Slashdot Mirror
KernelTrap Interviews Andrea Arcangeli
An anonymous reader writes "Andrea Arcangeli completely rewrote the 2.4 Linux kernel virtual memory subsystem several years ago, a surprising event during the evolution of a stable kernel series. A very intelligent 27-year-old from Italy, Andrea spoke with KernelTrap in great detail about the past, present and future of his Linux kernel efforts. An interesting interview ."
Linux hacker : Age 27 : lives with in parents house.
Who'd've thunk it, eh?
Actually, most italian men live with their parents until they get married.
He has higher education, he just doesn't have the degree. The article indicates that he learned a great deal from his college courses.
;-)
A degree gets you an interview. What you learned gets you the job. AA skipped the interview step
When I read this all I could think of was that I wish I had gone to his highschool.
Using the Paralell port with an ADC is actually fairly easy. You just connect the input to a neutral grounded source to prevent circuit noise, feed the input from the analog waveform source (a calibrated RF generator works nicely) and adjust the capacitance to normalize the digital output. Use the multiplexer to allow more than one input to be sampled, and code the assembly to sample nibbles in a round robin based on the matched timing of the data strobe from the ADC. Using the paralell port results in stepping of the data, as the maximum resolution of the scope is limited by the 8 bits of the port. However, with multiplexing, some of the multiplexed channels can be used to widen the bit count to 16, 24, or 32. I never tried it above 32, as the 8086 I was using was unable to sample fast enough beyond that, even in assembly. Process the port input as a large integer and convert that to a pixel position (and numeric readout if desired) in simple integer to fixed point math.
The hardest part is eliminating the circuit noise for millivolt readings. In larger waveforms (5v and higher) the noise is mostly drowned, but at the millivolt range, any circuit noise, or any unmatched grounding, causes jitters in the waveform being read. Calibration to the range is essential for any serious reading.
And for fun, hook an unamplified output from a portable CD player to the scope, and viola, instant waveform display of the sound signal. Great way to relax on a lazy sunday, listen to and watch the music. Of course, this was in the pre winamp days, so it was a Walkman.
[/Meandering off]
You can have it fast, accurate, or pretty. Pick any 2.