MIPT-MIPS is a pre-silicon simulator of MIPS CPU. It measures performance of program running on CPU, thus taking best features of RTL and common functional simulation:
- Precision. We provide cycle-accurate models of branch prediction unit and pipeline behavior.
- Customization. Cache size, branch prediction algorithms, and other parameters can be easily changed.
- Simplicity. Our source files are much more readable than RTL.
- Speed. Simulation frequency gets up to 0.7 MHz on i5-7300U.
MIPT-MIPS can be used for different purposes:
- Performance control of software optimizations: you may check IPC boosts of your programs
- Pathfinding of hardware optimizations: you may easily integrate some nice feature to CPU model
- Education: simulator is a nice experimental frog to study CPU internals
Features modeled:
- Configurable branch prediction unit with 5 prediction algorithms
- Configurable instruction cache with true-LRU policy
- Interstage data bypassing
More details about internals are available on Wiki
- Clone repository with submodules:
git clone --recursive https://github.com/MIPT-ILab/mipt-mips.git
- Install LibELF and Boost
- Build MIPS binutils if you need to build MIPS ELF binaries.
- Install CMake 3.8 or higher.
- Create build directory somewhere, then cd into it and run
cmake /path/to/mipt-mips/simulator && make
to get all the binaries in your build directory. Check our Wiki page to get more details about CMake build.
Users of IDE (Visual Studio, Eclipse, CodeBlocks etc.) may generate project files with CMake as well.
To run all unit tests, call ctest --verbose -C Release
from your build directory.
MIPT-MIPS uses C++17 features and Boost 1.61. Thus, you have to use compilers of these versions or newer:
- GCC 7
- Clang 5.0
- Apple LLVM 7.3.0
- MS Visual Studio 2017 (Boost 1.66 and CMake 3.10.2 are required)
-I
— modeled ISA, default option is "mips"-b <filename>
— provide path to ELF binary file to execute.-n <number>
— number of instructions to run. If omitted, simulation continues until halting system call or jump tonull
is executed.-f
— enables functional simulation only-d
— enables detailed output of each cycle
--bp-mode
— prediction mode. Check supported modes in manual.--bp-size
— branch prediction cache size (amount of tracked branch instructions)--bp-ways
— # of ways in branch prediction cache
--icache-size
— instruction cache size in bytes--icache-ways
— # of ways in instruction cache--icache-line-size
— line size of instruction cache
This project is a part of ILab activity at Moscow Institute of Physics and Technology (MIPT).
The main goal of the project is to teach the students the computer architecture through development of a microprocessor implementing the MIPS instruction set in both functional and performance simulators.
Yes, if you attend MIPT-MIPS lectures. See our CONTRIBUTING.md file for details.