Page 67 of 74 FirstFirst ... 17576263646566676869707172 ... LastLast
Results 1,321 to 1,340 of 1472
Like Tree4Likes
  1. #1321  
    has anybody else had random shutdowns while the phone is asleep with the new f-105? i dont know whats going on. is it possible that dropping the voltage too low on 500 could cause this? i dont see any rhyme or reason to it.
    thanks
  2. #1322  
    Quote Originally Posted by graffix31 View Post
    has anybody else had random shutdowns while the phone is asleep with the new f-105? i dont know whats going on. is it possible that dropping the voltage too low on 500 could cause this? i dont see any rhyme or reason to it.
    thanks
    I've had the shutdowns while on the Touchstone and sometimes when waking the phone up. This has occurred with with AV8B, F104A but not with F105 (running great for days now). I am not able to test SR71 as in the past crashed the phone (Not a Chuck Norris).
    Commodore Pet, Commodore 64&128, Amiga 4000, PC-7, Ubuntu, Treo 650. Treo 755p,Sprint Palm Pre Plus 2.2.4,2 Sprint Pre2 2.2.4, TouchPad 32gig.
  3. carrel's Avatar
    Posts
    425 Posts
    Global Posts
    426 Global Posts
    #1323  
    Psycho, my Pre- is running F104 and just locked up. I am reading up on how I can do better debug/analysis of what happened cuz right now I don't have much to provide. But I also was noticed that you recently made some locking changes to AV8B and F105. Are there any of those locking changes that might be needed in F104 too?? Just a quick thought. I'll see about getting better info for any future lockups.

    FWIW, I was testting some swap changes too so that may be an issue.
  4.    #1324  
    Quote Originally Posted by carrel View Post
    Psycho, my Pre- is running F104 and just locked up. I am reading up on how I can do better debug/analysis of what happened cuz right now I don't have much to provide. But I also was noticed that you recently made some locking changes to AV8B and F105. Are there any of those locking changes that might be needed in F104 too?? Just a quick thought. I'll see about getting better info for any future lockups.

    FWIW, I was testting some swap changes too so that may be an issue.
    The locking changes in the kernel have nothing to do with normal operation, it is for when Govnah sends sysfs changes at a rapid pace.

    If you feel there is a kernel problem then send detailed logs and how you can reproduce the lockup or crash.
    Live free or DIE!
  5. carrel's Avatar
    Posts
    425 Posts
    Global Posts
    426 Global Posts
    #1325  
    Quote Originally Posted by unixpsycho View Post
    The locking changes in the kernel have nothing to do with normal operation, it is for when Govnah sends sysfs changes at a rapid pace.

    If you feel there is a kernel problem then send detailed logs and how you can reproduce the lockup or crash.
    OK, thanks.

    I'm looking around to figure out what to do if there is a next time. It locked up solid mid key press while running an app. Screen was on with the water-drop shadow of the key-press left showing. Network and novacom were all dead.
  6. #1326  
    Quote Originally Posted by carrel View Post
    OK, thanks.

    I'm looking around to figure out what to do if there is a next time. It locked up solid mid key press while running an app. Screen was on with the water-drop shadow of the key-press left showing. Network and novacom were all dead.
    I've also experienced random freezes requiring battery pull. Seems to have been happening since the charger polling changes were made to F105. When it happens my phone also gets really hot...

    Really not sure what "detailed logs" means, any help on that would be appreciated.
  7.    #1327  
    Try lowering the Vdemand poll rate to 0.1 Sec and see if it still hangs.

    Post the contents of /var/palm/event.d/org.webosinternals.govnah-settings and org.webosinternals.govnah-compcache
    Live free or DIE!
  8. #1328  
    I changed the vdemand polling rate to 0.1 today. No lockups since.

