Bug 463557 - fingerprint reader gets hot
Summary: fingerprint reader gets hot
Status: RESOLVED FIXED
Alias: None
Product: openSUSE 11.1
Classification: openSUSE
Component: Mobile Devices (show other bugs)
Version: Final
Hardware: Other openSUSE 11.1
: P2 - High : Normal with 1 vote (vote)
Target Milestone: ---
Assignee: Timo Hoenig
QA Contact: E-mail List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2009-01-05 14:16 UTC by Uwe Winter
Modified: 2010-07-09 14:51 UTC (History)
3 users (show)

See Also:
Found By: ---
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---


Attachments
Updated g-s fingerprint reader patch to avoid warming (8.77 KB, patch)
2009-02-03 20:26 UTC, Timo Hoenig
Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Uwe Winter 2009-01-05 14:16:02 UTC
When using the fingerprint reader on the X41T it get's hot as hell. Don't think this is good for the hardware. Semms like the reader is always on.
Comment 1 Marcus Meissner 2009-01-05 14:28:05 UTC
the one on my x61 is also pretty hot, but I did not think much about it yet.
Comment 2 Timo Hoenig 2009-01-05 14:41:04 UTC
Known problem if you have enabled the UPEK device and don't log in using your finger but with your password.  Please confirm that this bug occurs under this circumstances.

The issue can not be solved entirely without specs from the vendor of the fingerprint reader device.  The vendor has refused to assist so far.

I've had a workaround added to ThinkFinger (now superseded by fprint/pam_fp) which I'll try to re-introduce when I get back from my vacation (12 January 2009). 
Comment 3 Uwe Winter 2009-01-06 09:39:28 UTC
The reader gets hot even with fingerprint login disabled (I had it enabled before).
I'll remove the library now as a temporary fix.
Comment 4 Timo Hoenig 2009-01-06 13:19:24 UTC
Uwe, no, not so quick please.  Please help to investigate under what conditions the heating occurs.  That will speed up my work on a fix.

Can you please elaborate what you mean by "gets hot even with fingerprint login disabled"?

Please skip the next question if that is what you've done before :-)

Can you please disable the fingerprint reader support using YaST (leaving pam_fp, libfprint installed)?

Thanks!
Comment 5 Uwe Winter 2009-01-06 14:29:17 UTC
You are right.
I once had activated fingerprint authentication activated in yast until I realized the reader gets hot. (remembering the bug the libthinkfinger once had)
The first thing I did was disabling fingerprint authentication in yast. As this didn't help I rebooted the machine but there is still the heating issue. 
Now the libs are removed, but i didn't reboot since. (Heating occurs)
The next thing I want to do is reboot the machine and look if the heating issue is gone. If not, remove the USB driver to check if the heat I feel is really from the Sensor itself. 

May sound a stupid idea filing a bug and then in the end checking if the heat really comes from the sensor. I do this to be absolutely 1000% ubersure since I didn't use the device for a longer time and it is next to the LCD-Cathode. After a night of sleep I think that it may also be possible that this heat comes from the cathode and I never realized it (but this is very unlikely).

It seems also be, that the heating up occurs randomly but I'll do a more scientific rather than random testing this evening.

I'll not run away from this bug, and I'll supply infos needed :)
Comment 6 Uwe Winter 2009-01-07 19:50:33 UTC
Sorry, for beeing one day to late. I had too much to do yesterday evening.
I can confirm that the fingerprint-reader get's hot (put finger on it 2 minutes afterwards) when logging in with password instead of finger. Swiping the finger over the Sensor makes him cool again.
Forget about random heating :)
Comment 7 Timo Hoenig 2009-01-08 11:14:03 UTC
Uwe, no worries for being late, I'm still on vacation ;-)

Thanks for verifying my assumptions.  I'll post updates once I have something ready for testing (ETA next week).

