Interpreting Data - HP E2444A User Manual

Interpreting Data

HP E2444A
80386DX/DXL Preprocessor Interface
Unless followed by a lower-case letter, all numeric output from the
inverse assembler is in hexadecimal format. Decimal values are
indicated by a lower-case "d" (as in the INT instruction).
Up to four instructions may be displayed for a single analyzer state,
because the 80386DX/DXL can fetch a double word with four
instruction bytes from program memory. If the least significant byte of
this double word contains a single-byte instruction, the next sequential
instruction begins in the next higher byte. This process continues from
the least significant byte to the most significant byte until all of the
bytes of the fetched double word are used. When a single state
contains more than one instruction, each instruction will be displayed
on a separate line. For example:
+ 0015 NEG ECX
MOV BYTE PTR [EBP][-1B].CL
+ 0016 xxxxxxDA READ MEM
Line number + 0015 displays two instructions from a double-word.
Since instructions may begin in any byte position, the last bytes of a
multiple-byte instruction may extend into the lower bytes of the next
double word fetched. In this instance, the next sequential instruction
begins in the next higher byte of the next double-word after the
previous instruction and operands. When interpreting a given state,
the inverse assembler will ignore bytes used by a previous instruction
and will only display the instructions that begin in that state. For
example:
Byte Position
3 2 1 0
01 20 BB 24 Single byte instruction MOV BX (BB) starts in byte 1.
10 BF D2 31 Double byte instruction XOR DX,DX (31D2) begins in byte 0
and continues into byte 1. Next instruction MOV DI (BF) begins
in byte 2.
F7 C8 8C 00 Double byte instruction DIVDI (F7F7) starts in byte 3 and
continues into byte 0 of next double word fetched.
30 CA 81 F7 Next instruction OR DX (81CA) begins in byte 2 immediately
after last instruction.
= OPCODE FETCH
=
= MEMORY READ
Analyzing the Intel 80386DX/DXL
2-13