Hey Nico,
So I've been seeing USB issues on my beagle xm board with the latest
linaro-2.6.38 tree, where when the system boots, lsusb will only show
the hubs, but no devices.
Andy reported seeing similar on his overo board, so I bisected the issue
down to:
9e64bb1e9f0613093b3e34ac5402fcfef0dcc35a is the first bad commit
commit 9e64bb1e9f0613093b3e34ac5402fcfef0dcc35a
Author: Keshava Munegowda <keshava_mgowda(a)ti.com>
Date: Tue Mar 1 20:08:19 2011 +0530
arm: omap: usb: Invoke usbhs core device initialization
The usbhs intialization is invoked by all omap3 and omap4
variant board files.
Signed-off-by: Keshava Munegowda <keshava_mgowda(a)ti.com>
Signed-off-by: Felipe Balbi <balbi(a)ti.com>
It looks like this patch is just converting the initialization over to
usbhs, so it doesn't really point out the specific issue, but it does
seem that something in the usbhs core isn't working properly.
Any tips here?
thanks
-john
The full bisect log is:
git bisect start
# good: [47dc59f22a6af00691a98975d05a8a1601714e1b] OMAP4: PandaBoard: remove unused power regulators
git bisect good 47dc59f22a6af00691a98975d05a8a1601714e1b
# bad: [8061f3a885ec3538bf405ff3957c205b1ab2aae4] mach-ux500: correct MMC/SDI parameters
git bisect bad 8061f3a885ec3538bf405ff3957c205b1ab2aae4
# bad: [1c49cb09b5666aa0d6df3b58f6c621df92c5dd44] Merge commit '05f6894' (omap-for-linus) into linaro-2.6.38
git bisect bad 1c49cb09b5666aa0d6df3b58f6c621df92c5dd44
# bad: [0cd3fd78e5951c1631ee60a2026603d63c671d98] Merge commit '9ced9f0' (rmk/devel-stable) into linaro-2.6.38
git bisect bad 0cd3fd78e5951c1631ee60a2026603d63c671d98
# bad: [23e0d1066f429ab44305e96fbff13f1793886277] usb: Refactor irq enabling out of usb_add_hcd()
git bisect bad 23e0d1066f429ab44305e96fbff13f1793886277
# good: [c9642374d0e969e8c17f4f31cd1a2bd111634227] USB: fix unsafe USB_SS_MAX_STREAMS() definition
git bisect good c9642374d0e969e8c17f4f31cd1a2bd111634227
# bad: [45d1b7ae205e39e95ec65747f8871661aaa105e4] usb-gadget: fix warning in ethernet
git bisect bad 45d1b7ae205e39e95ec65747f8871661aaa105e4
# good: [37db3af11f02c2ccdf44a788694da16062a0333c] usb: otg: TWL4030: Update the last_event variable.
git bisect good 37db3af11f02c2ccdf44a788694da16062a0333c
# good: [181b250cf53233a7a7c6d7e1e9df402506712e93] arm: omap: usb: create common enums and structures for ehci and ohci
git bisect good 181b250cf53233a7a7c6d7e1e9df402506712e93
# bad: [19403165c272cc4ed00c97973e7271714b009708] usb: host: omap: ehci and ohci simplification
git bisect bad 19403165c272cc4ed00c97973e7271714b009708
# good: [2236396d4d23828a0875a4d447103d0ab48aed0b] arm: omap: usb: usbhs core device initialization
git bisect good 2236396d4d23828a0875a4d447103d0ab48aed0b
# bad: [3b68ae73d8afa925807ebaae7eb14e2afd43f5b5] arm: omap: usb: cleanup ehci and ohci resources and devices
git bisect bad 3b68ae73d8afa925807ebaae7eb14e2afd43f5b5
# bad: [9e64bb1e9f0613093b3e34ac5402fcfef0dcc35a] arm: omap: usb: Invoke usbhs core device initialization
git bisect bad 9e64bb1e9f0613093b3e34ac5402fcfef0dcc35a
On Mon, Mar 21, 2011 at 03:25:16AM -0600, Grant Likely wrote:
> - Add packaging of .dtb files into linux-image-linaro-* packages.
> Loic and I discussed putting them under /lib/dtb/`uname -r`/, but
> thinking about it more, it might make more sense to share the modules
> directory and use /lib/modules/`uname -r`/dtbs. The dtc tool needed
As the .dtb files will be naturally generated in the same kernel
folder as kernel image sits, why do not we ship .dtb in the same
folder as kernel image /boot?
> to build the .dtbs is included with the kernel tree.
> - Add relevant dtb files to boot partition in linaro-image tools
The .dtb files will be generated and shipped with unique name, which
comes from .dts file name. But I intend to use the generic name,
maybe something like board.dtb along with l-m-c, just like we use
zImage and u-boot for all platforms in boot partition, so that l-m-c
does not need to encode platform specific dtb filename.
