Mailing List Archived Message #356

From: "Carl Gehr" <> Full Headers
Undecoded message
Subject: Re: [eCS T60/T61] 'Rapid Flashing' WiFi LED indicator on TPad T60
Date: Mon, 09 Mar 2009 22:35:13 -0400 (EDT)
To: "eCS ThinkPad T60/61 Mailing List" <>

On Mon, 09 Mar 2009 22:52:45 +0100 (CET), Frank Vos wrote:

>Hello Carl,
>>Except for possibly one time, I do not recall the 'flashing'
>>problem after the system has been idle for a while.  When it does
>>appear, it's usually after I've been using it for a while.  One
>>possible clue that just occurred to me:  When the system has been
>>idle, the air near the back left vents is usually cool.  Right now,
>>with the flashing having happened [and switched OFF/ON to get back
>>to a solid light] a few minutes ago, the air coming from the vent
>>is actually quite warm.  Two points this makes to me [not a h/w
>>1)  The fan is actually working.
>>2)  It could be an overheating problem.
>I like to add a possible culprit to the mix, that could also explain
>the overheating. From time to time my TPad R60 gets really hot and I begun
>using a utility (cpuspeed 1.4) to throttle down my CPU. Judging from the
>output of this utility I see that my CPU after some time is running at
>100% and the program is not able to trottle the CPU down anymore, heavy
>IRQ load.
>I tested different versions of ACPI this weekend, with version 3.13
>my WIFI led was flashing a lot and I also lost my connection. With
>version 3.11 my WIFI led isn't flashing, but from time to time I still
>have my CPU at 100%.
>What version of ACPI are you using? (bldlevel \os2\dll\acpi32.dll will
>tell you the version).

A couple of new points of interest in the last two weeks:

1)  These strange behavior patterns just seem to have started
    all of a sudden, with no clear explanation, in the last
    few months -- basically since mid-November or so.
2)  My system [except for the WiFi adapter and the system board
    that were changed, but have had no impact on the symptoms]
    has had no changes since last summer.  So, unless there is
    some kind of date sensitive code, I'm stumped about what
    could be causing this.
3)  If I do not flip the physical switch to stop the 'flashing'
    and go back to a solid on state, eventually the flashing
    begins to slow and ultimately the light goes out completely,
    and the xWLAN monitor shows:
        "Searching for a connection for the current profile."
    Usually, if I flip the physical switch, the connection comes
    back at near 100% and communications can be restarted.
    Occasionally, the communications activity will restart on its
    own [other than my flipping the switch].
4)  A relatively new wrinkle is that the sound processing appears
    to also have started erratic behavior.  e.g., After the
    system was sitting idle for approximately six hours, except
    for checking for mail every four minutes and sounding a beep
    if a message arrived, I found the system emitting a constant
    tone.  Pressing the mute button on the keyboard stopped the
    sound.  Then, pressing one of the high/low volume buttons
    returned the system to a normal state of sounding the beep
    only if a message arrived.  I also suspect sound incidents
    with recent total system lock ups.
5)  I just returned from a conference in Austin, TX.  While
    using my system via WiFi, I noticed a friend's T43
    running WinXP was exhibiting the same 'rapid flashing'.
    When I asked about it, my friend said he "...had never noticed
    that before."  So, add to the symptoms:  T43, WinXP and a
    totally different venue.  The one difference:  He was not
    seeing the wireless connection drop completely after some
    period of time.
6)  I guess I should mention that the system does run a bit 'warm'
    but I would not say the air feels 'hot.'

In response to the actual question you asked, here is the result
of the BLDLEVEL [slightly reformatted for readability]:

>[O:\]bldlevel \os2\dll\acpi32.dll
>Build Level Display Facility Version 6.12.675 Sep 25 2001
>(C) Copyright IBM Corporation 1993-2001
>Signature:       @#netlabs dot org:3.9#@ ACPI-CA support system.
>                  (c)  2005-2008
>Vendor:          netlabs dot org
>Revision:        3.09
>File Version:    3.9
>Description:     ACPI-CA support system. (c) 2005-2008

And, before I'm asked, here is the BLDLEVEL for UniAud32.  And, if
anyone wants more BLDLEVEL info, please just ask.

>[O:\mmos2]bldlevel uniaud32.sys
>Build Level Display Facility Version 6.12.675 Sep 25 2001
>(C) Copyright IBM Corporation 1993-2001
>Signature:       @#Netlabs:1.9#@##1## 1 Apr 2008 20:26:55
>                 mail.smedl:1.0.16 :::2::SVN r334@@Universal
>                 Audio Driver for OS/2 and eComStation (ACPI)
>Vendor:          Netlabs
>Revision:        1.09
>Date/Time:       1 Apr 2008 20:26:55
>Build Machine:   mail.smedl
>ASD Feature ID:  1.0.16
>FixPak Version:  SVN r334
>File Version:    1.9.2
>Description:     Universal Audio Driver for OS/2 and eComStation (ACPI)

I know this is more than you asked for, but this situation seems to be
getting more and more strange every day.

Thanks, Frank!


Subscribe: Feed, Digest, Index.
Mail to ListMaster