summaryrefslogtreecommitdiff
path: root/v4l2-apps/libv4l/libv4l2/libv4l2.c
AgeCommit message (Collapse)Author
2009-04-15libv4l: Don't add "fake" controls when not doing conversionhans@rhel5-devel.localdomain
From: Hans de Goede <hdegoede@redhat.com> Since all things fake controls enable (such as whitebalancing) depend upon libv4lconvert_convert being called, do not fake controls when conversion is disabled. Priority: normal Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2009-04-14libv4l: Only allow supported destination formats when doing processinghans@rhel5-devel.localdomain
From: Hans de Goede <hdegoede@redhat.com> Only report / allow supported destination formats in enum_fmt / try_fmt / g_fmt / s_fmt when processing, rotating or flipping. Priority: normal Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2009-04-13libv4l: add video processing: whitebalance and normalizehans@rhel5-devel.localdomain
From: Hans de Goede <hdegoede@redhat.com> As the version number shows this work is the basis for a beta release of the 0.6.x series, the big change here is the addition of video processing to libv4l currently this only does whitebalance and normalizing (which turns out to be useless for most cams) but the basic framework for doing video processing, and being able to control it through fake v4l2 controls using for example v4l2ucp is there. The initial version of this code was written by 3 of my computer science students: Elmar Kleijn, Sjoerd Piepenbrink and Radjnies Bhansingh. This initial hg commit is a cleaned up and somewhat bug fixed version of their code. Priority: normal Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2009-03-11libv4l: 0.5.7 releasehans@rhel5-devel.localdomain
From: Hans de Goede <hdegoede@redhat.com> * Fix a nasty (and stupid) bug in the special try_fmt handling for UVC cams * Add some more verbose logging of various calls when asking libv4l to log calls to a file, to assist in (future) debugging Priority: normal Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2009-03-11libv4l: Always do a s_fmt on uvc camshans@rhel5-devel.localdomain
From: Hans de Goede <hdegoede@redhat.com> Always do a s_fmt on uvc cams even if this changes nothing, as not doing the s_fmt triggers a bug in the uvcvideo driver in kernel <= 2.6.28 (with certain cams) Priority: normal Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2009-03-11libv4l: avoid try_fmt on UVC cams if possiblehans@rhel5-devel.localdomain
From: Hans de Goede <hdegoede@redhat.com> Avoid the use of try_fmt as much as possible on UVC cams, instead use the results of the enum_framesizes ioctl. This is because: 1) try_fmt actually causes IO with UVC cams making apps which do lot of querying of device capabilities slow (cheese) 2) some buggy cams don't like getting lots of UVC video probes and crash when they do Priority: normal Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2009-03-11libv4l: Don't report DQBUF errors when errno is EAGAINhans@rhel5-devel.localdomain
From: Hans de Goede <hdegoede@redhat.com> Don't report DQBUF errors when errno is EAGAIN, this fixes flooding the screen with errors when applications use non blocking mode Priority: normal Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2009-03-11libv4l: make s_fmt more robust part 2hans@rhel5-devel.localdomain
From: Hans de Goede <hdegoede@redhat.com> Check that s_fmt atleast gives us the width, height and pixelformat try_fmt promised us, and if not disable conversion Priority: normal Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2009-03-11libv4l: Make S_FMT handling more robusthans@rhel5-devel.localdomain
From: Hans de Goede <hdegoede@redhat.com> Some drivers (pwc) do not properly reflect what one really gets after a s_fmt in their try_fmt answer. So update dest format (which we report as result from s_fmt / g_fmt to the app) with all info from the src format not changed by conversion. Priority: normal Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2009-03-11libv4l: Add software cropping from CIF to VGA modes (fix skype)hans@rhel5-devel.localdomain
From: Hans de Goede <hdegoede@redhat.com> * Add support for software cropping from 352x288 -> 320x240 / 176x144 -> 160x120, so that apps which will only work with vga resolutions like 320x240 (Skype!) will work with cams/drivers which do not support cropping CIF resolutions to VGA resolutions in hardware. This makes all 2.6.27 gspca supported cams, except for the pac7302 which only does 640x480 (and skype wants 320x240), work with skype * The v4lconvert_convert function was becoming a bit of a mess, so split the functionailiy into seperate v4lconvert_convert_pixfmt, v4lconvert_rotate and v4lconvert_crop functions, and make v4lconvert_convert a frontend to these Priority: normal Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2009-02-13v4l2-apps: move libraries around to make the directory tree flatterHans Verkuil
From: Hans Verkuil <hverkuil@xs4all.nl> Instead of having libv4l2util in v4l2-apps/lib and libv4l in v4l2-apps/lib/libv4l, both are now moved to v4l2-apps/libv4l2util and v4l2-apps/libv4l. This is much cleaner and less confusing. Priority: normal Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>