- Period: (2011-04-22-2011-04-28)
PM: Mounir Bsaibes <email@example.com>
Burndown information : http://status.linaro.org/linaro-kernel-wg.html
Key Points for wider discussion
MX5: Keep L2 cache enabled before jump to kernel http://lists.denx.de/pipermail/u-boot/2011-April/090437.html
MX5: clock: Add clock config interface http://lists.denx.de/pipermail/u-boot/2011-April/090438.html
PMIC: Add dialog pmic support http://lists.denx.de/pipermail/u-boot/2011-April/090440.html
MX53: loco: Add power init support http://lists.denx.de/pipermail/u-boot/2011-April/090441.html
- On-going reviewing and queuing of ARM kprobes patches from Tixy.
- More investigation about the kernel decompressor acting funny with the LZMA decompressor, or even corrupting DTB when the DT append patch is used. Two separate issues were identified so far with cumulative effects: global variables marked static causing GOTOFF relocations instead of the expected GOT reference, and a misaligned stack in some circumstances reaching havoc with LDRD/STRD instructions. Those fixes plus a couple related cleanups were posted to LAK. One report indicates problems on a specific target may still exist.
- Initial review of tglx's sched_clock consolidation patchset.
- Review of Will Deacon's patch for SMP futex support.
- DMA memory management
- MMC patch review
- AXI bus driver
- Working RCU priority boosting bug: Cheng Xu encountered same bug with user-level program, so this is officially a scheduler bug. Fortunately, my workarounds still are working.
- Sent Sedat Dilek two sets of diagnostics thus far, working up a third set. Problem is in wake-up path for per-CPU kthread.
Linaro kernel activities
- Call for testing: OMAP4 extra support from Andy Green.
- Review of the proposed ST-Ericsson kernel repository. Sent comments to Lee Jones.
- Merged the following in linaro-2.6.38:
- Fix infinite loop in ARM user perf_event backtrace code
branch 'omap-edid' of git://git.linaro.org/people/doanac/linux-linaro-2.6.38
- a couple zImage fixes and cleanups from me described above.
Pushed out updated Linaro Android tree which resolves the omap3 USB issue and supports HDMI output on PandaBoard, allowing the Android UI to now function on Panda.
- Sent out the Posix Alarm Timers patch queue to lkml for review/inclusion.
Rebuilt the Android Trivial tree from latest Android tree: http://git.linaro.org/gitweb?p=people/jstultz/android-trivial.git;a=summary
- Sent out an MM patch for review from the Android Trivial Tree.
- Sent out six MMC patches for review from the Android Trivial Tree.
- Sent clocksource cleanup patch to Magnus Damm to follow up on a conversation we had at ELC.
- Submitted a LPC BoF discussion on Kconfig fragments for internal review.
- Discussing mmc nonblock implementation on mmc-mailing list.
- Added pre_req and post_req for SDHCI driver, but did not see performance gain on mx51 esdhc. No comments from Per Forlin on that so far.
- Added pre_req and post_req for for mxs-mmc driver and saw ~4% performance improvement at the best case. Per Forlin gave some comments saying pre_req and post_req were not added correctly. But I do not completely understand his comments and asked him for clarification. No further response yet so far.
- The .dtb support has been available in hwpack daily build and lmc v0.4.4. Tested it on mx51 babbage and it works out of box.
- With the help from Dave and Nico, the failure of John Bonesio's append-dtb patch on babbage was killed. It's caused by the use of static variable in uncompressed.h, which will not be relocated because it's not in .bss section.
- Caught the problem of babbage.dtb memory node, which misses device_type' property.
- Continuing migration of Samsung clock code to use common clock framework.
- Preparing for Linaro kernel board bring-up on Samsung's s5pv310 based low-cost board.
- Discussion with Thomas Gleixner, Andrew Morton, and Linus Torvalds on maintainership issue.
- Pre-Budapest definition/scheduling work, created UDS session blueprints
- Created a step by step instructions how to create Engineering blueprints, where to place the specification wiki page and how to attach TR blueprints to Engineering blue prints
- Close on UDS sessions schedule
Risks / Issues
- Not all UDS session leaders may be in Budapest, meed to make sure the sessions are covered
- New Team Lead Deepak Saxena
- Interviewed two candidates for Kernel WG, both are +1. Will set up interview with another candidate next week.
WorkingGroups/KernelArchived/Status/2011-04-28 (last modified 2013-01-14 19:40:38)