Parent


Attendees

  • David Rusling (Linaro)
  • Kiko (Linaro)
  • Roger Teague (ARM)
  • Andrea Gallo (ST-Ericsson)
  • Vijay Pasam (TI)
  • Yves Vandervennet (Freescale)

Absent

  • Paul McKenney (IBM)

  • Sree Kumar (Samsung)

Information

Guest Participants

  • Alexander Sack
  • Bobby Batacharia

Agenda

  • Review the run up to Linaro at UDS
    • TSC Meetings
      • 20th April - Multimedia
      • 27th April - Review technical requirements, actions, priorities, teams
        • Propose moving 27th April?
  • Review Platform and validation requirements

Minutes

  • Discussed the run up to Linaro Developer
    • Difficult for ARM as there are bank holidays in the way
    • Agreed to postpone the 27th April meeting to 4th May to allow more review time
      • Kiko to deputise engineering, Alexander suggested
  • TI
    • Vijay expressed concerns that Linaro can handle two distributions, Ubuntu and Android
    • We should evaluate to see if we should prioritise one over the other
      • May make sense to prioritise Android and deprioritise Ubuntu, need to discuss [ACTION: David to organise]
  • Presented the top level priorities
    • Talked to all the working group leads, asking what they need and thinking about how best to support them
    • Have merged that with management requests and with the TSC inputs
    • Platform not like a working group - is also about servicing the other teams, allowing them to be efficient
      • May see infrastructure work not in the TRs as they respond to requests / needs
    • Want to go to a more flexible, agile approach
    • ARM is very keen that the output of the working groups is upstreamed into an Android solution
      • Kernel is a good example, how do we ensure that the Android kernel gets back ported patches from the leading edge Kernel WG work
      • For priority 1 we should explicitly call this out [ACTION: Alexander will add this]
        • The idea of the LEB is to combine the best from the member's BSPs, Linaro WG's and open source
        • Look strongly at what Google might take upstream
        • TI recommends bumping up the priority of P1.2 to essential (all agreed)
        • ST Ericsson has concerns about the Linaro headless kernel becoming redundent, key to keep both Android and leading edge kernel in sync
    • Asked the members for feedback on the priorities
      • General agreement, although Freescale sees Ubuntu and Android as of equal priority
  • Worked through the TRs with Alexander explaining the scontents, some discussion on each of the requirements
    • P1.3, should it be high or medium. If it can be done whilst keeping P1.1 and P1.2 at essential, then happy to keep at medium.
      • Agreed to keep P1.3 as high.
    • P1.4. ARM happy with it being low [ACTION: all to feedback their use of Nano]
      • More effective to do this in the Platform team
    • P1.6. Medium versus high?
      • There is additional work to make the LEBs use device tree as this is not how it works today
      • Should be essential. Can bump it to high?
      • This task is for using it for the LEBs
      • We'd like to provide sizing information, but not mature enough in our process
        • Happens after LDS during the drafting
      • [ACTION: move the priority of P1.6 high]
      • If it comes to priorities and look at deprioritise P1.3 in favour of P1.6
      • If P1.1 is more important than P1.2, then what do we get if we don't get P1.2
        • We are assuming P1.1 and P1.2 have the same priority
    • P2 all about validation
      • all teams to integrate their output into the validation flow and use the validation outputs to steer their work
      • P2.7 should be ignored, it is part of P2.3
      • [ACTION: to clarify the work that is integrating output from the WGs versus overall validation testing]
      • P2.4 Have called out power explicitly as a goal to track.
        • If so important, why only medium? Cannot be higher than P2.3.
        • Happens in the Power Management WG, need to move it into platforms when the WG happens
        • What is the planning horizon?
          • Look again in 3 months
          • This reflects today's priorities, but we will revisit in 3 months
        • [ACTION: explicitly link the dependency with the PM WG output]
        • Worry that we are going too far with the LEB, it is not a product, it is a good way to access Linaro technology and improve time to market
          • Focus at the component level, but test in the full system
        • ARM would like to see P2.6 as a medium priority
        • ARM see P3.8 as essential, need to rewrite [ACTION: Alexander to rewrite]
        • ARM sees P2.5 and P2.6 (Android and Ubuntu UI testing) as low priority

Actions

Number

Action

Owner

State

2011-04-15-1

May make sense to prioritise Android and deprioritise Ubuntu, organise a discussion

David Rusling

2011-04-15-2

P2.7 Clarify the work that is integrating output from the WGs versus overall validation testing

Alexander

2011-04-15-3

move the priority of P1.6 high

Alexander

2011-04-15-4

Clarify the work that is integrating output from the WGs versus overall validation testing

Alexander

2011-04-15-5

P2.4 Explicitly link the dependency with the PM WG output

Alexander

2011-04-15-6

ARM see P3.8 as essential, need to rewrite

Alexander

Next Meeting

TSC/2011-04-13 (last modified 2011-04-15 18:15:15)