Author Topic: (book) Pipelined Multi-core MIPS, Implementation and Correctness  (Read 2284 times)

0 Members and 1 Guest are viewing this topic.

Offline DiTBhoTopic starter

  • Super Contributor
  • ***
  • Posts: 4227
  • Country: gb
  • A Pipelined Multi-core MIPS Machine: Hardware Implementation and Correctness Proof (Theoretical Computer Science and General Issues) 2014th Edition
    by Mikhail Kovalev (Author), Silvia M. Müller (Author), Wolfgang J. Paul (Author)
See here, on Amaozn dot com

Not yet bought, it sounds *very* interesting  :o :o :o
The opposite of courage is not cowardice, it is conformity. Even a dead fish can go with the flow
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9931
  • Country: us
Re: (book) Pipelined Multi-core MIPS, Implementation and Correctness
« Reply #1 on: August 02, 2024, 01:57:44 pm »
Bought it!  It should be an interesting read and perhaps it might be fun to synthesize on an FPGA.

Thanks for pointing it out!
 
The following users thanked this post: DiTBho

Offline ZaneKaminski

  • Contributor
  • Posts: 34
  • Country: us
Re: (book) Pipelined Multi-core MIPS, Implementation and Correctness
« Reply #2 on: August 09, 2024, 10:18:46 am »
Quite interested as well. If only formal proof was the currently accepted verification methodology and there were tools, etc. around it, then maybe we would be spending a lot less effort in verification. Very tempted to buy but it's a bit pricey and there is probably still a lot of work required to verify a real design in the way the title suggests
 

Offline DiTBhoTopic starter

  • Super Contributor
  • ***
  • Posts: 4227
  • Country: gb
Re: (book) Pipelined Multi-core MIPS, Implementation and Correctness
« Reply #3 on: August 11, 2024, 03:39:42 pm »
Quite interested as well. If only formal proof was the currently accepted verification methodology and there were tools, etc. around it, then maybe we would be spending a lot less effort in verification. Very tempted to buy but it's a bit pricey and there is probably still a lot of work required to verify a real design in the way the title suggests

I have the same doubts: a bit pricey, will it say something really useful, and therefore worth it, or will it just be an educational but sterile reading on a practical level?

Who knows ...  :-//

I designed and implemented  a toy-ISA, which has nothing interesting except for the C-like compiler I wrote, as it facilitates the machine level.
Now I am adding multi-processing and I need a way to validate things better, or rather, to spend less time than I have used so far.

The implementation I made has been carried forward and tested rigorously for modules, so, specific test-bench-units for each ALU circuit, COP0, COP1, load/store, register file (which has particular properties), etc.

Finally, the system integration test-benches.

I need to spend less effort in verification, so probably a better verification methodology.
The opposite of courage is not cowardice, it is conformity. Even a dead fish can go with the flow
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf