In my second week, time was an issue. Therefore, I attempted to finish up more this week (my third week) so that I can cover the workload from week two. Generally, I am supposed to finish the opcodes by the end of this week. However, like 99% of the things with programming, it always ends up as an underestimation.
My biggest underestimation was the complexity of the opcodes. While generally there are some opcodes that are easy, those tougher ones, such as bsra and bsrai definitely need more looking into than I expected. I will be working on writing test cases to test all the correctness of the functions and work on a setting up debugging logs which I think will be a very useful tool for future devs.
Besides that, I another underestimation was my understanding of the architecture. As I was working on the single thread, I set out to work on the registers first because they were most important. Then I had to think about how to design the structure for decoding the instructions. I wanted to make use of whatever information I have and that caused quite a delay for me. Luckily, I have found out an efficient way to handle the decoding, read and writes, branching and shifting. I will be cleaning up the code to reflect this better solution.
Besides that, as this will be a web project, I have to put in deep thoughts about the abstraction of all the modules and their related functions in order to reduce bigger change in the future. While I have finished the basic structures, I have yet to check my code for errors. I will therefore run it with JSLint to see what kind of possible errors you have written on your code and you change it to a better solution.
A quick update on my progress:
I have finished about 70 opcodes and edited the two main files, and I have added aeMB.js for the simulation of aeMB. I will continue to finish up as many opcodes as possible and start working on the simulator designs, code clean up and encapsulations. I think by Sunday I will be able to finish up with the opcodes also the structures of the website.
0 Comments