== Dave Martin <dmart> ==
=== General Activity ===
* Some kernel patch work:
* CPU info declaration macros -- A conclusion seems to have
been reached on the form these should take. Some
implementation work done; still need to finish, test and
post.
This should permit a fair number of lines to be removed
from arch/arm/mm, without any functionality impact.
* Will Deacon's Generic MMU off code -- helped fix an off-
by-one type bug which was preventing the temporary
stack's identity mapping from being installed properly.
* More activity on minor bugs in gas:
* http://sourceware.org/bugzilla/show_bug.cgi?id=12854 (ARM:
Nonsensical argument shift specifications accepted for
LSL/LSR/ASR/ROR mnemonics)
* Tracking down some userspace alignment faults:
* gtk-sharp2 -- https://bugs.launchpad.net/ubuntu/+source/gtk-
sharp2/+bug/798315
* firefox --
https://bugzilla.mozilla.org/show_bug.cgi?id=626635 (already
reported and fixed by other people upstream; fixed in 5.0 /
oneiric)
* Started a mail thread about whether we should be more
aggressive in getting alignment faults fixed, particularly
within Linaro.
=== linaro-kernel-o-standard-arch ===
* Testing of the linux-linaro-2.6.39 and linux-linaro-oneiric
trees with Thumb-2:
* Pinged John Rigby again -- he's agreed to try turning on
Thumb-2 for the vexpress packaged kernel.
* mx51evk status still to be determined
* Did some investigation into what mx51evk is using FIQ for.
Some compatibility issues for Thumb-2: still need to follow up
with the the imx guys.
=== Plans ===
* Check whether the new kernel trees work in Thumb-2 on mx51evk
and follow up with Freescale landing team.
* Follow up on Cortex-A15 model/platform availability.
== Highlights ==
* Tried to adapt OMAP I2C1 and twl driver into DT for panda board.
- Looks like device registration for omap drivers is done in different
way compared to other SoC's
- Adapting OMAP device drivers requires DT internals study and also OMAP
device registration might need some more changes.
* DT documentation study and looking at other SoC driver adaptation into DT
== Plan ==
* DT internals study for adapting OMAP drivers
* OMAP kexec - plan pending[need to sync up with Anand G]
-Manjunath
=== Device Tree ===
* Read device tree usage model document from Grant
* Studied the irq_domain translation and device name overridden
patches from Grant
* Based on above infrastructure code, worked out a basic device tree
kernel for imx51 babbage. With uart and fec supported added, the
dt kernel can boot into console with nfs root. It could be a good
base for starting playing dt and adding further device support.
The patches were posted to lkml and lakml, and had the good
feedback and valuable comments. Will respin for v2 soon.
=== ARM/imx consolidation ===
* Made some fixes on imx-dma device registration code, and cleaned
the code up to register dma device in per-soc-init function. The
fixing patches were picked up by Sascha, but the cleanup one has
not yet. Will ping him for that.
=== Misc ===
* [Bug 754254] imx51 randomly truncates serial input at 31
characters. Posted v2 of the patches that add card-detect
support for mx5, and acked by Wolfram.
http://thread.gmane.org/gmane.linux.kernel.mmc/8558/
So sent the pull request to Nicolas and he pulled it.
Heads-up for Nicolas: Arnaud Patard reported one issue with the
patches as below.
http://article.gmane.org/gmane.linux.ports.arm.kernel/120790
I will soon send a follow-up to get that fixed.
=== Plan ===
* v2 of the imx51-dt patch set
--
Regards,
Shawn
== Per Forlin <perfor> ==
=== Highlights ===
* Rebased mmc async request on Linux 3.0 rc3. Had a lot of stability
issue on rc3. Moved to rc3+ (torvalds HEAD) and then all tests pass.
=== Completed WIs===
* https://blueprints.edge.launchpad.net/linux-linaro/+spec/mmc-async-request
* rebase patches on top of linux 3.0, run all tests: DONE
* post linux 3.0 mmc async req support on mmc-list: DONE
=== Highlights ===
* Reviewed and commented on clocksource entropy idea on lkml
* Worked on a few RTC related bugs
* Sent hack to linaro-dev to fix DPI/DSI build issue
* Released a 2.6.39 based linaro+android kernel tree
* Attended Android sync meeting with Deepak and Zach
* Reviewed android-3.0 tree, noticed they have not integrated the
upstream alarm timers code, and sent a mail to Colin Cross to see if
there are any plans to do so.
* Took some initial looks to see how hard re-implimenting the android
alarmtimer code would be on top of my posix alarmtimers implementation.
* Worked on internal presentation
=== Plans ===
* Push current RTC/alarmtimer fixes upstream
* Continue working on internal presentation
* Continue looking at android/posix alarmtimers merging
* Generate linaro+android 11.06 freeze kernel
* Send out select from choice block patch
=== Issues ===
* NA
== Thomas Abraham <thomas-ab> ==
=== Highlights ===
* Completed re-run of tests with memory regions on smdkv310 board,
but there were no significant power savings.
* Discussed with Ankita, author of the memory regions patches,
about the test cases being used and the reasons about no
improvement in power savings.
* Started with the device tree work items for Exynos4 platform.
=== Plans ===
* .Complete the registration of platform devices
* Move board specific data to device trees.
Hello Andy,
I forward ported the following patches from you to linaro-2.6.39:
Author: Andy Green <andy(a)warmcat.com>
Date: Thu Mar 24 21:27:29 2011 +0000
OMAP2+: add cpu id register to MAC address helper
Introduce a generic helper function that can set a MAC address using
data from the OMAP unique CPU ID register.
For comparison purposes this produces a MAC address of
2e:40:70:f0:12:06
for the ethernet device on my Panda.
Signed-off-by: Andy Green <andy.green(a)linaro.org>
Acked-by: Arnd Bergmann <arnd(a)arndb.de>
Author: Andy Green <andy(a)warmcat.com>
Date: Thu Mar 24 21:27:38 2011 +0000
OMAP2+: PANDA: Fix up random or missing MAC addresses for eth0 and wlan0
This patch registers a network device notifier callback to set the mac
addresses for the onboard network assets of Panda correctly, despite the
drivers involved have used a random or all-zeros MAC address.
The technique was suggested by Alan Cox on lkml.
It works by device path so it corrects the MAC addresses even if the
drivers are in modules loaded in an order that changes their interface
name from usual (eg, the onboard module might be "wlan1" if there is a
USB wireless stick plugged in and its module is inserted first.)
Cc: Alan Cox <alan(a)lxorguk.ukuu.org.uk>
Signed-off-by: Andy Green <andy.green(a)linaro.org>
Acked-by: Arnd Bergmann <arnd(a)arndb.de>
Could you please follow up with the appropriate maintainer (probably
Tony Lindgren) to ensure those patches are merged upstream and not
forgotten about? You may add my ACK as well.
Thanks
Nicolas
=== Highlights ===
* Kprobe work items from blueprint "Finish Thumb2 Support"
* Modify ARM instruction emulation to use decode tables: DONE
* Decide on and implement handling of conditional instruction
probing: DONE
* Make ARM ALU and LDR instruction emulation interwork with Thumb:
INPROGRESS
* Tidyup and document kprobe changes: INPROGRESS
* Submitted an RFC PATCH to linaro-kernel containing the kprobes table
driven instruction decoding implementation.
=== Plans ===
* On holiday 2011-06-16 to 2011-06-26
