From: Shawn Guo <>


  • Sent a series to clean up basic clks of common clk framework, and
    • Mike will integrate them into his fixup branch.
  • Sent a patch to fix the issue that audio driver only works with SSI1
    • on imx5. It assigns dma peripheral type according to bus topology. (Merged by Mark)
  • Sent a couple of patches to fix sgtl5000 MIC Bias power and add
    • audio routing support for imx-sgtl5000 machine driver to get capture functional. (Merged by Mark)
  • Tested patch "ASoC: sgtl5000: Convert mic bias to a supply widget"
    • from Mark with imx-sgtl5000.
  • Added basic clock management for fsl_ssi driver. To make it compile
    • for both powerpc and imx, the powerpc dummy clock API needs to be built in. Sent a patch to linuxppc-dev list and waiting for comment.
  • Reviewed and merged anatop-regulator imx6q dts patch from Paul.
  • Reviewed and merged basic imx28 DT patch (booting into console with
    • nfsroot) from Aisheng.
  • Tested v3.4-rc1 on imx28 and imx6q, and sent a couple patches to fix
    • amba probe failure, section mismatch warning.

Deepak Saxena <dsaxena>


  • Worked on article on "Kernel Upstreaming Challenges" for executive

member update.

  • Ended up writing one version and completely throwing it away as it

was too introductory

  • but hope to use that to write a larger white-paper type of document

on the topic. Update

  • Self-review, engineer review. (Do we really need to do these every

3-4 months?)

  • Worked with Mounir on 12.04 blueprint updates
  • Submitted talk proposal on "One zImage to Rule Them All" to LinuxCon Japan

  • Regular 1:1s and meetings


  • Finish review cycle work
  • Finish patch tracking spreadsheet for upstream kernel work
  • Connect planning
  • Roadmap cards


  • No major ones, though still feeling behind on everything due to

being sick earlier

  • in the month. :/

Travel/Time Off

  • Off May 4th and following week, back to work on Monday the 13th. Will have no
    • interwebs and limited phone access.
  • Connect Q2, possibly with follow up

From: John Stultz <>


  • Sent out last bits of timekeeping changes for 3.4 & got them merged.

  • Spent a bit of time re-installing & reconfiguring my system

environment per new internal policies:

  • Sent out first pass at config fragment integration into the linaro

kernel. Got some good feedback & patches from Tixy and Tushar.

  • Ran Android Subteam phone meeting & set up recurring phone meeting


  • Reviewed Danial Walkers performance comparision between shm based and

android logger based logging. Got into a longish discussion off-list discussion with him about Android's design requirements. Hopefully will see another iteration from Daniel.

  • At Tixy's prodding, contacted Dima to see about the earlysuspend

removal in 3.3 and what fixes for the log flooding that caused are recommended. Got a quick response that they have a fix, and are looking to push it out soon. Tixy suggested a fix in the meantime.

  • Did a first pass fwd port of the Android 3.3 patch stack to 3.4.

Messed up some of the mmc items and will likely have to try again.

  • Got a first pass at reworking the fadvise volatile code to store tree

roots in a hash rather then bloat the address_space structure. It seems to be working, but I kind of don't believe it.


  • More testing on fadvise volatile hash rework & send to lkml

  • Get all of Tushar's changes merged into the config fragment tree
  • Take another shot at fwd porting the Android tree to 3.4


  • NA

From: Arnd Bergmann <>


  • Merge window, part two: everything we had queued is now in mainline
  • Discussed swap on flash with a few people, sent out first proposal
  • Continued DT discussion for spear, kirkwood, shmobile


  • more review for DT conversions
  • regression testing post merge window
  • getting my build server back running
  • dig out my CONFIG_NO_IOPORT patch set and get that into linux-next

From: Dave Martin <>

Activity Summary

  • Semihosting support for the big.LITTLE bootwrapper tested and

merged into gnu-build.

  • Experimental DT support added to the big.LITTLE bootwrapper.
  • Locally rebased model kernel support onto torvalds/master. I will

want for v3.4-rc1 to come out before pushing this.

  • Added cpufreq transition notifications to the reference switcher

cpufreq driver. This seems to correct the erroneous frequency values observed when reading cpufreq status from sysfs.

  • Followed up with the ARM guys on the switcher deadlock issue

observed last week. Some progress, but it's not fully fixed yet.

  • Kernel opcode injection patches reposted. These aren't urgent, so

I will probably wait until after the merge window before sending them to Russell's patch system.


  • Revisit hyp-entry series.

Work Items

no change this week


Fri 6 April - Fri 13 April

From: Rajendra Nayak <>


  • Worked on patches to make OMAP transition to common clk one
    • SoC family at a time and also maintaining git-bisect. Turned out much ugly than I initially thought. Shared the patches along with a few minor fixes with Mike.
  • Started working on optimizing (reducing) OMAP4 common clk data
    • on top of Mikes' initial work. Mostly done manual cleanups for now. Need to update the autogen scripts.
  • Started validating OMAP5 common clk using Mikes' framework changes
    • and autogen updates.


  • Autogen updates for OMAP4 data, OMAP5 validation and fixes to
    • framework changes.


  • Planning to take off on 2,9,10 April due to some personal
    • emergency travel.

From: Venkatraman S <>


  • Sent out pull request for UFS and eMMC features for Linaro Kernel release
  • Coding up eMMC foreground HPI code. Sent a rough first cut to Storage folks.
  • Studies UFS implementation. Plan to abstract out the PCI specific

part of driver files


  • More HPI coding and do some stress tests
  • Test eMMC Power off notify with OMAP.


  • Was on leave for 2 days

From: Saugata Das <>

Activity Summary

  • Tested cases for < 4KB sector and EFI partition, with native sector

support patch (

  • Power OFF notify patch submitted


  • Work on context ID patch


  • Submit a proposal on the implementation of large block size



  • None

From: Linus Walleij <>



  • Easter vacation april 2-5th.
    • expect next report on april 13th
  • Review incoming pinctrl patches.
  • Prepare to convert ux500 (gpio-nomadik) to pinctrl for the kernel 3.5 timeframe.
  • Attending the Embedded Systems Conference in San Jose 27th thru 30th of march.
  • Test the U9540 core patches on real hardware and rework as outlined by Arnd.
  • Test the PL08x patches on the Ericsson Research PB11MPCore and submit platform data for using pl08x DMA on that platform.
  • Look into other Ux500 stuff in need of mainlining... like - Ux500 clocks - the HWMON stuff.


  • N/A

Anton Vorontsov

Dong Aisheng

Mounir Bsaibes

Nicolas Pitre

Niklas Hernaeus

Thomas Abraham

WorkingGroups/KernelArchived/ActivityReports/2012-03-30 (last modified 2013-01-14 19:35:21)