OLPC System Software Teleconference Minutes

[Mitch Bradley]

OLPC System Software Teleconference Minutes, 2007-02-21

Present: Jim, Mitch, Zephaniah, Blizzard, Ball, j5, John Watlington, 
Andres, Marcelo, Jordan,
Richard, Ivan, Ronak, Michalis

Status
  * Machines are being shipped.  Will start arriving soon.
  * Expecting new keyboard samples
  * Anticipating move to Python 2.5 - benchmarks at 40% faster, plus 
opens fewer files
    - Will also have to go to FC7, which is not ready yet
    - Dependencies are too extensive to allow using experimental 
packages with FC6

USB performance problem
  * USB network interfaces cause a 30+% CPU slowdown due to a side effect
    of an errata workaround for a GX.  We are awaiting further elucidation
    from AMD experts.

New wireless problem
  * There is an iperf corruption wireless problem that has been showing up,
    that does not show up on LX or on other laptops.  We wonder if it might
    be related to 5536 errata number 44?

Fastboot firmware working, booting experimental kernel.

Admonition from Jim:  Use devel list instead of long CC lists.

Stable build

  * New stable build will only have new kernel and firmware; UI stuff is 
churning too fast.

OFW update for suspend/resume

  * Getting close, very close

Release notes.

  * Need stable release notes for new machines.

Priorities

  * Get kernel, firmware, updated release notes for new machines
  * Mesh networking performance problem

Battery management

  * EC microcode is getting better, but still not perfect

  * Supposedly the trickle charge works for battery recovery, but it
    may take a day or so to recover a battery

Wireless wakeup:

  * Need information about timing requirements of EC_WAKEUP signal.

  * Missing infrastructure for processor wakeup via SCI interrupts - 
wireless,
    ebook mode sensors, etc.



New action items:

AI: Mitch to try to duplicate the iperf problem with the USB analyzer
AI: Richard to work with Ronak to get him the wireless wakeup info he needs.

Previous action items:

* AI: cjb to relnote python performance
* AI: #668: jg to rewrite this bug to talk about rotation instead
* AI: Jordan and ajax to investigate xrandr segfault
* AI: dcbw and marcopg to investigate mesh/avahi not working.
      done, but we're seeing crashing reports