* Discussion to generalize the maintenance process of kernel branches
> for the periodic Linaro releases. This still needs documenting in the
> wiki.
>
> * Merged a set of patches put together by Anand Gadiyar to add display
> support to the PandaBoard.
>
> * Merged the following additional patches:
>
> * d983450 cpufreq: Add documentation for sampling_down_factor
> * 054fcd5 ARM: S5P: Fix end address in memory resource information for UART devices
> * bf47520 ARM: make SWP emulation explicit on !CPU_USE_DOMAINS
> * 459967a ARM: fixup SMP alternatives in modules
> * 7144fbc ARM: 6654/1: perf/oprofile: fix off-by-one in stack check
> * 067dfdf ARM: 6659/1: Thumb-2: Make CONFIG_OABI_COMPAT depend on !CONFIG_THUMB2_KERNEL
> * cc0f308 ARM: 6656/1: hw_breakpoint: avoid UNPREDICTABLE behaviour when reading DBGDSCR
> * 40ef21c ARM: 6657/1: hw_breakpoint: fix ptrace breakpoint advertising on unsupported arch
> * 9e97118 ARM: ptrace: remove single-step emulation code
>
== Upstream oriented activities ==
* Review of Arnd Bergmann's flash card article for LWN.
* Incorporation of feedback to the patch adding Thumb2 support to the
P2V branch.
* Another look at the Thumb-2 compatibility fixes for OMAP from Dave
Martin.
* Review of a patch series adding support for SDHCI v3.00.
* Posted patches:
* Rework of the kernel decompressor code to improve efficiency
* Removal of the 4x expansion presumption while decompressing the kernel
* kprobes insn decoding fix
* Ignore mdesc->boot_params if out of range
* Fold lookup_machine_type() into setup_machine()
== Linaro kernel activities ==
* Looked at some bugs:
* Bug 660811
* Bug 720055
* Merged 2.6.37.1 into linaro-2.6.37
* Merged Dave Martin's Thumb2 compatibility patches for OMAP.
* Merged core ARM ffixes from RMK.
* Merged OMAP fixes from Tony Lindgren.
* Opened up the linaro-2.6.38 branch.
Nicolas
== Dong Aisheng <DongAisheng> ==
=== Highlights ===
* Add basic DT support for pinmux-imx core driver (tested on MX6Q)
* Send out the patches for adding DT binding of pinmux mappings for pinctrl
core.
Received comments from Stephen and fixed a few issues, still on
discussing
the better way to bind pin maps due to some DT limitations.
Found a possible way which is similar as regulator and clock core code
do.
=== Plans ===
* Will try the way as regulator and clock framework to do pinmux map
binding.
=== Travel/Time Off ===
* Holiday vacation 1/1/12 - 1/3/12
* Holiday vacation 1/21/12 - 1/29/12
* Connect + ELC 02/05-02/18
== Saugata Das sdas ==
=== Activity Summary ===
* Published data tag patch
* Performed SD, MMC access pattern study during video recording and
USB mass storage
=== Plans ===
* Context ID implementation and internal review
=== Issues ===
* Not able to validate MMC-4.5 patches due to lack of MMC-4.5 device
on any board at my end
== Anton Vorontsov <cbou> ==
It seems that I haven't yet used to the weekly activity reports, but from
now on I'll try to set it right.
=== Highlights ===
* Got a new toy (almost a Christmas present): a pandaboard. :-) Played with it
a little bit: tried Linaro (w/ a serial console, no GUI), various kernels,
including linux-next.
* In addition to the pandaboard, set up a Linaro QEMU environment for quick
tests; especially useful when not having access to a real HW;
* Finally felt confident enough to send out my thoughts on the memory killer;
* From the discussion it seems that:
- Now mem_cg is a lot better on memory usage, but still not enough for
embedded use;
- cgroups has a big impact on the kernel size image;
- cgroups impacts performance in a bad way (using not so synthetic hackbench):
http://lkml.org/lkml/2011/12/21/487
(purely synthetic tests, like cyclictest, do not show any downgrade);
- there are still some demand for the dedicated memory notifiers;
So, it seems that mem_notify reincarnation is inevietable.