    Here are the contents for the files you specified:

    org.webosinternals.govnah
    description "Govnah Service"

    start on stopped finish

    exec /var/usr/sbin/org.webosinternals.govnah

    respawn
    org.webosinternals.govnah-settings
    description "Govnah Settings"

    start on stopped finish

    script

    [ "`/usr/bin/lunaprop -m com.palm.properties.prevBootPanicked`" = "false" ] || exit 0
    [ "`/usr/bin/lunaprop -m com.palm.properties.prevShutdownClean`" = "true" ] || exit 0
    [ "`/usr/bin/lunaprop -m -n com.palm.system last_umount_clean`" = "true" ] || exit 0

    echo -n 'screenstate-v2' > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
    echo -n '500000' > /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq
    echo -n '1005000' > /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq
    echo -n '2' > /sys/devices/system/cpu/cpu0/cpufreq/screenstate-v2/vdemand_factor
    echo -n '1' > /sys/devices/system/cpu/cpu0/cpufreq/screenstate-v2/vdemand_enable
    echo -n '0' > /sys/devices/system/cpu/cpu0/cpufreq/screenstate-v2/charger_override
    echo -n '60' > /sys/devices/system/cpu/cpu0/cpufreq/screenstate-v2/charger_poll_rate
    echo -n '100' > /sys/devices/system/cpu/cpu0/cpufreq/screenstate-v2/vdemand_poll_rate
    echo -n '63 67 65 45 34' > /sys/devices/system/cpu/cpu0/cpufreq/override/vdd1_vsel
    echo -n '47' > /sys/devices/system/cpu/cpu0/cpufreq/override/vdd2_vsel
    echo -n '50' > /sys/devices/system/cpu/cpu0/cpufreq/override/cpu_hightemp_alarm
    echo -n '45' > /sys/devices/system/cpu/cpu0/cpufreq/override/cpu_hightemp_reset
    echo -n '20' > /sys/devices/system/cpu/cpu0/cpufreq/override/battery_scaleback_percent
    echo -n '500000' > /sys/devices/system/cpu/cpu0/cpufreq/override/battery_scaleback_speed

    end script
    Also, i've disabled compcache completely using xanadu73's method from this post

    I'm going to wait a few days and see if the vdemand change fixes this. If not i will bump up my 500/1005 voltages by 1 point and see if that makes a difference. I've tested both 30 and 59 to be stable, so with a medium demand it should be ok. <shrug>
  9.    #1329  
    Quote Originally Posted by NickVTPre View Post
    I changed the vdemand polling rate to 0.1 today. No lockups since.

    Here are the contents for the files you specified:

    org.webosinternals.govnah


    org.webosinternals.govnah-settings


    Also, i've disabled compcache completely using xanadu73's method from this post

    I'm going to wait a few days and see if the vdemand change fixes this. If not i will bump up my 500/1005 voltages by 1 point and see if that makes a difference. I've tested both 30 and 59 to be stable, so with a medium demand it should be ok. <shrug>
    The voltage for 500MHz is extremely low. Also the voltage for 1GHz is low. Considering it does vdemand on that base value, I can see why it might hang sometimes. Increasing the polling will make it ramp up quicker, which is why you havent crashed yet.

    Keep in mind when you set the base voltages with using vdemand, it will scale down voltage as much as -8 depending on factor. So always choose a base voltage that is stable when using vdemand.
    Live free or DIE!
  10. #1330  
    Does F104 limit the compcache to 16 MB? I tried setting it to a higher value in Govnah and it always sits on 16 MB limit
    Quote Originally Posted by rwhitby View Post
    We always prefer that people donate in response to tangible items they can use today, rather than for intangible promises about the future that may or may not be possible to achieve.
  11.    #1331  
    Quote Originally Posted by Jakeeeee View Post
    Does F104 limit the compcache to 16 MB? I tried setting it to a higher value in Govnah and it always sits on 16 MB limit
    http://forums.precentral.net/webos-i...ml#post2652836

    http://forums.precentral.net/web-os-...ml#post2515625
    Live free or DIE!
  12. #1332  
    I did end up figuring out the problem, it wasn't that, because I did that twice. I had to restart the phone, but it works fine now thanks!
  13. #1333  
    Quote Originally Posted by unixpsycho View Post
    The voltage for 500MHz is extremely low. Also the voltage for 1GHz is low. Considering it does vdemand on that base value, I can see why it might hang sometimes. Increasing the polling will make it ramp up quicker, which is why you havent crashed yet.

