Readme: | Short: Disassembler for Intel 8080 processor Author: martin@fripost.org (Martin �berg) Uploader: polluks+aminet sdf lonestar org (Stefan Haubenthal) Type: dev/cross Architecture: m68k-amigaos; ppc-morphos
About =====
dis8080 is an Intel 8080 instruction disassembler. It reads machine code from standard input and writes the disassembly to standard output.
Usage =====
dis8080 [-d ADDR COUNT]
The option -d ADDR COUNT tells the disassembler to emit data definitions instead of instructions for COUNT bytes, starting at input offset ADDR. This option can be used multiple times.
Output from dis8080 can be assembled with an assembler such as vasm or asmx.
Build =====
Compile dis8080 with an ANSI C compiler:
cc dis8080.c -o dis8080
Examples ========
To disassemble the file example/add.bin, issue:
./dis8080 < example/add.bin
It generates the following output.
----------------------------------------------------- lxi b,L0009 lxi h,L000A jmp L000B L0009: mov d,d L000A: call L860A ret
L000B = $000B L860A = $860A
end -----------------------------------------------------
This does not look quite right, because: * The two "lxi" instructions hint that the values at label L0009 and L000A should be interpreted as data rather than instructions. * "jmp" is unconditional so the "mov" will not be executed. * The "jmp" target is inside the "call" instruction. * "mov d,d" is not a commonly used instruction.
What is happening here is that the disassembler does not know that following the "jmp" are two bytes of data, after which the program instruction stream continues at address $000B. In other words, the disassembler has become "out of phase". It is common to encounter cases like this in production code.
To get a more realistic disassembly, mark the addresses $0009 and $000A as data with the -d option:
./dis8080 < example/add.bin -d 0x0009 2
----------------------------------------------------- lxi b,L0009 lxi h,L000A jmp L000B L0009: db $52 ; 'R' L000A: db $cd L000B: ldax b add m ret
end -----------------------------------------------------
Now this looks better as locations $0009 and $000A are interpreted as data rather than instructions. At label L000B the two data bytes in memory are added together.
|