toshiba laptop powers itself down - acpi problem ? - Portable

This is a discussion on toshiba laptop powers itself down - acpi problem ? - Portable ; Hi group, I suspect this has to do with ACPI, but I have not been able to find any easy documentation. So, I'm trying the lazy way and ask here:-) I have an annoying problem with my Toshiba Satellite SA60-302 ...

+ Reply to Thread
Results 1 to 6 of 6

Thread: toshiba laptop powers itself down - acpi problem ?

  1. toshiba laptop powers itself down - acpi problem ?

    Hi group,

    I suspect this has to do with ACPI, but I have not been able to find any
    easy documentation. So, I'm trying the lazy way and ask here:-)

    I have an annoying problem with my Toshiba Satellite SA60-302 laptop
    running an LFS-6.0 based system. I have seen similar problems running
    Fedora Core (2 or 3, I don't remember which).

    The problem manifests itself like this:

    First time I boot the system, it runs fine for some time, 10min to 1h,
    then it suddenly shuts down. This also happens with the AC-adapter on.
    Another observation: The fan is silent.

    When I boot the system again (immediately after it has shut itself
    down), it starts up with the fan running. Now it runs stably with no
    problems.

    Some details about the system:

    Kernel is 2.6.11rc1. Output from dmesg contains:
    ....
    ACPI: AC Adapter [AC] (on-line)
    ACPI: Battery Slot [BAT1] (battery present)
    ACPI: Power Button (FF) [PWRF]
    ACPI: Lid Switch [LID]
    ACPI: Fan [FAN1] (off)
    ACPI: CPU0 (power states: C1[C1] C2[C2])
    ACPI: Thermal Zone [THZN] (82 C)
    ....
    ACPI wakeup devices:
    LID USB0 USB1 USB2 MC97 P2P LAN0
    ACPI: (supports S0 S3 S4 S5)
    ....
    This log is from the second boot, and the message above suggests that
    the system is too hot (82C).

    I don't have acpid running. I have the daemon, but I have no idea how
    to set it up:-( Any pointers to example configuration files would be
    great!!

    One peculiar thing about the hardware is that it does not have a BIOS
    setup program, so I can't adjust power management settings in the BIOS.

    Any ideas how I can solve this? Any help will be appreciated...

    Thanks
    Ivan

  2. Re: toshiba laptop powers itself down - acpi problem ?

    Ivan Riis Nielsen writes:

    >Hi group,


    >I suspect this has to do with ACPI, but I have not been able to find any
    >easy documentation. So, I'm trying the lazy way and ask here:-)


    >I have an annoying problem with my Toshiba Satellite SA60-302 laptop
    >running an LFS-6.0 based system. I have seen similar problems running
    >Fedora Core (2 or 3, I don't remember which).


    >The problem manifests itself like this:


    >First time I boot the system, it runs fine for some time, 10min to 1h,
    >then it suddenly shuts down. This also happens with the AC-adapter on.
    > Another observation: The fan is silent.


    Sounds like overheating. There is thermal shutdown.


    >When I boot the system again (immediately after it has shut itself
    >down), it starts up with the fan running. Now it runs stably with no
    >problems.


    >Some details about the system:


    >Kernel is 2.6.11rc1. Output from dmesg contains:
    >...
    >ACPI: AC Adapter [AC] (on-line)
    >ACPI: Battery Slot [BAT1] (battery present)
    >ACPI: Power Button (FF) [PWRF]
    >ACPI: Lid Switch [LID]
    >ACPI: Fan [FAN1] (off)
    >ACPI: CPU0 (power states: C1[C1] C2[C2])
    >ACPI: Thermal Zone [THZN] (82 C)


    Definitely too hot. (Mind you I had a desktop that went to 160C without
    shutting down-- bad situation. Bad motherboard/heatsink.)


    >...
    >ACPI wakeup devices:
    > LID USB0 USB1 USB2 MC97 P2P LAN0
    >ACPI: (supports S0 S3 S4 S5)
    >...
    >This log is from the second boot, and the message above suggests that
    >the system is too hot (82C).


    >I don't have acpid running. I have the daemon, but I have no idea how
    >to set it up:-( Any pointers to example configuration files would be
    >great!!


    I would disable acpi.
    acpi=off on bootup or in /etc/lilo.conf on the append= line


    >One peculiar thing about the hardware is that it does not have a BIOS
    >setup program, so I can't adjust power management settings in the BIOS.


    ???

  3. Re: toshiba laptop powers itself down - acpi problem ?

    Ivan Riis Nielsen wrote:

    > I suspect this has to do with ACPI, but I have not been able to find any
    > easy documentation. So, I'm trying the lazy way and ask here:-)
    > [ ... ]
    > First time I boot the system, it runs fine for some time, 10min to 1h,
    > then it suddenly shuts down. This also happens with the AC-adapter on.
    > Another observation: The fan is silent.


    > When I boot the system again (immediately after it has shut itself
    > down), it starts up with the fan running. Now it runs stably with no
    > problems.


