PATCH/RFC [1/3]: lis302dl-fix-suspend-with-open-device.patch

Andy Green andy at openmoko.com
Sat Nov 15 10:01:56 CET 2008


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

Somebody in the thread at some point said:
| Fix resuming of the device
|
| From: Simon Kagstrom <simon.kagstrom at gmail.com>
|
| Suspend/Resume with the device open broke with the threshold patch. This
| patch brings it to life again. When resuming with the device open, data
| collection is re-enabled through scheduled work after the rest of the
| resume.
|
| I'm not sure why this is needed, but perhaps an (edge) interrupt is
| missed during resume. I'd like input on this.

Something else is broken somewhere if we need a magic workqueue to come
along later and clear the interrupt source to get interrupts again.

I think what I would suggest for anything touching suspend-resume is
don't try to get it working on stable, only stable-tracking due to the
many races eliminated there (if not all).

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

iEYEARECAAYFAkkekAQACgkQOjLpvpq7dMqn9QCgkNxCsm8/M8CcmtZ7mern+T9q
SZUAn1vlNddhf1sP+ZVxxeD+N1WdHXAv
=cqtM
-----END PGP SIGNATURE-----



More information about the openmoko-kernel mailing list