Sunday 23 June 2013

Report 1 (June 17 - June 23)

This is the first of many weekly reports. I plan to post these every Sunday. They will give a rundown of what I've done during the week, list any problems I had, and what I plan to do the following week. I'll be posting the reports both to the beagle-gsoc Google Group and to my blog.

As this is my first report, I'm including links to all of the project resources:
Completed this week:
  • Rebased the skeleton code I had developed when writing my GSoC application.
  • Extracted the clock configuration code from libgpio and put it into it's own library, libclkconf. Merged.
  • Used libclkconf to configure the clocks for i2c. I2C_REV is now readable on all buses on the BeagleBone Black and BeagleBoard-xM.
  • Wrote code to configure the pins with libpadconf. Need to finish more of the driver to completely test this.
  • Removed duplicate code in libpadconf. Merged.
  • Implemented the i2c initialization procedure described in the TRM.
  • Began work on a read function. It isn't complete yet, but it can read the magic number and board name from the on board EEPROM on the BeagleBone Black. Screenshot.
  • Setup netbooting and figured out how to transfer files with ZMODEM. This will improve my productivity a lot. Documented procedure on the project blog (netboot, serial transfer) and Minix wiki (netboot, serial transfer).
  • Received a BeagleBoard-xM from TI. Thanks TI! Tried it out and made a blog post about running Minix on the BeagleBoard-xM. Will use it for developing/testing Minix i2c support for the dm37xx.
  • Received a Welcome Package from Google. Thanks Google!

Issues / Concerns / Challenges:
  • The files for the stage 1 bootloader and u-boot are downloaded from the Minix website when doing a fresh build of an SD card image. Once this week, the website was down for a few minutes and I couldn't do a build. To prevent this from happening again, I setup a local mirror on my computer.
  • Writing 2GB SD card images to the uSD card is time consuming. It takes about 7 minutes with my class 4 card. I'd prefer a much faster write, compile, test cycle. To overcome this issue, I setup a netboot environment and will use zmodem to transfer my changes (usually just 1 file).

Plan for next week:
  • Clean up and finish the read function. Get it working on the BeagleBoard-xM as well.
  • Implement the write function.
  • Put some EEPROMs on a bread board and test reading and writing data on all of the buses to confirm proper pin configuration and test read/write functions.
  • As time permits, begin implementing the /dev interface.

No comments:

Post a Comment