    Me too, completely different brand. I think it is related to this
    bug: http://bugzilla.kernel.org/show_bug.cgi?id=4124 . The symptoms
    are weirdness in handling of all sorts of events, like presses of
    the screen brightness keys. On my laptop, the fumbled events
    include starting the fan :-( If you also notice delays in your
    screen brightness controls, keep reading.

    I have a workaround, which is to keep reading files under /proc/acpi.
    For example reading the battery status periodically is enough to
    flush out the events. Maybe running a battery applet is enough.
    Another way is to poll the temperature,
    echo -n 1 > /proc/acpi/thermal_zone/THRM/polling_frequency

    The symptoms are reduced, but not eliminated, if you run a preemptible
    kernel (CONFIG_PREEMPT=y).


    > I don't have acpid running. I have the daemon, but I have no idea how
    > to set it up:-( Any pointers to example configuration files would be
    > great!!


    acpid runs programs of your choice when events occur. If you run it,
    you'll get entries in your syslog when it sees events and doesn't
    know what to do with them.

    I don't think acpid will help you; the problem is occurring at a
    lower level.

    Your laptop may work differently than mine and it is possible that
    none of the above applies. Can you run this command and post the
    output?

    find /proc/acpi -type f -perm -04 | xargs head


    --
    pa at panix dot com

  4. Re: toshiba laptop powers itself down - acpi problem ?

    Unruh wrote:
    > Ivan Riis Nielsen writes:
    >


    snip

    >
    >>Kernel is 2.6.11rc1. Output from dmesg contains:
    >>...
    >>ACPI: AC Adapter [AC] (on-line)
    >>ACPI: Battery Slot [BAT1] (battery present)
    >>ACPI: Power Button (FF) [PWRF]
    >>ACPI: Lid Switch [LID]
    >>ACPI: Fan [FAN1] (off)
    >>ACPI: CPU0 (power states: C1[C1] C2[C2])
    >>ACPI: Thermal Zone [THZN] (82 C)

    >
    >
    > Definitely too hot. (Mind you I had a desktop that went to 160C without
    > shutting down-- bad situation. Bad motherboard/heatsink.)
    >
    >
    >
    >>...
    >>ACPI wakeup devices:
    >> LID USB0 USB1 USB2 MC97 P2P LAN0
    >>ACPI: (supports S0 S3 S4 S5)
    >>...
    >>This log is from the second boot, and the message above suggests that
    >>the system is too hot (82C).

    >
    >
    >>I don't have acpid running. I have the daemon, but I have no idea how
    >>to set it up:-( Any pointers to example configuration files would be
    >>great!!

    >
    >
    > I would disable acpi.
    > acpi=off on bootup or in /etc/lilo.conf on the append= line


    Thanks for the tip. I went right ahead and tried that, but unfortunately this
    has a bad side effect: The system is not able to assign IRQs with acpi=off - it
    first suggests pci=usepirqmask which does not work, then it suggests pci=biosirq
    which doesn't work either. So, I'm stuck with acpi on:-(

    >
    >
    >
    >>One peculiar thing about the hardware is that it does not have a BIOS
    >>setup program, so I can't adjust power management settings in the BIOS.

    >
    >
    > ???


    Yeah, that's weird. Just today I tried hitting F2 during boot and voila,
    there's the BIOS setup program:-) Nothing about that on the screen. But then,
    who would need that, when the computer comes pre-installed with Windows which
    has a nice program for that... Something to experiment with.

    Thanks
    Ivan

  5. Re: toshiba laptop powers itself down - acpi problem ?

    Pierre Asselin wrote:
    > Ivan Riis Nielsen wrote:
    >


    snip

    >
    >
    > Me too, completely different brand. I think it is related to this
    > bug: http://bugzilla.kernel.org/show_bug.cgi?id=4124 . The symptoms
    > are weirdness in handling of all sorts of events, like presses of
    > the screen brightness keys. On my laptop, the fumbled events
    > include starting the fan :-( If you also notice delays in your
    > screen brightness controls, keep reading.
    >
    > I have a workaround, which is to keep reading files under /proc/acpi.
    > For example reading the battery status periodically is enough to
    > flush out the events. Maybe running a battery applet is enough.
    > Another way is to poll the temperature,
    > echo -n 1 > /proc/acpi/thermal_zone/THRM/polling_frequency
    >
    > The symptoms are reduced, but not eliminated, if you run a preemptible
    > kernel (CONFIG_PREEMPT=y).
    >


    Thanks a lot, this is something to work with. My kernel does have
    CONFIG_PREEMPT=y, but still shows the problem. I think I'll try out your
    work-around before trying to patch the kernel. As for the brightness control,
    this does not seem to work at all, but I actually experienced the fan starting
    and stopping automatically while I was playing around with the files in /proc/acpi.

    >
    >
    >>I don't have acpid running. I have the daemon, but I have no idea how
    >>to set it up:-( Any pointers to example configuration files would be
    >>great!!

