Tiny ELF binaries on ARM

Looks like breadbox didn't want to go down this rabbithole. So we'll have to do it instead.

• e_machine seems to be the only checked header field (e_entry alignment checks are normal, because if it wouldn't be aligned, the code would segfault on entry.)
• phdr parsing etc. is done architecture-independent, so the same tricks should be usable here as well.
  • However, on x86, we were using the way page mapping works to only have to specify a few flags, this probably can't be ported over.
• Each ARM opcode is 4 bytes long, and needs to be aligned. This kinda sucks for all the overlapping tricks. Also, arbitrary constants can't be loaded into registers easily.
  • Do we want to depend on Thumb-mode?
• Apparently the kernel doesn't look at the immediate field of swi instructions if it's configured as EABI-only.
• Dynamic linking stuff

A few questions on the target platform

• How many ARM Linux machines have thumb, half and fastmult? EABI or OABI?
  • Seems to be common enough. OABI is dead, everyone's on EABI now.
• Which CPU are we targetting? ARMv6T (RPI1)? ARMv5TE?

Minimal ELF Poc

Not that minimal :) (But it should be able to show you which fields can be bogus quite clearly.)

gcc -c -o tiny.o tiny.S
ld -nostdlib -nostartfiles -T tiny.ld -o tiny.elf tiny.o
objcopy -O binary tiny.elf tiny.bin # somehow ld --oformat=binary no worky?

(Of course, the toolchain should be arm-linux-gnueabi or sth.)

@ tiny.S
.arch armv5te ; @.cpu arm946e-s

.section .ehdr,"awx",%progbits
.align 4
.arm

#define ORG 0x08048000

ehdr:
        e_ident:
                .byte 0x7F; .ascii "ELF"
                .byte 'p' @ ELFCLASS32
                .byte 'c' @ ELFDATA2LSB
                .byte 'y' @ EV_CURRENT
                .byte '/' @ EI_OSABI_SYSV
                          @ EI_ABIVERSION
                .ascii "K2^TiTAN" @.byte 0,0,0,0,0,0,0 @ EI_PAD
        e_type: .2byte 2 @ ET_EXEC
        e_machine: .2byte 40 @ EM_ARM
        e_version: .4byte 1337
        e_entry: .4byte _start
        e_phoff: .4byte phdr - ehdr
        e_shoff: .4byte 31337
        e_flags:                                  
                @.4byte 0x2|0x4|0x40|0x80|0x00400000|0x05000000
                        @ 2: hasentry
                        @ 4: thumb interwork
                        @ 40: 8-bit struct alignment
                        @ 80: EABI
                        @ 00400000: little-endian AAPCS
                        @ 05000000: EABI v5
                .4byte 0xdeadbeef @ 0x05000000 @ EABIv5, no float stuff
        e_ehsize: .2byte e__end - ehdr
        e_phentsize: .2byte p__end - phdr
        e_phnum: .2byte 1
        e_shentsize: .2byte 1337
        e_shnum: .2byte 1337
        e_shstrndx: .2byte 1337
        e__end:
phdr:
        p_type: .4byte 1 @ PT_LOAD
        p_offset: .4byte 0
        p_vaddr: .4byte ORG
        p_paddr: .4byte 1337
        p_filesz: .4byte _start__end - _start + e__end - ehdr + p__end - phdr
        p_memsz: .4byte _start__end - _start + e__end - ehdr + p__end - phdr
        p_flags: .4byte 5 | (1337 << 8) @ R=4 W=2 X=1
        p_align: .4byte 0xDEADBEEF @ 0x1000
        p__end:

.global _start
_start:
        mov r7, #1 @ SYS_exit
        mov r0, #42
        swi #31337 @ literal can be nonsense
        @bkpt #1337 @ like x86 int3 @ literal can be nonsense
_start__end:

/* tiny.ld */
OUTPUT_FORMAT("elf32-littlearm","elf32-bigarm","elf32-littlearm")
OUTPUT_ARCH(arm)
ENTRY(_start)

SECTIONS {
        . = 0x08048000;

        .ehdr : {
                *(.ehdr*)
        }
}