=== i.MX6Q DT ===
* Submitted the v3 of imx6q patch series.
- Reworked the L2 register save/restore implementation per Lorenzo's
suggestion
- Rebased to Rob's latest GIC OF bindings series
- Rebased to Sascha's MULTI_IRQ_HANDLER support series
* imx6q fec, gpio and usdhc patches have been merged by sub-system
maintainers.
=== Plan ===
* Start looking at migrating imx to pinctrl sub-system
* Revisit iomuxc device tree bindings
=== Misc ===
* Getting cold, so will not attend the meeting tonight
--
Regards,
Shawn
=== Highlights ===
* Recovered from my head-cold.
* Continued lively email discussions with Mark Gross, Rafael Wysocki,
and Alan Stern on my and Mark's different wakelock proposals.
* Got very close to sending out very rough first draft of the patches
that illustrate my wakelock functionality idea (just one or two little
bugs to squish).
* Merged some patches from Vishal and Zach into the Linaro+Android tree.
* Updated, tested and released a Linaro+Android baseline kernel
linux-linaro-3.0-2011.09-0-android-1
* Met with Deepak for lunch.
* Resubmitted the config fragment merge script to lkml and got some
additional feedback.
* Maaaaybe pestered Dave Hansen enough about alternative ashmem idea
that he might start investigating it himself!
=== Plans ===
* Get first rough draft of wakelock idea reviewed by a few folks to try
to catch any glaringly terrible bits.
* Maybe submit wakelock patch queue to lkml for flame-bait/discussion?
* Integrate config fragment script feedback and resubmit.
=== Issues ===
* Need help to follow up on Linaro Connect expense funds getting
seemingly lost between Linaro's bank and mine.
Hi all,
We need to start planning for Linaro Connect/UDS in Orlando. For those
that have not attended one of these events before, the way they are
run is very different than a traditional conference. We will be running
summit sessions which are focused discussions on specific topics
with the goal of getting a list of concrete action items, both technical
and logistical, towards solving a problem. Unlike prior UDS events,
where Linaro was 100% focused on planning, we will split the day
with planning summits in the morning and hacking in the afternoon.
I'd like to gather a list of topics from everyone so that I can start creating
the session blueprints that will be used by the scheduler and the
notetaking system. My current list of topics:
- Single zImage (2 sessions?)
- Device Tree (2 sessions?)
- Blueprint/Work Item process
- eMMC 4.5/UFS
- Flash storage optimizations (follow up from last connect)
- Android Upstreaming
- Kernel CI Loop
Please provide any other ideas you may have.
Thanks,
~Deepak
Returning from a week of vacation followed by Linux Plumbers Conf, so
this contains bits from this week and last. Its good to be home!
=== Highlights ===
* Attended Linux Plumbers & presented my BOF on (K)Config fragments
* Sent out patches described in my BOF to lkml for review & feedback.
* At Plumbers, discussed wakelock alternative idea with Rafael J.
Wysocki (PM Maintainer), Amit Kucheria, Mark Gross (Intel) and others.
* Met up with Mark Gross in Portland to further discuss his ideas for
wakelock alternatives. Reviewed his initial prototype, and started some
off-list email discussions with other pm folks.
* Started prototyping my own wakelock alternative approach (very early
stages).
* Added SCHED_MC to Linaro+Android defconfigs
* Updated and pushed Linaro+Android tree for 11.09
* My 3.1 queue of fixes for RTC/AlarmTimers was merged into mainline
* My 3.2 queue of alarmtimer fixes has been merged into -tip
=== Plans ===
* Continue working on wakelock alternative prototyping and discussions
* Discuss config fragment tooling patches.
=== Issues ===
* Getting a cold.
== Venkatraman S svenkatr ==
(Out of office for 3 days last week due to TI India Vacation)
== Highlights ==
* Created dma unmap patches for OMAP which handles errors in async req
operations
(Posted to linux-mmc)
* Measured time spent on programming state for SD cards for implementing HPI.
Implemented an abort interface for OMAP HSMMC, which is precursor to HPI
* Creating elevator interface document to implement erase block support
== Plans ==
* Validate the abort interface on Pandaboard
* Post patches for adding erase block size parameter in block device
properties (struct hd_struct in gendisk
[Resend with update of l-a-k address in my alias to the correct
one on infradead].
