msm 7200A has: 1 apps processor, 1 modem
All the android code runs on the apps processor, so its a single core
system to the OS.
-- Mike
On Tue, Jun 22, 2010 at 9:59 PM, xlshe wrote:
> Hi,
>
> From the ADP2 specification, we know that the cpu in ADP2 is msm7200A.
> So I read the datasheet of
Two parts here, first part is for Frank, second is for the original
post (inline).
Applications:
Once all apps have released their wakelocks
App Framework:
Power manager service decides to suspend the device (if the user turns
off the screen)
by writing "mem" to /sys/power/state
Kernel:
early su
t
> know what it was up to at the time (whether it will misbehave
> connected via usb is another question...).
>
> "Build Number" = "voles-userdebug 2.2 FRFxx 3 test-keys"
>
> I did look in /data/dontpanic - the cupboard is bare
>
> Doug
>
> On Jun 1
Do you know if its a kernel reboot or a framework restart?
If its the kernel crashing you can get a persistent logs found:
For Nexus One:
You can check /proc/last_kmsg
On Droid:
/data/dontpanic/apanic_console
Make sure to check the timestamp of the file created or in the dmesg,
sometimes if the
On Droid we use the OMAP watchdog. I've found it useful to catch lockups in
the kernel suspend / resume path for drivers.
-- Mike
On May 4, 2010 10:18 AM, "Greg KH" wrote:
> On Tue, May 4, 2010 at 9:55 AM, Sudeep K N
> wrote:
>>
>>> Why is this a requirement for you to do, if you do not even kn
You can edit the .config file, then $ make oldconfig
-- Mike
On Wed, Mar 31, 2010 at 10:35 PM, xlshe wrote:
> Hi, all
> I need to configure the CPU frequency of Android phone. However, the
> current
> kernel doesn't support "userspace" governor. So I need to rebuild the
> kernel
> and turn on
essage "adbd cannot run as root in production builds". Can you
> tell me how to solve this problem? My phone is ADP2. Thank you very much~
>
>
> On Tue, Mar 23, 2010 at 3:22 PM, Mike Chan wrote:
>>
>> You need root access to change cpufreq values. I'm
s
> shows the problem that " Permission denied". Did you face similar problem
> when enable userspace governor? Would you please give me some advice on how
> to solve these problems? I will appreciate very very much for your help.
>
> On Mon, Mar 22, 2010 at 1:42 PM, Mike
Maybe try task_tgid_nr(current) ?
-- Mike
On Mon, Mar 22, 2010 at 7:27 PM, perumal316 wrote:
> Hi all,
>
> I have tried the following
>
> current->pid and current->comm to get the process ID and name of
> process from kernel module
>
> but what it is giving me is Binder Thread and an id which
What phone did you buy? If you're talking about the Nexus One the
userspace governor is already compiled into the kernel. Otherwise
you'll probably have to recompile the kernel with the userspace
governor.
We do DVFS using the ondemand governor currently.
To enable userspace governor:
adb shell e
ecursively.
>
> Signed-off-by: Chunqiu Wang
> Signed-off-by: Mike Chan
> ---
> arch/arm/plat-omap/include/plat/resource.h | 2 ++
> arch/arm/plat-omap/resource.c | 21 ++---
> 2 files changed, 12 insertions(+), 11 deletions(-)
>
> diff
What kernel are you on? Does this happen with you have a very large
number of wakelocks in the system?
-- Mike
On Thu, Mar 4, 2010 at 1:50 AM, Johnny Lin wrote:
> Hi Arve Hjønnevåg,
> I met a kernel panic that shows NULL pointer dereference while the PC
> is at wakelocks_read_proc function. Do
Currently we don't have CONFIG_FUSE_FS enabled so this is possible
unless you run a custom kernel. You will also have to setup the proper
permissions on the mount point.
-- Mike
On Mon, Mar 1, 2010 at 3:32 PM, Kabloc Vacvoa wrote:
> Hello there,
>
> I wonder know how to develop an application
nd the latest code in this git?
> Thanks,
> Arindam
>
> On Sat, Feb 13, 2010 at 1:34 AM, Mike Chan wrote:
>>
>> Take a look at
>> http://android.git.kernel.org/?p=kernel/common.git;a=summary
>>
>> -- Mike
>>
>>
>>
>> On Fri, Feb 12
Take a look at
http://android.git.kernel.org/?p=kernel/common.git;a=summary
-- Mike
On Fri, Feb 12, 2010 at 11:58 AM, scorproy wrote:
> Hello,
> I am trying to trace what all the changes have been made for Android,
> on top of the vanilla linux kernel.
