Assembly Challenges Programming Testing

Some Assembly Required No. 1

I’ve been working on some of the instruction tests in vial, and I wanted to test the implementation of LOOP variants. My objective was to make sure the vial version is identical to the real CPU version (as discussed here). To achieve this, I had to cover all of the essential behaviors of LOOP.

Well, using the framework Gil and I wrote, I hacked up some code that should cover the relevant cases:

code_template = """
mov edx, ecx ; control the start zf
mov ecx, eax ; number of iterations
mov eax, 0 ; will hold the result, also an iteration counter
    cmp eax, ebx    ; check if we need to change zf
    setz dh
    xor dh, dl      ; if required, invert zf
    inc eax         ; count the iteration
    cmp dh, 0       ; set zf
    loop%s loop_start
for loop_kind in ['','z','nz']:
    code_text = code_template % loop_kind
    c = FuncObject(code_text)
    for start_zf_value in [0,1]:
        for num_iters in [1,4,10]:
            for when_zf_changes in [1,2,15]:
                c(num_iters, when_zf_changes, start_zf_value)

Note that c(…) executes the code both on vial’s VM, and on the real cpu. c.check() compares their return value (EAX) and flags after the execution. I also wanted to avoid other kinds of jumps in this test.

To check that the code ran the same number of times, I returned EAX as the number of iterations.
All the games with edx are there to make sure that I’m testing different zf conditions.

The challenge for today:
Can you write a shorter assembly snippet that tests the same thing?

Algorithms Assembly decompilation Programming

First Code Transformation: Removing Flag Computations

Short introduction to code transformations

If you intend to write a decompiler, you’ll find yourself writing code transformations.
In our context, code transformations are operations that take as input an expression tree, and return an equivalent but different expression tree.

An equivalent expression tree is one that does the same thing as the original. More formally, it’s an expression tree that when run, will reach the same final state as the original expression tree.

Here’s a simple example, converting a while loop to a do-while loop:


while cond:


if cond:
    while cond

Code transformations are a very tricky business. You may find yourself changing the meaning of your expression trees without noticing. That’s one of the reasons we wrote the VM: we wanted to make sure that our code transformations will be correct.

Generally, you’ll want to transform your code to a somewhat ‘better’ version. ‘Better’ in what sense?
Well, the code may be:

  • More readable. A good purpose, as we are writing a decompiler for human beings. Unfortunately, this goal is not well defined. ‘More readable’ will have to be defined by heuristics.
  • Shorter/smaller. An easier goal to define, smaller code has the following advantages:
    • It might be faster to execute.
    • It might faster to process by other algorithms – after all, it is shorter.
    • It might be more readable.
  • Simpler, in the sense that it uses less types of operations. For example, take transforming a while to a do-while (shown above). If we wanted to analyze loops, it will allow us to write code to handle just do-while loops, instead of handling both cases.

There are many more possible goals for code transformations. The bottom line is that code transformations may do any of the following:
* Make the life of the human reader easier.
* Make the life of the decompiler programmer easier.
* Make the life of the decompiler easier.

An important aspect of code transformations is when to do them. Here is a good rule of thumb:

If it makes it harder for the programmer or the decompiler – it should be done as late as possible.
Otherwise – it should be done as early as possible.

Removing Flag Computations

This is the first code transformation we intend to write. Consider the following case:
sub eax, ebx
sub ebx, ecx

The equivalent expression tree for these two instructions looks something like this:

eax = eax - ebx
zf = eax == 0
ebx = ebx - ecx
zf = ebx == 0

Here I listed only the changes to zf and the actual registers. All the other flags (OF,CF,…) are changed as well. In a sequence of such instructions, all the modifications to the flags in the middle are actually dead code, and may be removed. This transformation will make the code shorter, faster to execute on our VM and faster to analyze by later stages.

This is a special case of dead code removal. This raises a valid question: why not let some general dead code removal transformation handle the flags?

There are a few reasons:

  • A general dead code removal might be more complex to write. Just writing exactly the same algorithm for general purpose registers will be more complicated, because of overlapping registers.
  • This transformation may remove up to 80% of many of the disassembled instructions. Flag calculations take up a lot nodes in the expression trees!
  • This transformation may be run as soon as the disassembly stage itself, while a general dead code removal transformation will not. This is so because of some assumptions we are allowed to make on the way flags behave. For example, an instruction template (“%0=%0-%1; ZF=%0==0” for sub) might have its parameters changed, but the flags affected stay the same.

The implementation of this transformation is not yet complete. When it is, I’ll talk a bit about the mechanics behind it. It is much more complicated than it seems!

Algorithms Assembly computer science Programming Projects Python

Issues in Writing a VM – Part 2

The Incredible Machine

Writing a VM capable of executing expression trees is different from writing a VM for executing assembly instructions. Here I’ll cover several issues stemming from this difference.

