Wednesday 01/09/2010

This month's meetings

<<MonthCalendar: execution failed [Argument "anniversary" must be a boolean value, not "workinggroups/powermanagement/meetingtemplate"] (see also the log)>>

Attendees

Name

Email

IRC Nick

Amit Kucheria

amit.kucheria@linaro.org

amitk

Amit Arora

amit.arora@linaro.org

amitarora

Vishwanath Sripathy

vishwanath.sripathy@linaro.org

vishwas

Bobby Bhatacharia

bobby.bhatacharia@arm.com

bobbyb_arm

Srinivas Kalaga

srinivas.kalaga@arm.com

?

Yong Shen

yong.shen@linaro.org

yongs

Agenda

  • Review action items from last meeting
  • Blueprint status

Blueprint

Assignee

Master blueprint for PM WG

amitk

PM Tools

amitarora

CPUIdle

vishwanath-bs

Cpufreq

amitk

  • Powerdebug visualisation (CTT tool from TI)

Action Items from this Meeting

  • Yong: test against 2.6.35 mainline tree + jeremy's patches on babbage 3.0 board
  • Vishwa: measure latency of SRAM functions to see if different cache behavious has any effect on context save and restore
  • Everyone: Help with testing of Linaro images
  • Everyone: Confirm availability for LDS in Orlando and if there are any blockers

Action Items from Previous Meeting

  • ACTIVE (Immediate):
    • ACTION: Yong to work with John Rigby and Ubuntu kernel team to make sure the Linaro kernel contains powertop kernel patches : Postponed
    • ACTION: Vishwa to verify powertop on 2.6.35 : Done (verified on zoom3)
    • ACTION: Vishwa to check with cpuidle/cpufreq experts in TI for verifying cpufreq behavior on multi-core OMAP: Postponed until implementation finalised and open-sourced
    • ACTION: Amit A to get the powertop patch integrated into Linaro/Ubuntu packages: in progress
    • ACTION: Yong to test common clk API patches on imx5 and help get it booting on babbage 3.0
  • DORMANT :
    • 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

Minutes

  • Link for TI tool for clock visualization : http://focus.ti.com/pdfs/wtbu/CTT-OMAP4430ES2.0-v1.7.0.2.zip

  • common clk api on imx5
    • jeremy's patches are for babbage 2.0, doesn't work on babbage 3.0
    • the patches are against mainline tree, not against freescale bsp
    • recommendation - test against 2.6.35 mainline tree + jeremy's patches on babbage 3.0 board
  • cpuidle latency tracking infrastructure
    • code to timestamp the ARM context save and restore (in SRAM, in assembly)
    • ARM context save and restore might be a constant time
      • So perhaps moving out the timestamping to C code (before entering assembly)
      • if it is found to take constant time, we could just use trace events and ftrace events instead of adding code to the idle path
  • powertop
    • upstream has low priority on the patch
    • powertop being modified to use libcpufreq by suse
    • Amit A will then rebase his patch
  • powerdebug

WorkingGroups/PowerManagement/Meetings/2010-09-01 (last modified 2010-09-01 16:33:24)