Re: [Announce] kernel-power unstable v49
 

nono
is stop bme then modprobe bq27x00, device stable, battery percentage can read.
The reboot is happen when rmmod and then start bme, after few mins, the phone reboot itself

Re: [Announce] kernel-power unstable v49
 

@pali

because I not too expert....I think I cant try that as you say is danger
the only thing I can do now is, when I go to charge after start bme, reboot the phone myself to prevent it reboot self

Re: [Announce] kernel-power unstable v49
 

@pali

you want this output?

Re: [Announce] kernel-power unstable v49
 

Yes, thanks. This is usefull. In log I found this error message:

So there is really problem with power_supply or bq27x00_battery module.

Can you test if this only happen if you rmmod bq27x00_battery module? And can you test if that happen too if you do not start bme (after rmmoding module)?

Re: [Announce] kernel-power unstable v49
 

any chance to bring the front camera to work with skype call again?

Re: [Announce] kernel-power unstable v49
 

@pali, I just finished the testing as you mentioned

1. If keeping using bq27x00, the phone never reboot self
2. When I change from using bq27x00 to BME, rmmod and start bme, the phone will reboot within 5 mins
3. If I only rmmod WITHOUT start BME, the phone still reboot within 5 mins
4. Last thing, If I rmmod and start BME, then I reboot myself, after the phone is finished booted up, fault message still pop up (An unexpected reboot has occurred, kernel cant be load)

Re: [Announce] kernel-power unstable v49
 

So If I understand correctly, problem is after module is unloaded. I will try to find problem

Re: [Announce] kernel-power unstable v49
 

yes, shoud be
but how come no one mentioned here....they got no problem at all@@?

Re: [Announce] kernel-power unstable v49
 

Now I'm testing bq27x00_battery module and I have same problem. It reboots device after time time after module is unloaded...

Re: [Announce] kernel-power unstable v49
 

Yeah! I discover a bug, you better thanks me, LOL

Re: [Announce] kernel-power unstable v49
 

maybe there was a bug in bq27x00-battery module earlier, and it just now popped up? perhaps it caused trouble before not only due to BME itself? i'm not an expert in drivers, but i hope that it may be long not discovered bug that when recovered from, it can help with something recently meant as impossible due to conflict.
(maybe it's one of reasons battery related crashes or unwanted scheduled shutdowns?)

PS: i don't touched bq module since i updated to v49, but general tasks of the phone works, hostmode works usb-networking too, compache works, so if you want me to test something another that those troubled battery module, maybe point me a feature to test?

Re: [Announce] kernel-power unstable v49
 

@pali: Did you disable compcache notifications in your 20111006 build? I wonder because:
- /dev/ramzswap0 creation and initialization works every time now, repeatable, which it never did on my device
- 'rzscontrol /dev/ramzswap0 -s' shows a value of '0' for every statistic given

I'm asking that caused by the empty statistics, also that freemangordon mentioned issues with notifications enabled quite some time ago (that could have been the root cause of my shutdowns observed with all previous v48 and v49 builds)

Also, if I have compcache enabled (96MB) I am able to open microB, Opera, Wordpress client, Evince, moNav, and several other applications.

But, if nothing besides conky is running, and I start Frogatto, it loads, then during the main menu some slight pulseaudio stuttering can be observed. If I load a level and start playing, it gets worse and the control input lags, up to the point of overall heavy frame drops and complete loss of audio.

When at this point I try to switch to the task manager using CTRL+BCKSP, the device almost freezes, inputs take at least 20 seconds up to several minutes to be registered, and from the minimized conky I can barely see that CPU usage is almost(?) 100%, on 805MHz in my case. General load is equally high. Only if I manage to close the game, after some while it gets back to normal behavior.

IIRC freemangordon wrote something about swap trashing; it looks like it...

Re: [Announce] kernel-power unstable v49
 

hey i experienced semi-random crashes during usb-networking
it happens if much of data is transferred at once, but not everytime.

i experienced it with kp48 too, but in kp49 it seems to trigger more frequent (or at least i wasn't so crazy to transfer nnn MB data at once, but i have done now)

to trigger that you have to transfer more than 100mb data in one data stream, like massive app install or huge up/download, it happens when one multifile transfer too.

i haven't checked but it may affect general networking not only usb ones (i have no wifi in home, so i have to use usb networking or to hunt for some not secured wifi in another place)

if it seems unconnected with kernel power, then sorry, but i feel i may have found a bug in KP, but if its really not it please point to where i have to search help.

notice that it doesn't happen when transferring huge amount of data while normal operation (ie. much of small chunks of data with many multi-milisecond time-breaks)

EDIT: this time i tried to transfer 500MB at once and 250MB at once, both failed once, and one recovered from filepart succesfully and one not

Re: [Announce] kernel-power unstable v49
 

Sorry for quoting me... but anyone?

