[09:59] <mchehab> hverkuil: pushed the Zoran patch series
[09:59] <mchehab> I ended fixing the remaining checkpatch issues... I had already a half-baked script here for fixing camel case
[10:07] <hverkuil> Nice.
[10:09] <hverkuil> I've posted a patch fixing the description of the V4L2_COLORSPACE_470_SYSTEM_BG documentation. You are right, it is not supersedes by 170M.
[11:44] <mchehab> hverkuil: thanks!
[11:45] <mchehab> hverkuil, syoung, sailus, pinchartl: can we start our weekly meeting a little earlier?
[11:45] * mchehab need to do some errands soon
[11:47] <hverkuil> sure. I don't have anything to report anyway :-)
[11:48] <mchehab> good!
[11:48] <mchehab> from my side, nothing to report either... everything to be merged for 5.10 are there already...
[11:49] <mchehab> (except for the Sphinx 3.x conversion patches, that I'll likely send together with other stuff, likely via the docs tree)
[11:49] <mchehab> as those require some changes at the docs building system
[11:50] <hverkuil> Note that I haven't looked into the dma-buf.h issues in the media compat build. I gather that it is broken for older kernels (<= 5.8?). I'm not really sure how to fix it, and testing it will be a pain as well.
[11:51] <mchehab> the compat build likely requires some extra care
[11:51] <hverkuil> I have to admit that it is very low on my priority list, but if someone wants to work on this, then that would be great.
[11:51] <mchehab> since CEC started to be part of builtin on some distros (Fedora, at least), using it can be somewhat peinful
[11:52] <mchehab> hverkuil: yeah, this has a lot lower priority than other stuff
[11:52] <mchehab> perhaps we could discuss it with the community
[11:52] <sailus> Heippa!
[11:52] <mchehab> btw, we should probably try to do a media summit this year
[11:53] <mchehab> perhaps an online event may attract other members of the community that usually don't travel a lot
[11:53] <mchehab> (it is probably a little late to talk about that, but maybe it would be easier for LF to allocate us some slot during ELCE --- I duno)
[11:54] <mchehab> duno -> dunno
[11:55] <mchehab> hverkuil, sailus, pinchartl: what do you think about having an online media summit this year?
[11:55] <mchehab> (being online, we don't really depend on LF - I guess)
[11:56] <hverkuil> Is there much to discuss? I don't think I have anything special from my side.
[11:56] <sailus> mchehab: Interesting idea. I don't think we'd depend on LF in that --- nor we do on face to face meetings either.
[11:57] <syoung> hi
[11:58] <syoung> I don't have anything for a media summit.
[11:59] <mchehab> hverkuil: well, if there's nothing to discuss, we could skip, but I would still try to check with the community and devels about the needs
[11:59] <syoung> I
[11:59] <mchehab> we didn't have a summit last year either, so IMO we should try to have something this year
[12:00] <syoung> I'm working on a BPF to IR compiler (just userspace) which should be done next year. Might be interesting to talk about once it's done.
[12:00] <mchehab> syoung: yeah, that sounds nice
[12:00] <hverkuil> I'm not opposed to it. I'm just not sure we have enough to discuss.
[12:01] <mchehab> we don't need to focus only on kernelspace... low-level userspace ones like BPF and libcamera are welcomed as well
[12:02] <mchehab> hverkuil: a few things we can discuss include media-build, criteria for removal of stuff at staging
[12:02] <mchehab> there are several drivers that are at staging for years
[12:03] <mchehab> IMO, we need a plan to solve the staging drivers issues
[12:06] <mchehab> anything else?
[12:06] <sailus> Yes.
[12:06] <sailus> I've been working on a MIPI CCS driver.
[12:07] <sailus> It's not actually an entirely new driver, but it updates the SMIA driver to the latest MIPI CCS standard.
[12:07] <sailus> I posted the patches to linux-media yesterday evening.
[12:07] <sailus> You can't miss it, it's 100 patches. :-)
[12:07] <mchehab> :-)
[12:08] <mchehab> it sounds I'll need to reserve at least hald a day for reviewing it :-)
[12:08] <mchehab> hald->half
[12:09] <sailus> I know some patches produce checkpatch.pl warnings; most of them are due to renaming things and _only_ renaming causes a lot of new lines being added and checkpatch.pl doesn't like some of them.
[12:10] <sailus> Some of them are also fixed later on in the series.
[12:10] <mchehab> please fix the remaining ones (that makes sense) at the end of the series
[12:11] <mchehab> weird that only renames would be generating warnings though
[12:13] <mchehab> well, function renames may misalign parenthesis - while the best would be to fix those together with the patch doing renames, I can work on having them fixed at the end
[12:13] <sailus> I believe I've fixed them already.
[12:13] <mchehab> anything else?
[12:14] <sailus> Yes.
[12:14] <sailus> The driver includes a library for the CCS static data format.
[12:14] <sailus> It's currently dual licensed (BSD + GPL v2) in the patches so it can be used elsewhere, e.g. user space as well.
[12:15] <sailus> Just FYI.
[12:19] <mchehab> ok. There's then the question if (BSD + GPLv2) is GPLv2 compatible
[12:19] <sailus> mchehab: It's one of the two.
[12:19] <mchehab> BSD | GPLv2 is GPLv2 compatible
[12:19] <sailus> Yes, that's what it is; not both at the same time.
[12:20] <mchehab> hmm... is it BSD OR GPLv2 or BSD AND GPLv2?
[12:20] <sailus> SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
[12:21] <mchehab> ah, OK!
[12:24] <mchehab> ok, anything else?
[12:25] <sailus> Not on my side.
[13:02] <pinchartl> hello
[13:02] <pinchartl> sorry, I'm late
[13:02] <pinchartl> just one question regarding reviews
[13:02] <pinchartl> I've posted a series that cleans up the pixel formats documentation for RGB and packed YUV formats
[13:03] <pinchartl> I'll continue with planar YUV, but I'd like feedback on the direction
[13:11] <hverkuil> pinchartl: it's on my todo list, and actually I have some time right now. I'll do a review.
[13:14] <pinchartl> thank you
[13:40] <hverkuil> pinchartl: reviewed.
[13:41] <pinchartl> hverkuil: thank you
[13:41] <pinchartl> have you looked at the generated documentation ? do you think the consolidation (grouping YUV packed formats in a table for instance) is a good way to go ?
[13:41] <pinchartl> I think it makes the doc much more readable
[13:42] <pinchartl> we had so much copy&paste with slight style differences here and there, it wasn't very consistent, and hard to navigate
[13:42] <hverkuil> I haven't checked the generated documentation, but I think it makes a lot of sense to consolidate this. Just a lot of history there, so it's good to clean it up.
[13:45] <pinchartl> thank you