A group of us met at Linux Plumber's Conf two weeks ago to
discuss struct clk and how we move forward with it. Several
of us had a follow up phone call last week, on Wed the 14th,
and what follows are meeting notes from this follow up
discussion (I've seem to have lost the notepad on which I
took notes during the LPC discussion).
Attendees:
Arnd Bergman
Stephen Boyd
Mark Brown
Thomas Gleixner
Shawn Guo
Saravana Kannan
Nicolas Pitre
Mike Turquette
Linus Walleij
Paul Walmsley
- Mike T.
- Remove set_parent and upstream clock arbitration
- Functionality is limited, but would work for OMAP as is
- We can add new features as needed
- Shawn
- Does not want to block new SoC support due to common clock
support not being ready. This is gating mx6 support being
merged upstream.
- Arnd
- Pre-req: Device Tree representation of struct clk before we
let these patches in.
- Mike T: Can we split the problems.
- Arnd: Yes?
- Thomas: Is not entirely separate.
Current patches are not enough to do actual representation of
building blocks. Need more than just the ops pointer structure.
Will look into what it would take the current patches and
then slowly add changes to building block.
- Stephen
- Want 1 tree lock instead of a framework lock:
+ Have 2 trees, 1 fast (1ms), 1 really slow (5ms)
- Thomas: Should be trivial to implement
- Once the traverse code is fixed, will be hard to change
the locking code, so need to get this right.
- Thomas: Need a clock tree base with its own lock and put
a struct clk tree into that tree base.
- Paul W.: How rate propagation would work across bases?
- Thomas: Should be doable with proper locking order, and
should not be too hard to solve. Non fast path.
- Can implement separate root clocks for right now
- tglx: separate struct clk from building block devices such
as frequency management.
- Paul W.
- Have multiple root clocks, but only one lock right now
- Figure out the scope of the patches to be.
Goal: get basic support for common struct clk upstream.
Get imx6 upstream w/o common struct clk
Has some bugs and races on existing code.
set_rate: Can parents be switched at same time?
operate under set_rate and set_parent as distinct operations.
clk_rate sample implementation: will only work in some limited
cases. Need to guarantee that clock is stable. There are lot
of hw specific that may not be possible to abstract.
Not true: "All parents are equal and should be treated the
same" during set_rate. From tglx: Looked at existing stuff
and noted that a lot of implementations share concepts...
walking a freq table or a divisor table. Paul: Agree,
can go into some sort of library code down the road.
in-tree vs out-of-tree: what's shipping on most device
is quite a bit more hacky and complex than what is
currently in mainline. Will need clock notifiers
before being able to use on real shipping devices.
Much code that calls clk_set_rate() assumes that code
will not be changed in the future.
Patch 1: good
Patch 2: needs a bit more thought
Patch 3: no comments yet
What do people think about having initial patches to convert
to using common clk struct but having sub-arch specific
set_rate and get_rate? General consensus: seems like a
good idea.
- Linus W.:
No major comments
- Nicolas:
Just listening in to understand the issues involved.
ACTION ITEMS:
- Thomas: will post updated patches by end of the week
- Others: Will post initial patches porting their platforms
to new patches as follow up.
- Deepak: Organize follow up call in two weeks and post to wider
audience to attend.
A group of us met at Linux Plumber's Conf two weeks ago to
discuss struct clk and how we move forward with it. Several
of us had a follow up phone call last week, on Wed the 14th,
and what follows are meeting notes from this follow up
discussion (I've seem to have lost the notepad on which I
took notes during the LPC discussion).
Attendees:
Arnd Bergman
Stephen Boyd
Mark Brown
Thomas Gleixner
Shawn Guo
Saravana Kannan
Nicolas Pitre
Mike Turquette
Linus Walleij
Paul Walmsley
- Mike T.
- Remove set_parent and upstream clock arbitration
- Functionality is limited, but would work for OMAP as is
- We can add new features as needed
- Shawn
- Does not want to block new SoC support due to common clock
support not being ready. This is gating mx6 support being
merged upstream.
- Arnd
- Pre-req: Device Tree representation of struct clk before we
let these patches in.
- Mike T: Can we split the problems.
- Arnd: Yes?
- Thomas: Is not entirely separate.
Current patches are not enough to do actual representation of
building blocks. Need more than just the ops pointer structure.
Will look into what it would take the current patches and
then slowly add changes to building block.