Re: [Announce] kernel-power unstable v49
 

So which files do I download to install 49?

[ ] kernel-power-bootimg_2.6.28-10power49_armel.deb
[ ] kernel-power-debug_2.6.28-10power49_armel.deb
[ ] kernel-power-flasher_2.6.28-10power49_armel.deb
[ ] kernel-power-headers_2.6.28-10power49_armel.deb
[ ] kernel-power-modules-debug_2.6.28-10power49_armel.deb
[ ] kernel-power-modules_2.6.28-10power49_armel.deb
[ ] kernel-power-source_2.6.28-10power49_all.deb
[ ] kernel-power_2.6.28-10power49_armel.deb 06-Oct-2011
[ ] linux-kernel-power-headers_2.6.28-10power49_armel.deb

dpkg -i *.deb?

Re: [Announce] kernel-power unstable v49
 

kernel-power_2.6.28-10power49_armel.deb
kernel-power-modules_2.6.28-10power49_armel.deb
kernel-power-flasher_2.6.28-10power49_armel.deb
kernel-power-bootimg_2.6.28-10power49_armel.deb (for multiboot)

Re: [Announce] kernel-power unstable v49
 

Thanks, done.

Re: [Announce] kernel-power unstable v49
 

Is order important?

Re: [Announce] kernel-power unstable v49
 

yes, of course and also you must install each package separatly.

Re: [Announce] kernel-power unstable v49
 

So it's kernel, then modules, then flasher?
Thanks

Re: [Announce] kernel-power unstable v49
 

Correct.

ten chars

Re: [Announce] kernel-power unstable v49
 

So, got some time and decided to give it a go again. After the reinstall, everything worked fine, including Skype. Installed KPv48 via appmanager, rebooted (verified that I was running KP, it was), running ok. The moment I start Skype, my device gets laggy again until I switch to offline (which by itself takes 5 minutes). Anyone having any thoughts? Thanks!

Re: [Announce] kernel-power unstable v49
 

There is nilfs2 support mentioned in changelog but it's not compiled in and there is no module. Both following commands return nothing:
Also attempt to mount a nilfs filesystem fails. Am I doing something wrong?

The kernel installed is kp49. I'm using it about half a day with compcache enabled, no bugs so far.

BTW, thanks, pali, for the great work!

Re: [Announce] kernel-power unstable v49
 

i have a problem after installing kernel 49 , skype calls stoped working , when i try to call any 1 i don't get ring for some time then i get General Error msg
thx for ur hard work pali and i hope u help me with this problem :)

Re: [Announce] kernel-power unstable v49
 

Confirming above bug (I don't use Skype, but tested just out of curiosity).

Re: [Announce] kernel-power unstable v49
 

Hi,

Sorry, I can't say if the KP v49 compiled by pali is the source of your problem, but the problem doesn't occur on my side, running KP v49 I have compiled myself, before pali come back to update v48 to v49.

I have just finished today a 37 minutes approx Skype call + switching video on during the session without any problems ...

If you want to check my version, you can find it here : My v49

Here is the corresponding changelog.

Hope it help ...

A++

Re: [Announce] kernel-power unstable v49
 

am on kp49, and skype works perfectly, so ur problem could b something else.

Re: [Announce] kernel-power unstable v49
 

To be honest, there are so many versions of kp49 floating around, that I may have lost orientation. We're going to get most complete one into extras-devel jsut after freemangordon conclude his work on 720 recording related things, yes?

Also, any progress on promoting *any* version of kp (for example, kp47) to extras-testing (and later, to extras), so we can use -devel in line with it's purpose, instead of development versions scattered all around servers? Not demanding anything, just asking.

/Estel

Re: [Announce] kernel-power unstable v49
 

Pali, I know this isn't ideal, but at this point I recommend you stop fighting with the repository and instead upload new kernel-power versions with a different internal package name. I.E. "kernel-power2-[whatever]" instead of "kernel-power-[whatever]".

Yeah it's an ugly work around and I know we both personally dislike those, but it's better to be able to promote your packages correctly and not have to deal with them getting stuck 'invisible' to the package interface or whatever. Starting with a new package name should get you a proper fresh start - or, at the worst, indicate that the problem is with your package somewhere, though I doubt that's the case.

Re: [Announce] kernel-power unstable v49
 

so thx
i reinstalled kernel and it's working now
idk where was the problem , but thx for help :)

Re: [Announce] kernel-power unstable v49
 

Hi to all,I just want to ask is solved problem with media player after install kernel-power v49

Re: [Announce] kernel-power unstable v49
 

With the latest kernel-power v49, sometime my device get stuck at 550 mhz despite the kernel configuration.

And if I use the ondemand governor, some processes like kondemand/0, events/0 appear at the top cpu time.

