User:Jpintozz

From REU@MU
Revision as of 16:04, 3 August 2009 by Jpintozz (Talk)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

About me

I'm Joe Pintozzi and I will be a senior at Marquette in the Fall of '09. I am currently working towards my bachelor's degree in Computer Engineering. I will be working on the "Xinu Team" this summer in lab 310.

Work

Week of May 18th

  • Created user accounts for both Wikis and Trac
  • Read through XINU wiki to get background information on projects I will be working on this summer
  • Awaiting access to the MSCS network since my account has been disabled
    • Temporary stand-alone account created on Szabo for the time being
  • Checked out trunk from subversion, compiled and ran successfully
  • Download and review Intel and AMD flash specification PDFs
    • Intel, 41-48
    • AMD, 30 - ?
  • Look over TTY code from /tools/console/unix/braches/zlund
  • Filled in comments in own tty-*.c code for further clarification

Week of May 25th

  • Tweaked tty-bridge code, tested on Morbius
    • Need to throw more data at the serial port, it isn't having a hard time keeping up. Must force write errors......
    • Decreased max bufsize to force tty-bridge to run out of buffer quickly.
      • Write auto-compensates by waiting until there is room in the buffer to write

Week of June 1st

  • Went over FOS paper
  • Attended weekly 2pm Thursday meeting
    • Afterwards discussed FlashRAM development
  • Soldered WRT54GL router
  • Made an AMD flash image of chub
  • Tested the HDM Debricker on chub
    • Did not break chub, but did not necessarily fix it either....

Week of June 8th

  • Modified the hostname of chub from "chub.xinu.mu.edu" to "joe.xinu.mu.edu"
  • Flashed chub to original image, if successful the hostname will be changed back to "chub.xinu.mu.edu"
    • Hostname still "joe.xinu.mu.edu". HDM did not work.
  • Went over FPGA paper
  • Killed chub
    • Searched for Debricking alternatives, HDM seems to be the only one, and v48 is the newest revision
  • Started entering AMD Command Codes into flash.h
  • Updated trunk-2.0 and created new file - physicalAMD.c

Week of June 15th

  • Went over Dingo paper
  • Removed Chub, placed Dask back into the JTAG group, tested HDM
    • HDM works! Hostname changed back after flashing. Failure with Chub may be due to an electrical short from a header pin.
  • Put Dask back and hook up Davros to the HDM, change hostname and flash

Week of June 22nd

  • Flashing did not brick the router, be neither did it change the hostname back
  • Attempted to manually input starting address and length of flash memory
    • Same result as before
  • Added more commands to the PhysicalAMDControl
  • Added more definitions in flash.h for the AMD commands

Week of June 29th

  • Presented my flash presentation to the REU students
  • Tried different HDM debricking techniques with Davros
    • Unsuccessful so far
  • Updated repository with my code
    • Tested to make sure there was compatibility first

Week of July 6th

  • Selected and presented the paper "Eliminating the Call Stack to Save RAM"

Week of July 13th

  • Went over Augmented Reality paper
  • Discovered HDM is not erasing blocks when flashing
  • Found branch of HDM called the Tornado MOD
  • Hypothesized about why HDM isn't working
  • Erased Davros with Tornado
    • Put image back on with HDM

Week of July 20th

  • Davros is unstable
  • Took most of the week of (1/2 of Tuesday and on)
  • Went to Vegas
  • Came back with MUCH less than I left with
  • Flew on a plane with Brian Rancor

Week of July 27th

  • Ryan prepped a JTAG on Mandragora
  • Attempted to flash Mandragora after changing the host name, the host name has not changed leading me to believe it was never erased
  • Erasing and flashing separately, awaiting results
  • Built 'screen' and 'urjtag' on Aneth

Week of August 3rd

  • Came in early Monday
  • Mandragora does not stop in the same place every time.
    • Bad cable? Defective buffer?