERP       0x00000003 /* Interpreter information */
#define PT_NOTE         0x00000004 /* Auxiliary information */
#define PT_SHLIB        0x00000005 /* Reserved */
#define PT_PHDR         0x00000006 /* Segment containing program header table itself */
#define PT_TLS          0x00000007 /* Thread-Local Storage template */
#define PT_NUM          0x00000008 /* Number of defined types */
#define PT_LOOS         0x60000000 /* Reserved inclusive range. Operating system specific: */
#define PT_GNU_EH_FRAME	0x6474e550	/* GCC .eh_frame_hdr segment */
#define PT_GNU_STACK	0x6474e551	/* Indicates stack executability */
#define PT_GNU_RELRO	0x6474e552	/* Read-only after relocation */
#define PT_LOSUNW	    0x6ffffffa
#define PT_SUNWBSS	    0x6ffffffa	/* Sun Specific segment */
#define PT_SUNWSTACK	0x6ffffffb	/* Stack segment */
#define PT_HISUNW	    0x6fffffff
#define PT_HIOS         0x6FFFFFFF /* Reserved inclusive range. Operating system specific end*/
#define PT_LOPROC       0x70000000 /* Reserved inclusive range. Processor specific */
#define PT_HIPROC       0x7FFFFFFF /* Reserved inclusive range. Processor specific */

/* Flags, p_flags 64-bit field */
#define PF_X        0x1 /* Executable segment */
#define PF_W        0x2 /* Writeable segment */
#define PF_R        0x4 /* Readable segment */
#define PF_MASKOS   0x0FF00000 /* OS-specific */
#define PF_MASKPROC 0xF0000000 /* Processor-specific */

/* Other Program Header fields:
    p_offset: Offset of the segment in the file image.
    p_vaddr: Virtual address of the segment in memory.
    p_paddr: On systems where physical address is relevant, reserved for segment's physical address.
    p_filesz: Size in bytes of the segment in the file image. May be 0.
    p_memsz: Size in bytes of the segment in memory. May be 0.
    p_flags: 32-bit flags (don't exist on 64 bit header)
    p_align: 0 and 1 specify no alignment. Otherwise should be a positive, integral power of 2, with p_vaddr equating p_offset modules p_align
*/

/* Elf64 Program Header */
struct Elf64_Phdr {
    uint32_t p_type;
    uint32_t p_flags;
    uint64_t p_offset;
    uint64_t p_vaddr;
    uint64_t p_paddr;
    uint64_t p_filesz;
    uint64_t p_memsz;
    uint64_t p_align;
};

// ============================================================================
// =================================== Section Header =============================
// ============================================================================

/* Section Header Type, sh_type field  */
#define SHT_NULL            0x0 /* Section header table entry unused. First entry in the section table must be SHT_NULL according the stanard */
#define SHT_PROGBITS        0x1 /* Program data */
#define SHT_SYMTAB          0x2 /* Symbol table */
#define SHT_STRTAB          0x3 /* String table */
#define SHT_RELA            0x4 /* Relocation entries with addends */
#define SHT_HASH            0x5 /* Symbol hash table */
#define SHT_DYNAMIC         0x6 /* Dynamic linking information */
#define SHT_NOTE            0x7 /* Notes */
#define SHT_NOBITS          0x8 /* Program space with no data (bss) */
#define SHT_REL             0x9 /* Relocation entries, no addends */
#define SHT_SHLIB           0x0A /* Reserved */
#define SHT_DYNSYM          0x0B /* Dynamic linker symbol table */
#define SHT_INIT_ARRAY      0x0E /* Array of constructors */
#define SHT_FINI_ARRAY      0x0F /* Array of destructors */
#define SHT_PREINIT_ARRAY   0x10 /* Array of pre-constructors */
#define SHT_GROUP           0x11 /* Section group */
#define SHT_SYMTAB_SHNDX    0x12 /* Extended section indices */
#define SHT_NUM             0x13 /* Number of defined types */
#define SHT_LOOS            0x60000000 /* Start OS-specific */
#define SHT_GNU_ATTRIBUTES  0x6ffffff5	/* Object attributes.  */
#define SHT_GNU_HASH	    0x6ffffff6	/* GNU-style hash table.  */
#define SHT_GNU_LIBLIST	    0x6ffffff7	/* Prelink library list */
#define SHT_CHECKSUM	    0x6ffffff8	/* Checksum for DSO content.  */
#define SHT_LOSUNW	        0x6ffffffa	/* Sun-specific low bound.  */
#define SHT_SUNW_move	    0x6ffffffa
#define SHT_SUNW_COMDAT     0x6ffffffb
#define SHT_SUNW_syminfo    0x6ffffffc
#define SHT_GNU_verdef	    0x6ffffffd	/* Version definition section.  */
#define SHT_GNU_verneed	    0x6ffffffe	/* Version needs section.  */
#define SHT_GNU_versym	    0x6fffffff	/* Version symbol table.  */
#define SHT_HISUNW	        0x6fffffff	/* Sun-specific high bound.  */
#define SHT_HIOS	        0x6fffffff	/* End OS-specific type */
#define SHT_LOPROC	        0x70000000	/* Start of processor-specific */
#define SHT_HIPROC	        0x7fffffff	/* End of processor-specific */
#define SHT_LOUSER	        0x80000000	/* Start of application-specific */
#define SHT_HIUSER	        0x8fffffff	/* End of application-specific */