Thoughts?
--
Regards,
Shawn
Greetings,
We are pleased to announce that we now have a new blog engine
(WordPress, so much better than the previous one), and also a Planet
Linaro RSS aggregator.
See
http://www.linaro.org/linaro-blog/2011/03/31/wordpress-planet-linaro/
for details.
Many thanks to Ian Davenport for the blog upgrade and to James Westby
and Canonical IS for Planet Linaro.
Your comments and suggestions are most welcome, as these have just been
put on line.
Cheers,
Michael.
--
Michael Opdenacker - Community Manager
Linaro, http://linaro.org
Cell: +33 621 604 642
IRC: 'opm' in #linaro on irc.freenode.net
Hi,
Linaro is pleased to announce that the 11.05 Beta Ubuntu images are now
available to download.
After much blood, sweat and tears we now have a total of 10 different
boards supported (in our own unique hardware pack and board-neutral
rootfs architecture) along with a more focused 4 different images to try
out including the much coveted Ubuntu Unity interface on the Ubuntu Desktop
image. This is in addition to the small nano image, the tools rich
Developer image and the ARM Internet Platform (ALIP) image. A 2.6.38 kernel,
state-of-the-art Linaro toolchain and a whole host of ARM-related
improvements make for a thrilling release. What are you waiting for, go
download it now!
As always, if you have supported hardware, as found on:
http://releases.linaro.org/platform/linaro-n/hwpacks/beta/
please help our initiative by testing the official Linaro Evaluation
Build (LEB):
Ubuntu Desktop:
http://releases.linaro.org/platform/linaro-n/ubuntu-desktop/beta/
and our Developer images:
Nano:
http://releases.linaro.org/platform/linaro-n/nano/beta/
ALIP:
http://releases.linaro.org/platform/linaro-n/alip/beta/
Developer Tools:
http://releases.linaro.org/platform/linaro-n/developer/beta/
As a side note, hwpacks that have an -lt- in their name are outputs from
the Linaro Landing teams, using some of their components.
Make your way to:
http://wiki.linaro.org/Releases/MilestoneBuilds
for an explanation on how to test and submit your results to the QA
tracker at:
http://qatracker.linaro.org
For an explanation of how to use the qatracker please see:
https://wiki.linaro.org/QA/QATracker
Regards,
Jamie.
--
Linaro Release Manager | Platform Project Manager
This patch set is to add full platform dt clock support for mx51
babbage, based on Grant's devicetree/test tree. All mx51 non-dt
clocks in clock-mx51-mx53.c are translated to dt ones.
Regards,
Shawn
Changes since v1:
* unify the references to clocks that could be created statically
and dynamically, so that we do not have to duplicate functions
* use for_each_matching_node suggested by Grant to match clock
creating functions
* reserve 'data' of device_node structure and use new one 'clk'
* change macro ALLOC_CLK_LOOKUP and ADD_CLK_LOOKUP to functions
* fix codes that will break existing board support
* use clock provider 'src-clock' and 'dep-clock'
* fix up aips and spba node name to remove address info (suggestion
given by Grant on Jason's patch)
Shawn Guo (6):
arm/dts: babbage: add all available clock nodes
arm/mxc: add clk member 'rate' to ease dt fixed-clock support
dt: add new member 'clk' into device_node
arm/dt: mx51: dynamically add clocks per dt nodes
arm/dt: mx5: change timer init function to dt clock way
of/clock: eliminate function __of_clk_get_from_provider
arch/arm/boot/dts/babbage.dts | 501 +++++++++++-
arch/arm/mach-mx5/Makefile | 2 +-
arch/arm/mach-mx5/board-dt.c | 9 +-
arch/arm/mach-mx5/clock-dt.c | 52 --
arch/arm/mach-mx5/clock-mx51-mx53.c | 1433 +++++++++++++++++++++++++++++++-
arch/arm/plat-mxc/include/mach/clock.h | 4 +
drivers/of/clock.c | 23 +-
include/linux/of.h | 1 +
8 files changed, 1913 insertions(+), 112 deletions(-)
I have an Overo Tide that is capable of running at 720Mhz. It defaults
to 600Mhz, but the mpurate kernel command line argument allows it to
overridden.
The problem is that Linaro kernels (with the Linaro .config) always show
499.92 BogoMIPS in /proc/cpuinfo regardless of what I pass as the mpurate.
I've discovered that by disabling CONFIG_SMP (Linaro defaults this to
'y') the problem goes away.
I'm not sure how to go about fixing this, but its been noticed by the
Gumstix mailing list that Linaro builds for Overo seem sluggish.
Does anyone have any ideas what might cause this, or any ideas where I
might start looking?
FYI - the LP bug is: 739582
thanks,
-andy
