Release 3.2 kernel issue tracking thread


  • Charging via USB: Works for me.

Seemed to work for me as well at first, but now it's just stopping intermittently. Charged from 44% to 56%, then stopped. I unplugged the USB cable, and after putting it back in, the system charged to 67% and has stopped again. Now it just charges for a couple of minutes after I replace the USB cable and stays around 67%.
 
Experienced the unreliable LED status again: Device turned into low power mode (with about 80% battery capacity). At first only the Power-LED shone green. But then the Charging-LED shone too, allthough the device was not charging, nor with critical battery capacity (< 5%).
 
A revised (added some error checking and fixed some formatting) version of the patch is on the linux-bluetooth mailing list now. We'll see what they have to say about it... (PS. Its way too hard to behave correctly on mailing lists with gmail... Forgot to click on text only? You must be a spammer!)
 
I found screen blanking (both spontaneous and due to time change) causes system freeze (3.2 kernel of beta 2). Tested a few times swithing between standard (everything ok) and experimental kernel (not ok)
No such problem here, do you have anything running when this happens?

nope - I have experimented a bit more and it doesn't happen always. I just tried it again rebooting into 3.2


-> LCD blanking on -> set time 1 hour forward -> everythin ok, then I tried rebooting into 3.2 after


pulling the battery out, then -> set time 1 hour forward -> system freezes (with screen on)... :huh:
 
Seems my mmc errors are not caused by suspend, happen with some I/O load all the time:



Code:
[ 1381.084930] mmcblk0: error -84 transferring data, sector 9332712, nr 8, cmd response 0x900, card status 0xb00

[ 1381.084991] mmcblk0: retrying using single block read
 
Those happen when you transfer using both SD cards at the same time - i think it is the same multiblock errata that we had with 2.6.27, but the 3.2 code attempts to retry using single block read and succeeds (thats why there's no follow-on error for the single block read error sector).
 
Those happen when you transfer using both SD cards at the same time - i think it is the same multiblock errata that we had with 2.6.27, but the 3.2 code attempts to retry using single block read and succeeds (thats why there's no follow-on error for the single block read error sector).
I have just one SD card active. If it's non-critical, I'm not particularly bothered though.
 
It has never happened for me with only one SD card attached IIRC. (if it does that then it is notable.) I just remember that error from when I tested some new kernel over a year ago. Back then the old 2.6.27 would I/O error on SD-SD copies and new kernel (still older than 3.2 IIRC... ) would just spew errors but still manage to do the job. I guess it is on the long list of things to properly fix...
 
The LED unreliability has its harms...


WiFi definitely is always on after booting with kernel 3.x, but not necessarily the WiFi LED! Checked with uname -a, to be sure I booted with 3.x.


If you are in MiniMenu you don't notice the actual WiFi activity, only in XFCE the WiFi-Status icon in the bottom bar reminds you, that it's on!
 
During my test (which particular keys wake the Pandora from low power mode), once after my Pandora's wake-up ended up in CLI mode, with this error display:


[ 1438.094879] Could not enter target state in pm_suspend


[16115.086364] Could not enter target state in pm_suspend


By pressing <Ctrl><Alt><F7> I luckily got back to the GUI again, but it was broken:

  • Closing the lid had no effect
  • Power slider had no effect
  • None of my 2 SD cards was mounted
  • Thus had to save my dmesg debug log to NAND


Did a hard reset (Pandora+PowerSwitch). Log attached.
 

Attachments

  • ring-buffer.txt
    15.3 KB · Views: 150
Last edited by a moderator:
SD Mass Storage: Sudden unexpected disconnection of mounted SD partition


Had my FAT32 mounted to my Mac and ran rsync (as dozen times before without problems).


After about 2 minutes it disconnected from my Mac. Pandora system was frozen (time counter stuck).


Only hard reset helped.
 
Next attempt with SD Mass Storage:

  • Hosted to Mac, finished rsync without problems.
  • Unmounted from Mac. No problems on Mac or Pandora (pointer reacts normal)
  • In the moment I end the mass storage mode on the Pandora by "OK", its screen freezes (pointer + clock stuck).
  • Interestingly on the Mac in the System Profiler I still see the USB device, also minutes later after the Pandora system freeze, as:


CDC Composite Gadget:


Product ID: 0xa4aa


Vendor ID: 0x0525 (PLX Technology, Inc.)


Version: 3.16


Speed: Up to 480 Mb/sec


Manufacturer: Linux 3.2.1-00104-g5983d59 with musb-hdrc


Location ID: 0xfd100000 / 4


Current Available (mA): 500


Current Required (mA): 2

  • As soon as I unplug the USB cable, it disappears from Mac's System Profiler.
 