    Keep in mind when you set the base voltages with using vdemand, it will scale down voltage as much as -8 depending on factor. So always choose a base voltage that is stable when using vdemand.
    Thanks. That confirms what i had thought.
    I've tested as low as 30 and 59 to be stable for 500/1005 speeds, but perhaps didn't test extensively enough. I'll bump those if problems persist.
  14. RRaburn's Avatar
    Posts
    53 Posts
    Global Posts
    102 Global Posts
    #1334  
    Latest F105 is stable on Pre- 1.4.5 for the past 24 hrs with the following settings:

    1.005 GHZ 1425 mV(66)
    500 GHZ 1125 mV(42)
  15. #1335  
    is f104 available for 1.4.5?
  16.    #1336  
    Quote Originally Posted by rigo0523 View Post
    is f104 available for 1.4.5?
    Don't mean to be a pain, but, did you try looking? F104 is one of the 1st kernels I ever did. Namely for 1.4.x.
    Live free or DIE!
  17. #1337  
    Quote Originally Posted by unixpsycho;
    Keep in mind when you set the base voltages with using vdemand, it will scale down voltage as much as -8 depending on factor. So always choose a base voltage that is stable when using vdemand.
    Just to get a better understanding of vdemand. I have intensively tested f105 screenstate with vdemand off. i have reliable voltages for every frequency.
    basically you say that vdemand reduces the voltage by some points and then increases it when required (i assume to my set voltages).
    is this thought correct?
    if i increase my stable voltages by a certain amount then I am able to use vdemand and when the phone is unused it will fall back to a lower voltage.

    if correct: how much is the voltage dropped in vdemand low / medium / high? is it a fixed value? I'd like to know these values if there are so i can experiment with my voltages and vdemand.

    Thanks beforehand,

    Heino
  18. #1338  
    Quote Originally Posted by heino View Post
    Just to get a better understanding of vdemand. I have intensively tested f105 screenstate with vdemand off. i have reliable voltages for every frequency.
    basically you say that vdemand reduces the voltage by some points and then increases it when required (i assume to my set voltages).
    is this thought correct?
    if i increase my stable voltages by a certain amount then I am able to use vdemand and when the phone is unused it will fall back to a lower voltage.

    if correct: how much is the voltage dropped in vdemand low / medium / high? is it a fixed value? I'd like to know these values if there are so i can experiment with my voltages and vdemand.

    Thanks beforehand,

    Heino
    This is covered in previous posts in this thread.

    Yes, the voltages you set are the "base voltages", and those will not be increased. vdemand drops it from that base voltage. Low=2 Med=4 High=6 and OMG=8 i believe. So if you want to use Med set you voltages at least 4 higher than what is stable, 5 is probably better.
  19. carrel's Avatar
    Posts
    425 Posts
    Global Posts
    426 Global Posts
    #1339  
    Psycho,

    I have a reliable kernel crash. I'd like to help track it down if you can help me figure out how. The scenario for repeating it is rather easy:
    Code:
    dd if=/dev/zero of=/medial/internal/swapfile bs=1024 count=20480
    mkswap /media/internal/swapfile
    swapon /media/internal/swapfile
    That should simply create a 20M swap file and use it for swap. But the swapon crashes. swapon works fine with partitions, but crashes every time with files. (Well, the kernel crashes, not swapon.) swapon claims (swapon -h) that it knows about swap files.

    How can I get a crash log or crashdump?

    I have tried this on other partitions, so it is not VFAT related. I've tried much bigger and smaller files too. (No I wasn't wasting my time with a 20M swapfile, but I used that above to save time since it crashes just the same.)

    I am running current F104A-59 on my 2.1.0 Pre-.
  20.    #1340  
    Quote Originally Posted by carrel View Post
    Psycho,

    I have a reliable kernel crash. I'd like to help track it down if you can help me figure out how. The scenario for repeating it is rather easy:
    Code:
    dd if=/dev/zero of=/medial/internal/swapfile bs=1024 count=20480
    mkswap /media/internal/swapfile
    swapon /media/internal/swapfile
    That should simply create a 20M swap file and use it for swap. But the swapon crashes. swapon works fine with partitions, but crashes every time with files. (Well, the kernel crashes, not swapon.) swapon claims (swapon -h) that it knows about swap files.

    How can I get a crash log or crashdump?

    I have tried this on other partitions, so it is not VFAT related. I've tried much bigger and smaller files too. (No I wasn't wasting my time with a 20M swapfile, but I used that above to save time since it crashes just the same.)

    I am running current F104A-59 on my 2.1.0 Pre-.
    Without pulling the battery, plug in the USB cable and put the phone into recovery mode, then run "novacom get file://klog".

    Also post relevant config files mentioned on the previous page of this thread.
    Live free or DIE!

Tags for this Thread

Posting Permissions