Developers of the Mill, a clean-sheet rethink of general-purpose CPU architectures
Faster, Cooler, Safer Computing.
For the existing, portable code in the world, a re-compiled program will run faster, cooler and safer.
Most existing code is single thread, so the Mill is designed to speed up the execution of each thread by being capable of many more operations per clock.
For multi-thread code, the Mill is designed with hardware support for the rapid spawning, synchronization and termination of threads.
The Mill also has hardware support for rapidly switching context for interrupts, protection domains, thread states, function call and return.
The Mill uses low power statically scheduled hardware to achieve performance similar to high power dynamically scheduled hardware.
The Mill routes most results directly from operation to operation, reducing high speed register file power.
The Mill uses informed memory management hardware to keep most temporary data on chip, reducing external main memory power.
The Mill uses a separate machine state stack to stop buffer overflows from installing hostile code and Return-Oriented Programming attacks.
The Mill function call and return mechanism hides intermediate data that are not explicitly passed, preventing data leakage across either a call or a return.
The Mill memory allocation hardware prevents data leakage between processes, with zero power and latency overhead.
|announcements –||See and discuss them in our Forum.|
|events –||Find out about any future or recent ones.|
|in the press –||Press articles mentioning Mill Computing.|
|links –||Places on the web where the Mill Architecture or Mill Computing have been mentioned.|
|news email list –||Subscribe to our Mill Computing email news list.|
Who are we?
In February 2003, the original founders saw that key problems with general purpose processor architectures simply were not being addressed adequately by any contemporary CPU, and so rethinking CPU architecture was something that “just needed doing.” The Mill CPU project was started knowing full well that delivering processors to customers also means delivering a lot of support software. Several of us were serial entrepreneurs, and we decided to form a company, one that would avoid some of the traps that cause young companies to fail. We sought advice from several well-known people in the microprocessor industry.
In January 2004 Out-of-the-Box Computing came into existence as a “company in formation.”
In March 2014 after a decade of work Out-of-the-Box Computing was incorporated as Mill Computing, Inc. in order to raise funding to begin filing patent applications and move on to the next phase. Our fundraising from angel investors has gone well, and we now have over twenty patents filed on all aspects of the Mill.
As of June 27, 2017, 6 of our patents have been granted.
What do you get by joining us?
In the beginning we were a sweat equity organization; no one received a salary; instead, contributors received units that converted to stock when we incorporated. At incorporation 45 people had worked on the Mill and became shareholders. After incorporation we are still a sweat equity organization; we now use a stock option system for sweat equity, and we still pay no salaries. Reward for work today is comparable to what it was before incorporation.
Where do we work?
We are a distributed company with individual contributors working from their homes. While the founders are in Silicon Valley, some of us live in Europe, and for a while one even lived in Borneo!
Why are we looking for talent now?
Now that we are well into implementation we can benefit by having a larger team. Earlier on, a larger team would have just slowed us down.
What work needs to be done?
- Writing compiler toolchain code, including working on our own code generator back end and modifying LLVM to support some unusual capabilities of the Mill such as quad precision, overflow detection and decimal floating point
- Writing test sequence generators in C++ for individual operations
- Modifying and/or writing logic generators in C++ for hardware functional units
- Porting libc and libc++
- Porting an open-source BIOS
- Writing a micro-kernel similar to L4
- Porting Linux
- Writing white papers, technical web sites, and user guides
What skills do contributors need?
- All code contributors should be comfortable writing and enhancing sophisticated C++ code and navigating large C++ code bases.
- Software toolchain contributors will need familiarity with LLVM and/or ELF.
- BIOS contributors will need familiarity with programming for hardware such as USB, PCI and DRAM controllers.
- Hardware contributors will need to know Verilog in addition to C++ and must be familiar with standard cell design techniques.
- Technical writing contributors need to be familiar with computer architecture and systems software.
If you are interested in joining us, please email
Investor Email List Signup
This list is for those wishing to invest in Mill Computing, Inc., when and if opportunities to do so become available. Future opportunities, if any, will be announced only on this list.
By joining this list you are certifying that you are an Accredited Investor under Regulation D of the Securities Act of 1933 or are otherwise exempt from the registration requirements. Typically an Accredited Investor will have a net worth of $1 million or more excluding the primary residence, or an income of $200,000 or more in the current and prior two years. The full list of requirements is at http://www.sec.gov/answers/accred.htm. See also http://www.sec.gov/answers/rule504.htm.
I qualify as a Regulation D Accredited investor. Sign up for investor list.
I do not qualify. Return to front page.