Archives

You are currently viewing archive for March 2008
Category: General
Posted by: afeldstein
I modified the interface to Class A's only constructor, increasing the length of it's argument list from two to three. Class B instantiates Class A (i.e. B depends on A), still calling it's constructor with only two arguments. Since Class A changed, NetBeans IDE 6.1 should have recompiled Class B and found the fault in its Class A constructor call. However, there was no failure until run time.

NetBeans Blogging Contest
Category: General
Posted by: afeldstein
In the NetBeans IDE 6.1 editor, unused imports are marked. This is a useful feature.

NetBeans Blogging Contest

2008-03-25: The 29 GHz Barrier

Category: General
Posted by: afeldstein
No legal SPARC-V9 implementation can be clocked faster than 29 GHz. We know that the TICK register represents CPU clock cycles. But IMPL. DEP. #105 states that "TICK.counter ... must be able to count for at least 10 years without overflowing." If you do the math, you'll see that that imposes a 29 GHz limit on clock rate.

What if I want a faster SPARC? Is anyone working on SPARC-V10?
Category: General
Posted by: afeldstein
I had a few minutes to work on this, so I thought I'd at least kick off a download. It wasn't clear how I was to get what I wanted, and I ended up with the wrong thing. I want those few minutes of my life back ...

» Read More

2008-03-13: Sashimi anyone?

Category: General
Posted by: afeldstein
Using FSS to debug Sputnik's integer multiplier has not allowed me to find the root cause of the current problem (the cycle when product_valid is first asserted is not consistent, even with the same operands). I'm asking myself how FSS could have provided all of the necessary debug features, because you will require those. Cheating by using a source-level debugger, I found an unexpected intra-cycle transition. But the Software Requirements Specification requires that FSS be a cycle-based simulator.

In order to allow even minor changes to the SRS, I don't think the Waterfall model is going to work for me anymore. The Sashimi approach looks promising ...

» Read More