Creates a new code relocator for copying x86 instructions from one memory location to another, taking care to adjust position-dependent instructions accordingly.
Source address to copy instructions from.
X86Writer pointed at the desired target memory address.
Indicates whether end-of-block has been reached, i.e. we've reached a branch of any kind, like CALL, JMP, BL, RET.
Indicates whether end-of-input has been reached, e.g. we've reached JMP/B/RET, an instruction after which there may or may not be valid code.
Latest Instruction read so far. Starts out null and changes
on every call to readOne().
Eagerly cleans up memory.
Peeks at the next Instruction to be written or skipped.
Peeks at the address of the next instruction to be written or skipped.
Reads the next instruction into the relocator's internal buffer and returns the number of bytes read so far, including previous calls.
You may keep calling this method to keep buffering, or immediately
call either writeOne() or skipOne(). Or, you can buffer up
until the desired point and then call writeAll().
Returns zero when end-of-input is reached, which means the eoi
property is now true.
Skips the instruction that would have been written next.
Skips the instruction that would have been written next, but without a label for internal use. This breaks relocation of branches to locations inside the relocated range, and is an optimization for use-cases where all branches are rewritten (e.g. Frida's Stalker).
Writes all buffered instructions.
Writes the next buffered instruction.
Writes the next buffered instruction, but without a label for internal use. This breaks relocation of branches to locations inside the relocated range, and is an optimization for use-cases where all branches are rewritten (e.g. Frida's Stalker).
Relocates machine code for x86.