PM WG weekly status meeting

Wednesday, 22-12-2010

Attendees

Name

Email

IRC Nick

Vishwanath Sripathy

vishwanath.sripathy@linaro.org

vishwabs

Yong Shen

yong.shen@linaro.org

yongs

Vincent Guittot

vincent.guittot@linaro.org

vingu

Torez Smith

torez.smith@linaro.org

torez

Agenda

New Action items

  • Post Hotplug latency measurement patch to lkml and laml - Vincent
  • Start the discussion on handling dynamic updates as part of debugfs - Vincent

Old Action Items

  • Vincent: Talk to partners about what their plans for hotplug are - PENDING
  • ACTION: Srinivas: Check if Versatile Express has any thermal sensors: Pending
  • ACTION: Srinivas: Enquire about status of talking to community about the causality tool for power profiles: Pending
  • ACTION: Amitk: Paraphrase our requirements for PM_QoS for dealing with application constraints vs. system constraints: Pending
  • ACTION: Robin: Get back with details on how to split access into secure and non-secure context: PENDING
  • ACTION: amitdaniel: Add work item to investigate a simpler governor that cares only about wakeup latency w/o the prediction logic: PENDING
  • ACTION: amitdaniel: Add work item to investigate how the governor can mask certain c-states based on latency constraints: PENDING
  • DORMANT :
    • ACTION: ARM: results of using PMU counters for correlation to PM: POSTPONED
    • ACTION: ARM to share internal instrumentation flow (BAB: we might also align with Linaro on workload discussions): POSTPONED
    • ACTION: Srinivas to provide details of where he believes userspace - kernel interaction is required. (low prio): INPROGRESS

Minutes

  • Vincent posted patch for measuring cpu hotplug latency patch. He also developed plugin for pytimechart to display various cpuhotplug events.
  • This Will help in analysing cpuidle pattern in a SMP system.
  • Plans to post hotplug patch to lkml and plugin to pie timechart mailing list.
  • Torez plans to work on testcases for cpufreq for next release cycle. cpuidle is not yet common acrosss ARM SOC platoforms (C state definitions) and hence cannot have common testcases. cpuidle testcases can be taken up once C state definition is standardised across all ARM SOCs.
  • Vincent has question on how does clock debugfs handle scenarios where parent clock of a given clock node changes dynamically. Yong to check this.

WorkingGroups/PowerManagement/Meetings/2010-12-22 (last modified 2010-12-24 08:24:32)