    >
    >
    > acpid runs programs of your choice when events occur. If you run it,
    > you'll get entries in your syslog when it sees events and doesn't
    > know what to do with them.
    >
    > I don't think acpid will help you; the problem is occurring at a
    > lower level.
    >
    > Your laptop may work differently than mine and it is possible that
    > none of the above applies. Can you run this command and post the
    > output?
    >
    > find /proc/acpi -type f -perm -04 | xargs head
    >


    Sure, here it comes:

    ==> /proc/acpi/wakeup <==
    Device Sleep state Status
    LID 4 *enabled
    USB0 3 disabled
    USB1 3 disabled
    USB2 3 disabled
    MC97 3 disabled
    P2P 4 disabled
    LAN0 4 disabled

    ==> /proc/acpi/alarm <==
    2005-03-00 00:00:00

    ==> /proc/acpi/sleep <==
    S0 S3 S4 S5

    ==> /proc/acpi/thermal_zone/THZN/polling_frequency <==


    ==> /proc/acpi/thermal_zone/THZN/cooling_mode <==
    cooling mode: active

    ==> /proc/acpi/thermal_zone/THZN/trip_points <==
    critical (S5): 105 C

    ==> /proc/acpi/thermal_zone/THZN/temperature <==
    temperature: 39 C

    ==> /proc/acpi/thermal_zone/THZN/state <==
    state: ok

    ==> /proc/acpi/processor/CPU0/power <==
    active state: C2
    max_cstate: C8
    bus master activity: 00000000
    states:
    C1: type[C1] promotion[C2] demotion[--] latency[001]
    usage[00003020]

    ==> /proc/acpi/processor/CPU0/limit <==


    ==> /proc/acpi/processor/CPU0/throttling <==


    ==> /proc/acpi/processor/CPU0/info <==
    processor id: 0
    acpi id: 0
    bus mastering control: yes
    power management: yes
    throttling control: no
    limit interface: no

    ==> /proc/acpi/fan/FAN1/state <==
    status: off

    ==> /proc/acpi/button/lid/LID/state <==
    state: open

    ==> /proc/acpi/button/lid/LID/info <==
    type: Lid Switch

    ==> /proc/acpi/button/power/PWRF/info <==
    type: Power Button (FF)

    ==> /proc/acpi/battery/BAT1/alarm <==
    alarm: unsupported

    ==> /proc/acpi/battery/BAT1/state <==
    present: yes
    capacity state: ok
    charging state: charged
    present rate: 0 mA
    remaining capacity: 4137 mAh
    present voltage: 15000 mV

    ==> /proc/acpi/battery/BAT1/info <==
    present: yes
    design capacity: 4300 mAh
    last full capacity: 4127 mAh
    battery technology: rechargeable
    design voltage: 15000 mV
    design capacity warning: 0 mAh
    design capacity low: 0 mAh
    capacity granularity 1: 0 mAh
    capacity granularity 2: 4127 mAh
    model number: PA3384-BAS/BRS

    ==> /proc/acpi/ac_adapter/AC/state <==
    state: on-line

    ==> /proc/acpi/info <==
    version: 20041210

    ==> /proc/acpi/power_resource/PFA1/state <==
    state: off
    system level: S0
    order: 0
    reference count: 0

    ==> /proc/acpi/power_resource/PUT2/state <==
    state: on
    system level: S0
    order: 0
    reference count: 0

    ==> /proc/acpi/embedded_controller/EC0/info <==
    gpe bit: 0x07
    ports: 0x66, 0x62
    use global lock: no



    Thanks
    Ivan

  6. Re: toshiba laptop powers itself down - acpi problem ?

    Ivan Riis Nielsen wrote:
    > Pierre Asselin wrote:


    > Thanks a lot, this is something to work with. My kernel does have
    > CONFIG_PREEMPT=y, but still shows the problem. I think I'll try out your
    > work-around before trying to patch the kernel. As for the brightness control,
    > this does not seem to work at all,



    Do you mean that the brightness buttons exhibit no problems, or
    that they don't work at all ? if the latter, it's always a good
    idea to look at http://acpi.sourceforge.net/dsdt/view.php
    to see if there is a known problem with your model's firmware.
    To use the DSDT's therein, you pretty much have to recompile
    your kernel.

    If it's the former, i.e. the buttons work, you may still have the bug,
    but your CONFIG_PREEMPT=y turns it from a fairly reproducible bug
    into an intermittent one :-O


    > but I actually experienced the fan starting
    > and stopping automatically while I was playing around with the files in /proc/acpi.


    and you're never sure if it was a coincidence, right? Sounds familiar.


    > > find /proc/acpi -type f -perm -04 | xargs head

    > Sure, here it comes:


    > [ ... ]
    > ==> /proc/acpi/thermal_zone/THZN/cooling_mode <==
    > cooling mode: active
    > [ ... ]
    > ==> /proc/acpi/fan/FAN1/state <==
    > status: off


    Nothing unsual. Linux is supposed to turn on a fan when
    the temperature gets too high, and you do have a fan.

    Everything is still consistent with the bug I referenced earlier.
    If GPE's are delayed, the fan control will be delayed.

    Try echoing a "1" into /proc/acpi/thermal_zone/THZN/polling_frequency.
    Keep us posted.

    --
    pa at panix dot com

+ Reply to Thread