Disassembly of Other Sections  Implementation Details of the  First Pass of Disassembly

Second Pass of Disassembly

The objective of this pass is to generate assembly language instruction from their binary counterpart. Since the address ranges of valid code have been identified in the first pass, we only need to disassemble the instructions in each address range. The instruction disassembly is carried out in the following steps.

• Symbol Generation : To perform the symbolic disassembly, at the beginning of the disassembly of an instruction, it is checked whether a symbol is associated with the address of the current instruction and if so, the symbol type (function name or just a label) is also extracted. If the symbol refers to a function, further information regarding the type and size of the function is also extracted.
• Instruction Generation : An instruction is matched using the instruction matching algorithm (as described in section ) and corresponding assembly language instruction is output with appropriate parameters. If the instruction is a control transfer instruction, a symbol is found from the symbol table constructed during the first pass and corresponding symbol name is used in the disassembled instruction. Further, memory references to the .data, .rodata and other such sections are found. In case of such a reference, we try to resolve the address. The size of the operand and the symbol name is stored for the reference.

This procedure is repeated for each instruction in all address ranges. The gaps within the .text section are overlooked for the purpose of text disassembly. One possibility is to simply ignore the bytes in the gaps and change the current location counter so that it reaches the beginning of the next valid address range. However, it is possible that these gaps contain initialized data (which are not referenced by the normal methods, for instance using register indirection instead of symbols, otherwise it would have been discovered during the first pass). In such a case, ignoring them might break the intended equivalence between the relocatable object file and the generated assembly code. Therefore, we simply output bytes in the gaps as data and generate appropriate pseudo-ops to glue the code.

  
 Disassembly of Other Sections  Implementation Details of the  First Pass of Disassembly
Fri Jan 15 11:17:08 IST 1999