<!-- 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>arnd</u>: ping
arnd: hi hverkuil
<br> <u>hverkuil</u>: I'm sure there was a patch I was meant to follow up on, let me check
hverkuil: <u>arnd</u>: good morning!
<br> https://patchwork.linuxtv.org/project/linux-media/patch/20210318134334.2933141-1-arnd@kernel.org/
<br> that's the one :-)
arnd: right. I had applied it on my randconfig series, which has a couple of other stuff applied, including an older global cleanup for time32 that I never sent, I'll rebase and resend the fix
hverkuil: Thanks!
<br> <u>ezequielg</u>: ping
sinajit: Is it possible for a capability retrieved by VIDIOC_QUERYCAP to be supported only for some image formats?
<br> For some reason VIDIOC_STREAMON fails on my device for RX24 but it works for RGB3/BGR3
hverkuil: <u>sinajit</u>: what driver are you using?
sinajit: Don't know, how would I check that?
hverkuil: v4l2-ctl -D
<br> (QUERYCAP)
sinajit: bm2835 mmal
<br> Ah nevermind, it appears it was a memory issue.
hverkuil: That's one reason why it can fail :-)
ezequielg: <u>hverkuil</u>: hey Hans.
<br> <u>hverkuil</u>: thanks for reviewing the mpeg-2 series.
hverkuil: <u>ezequielg</u>: no problem.
<br> <u>ezequielg</u>: I replied to "NULL pointer dereference at v4l2_ctrl_request_complete" a few days ago. Were you able to test with the provided patch so I can get a bit more information about what is going on?
ezequielg: <u>hverkuil</u>: Not yet. I wasn't at the office these day. It's a high priority issue for me, so I'll run tests as soon as I get back to work.
hverkuil: <u>ezequielg</u>: ah, good. I'll wait for feedback then.
ezequielg: (BTW, I had the same printks, keep in mind it gets super spammy, because it takes time to reproduce).
<br> I have the feeling it happens with those controls that aren't explicitly set with S_EXT_CTRL and a request submit. For instance, MPEG2_SEQUENCE, or H264 decoding mode control.
hverkuil: how long does it typically take to reproduce?
ezequielg: a few seconds, with 3 pipelines.
<br> but enough to spam the logs.
hverkuil: <u>ezequielg</u>: when will you resume work on this? I'll make sure I'm available at the time so we can dig into this together.
-: ezequielg how about tomorrow 8am argentina?
ezequielg: <u>hverkuil</u>: ^
hverkuil: sounds good.
<br> <u>mchehab</u>: this PR https://patchwork.linuxtv.org/project/linux-media/patch/b2a45eb6-df13-05e3-c24b-6025bb41fd04@xs4all.nl/ is marked as Accepted today, but I don't see the patches in our master branch. Did something go wrong?
Phill93: Hello, I have a Startech Framegrabber (SVID2USB232) this has the chip EM28281 installed. The em28xx driver does not recognize this. Can someone help me to get it running?
mchehab: <u>hverkuil</u>: I remember merging sometthing like this today: vivid: update EDID
hverkuil: But none of the others arrived.
mchehab: commit 443ec4bbc6116f6f492a7a1282bfd8422c862158
<br> <u>Author</u>: Hans Verkuil <hverkuil-cisco@xs4all.nl>
<br> <u>AuthorDate</u>: Thu Mar 25 08:48:21 2021 +0100
<br> <u>Commit</u>: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
<br> <u>CommitDate</u>: Tue Apr 6 10:37:18 2021 +0200
<br> <u>media</u>: vivid: update EDID
<br> funny
<br> <u>hverkuil</u>: yeah, I suspect it was marked as accepted by mistake - not sure if it was due to script or not...
<br> Tue Apr 6 10:38:20 2021: pwclient update -s 'Accepted' 72624 # patches/lmml_72624_git_pull_for_v5_13_various_fixes.patch
<br> Tue Apr 6 10:38:20 2021: pwclient update -s 'Accepted' 72675 # patches/lmml_72675_git_pull_for_v5_13_various_fixes.patch
<br> there were two PRs with the same name, one after the other
<br> I'm applying right now the second PR
<br> next time, it could be worth numerating e-mails with the same subject - or use a different name for each
<br> anyway, problem solved. patches applied
hverkuil: much appreciated!
mchehab: you're welcome
<br> I've been more aggressive those days with a new cleanup script set
<br> that tries harder to identify duplicated patches
<br> maybe the script is the one to blame, as both PRs were "accepted" at the same second
hverkuil: Ah, that same patch was actually in two PRs somehow.
<br> I'm not sure how that happened, but that clearly was my mistake.