Accelerometers losing values

Andy Green andy at openmoko.com
Wed Feb 11 15:18:35 CET 2009


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Somebody in the thread at some point said:
| When I use the attached python script in SHR , I get the following
error in
| every Kernel. Sometimes 16 or 32 values get lost as you can see in the
log.
| This doesn't happen if I set the sample_rate to 100 Hz. Is this a known
| hardware/kernel/python problem ?
|
| What does the number in
| /sys/class/i2c-adapter/i2c-0/0-0073/lis302dl.1/overruns
| in the new kernels mean?
|
|
| Log:
|
| counter | ms | diff in ms | x | y | z
|
| 41 263.997 2.388 18 90 882
| 42 347.221 83.224 18 90 918   <- losing ~32 values

The overruns /sys thing is the number of times we came to collect a
sample set from the accelerometer, but when we did it told us we were
late and missed one or more samples that got lost before this one.

- -Andy

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmS3joACgkQOjLpvpq7dMo11wCffGNJ+W1rbu14A5Z0RtzD7dcb
EggAnjehjLCJRZDpoug0n7HyXrEnLxeH
=s7g6
-----END PGP SIGNATURE-----



More information about the openmoko-kernel mailing list