On Sunday 19 October 2008 10:45:49 Hans Verkuil wrote: > On Saturday 18 October 2008 19:04:12 Hans Verkuil wrote: > > Mailed again, this time with a CC to Sakari. > > > > Regards, > > > > Hans > > > > On Saturday 18 October 2008 19:02:14 Hans Verkuil wrote: > > > Hi Mauro, > > > > > > Please pull from http://www.linuxtv.org/hg/~hverkuil/v4l-dvb for > > > the following: > > > > > > - v4l2-int-if: Add command to get slave private data. > > > - v4l2-int-if: Add cropcap, g_crop and s_crop commands. > > > - v4l2: Add 10-bit RAW Bayer formats > > > - v4l2-int-if: Define new power state changes > > > - v4l2-int-if: Export more interfaces to modules > > > - v4l2-int-if: Add enum_framesizes and enum_frameintervals > > > ioctls. - v4l2: add video_ioctl2_unlocked for unlocked_ioctl > > > support. - ivtv: switch to unlocked_ioctl. > > > > > > These are the patches from Sakari Ailus, together with an > > > implementation for video_ioctl2_unlocked (also based on an old > > > patch from him). I've also moved ivtv to start using > > > unlocked_ioctl. > > > > > > These are all for 2.6.28. > > > > > > Thanks, > > > > > > Hans > > Hi Mauro, > > Two new patches added: > > - ivtv: avoid green flashing when loading ivtv > - bt87x: backport kernel changes > > The last one fixes this compile warning: > > bt87x.c: In function 'snd_bt87x_create_risc': > bt87x.c:278: warning: passing argument 1 of 'snd_sgbuf_get_addr' from > incompatible pointer type > > Your changeset 9136 ('backport changes on files maintained > elsewhere') broke this driver. I compared it against the latest > 2.6.27 driver and fixed it. Hi Mauro, And another two patches: - v4l: use video_device.num instead of minor in video%d - ivtvfb: FB_BLANK_POWERDOWN turns off video output The patch 'v4l: use video_device.num instead of minor in video%d' MUST go to 2.6.28. When I made the minor numbers independent from the kernel numbering for v4l2 device nodes (i.e. videoX, radioX, etc.) I forgot to go through all the drivers and modify all places where 'minor' was used to construct a device name (sprintf("video%d", vfd->minor);) to use the new 'num' field instead. Without this change you get can strange logging such as: vivi: V4L2 device registered as /dev/video9 when in fact the device is /dev/video1. But the minor was indeed 9 :-) I've CC-ed those whose drivers are affected by this change as well: it's important to know if you maintain an out-of-tree version of the driver as well. I hope I listed everyone. Regards, Hans _______________________________________________ v4l-dvb-maintainer mailing list v4l-dvb-maintainer@xxxxxxxxxxx http://www.linuxtv.org/cgi-bin/mailman/listinfo/v4l-dvb-maintainer