=== Plans ===
* I haven't yet able to boot Linaro/Android on the pandabord, need a HDMI-DVI
cable, I'll get it soon; I've tried to boot Android on the QEMU/VExpress,
but it took ages (4+ hours) only to get it to the welcome screen (I was
trying on my travel x120e notebook, 1.6 GHz. But I will try on a more
powerful machine soon, just out of curious).
* I'll take some side project, also Android related. This is to have some
"rest" from the lowmemory notifiers/killer;
* Address some community comments on the current lowmemorykiller driver;
* Start implementing "low memory notify" driver.
=== Highlights ===
* Got a "non-objection" from Arve for pushing the ashmem patches to
staging. Wanted an Ack, but he said he wouldn't have time to look at
them until after the holidays, so I should go ahead and post them. Will
sync with him after the holdiays to make sure nothing in staging looks
out of place. The good news though is that ashmem is in staging, and
that allows us to be very close to a functional ICS with the vanilla
mainline kernel.
* Sent Android evdev patch to input list for feedback. Got a mail from a
ChromeOS guy who had tried to push a similar patch and ran into ABI
issues (legacy applications expect CLOCK_REALTIME timestamps, and
Android expect CLOCK_MONOTONIC). This is problematic because it means
Android ICS depends on breaking Linux' input ABI semantics slightly.
Generated a patch to allow the different ABI semantics to be selectable
via an IOCTL and sent a rough draft to list for feedback. Spent some
time reworking and improving the draft so that I can resend after the
holidays.
* Created a new blueprint to cover the work above:
https://blueprints.launchpad.net/linux-linaro/+spec/android-evdev-upstreami…
* Sent links to Andy's stgit workflow and code to the Android team.
Hopefully they will be able to review it over the break.
* Bunch of discussion around the wider Android upstreaming effort Tim
Bird is trying to organize.
* Piped in somewhat on the Android Logger patch discussion Tim started.
* Reworked fadvise volatile patch to use its own LRU and shrinker
methods which is falling back to being more similar to Android's ashmem.
Still having some locking difficulties to figure out between allocation
and freeing paths.
* Had initial meeting with Anton and made sure he had enough work on his
plate.
=== Plans ===
* Holidays. Will be back in the new year.
* Look into shmem_mmap alternative to using shmem_set_file in ashmem.
* More fadvise work
* Try to sort out something for an ELC talk
* Need to sort out plans dealing with some scheduled extended downtime
for my main development box next year.
=== Issues ===
* None.
== Deepak Saxena <dsaxena> ==
=== Highlights ===
* Had kick off meetings for pinmux and storage sub teams
* Released Linaro 11.12 kernel
* Worked up updating and cleaning up KWG blueprints for 12.01 work
=== Plans ===
* Work on kernel wiki page updates
* Distribute monthly team highlights
* Connect session planning
* Cleanup remaining unfinished KWG roadmap cards
=== Travel/Time Off ===
* Will be out of contact mid-day Tuesday due to veterinary appointment
* Holiday vacation 12/24/11 - 01/09/12
* Connect + ELC 02/04-02/18
== Venkatraman S svenkatr ==
=== Highlights ===
* Review of Data Tag implementation patches
* Review and forward porting of eMMC HPI implementation for data path.
(I need to break this down to changes for each layer in MMC stack)
* Continued study on deadline scheduler and characterization.
=== Plans ===
* eMMC HPI forward porting
* Draft UFS validation plan
== Saugata Das sdas ==
=== Activity Summary ===
* Prepared data tag patch. Internal review ongoing
* Continued review and follow up on MMC-4.5 patches within community.
Helping on cache control, power class selection, HS200 patches.
=== Plans ===* Study the MMC, SD block access pattern during "write
intensive" use cases like video recording and USB mass storage
transfer.
