John Martin wrote: > Dan Mick wrote: >> I can't find bugs about these on d.o.o, so let me start by asking: >> >> The Toshiba M10 has been reported to work with usbvc/v4l2 on >> OpenSolaris; see >> http://blogs.sun.com/PotstickerGuru/entry/javaone_communityonne_opensolaris_2009_06, >> >> for instance, and bugs that indicate cheese has run as well, because >> it stopped suspend/resume from working on the M10. >> >> But when I try Cheese, it doesn't touch usbvc (according to dtrace -m >> usbvc), and the "preferences" doesn't show any selectable devices >> (null/null is all I get). When I try Ekiga, it does poke at the usbvc >> driver, but coredumps in some >> gconf-related string handling. More details on both are available. >> >> I tried the code at >> http://blogs.sun.com/madlab/entry/getting_started_with_video4linux2_usbvc, >> and >> I get reasonable output; it seems to respond to the query >> successfully, and >> reports that it supports capture, r/w, streaming. >> >> Any diagnostic output I can turn on that would be useful for ekiga or >> cheese? >> Anyone else experienced similar problems? > Broken in b117 or greater: > > http://defect.opensolaris.org/bz/show_bug.cgi?id=9782 > > Target for fix is b121.
The gstreamer-properties test in that bug works both in 09.06 and my BFU-updated copy (with a kernel ca. 30 Jul). However, cheese does *not* work in the 7/30 BFU, while it does still work in 09.06. It doesn't poll /dev/video0 in 09.96, but it never even gets to open /dev/video0 in the BFU'ed copy, despite the fact that the driver is there and attached (as the gstreamer-properties test proves as well). The ekiga coredump seems like something completely different as well. However, it also is resolved by running the stock 09.06. No idea what that might arise from. So the upshot is I don't think this is the source of either problem I'm having.