/* Section Header flags, sh_flags field */
#define SHF_WRITE               0x1 /*Writable*/
#define SHF_ALLOC               0x2 /*Occupies memory during execution*/
#define SHF_EXECINSTR           0x4 /*Executable*/
#define SHF_MERGE               0x10 /*Might be merged*/
#define SHF_STRINGS             0x20 /*Contains null-terminated strings*/
#define SHF_INFO_LINK           0x40 /*'sh_info' contains SHT index*/
#define SHF_LINK_ORDER          0x80 /*Preserve order after combining*/
#define SHF_OS_NONCONFORMING    0x100 /*Non-standard OS specific handling required*/
#define SHF_GROUP               0x200 /*Section is member of a group*/
#define SHF_TLS                 0x400 /*Section hold thread-local data*/
#define SHF_COMPRESSED          0x800 /*Section with compressed data*/
#define SHF_MASKOS              0x0FF00000 /*OS-specific*/
#define SHF_MASKPROC            0xF0000000 /*Processor-specific*/
#define SHF_ORDERED             0x4000000 /*Special ordering requirement (Solaris)*/
#define SHF_EXCLUDE             0x8000000 /*Section is excluded unless referenced or allocated (Solaris)*/

/* Other Section Header fields:
    sh_name: An offset to a string in the .shstrab section that represents the name of this section. Zero means no name.
    sh_addr: Virtual address of the section in memory, for sections that are loaded.
    sh_offset: Offset of the section in the file image.
    sh_size: Size in bytes of the section. May be 0.
    sh_link: Contains the section index of an associated section. This field is used for several purposes, depending on the type of section.
    sh_info: Contains extra information about the section. This field is used for several purposes, depending on the type of section.
    sh_addralign: Contains the required alignment of the section. This field must be a power of two.
    sh_entsize: Contains the size, in bytes, of each entry, for sections that contain fixed-size entries. Otherwise, this field contains zero.
*/
/*
    The most common ELF sections:
    .text - contains executable code. Packed with Read and Execute flags. Load only one times. Can't be changed.
    .data - initalized data. Read and Write flags.
    .rodata - initalized read only data. Read flag.
    .bss - unititlized data. Read and Write flags.
    other sections you can see here: https://gist.github.com/x0nu11byt3/bcb35c3de461e5fb66173071a2379779#sections

    Also, there is something like "group of sections" (readelf -g), but it's very rare.
*/

/* Elf64 Section Header */
struct Elf64_Shdr {
    uint32_t sh_name;
    uint32_t sh_type;
    uint64_t sh_flags;
    uint64_t sh_addr;
    uint64_t sh_offset;
    uint64_t sh_size;
    uint32_t sh_link;
    uint32_t sh_info;
    uint64_t sh_addralign;
    uint64_t sh_entsize;
};

// ============================================================================
// =================================== Symbols ================================
// ============================================================================
// https://docs.oracle.com/cd/E19683-01/816-1386/chapter6-79797/index.html
#pragma message("TODO: do")

/*  Symbol fields:
    st_name: Symbol name
    st_info: Symbol type and binding. It is calculated using macros
    st_other: Symbol visibility.
    st_shndx: Section index.
    st_value: Symbol value.
    st_size: Symbol size;
*/

/* Symbol struct */
struct Elf64_Sym {
    uint32_t st_name;
    uint8_t st_info;
    uint8_t st_other;
    uint16_t st_shndx;
    uint64_t st_value;
    uint64_t st_size;
};


// ============================================================================
// =================================== ELF Page size ==========================
// ============================================================================
#define ELF_PAGE_SIZE   0x1000
#define ELF_PAGE_MASK   (ELF_PAGE_SIZE - 1)

// ============================================================================
// =================================== ELF Load Result ========================
// ============================================================================
// ATM it uses in efi_elf_parser.cpp

/* Error codes for ElfLoadResult.error */
#define ELF_ERR_NONE           0  /* success */
#define ELF_ERR_FILE_TOO_SMALL 1 /* file smaller than ELF header */
#define ELF_ERR_INVALID_MAGIC  2 /* bad magic */
#define ELF_ERR_INVALID_CLASS  3 /* bad class */
#define ELF_ERR_INVALID_IDENT  4 /* bad ident */
#define ELF_ERR_INVALID_ARCH   5 /* bad machine */
#define ELF_ERR_INVALID_TYPE   6 /* bad type */
#define ELF_ERR_NO_LOAD_SEGS   7 /* no PT_LOAD segments found */
#define ELF_ERR_ALLOC_FAILED   8 /* AllocatePages failed (check efi_alloc_status) */

/* Elf Load Result object */
struct ElfLoadResult {
    uint64_t entry_point;
    uint64_t phys_base;
    uint64_t virt_base;
    uint64_t total_size;
    uint64_t efi_alloc_status; /* EFI_STATUS from AllocatePages, valid when error == ELF_ERR_ALLOC_FAILED */
    uint8_t  error;            /* ELF_ERR_* code above */
    bool     success;
};