- Giving Acknowledments to TA Help hours for support in:
- starting a sound structure of UniversalMachine
- connecting disassembled instructions to my um
- Helping fix my div, mult, add, etc. The operations. I kept getting a 'divisor of zero' panic.
- Credits to Professor Daniels for rumdump lab as this made me understand it a lot more and borrowed methods from that lab (load, opcode enums).
Goal of this assingment is to understand virtual machine code by programming the Rust Universal Machine (RUM), which a is a simple virtual machine implemented in Rust.
I believe my implementation correctly:
- Uses bitshift to parse opcodes, registers, and values
- represents the state of Universal Machine
- follows the implementation for each seperate instruction
As far as departures from design, just probable modified some names for modules and methods I had initially.
-
main.rs This is where the program starts and just runs. Simply used to for argument parsing and getting the instructions to interact with the state of the machine.
-
load.rs Boilerplate pretty much. Used for getting the binary of a um program. Ouputs a Vec/instructions which is read in by my `UniversalMachine`` struct. (taken from rumdump lab)
-
rumdis.rs rumdis.rs is responsible for most of the opcode parsing from an instruction.
Run()interacts withUniversalMachineto load up the initial instruction and then continue to parse the rest.disassemble()uses fields and some bitshifting helper methods to to gather the correct registers and pass them to corresponding instruction methods (in `state.rs``). -
state.rs (invariants described) Module is used to hold the representation for my UniversalMachine which has the following data representation:
-
Registers The machine is equipped with eight general-purpose registers, represented as a vector of u32.
-
Mapped Memory (Segmented) Memory in the RUM is organized into segments and offsets, forming a two-dimensional vector of u32. The first index represents a memory segment, while the second index represents an offset within that segment.
-
Unmapped memory (free segment) The UM has a vector called
unmapped_memory, where each index represents an unmapped memory segment. Used to allocate and deallocate memory as needed during the execution of programs. -
Opcode Instructions The UM has set of 14 opcode instructions that is disassembled in rumdis.rs and then calling the appropriate method to that instruction in the struct:
-
Conditional Move
-
Load
-
Store
-
Add
-
Multiply
-
Division
-
Bitwise NAND
-
Halt
-
Map Segment
-
Unmap Segment
-
Output
-
Input
-
Load Program
-
Load Value
My Laptop CPU: Intel(R) Core(TM) i5-1035G1 CPU @ 1.00GHz, 1190 Mhz, 4 Core(s), 8 Logical Processor(s)
Based on the 2113497561 instructions ran from sandmark.
Time(s):
| attempt | time (seconds) |
|---|---|
| 1 | 25.677 |
| 2 | 32.663 |
| 3 | 27.135 |
| 4 | 28.095 |
| 5 | 25.664 |
On average of 5 runs, rum takes 27.8468s for 2113497561 instructions. Therefore, I do 27.8468 / ( 2113497561/50000000 ) to get the time for 50 million instructions: approxiamtely 0.6588s.
- I spent around 8-10 hours analyzing and then probably another 6-8 designing UM.
- After analysis, I spent a decent amount of nights trying to program so probably 15-20 hours