After reading in all the parameters and setting all neccesary environments, compile_script will call the following scripts to compile benchmark programs step by step:

1. compile_bench:
Under trimaran/impact/scripts/compile_bench
This script including a dozen scripts that front-end IMPACT processes from source code to various intermediate representations (including profiling).

2. gen_elcor -> .0_el file

3. gen_EtoC -> .c file used to simulate execution

4. gen_CtoO -> executable of the simulator

Directional magnetic recharging

Daydreaming in the restroom just now, I was thinking how wonderful this world has become because of the discovery of electrocity. Electrical power captivated inside small batteries enables countless portable and handy applications, yet charging batteries frequently is really a headache and more and more becomes a bottleneck for all portable devices. I was wondering if there is any way to make recharging easier... here comes the idea. As we all know, electric power and magnetic power are interchangable. While electrocity conduction usually needs solid/liquid media like copper wire, the conduction of magnetic power can be done in the thin air. According to this simple fact, one can construct a pair of transmitter-reciever devices to transmit power from the power grid to charge the consumer device. The transmitter connected to the power grid convert electric power into a highly directional and focused magnetic beam, and the reciever obsorbs the magnetic beam entirely and convert it back to e-power to charge the device battery. Apparently one prerequisit needs to be kept: there mustn't be any living obstacles between the transimitter and reciever, to avoid harmful effects due to human body exposed to strong magnetic field.

Imagine a fuel free country with wirelessly rechargable transportations. Think of recharging recharging ur laptop on a bus, in ur car, etc. How amazing things would be.

Sth about IP

A Integer Programming without a objective function is a decision problem, instead of an optimization problem. Although the computational complexity is the same for both of them, decision problem is faster to solve, for it only requires one feasible solution (rather than the optimal one). In case an optimal solution is not neccesary, and bounds are given for a feasible one, such formulation is helpful.

Being science

Is the thrill gone?

Test from thinkfree posting to blogger

It's cool to use thinkfree office, which is based on java applets, meaning you can use it on almost all platforms with (or without) a browser. I've been editing ppt with it, and it seems to be able to accomplish most frequent tasks perfectly. It mimics MS office graphic interface very well (at least GUI compatible). All in all, it's a cool service, for free (this I doubt).