Wednesday, 22-09-2010




IRC Nick

Amit Kucheria


Amit Arora


Vishwanath Sripathy


Bobby Bhatacharia


Srinivas Kalaga


Yong Shen


Robin Randhawa



  • Review action items from last meeting
  • Blueprint status



Master blueprint for PM WG


PM Tools






Action Items from this Meeting

  • Vishwa: working to refactor assembly routines for save/restore/wfi
  • Vishwa: split up work items to be more granular
  • Yong: talk to FSL u-boot maintainer about missing features and ways to get them into mainline u-boot and linaro u-boot
  • Everyone: help testing linaro images

Action Items from Previous Meeting

  • ACTIVE (Immediate):
    • Everyone: help testing linaro images
    • Yong: talk to FSL u-boot maintainer about missing features and ways to get them into mainline u-boot and linaro u-boot: INPROGRESS
    • Amit A: send out email for testing of powertop once it hits ubuntu packages: DONE
    • Vishwa: working to refactor assembly routines for save/restore/wfi: INPROGRESS
    • Amit K: give overview of work for next cycle: Done
    • ACTION: Amit K to spend some time on usecase to reproduce ondemand governor problems: POSTPONED
    • Amit K (added post meeting): looking into merging linux-pm branch into Linaro
    • ACTION: ARM to share internal instrumentation flow (BAB: we might also align with Linaro on workload discussions)
      • Might take couple of months
    • ACTION: Amit K to talk to jeremy about power domain framework: DONE
      • Jeremy needs help, will revisit in a few weeks
    • ACTION: Srinivas to provide details of where he believes userspace - kernel interaction is required. (low prio)
    • ACTION: Bobby to check on multi-core boards availability (request open)
    • ACTION: ARM to discuss giving out internal Eclipse based tool (similar to powertop) (no ETA as of now)
    • ACTION: Amit Kucheria and Vishwa to get inputs from community on the issues related to CPUIDLE governor: POSTPONED until instrumentation work


  • Testing Linaro images
    • test entire image, not just omap kernel
    • test on BB and then on Zoom3
  • The next 6 months
    • Thermal management with cpufreq
      • Thermal Abstraction to register various temp. sensors, HW monitors (existing thermal framework useful?)
      • API for cpufreq to read from this abstraction and make a decision?
      • Should cpufreq make decision or should it get a more refined input
        • Raw degrees from driver and then intelligence in platform-specific cpufreq driver?
        • OR get a suggestion from thermal abstraction to reduce frequency(how will thermal abstraction know of platform-specific details?)
      • Thermal info can be thought of as a 'constraint', can it be modeled into PM-QoS?
      • Look at thermal framework and how intel_idle driver takes thermal into account
    • ARM idle state -> C/P-state mapping

      • ARM needs to work with various vendors to arrive at common definitions for various states
      • Tools wouldn't be as useful if the states mean different things on different platforms
    • "tuning for idle"
      • split out the per core wakeups in powertop?
      • nmon (look for ideas for displaying)
      • Documentation about howto debug low-power usecases
    • ARM cpu hotplug
      • hotplug very expensive
        • not different from putting core into WFI
      • consider looking at containers or intelligent scheduling?
        • schedule processes on first core until a threshold is reached, only then activate second core?

WorkingGroups/PowerManagement/Meetings/2010-09-22 (last modified 2010-09-22 09:44:52)