For both problems, a reboot fix it but it appears again sometime later
.
Anybody experienced the above issues?

Re: [Announce] kernel-power unstable v49
 

Same here... but the frequency is 500 Mhz.
Usually it happens after a recharge.

After this, very often the battery drain increases about 3 times until next shutdown.

I have verified everything, but also creating a dbus script executed at the end of charge to fix the min max limits of cpu clock the issue remains.
Often, in locked mode i can find:
min=125
max=250
current frequency=500 ?!?! :eek:

Re: [Announce] kernel-power unstable v49
 

I've got a similar issue on kp48, haven't tried kp49 yet.
I use Qt Mobile Hotspot for routing internet through USB. Some weeks ago, the reboot occurred many times when my N900 was downloading data at once through the USB connection, the N900 froze and rebooted although downloading only some megs. I reflashed the N900 before that. I can use WinSCP without a problem.

I've seen it a couple of times when a website starts to load on PC, the N900 reboots. But it has never rebooted when downloading files even larger than 500MB on PC. Neither has it rebooted when listening to internet radio on N900. It must be the USB networking causing it. When it does reboot, the boot reason is always sw_rst. I didn't get it to reboot on a Wifi network.

I don't know if this is related, but at the time I have reflashed the N900 and installed kp48 with FAM. Somehow the BFS kernel got installed too since I saw it as the first entry in Multiboot. I had no idea what it was and, of course, booted into the kernel. I was then amazed how can it be so blazing fast without overclock. Being doubtful I checked "kernel-config show" and what the hell, it was at 1.15GHz. I loaded my custom 805MHz profile and after some minutes the N900 rebooted. Then it was back at 1.15GHz and the boot reason was 32wd_to. I immediately reflashed the device and installed kp48 without any problem. I know that's strange. I never wanted to overclock higher than 805MHz. Too bad I didn't take a screenshot of the voltages. I have a feeling the reboot problem started after that, although I may be wrong. I was just worried that the extreme overclock would've caused it. This guy seems to have a similar problem: http://talk.maemo.org/showthread.php?t=79350

Ok so I'm not alone with this USB networking problem, what can be done?

majaczek, does your N900 still reboot on its own?

Any help is appreciated.

Re: [Announce] kernel-power unstable v49
 

Finally problem is fixed, thanks to freemangordon :)

New builds with this fix + new dsp bridge driver by freemangordon is on standard place.

Re: [Announce] kernel-power unstable v49
 

Wow new update! Any updates with the u-boot versions pali?

Re: [Announce] kernel-power unstable v49
 

Just tried out the new kernel. Now device constantly reboots on default profile.
Anybody experienced the same?

Edit: Now won't boot. Corrupted Optfs. Cannot mount. What should I do now?

Re: [Announce] kernel-power unstable v49
 

diff --git a/kernel-power-settings/config/default b/kernel-power-settings/config/default

index 68b8160..e369af5 100644 (file)


--- a/kernel-power-settings/config/default
+++ b/kernel-power-settings/config/default
@@ -4,9 +4,9 @@ MINFREQ=250
# maximum frequency to use
MAXFREQ=600
# list of frequency configurations: each "frequency:volt,dsprate"
-FREQS="0:30,90 125:30,90 250:38,180 500:48,360 550:54,400 600:60,430 700:60,430 750:60,430 805:60,430 850:60,500 900:60,500 950:60,500 1000:60,500 1100:72,520 1150:72,520"
-SMARTREFLEX_VDD1=0
-SMARTREFLEX_VDD2=0
+FREQS="0:30,90 125:30,90 250:38,180 500:48,360 550:54,400 600:60,430 720:60,430 805:60,430 850:60,500 900:60,500 950:60,500 1000:60,500 1100:72,520 1150:72,520"
+SMARTREFLEX_VDD1=1
+SMARTREFLEX_VDD2=1
GOVERNOR=ondemand
UP_THRESHOLD=95
SAMPLING_RATE=300000

You could see there the default values of kp-settings was altered I'm not sure yet but most probably it's a case to case problem. Maybe you could try loading lv profile instead? Hmmm wonder why they changed the default values :confused: I think default should always be default nomatter what and any changes should be applied seperately

Re: [Announce] kernel-power unstable v49
 

I was unable to boot. And everything messed up now. After trying to repair the optfs file system, it still won't boot. Now I officially lost /home. And I didn't back things up. FML.


Can anyone backup the rootfs and optfs with Backupmenu and share me the files? Thank you.

Re: [Announce] kernel-power unstable v49
 

I don't have any backups on my n900 as I do a full reflash if I mess up anything so sorry can't give you any files from backupmenu... But how did optfs got included with the problem? As I can see in the diff from the oct 12 version to the nov 1 version only bq module & dsp dsp codes were modified and yeah + some from kp-settings which I presumed caused your reboots.