> I am specifically looking what is the d
?
>
> On 14-Jan-2010 12:02am, Mike Chan wrote:
>> That message indicates that the device is going into sleep. State 0 is
>>
>>
>> on, I'm not sure what state 2 is, as android should only use states 0
>>
>>
>> and 3 (3 is suspend). This is bein
That message indicates that the device is going into sleep. State 0 is
on, I'm not sure what state 2 is, as android should only use states 0
and 3 (3 is suspend). This is being triggered by userspace so most
likely the framework is telling the kernel to sleep (probably screen
timeout).
What kernel
you should look at http://android.git.kernel.org/
check out the common tree, and the msm / omap tree if you want device specifics
-- Mike
On Thu, Dec 31, 2009 at 1:36 AM, noamtm wrote:
> For some unknown reason, I wasn't able to find information regarding
> the changes and configuration done
We are looking at going to 32 or 33 in the next few months
-- Mike
On Fri, Dec 11, 2009 at 2:10 AM, journeyer wrote:
> Hi, List
>
> I am working with Android-x86 project.
> The version of it is 2.6.29 (on donut).
>
> I installed android-x86(1.6 donut) on several PC and tested.
> And I found In
You are certain this is a valid speed you are writing?
cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies
Also make sure the speed you are writing to are within the
scaling_min_freq / scaling_max_freq limits.
-- Mike
On Wed, Dec 9, 2009 at 12:59 AM, tarek attia wrote:
> H
The actual power consumption of the device should lower if you do not
hold the wakelock, what you call (power save mode).
The reason why you see a sudden spike in the /proc/cpuload is because
when you wake up the device the cpuload timers are queued up so
CALC_LOAD is called muitple times for the
ot;Fix 2.6.29 build".
> Is the Android kernel upgraded to 2.6.29? Where can I get it?
>
> Thanks
> Weihua Wu
>
> >
>
--
MIke Chan
--~--~-~--~~~---~--~~
unsubscribe: android-kernel+unsubscr...@googlegroups.com
website: http://groups.google.com/group/android-kernel
-~--~~~~--~~--~--~---
I want to test rild function and it seems command "rild" can test it.
> But I don't know how to use the command.
>
> Any reference about how to use those commands?
>
> By the way, is it right to post here?
>
> Be
power.c in 2.6.25 but I could not find it in
> kernel2.6.27.
>
> Thanks
> Jackie (Weihua) Wu
>
> >
>
--
MIke Chan
--~--~-~--~~~---~--~~
unsubscribe: android-kernel+unsubscr...@googlegroups.com
website: http://groups.google.com/group/android-kernel
-~--~~~~--~~--~--~---
Hopefully very soon, the team is just getting back from the holiday season.
On Wed, Dec 24, 2008 at 10:56 PM, jinyan wrote:
>
> It seems the emulator cant boot current release still, right ? Is
> there any plan to push goldfish ?
>
> thanks
>
> On Nov 26, 4:48 am, "Mi
tialized)
>
> i think it will be efficient if i have a debugger
> rather than just seeing printk by dmesg
>
> thanks
>
> On Nov 22, 4:36 am, "Mike Chan" <[EMAIL PROTECTED]> wrote:
> > Perhaps you are referring to the msm_serial debugger? Whi
wrote:
>
> Don't see any commits of goldfish code in the android-2.6.27 branch of
> kernel/common.git, yet. Will the emulator be supported in 2.6.27?
>
> thanks,
> D
>
> >
>
--
MIke Chan
--~--~-~--~~~---~--~~
unsubscribe: [
hing about that ? If I
> file a bug, is it expected to be fixed ? Am I supposed to send the
> fix ? Would it be fixed. ? Another related question is: is the
> expected behavior of core applications specified ? Should it be ?
>
> >
>
--
MIke Chan
--~--~-~--~~---
ound other enhancements like alarm, ashmem, binder , etc
> > > but anyway i cannot find kernel debugger
> >
> > > and how should i do to do kernel debugging?
> > > just take qemu for example
> > > Android has removed the -S option which freezes CPU a
e new kernel ?
> - I would like to do this port as a final project for my OS classes,
> so the deadline is short (two weeks). Should I take the changes from
> 2.6.25 and create my own port for 2.6.27, so that the rest of the
> system still works ?
>
> Thanks,
>
>
30 matches
Mail list logo
