System Software Telecon minutes, 2006-10-10

[Mitch Bradley]

System SW Meeting Minutes
2006-10-10

Jim, Chris Blizzard, Mitch, Ray, Jordan, Chris Ball, Andres

AI: Mitch: supply Open Firmware for people to start testing
AI: Mitch: supply Open Firmware test for B1 mouse
AI: Jim: look into what it takes to add verified state to trac
AI: Jordan: add gamma support to gx
AI: Chris Blizzard: ping Marcello about #42 and #147


Decided to use Open Firmware as the bootloader.  Key decision factor
was the ability to include diagnostics.  Will still use LinuxBIOS,
but the payload will be OFW instead of Linux-as-bootloader.
New OFW USB stack is working with mass storage, keyboards.
OFW Marvell wireless driver is written but not tested.
(5536) NAND FLASH driver, JFFS2 driver (R/O), SMbus driver are working.
No IPV6 support yet.  No consensus about whether we need IPV6 in firmware.
May wait until after B1 build to deploy Open Firmware.
Need to get people started on testing.

Do we need a verified state in Trac.  General consensus is yes.

Do we need another build?
  * Xorg 7.2 is drawing funny colors in 16-bit mode.
  * sysprof and memprof are not in the build yet, but they are not blockers.
  * Jim wants to get the 7.2 server with input working into people's hands.
  * The driver is usable in 24-bit mode.  The down side is that it uses
    a lot of memory
  * evdev is included, but not config'ed in

LB interrupt routing for CaFe.  Pin A needs to be directed to a real pin.

Richard has an offer from OLPC.

Priorities:
  - Get all the hardware working
  - Get a network manager interface for Sugar, get the browser working well
  - Something to demonstrate the camera
  - Stabilize the X setup

Chris Ball to get closure on the buildrom sandbox.

Driver/integration:
  - DCON being integrated into 2.6 git tree
  - CaFe drivers - all integrated into 2.6 tree, can't test because
    of BIOS setting.
  - We think it is possible to use the CaFe drivers using class id
    matches.

The NAND section of the Altera FPGA CaFe is not yet working.  Need
to ping Mark to see if this is a schedule risk.

soren doesn't actually own memprof.  No more progress on that front.

7.2 X server is in the tree.

Jordan and Adam are talking about X, but need to have more regular 
discussions.

tmpfs in RAM.  Marcello has opinion

Mitch:
  B1 build will have 256M.  B2 build will have an assortment of 128M.
  Mark asked Mitch to send the failed board back to Quanta.

X config that "just works": Zephaniah and Chris Ball have been working
on it, sent Chris Blizzard a config file for testing.

Things are acting strangely on the B test board.  Not seeing proper 
interrupts.
(The EC is different.)

Jordan:
  PRS - promises to attack it tomorrow.
  Needs to open a trac bug about the speed problem.
  Need to talk to Tom about NMI and oprofile
  DCON is in the kernel.  Backlight, color, but not gamma or dcon sleep

Ray:
  Today will ship out battery to David W. et al
  Update EC code over the weekend so we can start on battery driver

Andres:
  Touchpad calibration - imported Z's work and added calibration stuff