== Per Forlin <perfor> ==
=== Highlights ===
* integrated mmc none blocking in our internal tree at ST-Ericsson to
be able to piggy back on testing.
* started to investigate (specs for next cycle) how reads propagates
from user space down to mmc framework through io-scheduler and block
layer.
=== Absence ===
Friday
=== Plans ===
* Specs for next cycle
* Refactor "mmc none blocking" based on feedback
== John Stultz <jstultz_vm> ==
=== Highlights ===
* Gave my Presentation "Android OS on Servers?" at ELC
* Got to have a number of really useful conversations at ELC
* Attended a meeting with Google Android developers
=== Plans ===
* Catching up from being out last week and summarize my ELC experience
* Get inject_sleep_time patch sent to lkml/google devs for review.
* Get Linaro Android kernel updated, tested and pushed out for consumption
* Try to respin and push out the Android trivial-tree mmc patches for discussion
=== Issues ===
* Not enough time! :)
[ACTIVITY] (Thomas Abraham) 2011-04-12 - 2011-04-18
== Thomas Abraham <thomas_ab> ==
=== Highlights ===
* Continued testing of kexec-reboot failure of smdkv310 board
- Does not seem to be a issue with L2 cache enabled.
- Found with jtag that uart stops transmitting data after L2 cache is
disabled and hence no more console display.
- Need further debugging on this.
* Not much progress with kexec-crashdump on s5pv310 when using slub allocator.
- Works fine when using slub allocator.
* Working on converting Samsung clock code to use common clock framework.
- works on s5pv310 platform but need to extend it to all other Samsung platforms
* Working on extending clkdev support to all other Samsung platforms.
* Submitted third version of clkdev patch series for s5pv310 platform.
- http://www.mail-archive.com/linux-samsung-soc@vger.kernel.org/msg04529.html
* Submitted bug fixes to to solve ethernet and sd/mmc card detection
issues on smdkv310 board.
=== Plans ===
* Complete clkdev and common clock support for all Samsung platforms.
=== Issues ===
* None.
== Conferences ==
* Talked at Linux Filesystem and Storage summit about flash drive problems
and how we can solve them
* Participated in the kernel developer panel at Collaboration Summit:
http://lwn.net/Articles/438155/
* Presented my research at Embedded Linux Conference
* Gave a keynote at Embedded Linux Conference
== Kernel ==
* Only few reviews this time, due to traveling:
u8500-hsem, spear pcie, sysfs-soc representation, mmc
== Other ==
* Vacation between the conferences
== Device Tree ==
* The make-dtbs patch has been merged by John Rigby on the
linux-linaro-natty tree.
* After a couple rounds of review given by Loic, the lmc dtb support
on branch lp:~shawnguo/linaro-image-tools/add-dtb has been
approved by Loic for merging lmc trunk.
== Misc ==
* Played Per Forlin's non-blocking mmc patch set on mx51 esdhc and
mx28 mxs-mmc, sent the patches of adding pre_req and post_req
support on these two controllers for reviewing.
* Reviewed the mxs gpmi device patch and a few i.mx consolidation
patches posted by Uwe.
--
Regards,
Shawn
== Paul E. McKenney <paulmck> ==
=== Highlights ==
* Working RCU priority boosting bug: found workaround that passes testing!!!
* Chasing the above bug suggested a number of possible simplifications, some of which might actually be worthwhile
* Sedat Dilek found a bug in a related patch, will reorganize tree to send him diagnostics along with that particular patch
* Found and fixed kexec bug in Power, but someone else beat me to it by one day
* Attended ELC: Google Android discussions and ARM maintainership issue
== Miscellaneous ==
* We have a couple of possible solutions for Lai Jiangshan's RCU_PREEMPT inlining patches
== Dave Martin <dmart> ==
=== Activity ===
==== hardware-n-kernel-standard-architecture ====
* Reposted VFP coredumps patch -- no further traffic.
* Investigated the status of ftrace with regard to Thumb-2.
The implementation appears already to be there and seems to
work, including dymamic ftrace.
==== miscellaneous ====
* Ongiong activity helping support linaro validation farm
activities (network, u8500 bringup). Got bootloader
reflashing working, but we don't have a working bootloader
image yet.
=== Plans ===
* Ping upstream re outstanding patches.
* Resume imx51 Thumb-2 investigations
=== Absences ===
* Apr 21 2011 - May 2 2011
---------- Forwarded message ----------
From: Dave Martin <dave.martin(a)linaro.org>
Date: Thu, Apr 14, 2011 at 5:01 PM
Subject: Re: Is VFP or Neon supported in kernel-side code? <EOM>
To: Tixy <tixy(a)yxit.co.uk>
> Is VFP or Neon supported in kernel-side code?
Currently, no.
There is some possible future work to see if we can enable this, to
allow acceleration of some number-crunching algorithms in the kernel,
but it's not implemented yet.
For now, I think it's reasonable simply to reject attempts to insert
kprobes on VFP/NEON instructions (as we must for cp15). Â I think that
these instructions are never used at the entry/exit points of
compiler-generated functions -- in any case, we build the kernel with
-msoft-float, so actually the compiler should never implicitly
generate these instructions at all.
Cheers
---Dave
On Thu, Apr 14, 2011 at 2:21 PM, Tixy <tixy(a)yxit.co.uk> wrote:
>
>
>
> _______________________________________________
> linaro-kernel mailing list
> linaro-kernel(a)lists.linaro.org
> http://lists.linaro.org/mailman/listinfo/linaro-kernel
>
On recent make, commit 1449fb96, "arm/dt: Add a make rule to build
dtb for enabled boards" breaks build of the kernel immediately:
/projects/linaro/linux-2.6/arch/arm/Makefile:292: *** mixed implicit
and normal rules. Stop.
> $ rpm -q make
> make-3.82-4.fc15.x86_64
>
> ''The previous behavior of mixing implicit and pattern rules in one line
> like that was never supported or documented, but due to a "hole" in the
> parser it worked for certain specific cases (for example, reversing that
> line so the pattern comes first would not work, nor would adding extra
> explicit targets after the pattern target).
>
> In 3.82, the parser was tightened in this respect and that "hole" was
> closed, hence you see this error.''
This patch splits the rule to fix the bug.
Reported-by: Andy Green <andy.green(a)linaro.org>
Signed-off-by: Grant Likely <grant.likely(a)secretlab.ca>
---
arch/arm/Makefile | 6 +++++-
1 files changed, 5 insertions(+), 1 deletions(-)
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 1be856a..4570ca7 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -289,7 +289,10 @@ zImage Image xipImage bootpImage uImage: vmlinux
zinstall uinstall install: vmlinux
$(Q)$(MAKE) $(build)=$(boot) MACHINE=$(MACHINE) $@
-dtbs %.dtb:
+%.dtb:
+ $(Q)$(MAKE) $(build)=$(boot) MACHINE=$(MACHINE) $(boot)/$@
+
+dtbs:
$(Q)$(MAKE) $(build)=$(boot) MACHINE=$(MACHINE) $(boot)/$@
# We use MRPROPER_FILES and CLEAN_FILES now
@@ -308,6 +311,7 @@ define archhelp
echo ' uImage - U-Boot wrapped zImage'
echo ' bootpImage - Combined zImage and initial RAM disk'
echo ' (supply initrd image via make variable INITRD=<path>)'
+ echo ' dtbs - Build device tree blobs for enabled boards'
echo ' install - Install uncompressed kernel'
echo ' zinstall - Install compressed kernel'
echo ' uinstall - Install U-Boot wrapped compressed kernel'
