<!-- 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> posciak: <u>pinchartl</u>: ^ pinchartl: <u>posciak</u>: could you point me to it ? :-) <br> <u>javier__</u>: thanks <br> (I should read all messages before starting to reply :-)) mszyprow: <u>hverkuil</u>: do you have a branch with vb2_alloc_ctx removal patches? hverkuil: <u>mszyprow</u>: http://git.linuxtv.org/hverkuil/media_tree.git/log/?h=context3 mszyprow: <u>hverkuil</u>: thx ***: tizbac has quit IRC (Ping timeout: 250 seconds) <br> jmleo has quit IRC (Quit: Disconnecting from stoned server.) <br> benjiG has left bparrot: hverkuil, ping on qdisp ***: pfallenop has quit IRC (Read error: Connection reset by peer) groot: Hi! I am learning about v4l2 and drm/kms, I am trying to understand it's differences, use cases and functionalities. I read the v4l2 focus more on video capturing and drm on manipulating frame buffers data with some graphic card to output the image on a display. But v4l2 has also API for video output. My question is: is it possible to use v4l2 with an X server instead of drm? is v4l2 and drm overlaping in functionalities? <br> If they (v4l2 and drm) are overlapping in functionalities, why they co-exist? For historical reasons? Or v4l2 output API is more suitable for some application? larsc: they have grown into it <br> if you'd start from scratch today you'd have a common framework groot: larsc, so they are overlapping some how, is there any application that I would prefer using v4l for output instead of drm now a days? Or it is just a matter of taste? larsc: it's more targeted to streaming output switching frames on vsync <br> but DRM supports that pretty much as well theses days <br> I'd say DRM is pretty much the way to go unless you have a legacy codebase you want to use shuah: <u>mchehab</u>: did you send patch to set cdev parent kobj? If so, could you send me the link <br> I recall you sent a patch for cev - not sure if that is allocating cdev or just setting the parent kobj mchehab: <u>shuah</u>: I sent a cdev patch, but using a different approach <br> allocating cdev <br> cdev parent is a better (and simpler) change groot: <u>larsc</u>: thank you for your answer. I was just wondering why v4l still keeps the output API larsc: <u>mchehab</u>: it's not only simpler, it's also correct <br> there are two refcounted structures inside the struct media_devnode <br> the struct device and the struct cdev <br> dynamically allocating the cdev only solves one of the issues mchehab: yes, I know shuah: <u>mchehab</u>: okay I will continue my cdev work