=== Issues ===* Not able to validate MMC-4.5 patches due to lack of
MMC-4.5 device on any board at my end
=== Highlights ===
* Reworked ashmem patches that I've preped for staging per Arve's
feedback and resent the queue for acks. As soon as I get an ack, I'll
fold the patches down and send to Greg.
* Sent a small and simple evdev patch from Android to input maintainer
for feedback. This is the one patch outside of the ashmem patches I have
queued for staging that allows ICS to run on a staging-next kernel.
* Reworked the Android alarm-timer driver so it would be ready for
possible inclusion into staging. Its a little more awkward because its
very dependent on wakelocks. Sent out for feedback to Android developers
and Greg.
* Synced with Greg and others on what remains in common.git that has the
potential to go into staging.
* Re-sent Rafael the split out Android wakelock queue.
* Re-adding CONFIG_SCHED_MC options to 3.2 based Linaro+Android kernel.
* Helped Jesse with some git usage.
* Queued two RTC fixes for 3.2 and sent them to Thomas.
* Discussed vdso clock_gettime(CLOCK_THREAD_CPUTIME,...) implementation
issues with some of the facebook folks. This took a tangent into larger
discussion about timekeeping items like future support for CLOCK_TAI and
other issues like improved leap-second handling.
* Got a mail from Tim Bird on setting up a wider Android upstreaming
effort, trying to get a handle on who all is doing what in the
community. Sent him updates on what the Linaro team is doing and
GregKH's staging tree plans. Plans for a meeting at Connect including
non-linaro folks working on these issues is in progress.
* Spent a little bit of time on the fadvise volatile patches, and ran
into some potential ugly locking issues that I need to resolve.
Considering falling back to the lru approach the ashmem patch uses.
=== Plans ===
* Hopefully get final ack from Arve on ashmem patches and push those to
Greg.
* Follow up on alarmtimer mending patches.
* Look into shmem_mmap alternative to using shmem_set_file in ashmem.
* More work on fadvise volatile implementation.
* Prep for the holidays
* Start actually acting as sub-lead for Kernel Android work (I've been
bad the last two weeks) and get a regular meeting planned.
=== Issues ===
* None.
== Linus Walleij linusw ==
=== Highlights ===
* Located and fixed a nasty ux500 iotable bug reported by
Daniel Lezcano (PM) on linux-next:
http://www.arm.linux.org.uk/developer/patches/viewpatch.php?id=7233/1
* Located a clock drift problem on ux500 found by
Daniel Lezcano (PM) as well. Thinking about a patch...
* Various pinctrl patches being applied and discussed among them
dummy pinmuxes akin to dummy regulators, we need to
see if this is a good idea or not.
* Blueprints and stuff updates with assitance from Mounir
(thanks!) we're using pinctrl as some test
case for the process, which is both fun and painful.
* Reviewed Samsung and Freescale and PXA pinctrl
(mainly mux) implementations, it actually looks like these
drivers will be mainly a smooth ride.
* Discussed DT mappings to pin controllers again, now
Dong Aisheng is taking it a step further since he has a
clear idea on how this should work and will write and
document pinctrl bindings for Freescale, patches already
exist for iMX.6 this is looking real good.
* Ux500 mainlining since too little is happening:
- core ASIC patches and dropping prototype silicon patches
sent out.
- Helping our ALSA SoC authors to get started to get their
stuff upstream, some problem with the usual - clocks,
regulators, DMA, but it's looking OK.
* Helped Sjur Brændeland to get M7400 XSHM patches
into linux-next and will help him to establish a
k.org repository.
=== Plans ===
* Vacation 2011-12-19 thru 2012-01-06
Expect next report at the end of my first working week
2012-01-13.
* Drive generalization of Nomadik GPIO
by using the pinctrl framework.
drivers/gpio/gpio-nomadik.c
* 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
- the HWMON stuff.
- DB8500 PRCMU updates
=== Issues ===
* Snowball/Igloo may need help with a USB issue, but
I won't be in any capacity to help them :-(
Thanks,
Linus Walleij
