On Sun, 29 Mar 2009 14:22:08 +0200, Hans Verkuil wrote: > Hi Mauro, > > For your information, this is the current status: > > - Steve Toth tested the HVR-1800 for me, so I've posted the pull request for > the converted cx23885 driver. > > - The cx88 driver is also finished. I'm waiting for a last test by Jean > Delvare before I post that one as well. Testing complete, it seems to work just fine. Thanks! > - Added support for saa6588 to saa7134: needed to drop the legacy i2c API > from saa6588. This is in my pull request for my v4l-dvb tree. > > - Douglas has almost finished the em28xx driver conversion. > > - Jean Delvare is working on the ir-kbd-i2c conversion. > > That last conversion is stand-alone (i.e. has no impact on the internal v4l > API) and I don't think it prevents pushing our v4l-dvb changes to 2.6.30. > > When the first four items are finished and merged into v4l-dvb, then I have > a to do a few final cleanup actions to make everything ready for the 2.6.30 > merge: > > - Remove v4l2-i2c-drv-legacy.h from the remaining i2c drivers. Remove the > files v4l2-i2c-drv-legacy.h and v4l2-common.c since these are no longer > used. Cleanup v4l2-common.h since the internal ioctls are no longer needed. > Update v4l2-framework.txt, removing any references to the legacy behavior. > > - Fix two subdev callbacks that are in the wrong place (s_std belongs to the > video ops, s_standby belongs to the tuner ops). > > - Add a load_fw callback to the core ops and use that were appropriate > instead of the init callback. Analyze whether the init callback can be > removed altogether. > > - Analyze how the probe addresses are used in the v4l2 drivers and move > those lists over to the appropriate i2c driver headers. > > - Add enum_frameintervals and enum_framesizes callbacks for use with omap. > > - Check for any remaining uses of I2C_DRIVERID and remove them. Note that this isn't urgent as far as I am concerned. Driver IDs can stay as long as they are needed. I _do_ expect them to become useless thanks to the improved design, and then we can remove the structure fields, but we can wait and see if I am correct, no need to rush. > This shouldn't take much time to implement. Thanks for all your work Hans. -- Jean Delvare -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