- Stephen
- Want 1 tree lock instead of a framework lock:
+ Have 2 trees, 1 fast (1ms), 1 really slow (5ms)
- Thomas: Should be trivial to implement
- Once the traverse code is fixed, will be hard to change
the locking code, so need to get this right.
- Thomas: Need a clock tree base with its own lock and put
a struct clk tree into that tree base.
- Paul W.: How rate propagation would work across bases?
- Thomas: Should be doable with proper locking order, and
should not be too hard to solve. Non fast path.
- Can implement separate root clocks for right now
- tglx: separate struct clk from building block devices such
as frequency management.
- Paul W.
- Have multiple root clocks, but only one lock right now
- Figure out the scope of the patches to be.
Goal: get basic support for common struct clk upstream.
Get imx6 upstream w/o common struct clk
Has some bugs and races on existing code.
set_rate: Can parents be switched at same time?
operate under set_rate and set_parent as distinct operations.
clk_rate sample implementation: will only work in some limited
cases. Need to guarantee that clock is stable. There are lot
of hw specific that may not be possible to abstract.
Not true: "All parents are equal and should be treated the
same" during set_rate. From tglx: Looked at existing stuff
and noted that a lot of implementations share concepts...
walking a freq table or a divisor table. Paul: Agree,
can go into some sort of library code down the road.
in-tree vs out-of-tree: what's shipping on most device
is quite a bit more hacky and complex than what is
currently in mainline. Will need clock notifiers
before being able to use on real shipping devices.
Much code that calls clk_set_rate() assumes that code
will not be changed in the future.
Patch 1: good
Patch 2: needs a bit more thought
Patch 3: no comments yet
What do people think about having initial patches to convert
to using common clk struct but having sub-arch specific
set_rate and get_rate? General consensus: seems like a
good idea.
- Linus W.:
No major comments
- Nicolas:
Just listening in to understand the issues involved.
ACTION ITEMS:
- Thomas: will post updated patches by end of the week
- Others: Will post initial patches porting their platforms
to new patches as follow up.
- Deepak: Organize follow up call in two weeks and post to wider
audience to attend.
=== Highlights ===
* Lots of 1:1s, went over reviews
* Facilitated struct_clk discussion (will post notes this week, I managed
to loose track of them)
* Thinking about post 11.11 tasks
* Had a good CI loop UI discussions with Asac, Paul L., David Z. and
need to continue looking into what's needed to make this meet both
needs of developers and from POV of showing off Linaro value-add.
* Good meeting with Grant to talk about DT tasks
=== Plans for This Week ===
* 1:1s, finish reviews
* Fill out my review/objectives
* Discuss kernel ci loop platform bug fixing/triaging with Scott
* Pick out and fill out two requirements in papyrs
* Various wiki updates
=== Upcoming Travel/Time Off ===
* Time off Sept 26th/27th. Will be available for KWG meeting.
* Travel to Arm Summit/ELC-E/LinuxCon-E Oct 22-28.
* Linaro Connect Oct 31st - Nov 4.
* Tentative bay area travel week of Nov 7th.
=== i.MX6Q DT ===
* Submitted the v2 of imx6q patch series with comments on v1 addressed,
and rebased to v3.1-rc6.
* Rebased imx6q patch series to Rob's latest GIC binding patches.
* Tested nico's zImage-dtb-append patch series on imx6q.
* Posted the patches for imx6q fec, gpio and usdhc initial support.
=== Misc ===
* Attended the conference call for clock framework discussion.
* Helped review imx cpuidle patches internally.
--
Regards,
Shawn
== Thomas Abraham <thomas-ab> ==
=== Highlights ===
* Submitted third version of i2c device tree patches. Ben Dooks has
merged these patches to his tree
* Submitted fourth version of pl330 dma controller dt support patches
with changes suggested by Grant.
* Submitted third version of dt support for Exynos4 keypad controller
driver with changes suggested by Grant.
* Submitted second version of patches to remove cfg_card helper
callback from sdhci-s3c driver.
* Tested Rob Herring's gic dt support patches with Exynos4 dt support.
Will require changes in Exynos4 interrupt mapping to use Rob's
patches.
=== Plans ===
* Submit dt support patches for sdhci driver with no dependency on
platform data.
* Submit single dt-enabled board file for smdkv310 board and origen boards.
* Rework Exynos4 interrupt mapping to use Rob's GIC dt support patches.
