Agenda

Owner

Topic

Ilias

Small retrospective for 1109: we delivered few of the originally targeted work items - this is the 2nd month in a row when work items are moved in block to the next month. Do we need to change them? How?

Actions from previous meeting

  • CARRIEDOVER: Paul to write up the current capabilities of the scheduler INPROGRESS
  • CARRIEDOVER: Write down thoughts based on the future topics INPROGRESS
  • CARRIEDOVER: Ilias : to setup LCQ4.11 OCTO team meeting - DONE
  • CARRIEDOVER: Ilias to organize the STB summit - INPROGRESS
    • Ilias: Agenda created and shared also some more folks were contacted.
  • CARRIEDOVER: Ilias to organize the IVI discussion in Linaro Connect - DONE

Attendees

  • Loïc Minier
  • Steve McIntyre

  • Grant Likely
  • Ilias Biris

Minutes

  • For BA - comments from Rob Herring (also copied to the OCTO list by David) were discussed briefly. Incidentally this was also discussed in the boot architecture weekly meeting, and apart from Grant's suggestion which was to focus on ARM Server side, we also decided to create a backlog composing of short term work to cover the pain points which were identified, as well as long term issues.
  • Papyrs to be used to track the roadmap items and requirements which need to be created before discussing them with the TSC. There are currently some requirements for OCTO (check with Joey Stanford if you do not have access to Papyrs yet):
  • Connect
  • Small retrospective for 1109: what to do with the blueprints, since it is a recurring pattern of having blueprints which are planned to be done within a month but at the end of the month either no work item is ready to be marked done, or the work items for the month have been partially completed. Is there something we should be changing - plan less scope, or change the structure of the work items in order to handle the work a bit better?
    • July/August/September were a bit awkward due to leaves
    • OCTO is often looking at new things coming up which are not really planned in advance so they become distractions or variations in what we are doing - quite difficult to stick to an idea of what was planned some time ago to deliver
    • Having blueprints in OCTO for DT/BA does not make much sense - for actual deliverables it makes more sense to have the BPs in the actual working groups (Eg kernel in this case)
    • AGENDA FOR NEXT MEETING: We should defer to next meeting, and possibly look at the points to clarify (if any) at Connect
  • Grant reported from his side. Boot architecture is really about server class machines, and commodity machines. For the embedded systems DT support is enough.
    • One thing that came up in the last meeting for boot-architecture: runtime services from UEFI. It is like calling into the BIOS to get information about low level parts like FW. Could it be completely abstracted in the UEFI so that it does not matter what provides the service - eg DT or anything else? It is doubtful, as it will require more features from FW. For instance this would be useful to eg handle both ACPI and DT.
    • Grant will look into the list of requirements to make a shift in the attention for boot architecture work towards server class ARM devices
    • DT support is a given and continued target for embedded systems

Actions from this meeting

  • CARRIEDOVER: Paul to write up the current capabilities of the scheduler INPROGRESS
  • CARRIEDOVER: Write down thoughts based on the future topics INPROGRESS
  • CARRIEDOVER: Ilias to organize the STB summit - INPROGRESS
    • Ilias: Agenda created and shared also some more folks were contacted.
  • CARRIEDOVER: Ilias to organize the IVI discussion in Linaro Connect - INPROGRESS

OfficeofCTO/2011-10-04 (last modified 2011-10-07 10:12:32)