Which input devices wake the Pandora from low power mode?


As of SuperZaxxon, Release 1.5, Beta 2.


I tried to determine this by trial and error. These are my findings:


Cause no wake event: A/B/X/Y, ArrowKeys, Start, Select, L-shoulder, L+R nub


Cause wake event: A-Z, 0-9, Fn, Pandora, Shift, Enter, Backspace, Space


Where is this defined?


Is there only a technical implementation/definition, or also a human interface guideline behind this?
 
USB charging definitely does not work on my unit with the new kernel. Plugged USB in at 20%, charge led on. Came back a couple of hours later just in time to see the system shutting down because of no power.


Now charging up from 0% with the 2.6 kernel; seems fine up to now.


[edit] strike that, similar problem with old kernel on SuperZaxxon. I'll try a previous firmware and see how it behaves there.
 
Last edited by a moderator:
^ Couple of hours? From 20% with the screen off and idle my system would have 18 hours of runtime (with the 3.2 kernel we're talking about). In reality some hours less because it doesnt start shutdown at 0%, but some nonzero percentage (cant remember)...


To the point, there is some problem with my current bluetooth setup. Things work for a while (sometimes like for 8 hours), but eventually it seems to error out (to me that seems like the serial receive goes out of sync, but I'm just guessing) (and connections just freeze until I reset the bt, then its fine again): there's some log plus hciconfig after catching that error list: http://urjaman.dyndn...fo/bt-error.txt


EDIT: I'm thinking that this might have something to do with this: http://git.openpandora.org/cgi-bin/gitweb.cgi?p=pandora-kernel.git;a=commit;h=000ff8eac2cd7a08235c24e38118c9a053c479a5 . So I added "echo 0 > /sys/devices/platform/omap/omap_uart.0/sleep_timeout" to my btup script, and now I'm just seeing if this happens again.
 
Last edited by a moderator:
^ Couple of hours? From 20% with the screen off and idle my system would have 18 hours of runtime (with the 3.2 kernel we're talking about). In reality some hours less because it doesnt start shutdown at 0%, but some nonzero percentage (cant remember)...
System was under load (trying to bootstrap a current gcc, but it turns out the cdevtools pnd doesn't work as a base for that because of library differences - different topic ;) ...).
 
To the point, there is some problem with my current bluetooth setup. Things work for a while (sometimes like for 8 hours), but eventually it seems to error out (to me that seems like the serial receive goes out of sync, but I'm just guessing) (and connections just freeze until I reset the bt, then its fine again): there's some log plus hciconfig after catching that error list: http://urjaman.dyndn...fo/bt-error.txt


EDIT: I'm thinking that this might have something to do with this: http://git.openpando...118c9a053c479a5 . So I added "echo 0 > /sys/devices/platform/omap/omap_uart.0/sleep_timeout" to my btup script, and now I'm just seeing if this happens again.
It's a shame if that doesn't work, BT chip has full flow control connected and should be able to wake the OMAP when needed without data loss by asserting CTS/RTS. Maybe it needs tcgetattr() call, setting CRTSCTS flag to enable flow control, and then tcsetattr() (would be good to first check if CRTSCTS is set by default).
 
Last edited by a moderator:
I have Super Zaxxon Beta 2 on my Pandora, and if i run the normal Kernel, and try to shutdown the unit, the screen seems to stuck on the shutdown screen and the pandora ditnt seems to gos out, i have to reset it and made a forced shutdown via the boot menü ..


EDIT: This also happens whit the new kernel, but wifi seems to work, and somethimes it works but the wifi led isnt on.., tryed it whit the webradio app.., wifi dit work, even if i the led ditnt shine..
 
Last edited by a moderator:
Ok, I want to have a long idle time and so I tested SuperZaxxon with Kernel 3 but without luck.


Whenever the Pandora gets idle, it hangs.


I have tried the minimal case:


- no sd cards


- no usb


- no wlan


- just booted kernel 3 with xfce


- started top


- started an while/dmesg|tailloop


- waited


- freezed pandora


The last I could see in top was 100 processes in sleep staed and 1 running. 12 Minutes uptime.


When the Pandora hangs no key has any effect, tried to switch to console, power button, Pandora-Button. I have to remove battery to get it going again.


Any ideas on how to debug this?


greetings,


sirtobi
 
Did you re-downloaded and re-flashed ?


May sounds a bit heavy, but it's necessary to be 100% accurate.


What was your battery level ?


Pandora button + power button didn't rebooted ?
 
Back
Top