<!-- Some styling for better description lists --><style type='text/css'>dt { font-weight: bold;float: left;display:inline;margin-right: 1em} dd { display:block; margin-left: 2em}</style>

   jmleo: basically, I think it should go through subdev3... ?
   pinchartl: <u>jmleo</u>: I'm afraid I don't have enough motivation at the moment to answer that question
   hverkuil: <u>jmleo</u>: it should go through subdev3 or possibly even the bridge. Strictly speaking subdev1 doesn't know anything about subdev2 and vice versa.
   jmleo: <u>hverkuil</u>: ok, so how can subdev2 send an event to subdev3 ? AFAIK, you can send an event to the bridge from a subdev, but not between subdevs ?
   hverkuil: <u>jmleo</u>: the only option available today is to use the core 'ioctl' op.
   <br> sounds all very hackish
   jmleo: clearly
   sailus: <u>javier__</u>: Applied to my tree.
   javier__: <u>sailus</u>: great, thanks a lot for your help
   sailus: <u>javier__</u>: No problem. And my apologies for the delay. I saw the set, but was awaiting for some discussion. Then I forgot it. :-I
   javier__: <u>sailus</u>: no worries
   ***: headless_ changes his alias to headless
   tasslehoff: I'm trying to capture the camera on my Jetson with gstreamer/v4l2src. I set Width/Height to supported settings, but my .try_mbus_fmt is called with weird values like 1x1, 2x0, 32775x35769.
   mchehab: pinchartl, sailus, hverkuil: FYI, that's the omap3isp diagram using G_TOPOLOGY generated with:
   <br> mc_nextgen_test --dot
   <br> <u>javier__</u>: thanks for that!
   <br> https://mchehab.fedorapeople.org/mc-next-gen/igepv2_omap3isp.png
   <br> <u>javier__</u>: I added the diagrams you pointed me to the places where I'm storing other diagrams generated with mc_nextgen_test, for them to be all at the same place
   <br> I hope that's ok for you ;)
   javier__: <u>mchehab</u>: here are the output of mc_nextgen_test -e -i -I -l and media-ctl -p as fwiw
   <br> http://hastebin.com/bemafalage.m and http://hastebin.com/aririkigow.coffee
   mchehab: funny how hastebin mangled the colors ;)
   <br> those :
   <br> entity entity#13: 'unknown V4L2 subdev' OMAP3 ISP AEWB, 1 pad(s), 1 sink(s)
   <br> "unknown V4L2 subdev" looks really weird at entity level
   <br> perhaps we should call them, instead, as "unknown entity type"
   <br> while this doesn't get fixed at the drivers
   <br> --- a/contrib/test/mc_nextgen_test.c
   <br> +++ b/contrib/test/mc_nextgen_test.c
   <br> @@ -230 +229,0 @@ static inline const char *ent_function(uint32_t function)
   <br> - return "unknown V4L2 subdev";
   <br> @@ -232 +231 @@ static inline const char *ent_function(uint32_t function)
   <br> - return "unknown";
   <br> + return "unknown entity type";
   <br> what do you think?
   javier__: <u>mchehab</u>: sounds better to me
   mchehab: @@ -154 +154 @@ static inline const char *gobj_type(uint32_t id)
   <br> - return "unknown";
   <br> + return "unknown intf type";
   <br> would also make sense
   <br> for unknown interfaces
   <br> (none should display this message ATM, as both userspace and Kernel are in sync)
   <br> but that would actually happen when running an older version of the tool on some new Kernel with newer interfaces on it
   <br> ok, patch pushed to my v4l-utils experimental branch
   ***: stormer' changes his alias to stormer
   <br> _jmleo changes his alias to jmleo
   abzman: so, bttv card=98 has 2 things wrong, first is the comment in the code calls it 0x4f, which it no longer is (0x62 I think) and it has inputs set to 2 and not 4
   <br> I'm gonna try cloning card=77
   <br> because that looks closer to what I actually need for my card
   <br> (the one myself has been talking about)
   myself: Oh btw, some more plot-thickening: http://www.provideo.com.tw/web/DVR%20Card_PV-981A.htm and http://www.cctvforum.com/viewtopic.php?f=2&amp;t=43131 and http://comments.gmane.org/gmane.linux.drivers.video-input-infrastructure/16659
   <br> It looks like the PV-150 designation is just flat-out bogus. I just looked back at the order confirmation and what I bought was listed as a PV-981. I'll update the wiki page in a minute.