The first group of issues involve generality. Supporting a specific instruction set is a straightforward job, even if a hard one. Vial has to support multiple platforms and that’s a little tricky. These are just a few of the problems:

  1. Different registers for each instruction set. This one is easy enough to solve. Just have a node in the expression tree with some value to indicate the register.
  2. Register overlaps. A change to one register implies a change to its parents and its children. Consider RAX->EAX->AX->AH,AL. Changing EAX will affect all of the registers in the hierarchy.
    To handle this, we wrote a CPU class to keep all the info about the platform, including register overlaps.
  3. Delayed branches. Some platforms have branch delay slots. This means that after any branch instruction, right before the branch is taken, instructions in the delayed slots are executed anyway. For instance, SPARC has three delay slots, while MIPS has just one. This isn’t an easy issue to solve, and for now we didn’t tackle it. We’ve got a few ideas though.

To make sure that our implementation is generic enough, we decided to write a skeleton disassembler implementation for MIPS as well.

The second group of issues involve the nature of expression trees versus instructions:

  1. Stepping over statements or instructions? Each expression tree for an instruction usually holds more than one statement. For example, dec eax changes eax as well as the zero flag. Since some instructions like rep stosd may contain a long loop, being able to step over statements instead of expressions is preferable.

    The problem is, executing expression trees is done with a DFS-like walk. If implemented with recursion it makes pausing the walk for each step a bit complicated. However, recursion is the clearest way to write the execution code, and I’d rather not give it up.

    My solution was to use Python generators. Each time a statement was executed, the function would yield, thus returning control to the calling function, while keeping its state intact.

  2. Instruction Pointer changes. The lower-level disassembler returns expression trees for each instruction. However, it does not return the jump to the next instruction as well. While this is the required behavior, it means that the VM should change the instruction pointer after executing each instruction.

    This is easier said than done: What should we do after jumps? Several possible solutions emerged.

    The first was to append an IP change to each instruction’s expression. Those changes will have to be appended only where needed.
    A second solution was to check if the IP was changed, and if it was not, change it. This solution however will not support instructions that jump to themselves.
    The last and our preferred solution was to check if the IP was touched. This solution is simple and very straightforward.

There are many more issues I didn’t write about, for example, self modifying code. I’ll leave those issues for future articles.

Assembly computer science Programming Projects Testing

Issues in writing a VM – Part 1

Arkon and I decided to write a VM for vial. First though, a short explanation on what is vial:
vial is a project aimed at writing a general disassembler that outputs expression trees instead of text. On top of vial, we intend to write various code-analysis tools. The expression trees in the output should be an accurate description of the all of the code’s actions.
(note: the x86 disassembler behind vial is Arkon’s diStorm.)

So why do we need a VM? Apart from it being ‘nice and all’, it is critical for testing.

Some time ago, I described writing a VM to test a compiler I wrote as university homework. It is a similar issue here.
The disassembler is written according to the x86 specification. If we just check its output against this specification, we are not doing much to verify the code’s correctness. This is evident when you try to implement such a testing module – you end up writing another disassembler, and testing it against the original one. There has to be a different test method, one that does not directly rely on the specification.

Enter the VM. If you write a program, you can disassemble it, and then try to execute the disassembly. If it yields the same output as the original program – your test passed.
This is a good testing method, because it can be easily automated, reach good code coverage, and it tests against known values.
Consider the following illustration:

Testing Process

We are testing here a complete process on the left hand, against a known valid value, the original program’s output, on the right hand. All of the boxes on the left hand are tested along the way. Of course, one test may miss. For example, both the VM and the disassembler may generate wrong output for register overflows. We can try to cover as many such cases as possible by writing good tests for this testing framework. In this case, good tests are either c programs, or binary programs. This is essentially what I was doing when I manually fuzzed my own compiler.

Once the VM is finished, we can start writing various optimizations for the disassembler’s generated output. We can test these optimizations by checking the VM’s output on the optimized code against the output on the original code. This makes the VM a critical milestone on the road ahead.

Assembly Compilation computer science Personal Programming Python

Writing a Quad Interpreter

My compilation homework is writing a compiler from simplified CPL to Quad, a made up assembly language. The compiler was going well, and when it started to emit code, I naturally wanted to test it.

The problem is, I work on Ubuntu Linux, and the only available Quad interpreter was for Windows. Trying to run it with Wine didn’t work, and I gave up on it pretty quickly. I also didn’t want to start working on Windows. So I sat down for half an hour and wrote a Quad interpreter of my own, in Python.

Afterwards, I thought of publishing it, and sat down to write a step-by-step mode as well. I didn’t implement breakpoints, although it is quite easy.

The compiler is working now, although I do have to do the finish. Since it is a homework assignment, I will not be publishing it, but I will publish the Quad Interpreter. I hope it benefits you.

A note of caution though: I didn’t test it thoroughly . The most I did was running it with some toy programs and the output of my compiler to see if it works. It does, and that’s enough for me.