<!-- 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>

   hverkuil: <u>ezequielg</u>: ping
   ***: hfr has quit IRC (Read error: Connection reset by peer)
   <br> benjiG has left
   ezequielg: <u>hverkuil</u>: ey
   <br> how goes?
   hverkuil: <u>ezequielg</u>: I remember that someone (I think you) mentioned a utility that could be used to do e.g. regression tests on a newly kernel. Unfortunately, I can't remember the name.
   ezequielg: other than kernelci?
   hverkuil: <u>sailus</u>: this is delegated to you: https://patchwork.linuxtv.org/patch/53574/ But it is quite old and still unreviewed (Marco complained).
   <br> <u>ezequielg</u>: yes. Apparently it is something that can be used to quickly build a qemu image that you can run. (I think it was qemu...)
   <br> Really annoying, it is at the tip of my tongue, but I just can't remember it.
   ezequielg: hm
   <br> I use virtme as a qemu wrapper.
   <br> and then I combine that with gst-build, to check regressions.
   <br> and I have played with travis-ci, but I don't think I've mentioned that to you.
   hverkuil: I think it was virtme! That starts to ring a bell.
   ezequielg: virtme is gold
   <br> https://www.collabora.com/news-and-blog/blog/2018/09/18/virtme-the-kernel-developers-best-friend/
   hverkuil: yep, that's the one.
   <br> I'm going to play with that, I want to have my daily build run the media regression tests automatically.
   ezequielg: nice
   <br> <u>hverkuil</u>: mchehab: mripard and I would like to press fast-forward on the H264 interface.
   <br> if you are OK with the PIX_FMT proposal we've made, perhaps Maxime can post another round and start converging?
   <br> <u>tfiga</u>: you around ?
   <br> maybe we can talk about the ref pic lists?
   <br> (and about the long term flag as well)
   hverkuil: I have no objections. Spin a v4 and I will give it another review. It is likely too late for 5.1, but early 5.2 should be eminently doable.
   <br> Make sure the pixelformats are clearly documented.
   sailus: <u>hverkuil</u>: Thanks for the ping. I'll try to check that out tomorrow.
   ndufresne: mripard, ezequielg: Just to add up, having specific formats for slice_raw and slice_annex_b will help be in making sure we do the minimal amount of bitstream conversion in gstreamer
   ezequielg: ... and all the other parsers, of course ;)
   ***: APic has quit IRC (Ping timeout: 240 seconds)
   <br> syoung has quit IRC (Ping timeout: 240 seconds)
   <br> TFKyle has quit IRC (Ping timeout: 240 seconds)
   <br> mjourdan has left