Thanks again! 
Comment 9 Hu Jiaqi 2009-01-21 13:45:37 UTC
My AuthenTec AES2501A of a hp dv2500 also gets hot after it's enabled.
Comment 10 Timo Hoenig 2009-01-21 15:15:21 UTC
Hu, can you please state if there is a difference in temperature in the following two use cases:

1.) boot the system, log into it using your finger
2.) boot the system, log into it using your password

Thanks!
Comment 11 Hu Jiaqi 2009-01-22 04:52:00 UTC
1.)Using finger does not make it hot.
2.)Actually I am not able to login with password in kdm, kdm just freeze after I enter my password.
  However I login at a console with password and then startx. After a while the temperature increased.
Comment 13 Timo Hoenig 2009-01-26 13:16:41 UTC
Uwe:

I have a preliminary fix for the UPEK driver.  However, I have to rely on your testing as my hardware doesn't show this bug.

I'll post some packages this afternoon which I'd like you to test.


Hu:

I've been investigating this with an AES2501 fingerprint reader but can't get it hot either.

Can you please test the following:

* boot your system
* login using your password
* wait some time
* check temperature

Important:  Do not start X.  Just use the console.
Comment 15 Timo Hoenig 2009-01-26 19:44:37 UTC
The following test packages are only object to fix the warming of the UPEK device.

Uwe, please test and report if this is true. 

After the sync cronjob has kicked in they will be available at:

* http://beta.suse.com/private/thoenig/FACTORY

and

* http://beta.suse.com/private/thoenig/openSUSE-11.1

Hu, please answer my questions from comment #13 so I can think about an solution for the AuthenTec device.

Thanks!
Comment 16 Tomasz Sot 2009-01-27 08:20:25 UTC
Timo:

I've installed your new packages. Reader is no longer hot. I'm using x86_64 version of Opensuse 11.1 on Thinkpad T61 with UPEK reader.

Thanks for the fix.
Comment 17 Timo Hoenig 2009-01-27 08:25:44 UTC
Tomasz, thanks for testing and the prompt reply.  I'll try to push out an online update with the fixed packages for 11.1.
Comment 18 Timo Hoenig 2009-01-27 13:08:13 UTC
Submitted to SLE11 and FACTORY.  Leaving this bug open for further analysis, lowering importance.

This should be provided as an online update for OS11.1

Adjusting summary.
Comment 19 Hu Jiaqi 2009-01-28 07:32:03 UTC
I followed this:
> * boot your system
> * login using your password
> * wait some time
> * check temperature
and now temperature seems normal.

but why start X will cause the problem?
Comment 20 Timo Hoenig 2009-01-28 10:17:39 UTC
Hu, I'm not completely sure.  Actually it was just a shot in the dark from my side to see if X is causing problems.

Can you now please test if the new packages also help if X kicks in?

* boot your system
* login using your password
* start X
* wait some time
* check temperature

Thanks for helping out -- as I can not reproduce this issue with my AES2501 I'm depending on your testing.
Comment 21 Tomasz Sot 2009-02-02 19:30:03 UTC
(In reply to comment #17)
> Tomasz, thanks for testing and the prompt reply.  I'll try to push out an
> online update with the fixed packages for 11.1.

I found, that when I unlock  gnome screensaver with password, then reader still gets hot.
Comment 22 Timo Hoenig 2009-02-02 20:19:39 UTC
Good catch.  gnome-screensaver takes another code path.  Will try to cook a patch and will provide packages.

Tomasz: I hope you can test those packages as I won't be able to reproduce the issue with my fingerprint reader device.
Comment 23 Timo Hoenig 2009-02-03 14:16:59 UTC
Tomasz:  Once synced you will find updated packages for gnome-screensaver on

* http://beta.suse.com/private/thoenig/openSUSE-11.1/

Please give them a spin and report if the issue is fixed.

Thank you!
Comment 24 Tomasz Sot 2009-02-03 20:09:12 UTC
Timo: I've installed new gnome-screensaver package and tested it. The reader doesn't get hot anymore.

Thanks for the fix.
Comment 25 Timo Hoenig 2009-02-03 20:26:31 UTC
Created attachment 269769 [details]
Updated g-s fingerprint reader patch to avoid warming

Very well, Tomasz.  Thanks for testing this so quickly!

Rodrigo, this is the updated patch.

I've re-arranged some of your code and gave it a few test runs.  I'll submit this to SLE11 and also provide this as an update for OS11.1.
Comment 26 Timo Hoenig 2010-07-09 14:51:34 UTC
Closing.