Hi,
Just tested the imx51 Thumb-2 kernel on efikamx here -- it boots fine,
but we only went as far as the initramfs.
Is anyone currently relying on CONFIG_LATENCYTOP? If so, we should
investigate what the incompatibility is...
Cheers
---Dave
On Wed, Feb 23, 2011 at 6:11 PM, Dave Martin <dave.martin(a)linaro.org> wrote:
> Hi there,
>
> Talking with Loic just now, we thought it could be worth turning on
> Thumb-2 in the main omap kernel package after all, since we've at
> least seen it work on a few different boards now. Â It feels ready for
> wider testing.
>
> Is it possible for you to move to a newer snapshot of the 2.6.38
> linaro tree, or will that be problematic at this stage?
>
>
> Since the imx51 kernel at least seems to build and boot in Thumb, it
> could be worth enabling a separate build in a PPA, but it's may not
> mature enough for everyone to use as the main kernel just yet. Â If you
> want to try this, you need CONFIG_ARM_UNWIND=y, CONFIG_LATENCYTOP=n
> and CONFIG_THUMB2_KERNEL=y
>
> Cheers
> ---Dave
>
== Aneesh V <aneeshv> ==
=== Highlights ===
== U-Boot SPL ==
* Debugged and fixed issues around SDRAM initialization for different
revisions of OMAP4
* More testing and cleanup
* Patches ready for submission - work complete for now
== U-Boot Thumb2 Build ==
* Built U-Boot for OMAP4 in Thumb2
* 23 % size reduction.
* No noticable change in boot-time
=== Plans ===
* Continue working on v2 of U-Boot cache series
* Work on dynamically selecting ARM/Thumb2 in Makefile based on
tool-chain capability
=== Issues ===
* None
== Device Tree ==
* Continue to comment the possible problem found in mx51 dt patch set
* Had some very basic code to prove that the dynamically creating and
registering 'struct clk' per dt clock node may work, but there are
still further details to work out, especially on the relationship
of 'struct clk'
* Needs help on dts specification, especially on clock binding
== Misc ==
* mxs-mmc was merged
* mxs-dma is ready for merging, Vinod will do once he gets to work
* Tested the esdhc write-protect and card-detect patch set
--
Regards,
Shawn
== Device Tree ==
* Modified s5pv310 uart driver to keep a copy of platform data and use
it. This was required for adding device tree support in uart driver
* Trying to make clock lookup (clk_get) independent of the id in platform_device
== Misc ==
* Will on vacation from 1st March to 6th March 2011 (not planned, but
have to take leave for personal reasons).
== Dave Martin <dmart> ==
=== Activity ===
==== hardware-n-kernel-standard-architecture ====
* Upstream link-time discard versus SMP_ON_UP problem has now
been fixed by a patch merged by Russell upstream.
* Still waiting for acks/tested-bys on omap3/4 Thumb-2
compatiblity patches. Will re-prompt people this week.
* Tried building the linaro kernel for iMX51. The resulting
kernel seems to work OK on mx51evk; also booted the same
kernel to a prompt on a Genesi efikamx (though without the
hardware-specific drivers etc.)
* Suggested to jcrigby to start building the linaro kernel in
Thumb-2 for OMAP, but he encountered a problem with module
loading. The issue may be a tools bug
<https://bugs.launchpad.net/binutils-linaro/+bug/725126>
* An external developer expressed interest in porting kprobes
to Thumb-2. Waiting for him to confirm whether to go ahead
with it.
==== miscellaneous ====
* Attended the Emdebian sprint in ARM Cambridge. Mostly I
wasn't directly involved, but it was a good opportunity to
meet Debian community guys.
* Konstantinos Margaritis expressed interest in getting a
Thumb-2 kernel on the Genesi platforms.
=== Plans ===
* Discuss prioritisation for outstanding kernel tasks.
* Write up the output from the Freescale i.MX BSP review
discussion, and post for comment.
=== Issues ===
* (none)
=== Absences ===
* Possible absence of a couple of days in early March
(unconfirmed, possibly around the weekend of the 12th).
== Jason Liu <Jasnliu> ==
=== Highlights ===
====DT====
* Check the review comments for the patch set: Add MX51 basic DT
support, prepare for the new version,
* Investigate to add new device support with MX51 babbage board such
as FEC, I2C, GPIO, MMC, etc.
====Kernel configuration ====
* Workaround the set value and unset issue by adding one config item
into the mx51.Kconfig file.
* Complete the full Kconfig fragment config file for mx51 babbage
which can ouput the same .config file with as mx51_defconfig.
====Misc====
* MX53 LOCO board support, v2 patch send out to uboot ML, no comments till now.
=== Plans ===
* DT support for mx51 babbage board and to add more device support
=== Issues ===
* How to handle the platform data by DT?
== Storage performance mmc async request ==
* Implemented test cases in mmc_test.c to compare synchronous and
asynchronous requests.
* Run IOZone test and block device test on newly upgraded mmci driver
to compare asynchronous request vs synchronous request.
* Created a wiki page to present results from MMC async requests here
https://wiki.linaro.org/WorkingGroups/KernelConsolidation/Specs/StoragePerf….
So far there are only results from one board but hopefully more will
come.
== Kernel development ==
* Continued multiple code reviews
* Bug 712175, partial analysis
* Discussed ST-Ericsson MCDE driver design
== Flash drive performance ==
* Discussed remapper design with Jörn Engel (logfs developer) and others
* More measurements, discovered a card that does log-structured writes
across multiple segments
* Helped other people start using flashbench
== Vacation ==
* Took one day off
== RCU priority boosting ==
* RCU priority boosting passes moderate testing
* Posted RCU priority boosting patches to LKML, good feedback and fixes
* Continuing testing
== Processes ==
* Completed a TSC license request.
== Misc ==
* Continued chasing additional Power bugs to enable tests on larger systems.
* Did some testing of the John's and Jason's kernel-configuration management work
