[04:04] *** mchehab has quit IRC (*.net *.split)
[04:05] *** sailus has quit IRC (*.net *.split)
[04:05] *** pinchartl has quit IRC (*.net *.split)
[04:06] *** syoung has quit IRC (*.net *.split)
[04:07] *** hverkuil has quit IRC (*.net *.split)
[04:07] *** hitchcock.freenode.net sets mode: +v hverkuil
[04:10] *** hitchcock.freenode.net sets mode: +v mchehab
[04:11] *** hitchcock.freenode.net sets mode: +v pinchartl
[09:33] <hverkuil> pinchartl: ping
[11:14] <pinchartl> hverkuil: pong
[11:15] <hverkuil> Hi Laurent! I have a few questions about the ti-vpe cal series:
[11:16] <hverkuil> 1) Can you run v4l2-compliance for this device? Both video centric and MC centric (using the -m option). Just to check that there are no regressions compared to the current video centric driver.
[11:16] <hverkuil> 2) Is there an affordable board that can be used to test this driver?
[11:17] <pinchartl> I've run v4l2-compliance, and there were no regressions. for the MC API there can't be any regression as it's new :-)
[11:18] <hverkuil> I'll also comment on two patches (57 and 73). Take a look at my comments (will post in a minute) and if you are OK with it, then I can make the change for you.
[11:18] <pinchartl> I can provide the output, but the project is on hold right now due to administrative issues, so it would have to wait a little bit
[11:18] <pinchartl> as for an affordable board, I don't think so
[11:18] <pinchartl> I use the AM65x EVM and the DRA76x EVM
[11:18] <pinchartl> they're not affordable
[11:19] <hverkuil> OK, you've run the compliance test for the video centric case, that's good enough for me.
[11:21] <hverkuil> It's actually patch 57 and 69 for which I have small comments.
[11:22] <hverkuil> posted both comments.
[11:22] <pinchartl> I'll read that, thanks
[11:23] <hverkuil> If you are OK with those changes, then I would like to make a PR for patches 01-84. I prefer to wait one more week before I take the remainder of the series.
[11:32] <hverkuil> pinchartl: ah, OK, those changes you put in 57 and 69 are needed later. Never mind then, I'll take it as-is. It's a bit confusing when reviewing, but not worth the effort to change this.
[11:49] <pinchartl> hverkuil: I've replied to both. they're valid points, but the code gets further reworked in the series, and both constructs that you pointed to as not needed would have to be added back. I'm not sure it's worth it removing them in 57 and 69
[12:21] <hverkuil> mchehab: you mentioned before that there was a limit to the number of patches in a PR, was that 50 or 100? And does patchwork check for that? Because I see that my PR for the 84-patch series doesn't appear in patchwork. If the limit is 50, then I'll have to split it up.
[13:10] <hverkuil> mchehab: it looks like patchwork is not longer updating (that might be why I didn't see my PR). Can you check that it is receiving mails from the mailinglist?
[13:18] <mchehab> patchwork doesn't have any limit
[13:18] <mchehab> the limig is somewhere between 50 and 100 (don't remember the exact number) at the mailbomb script
[13:19] <mchehab> when someone pushes stuff at the git trees on linuxtv.org
[13:19] <mchehab> (the ones which generate notification e-mails)
[13:19] <mchehab> it affects only the notification e-mails
[13:20] <hverkuil> The real issue is that new emails do not seem to arrive in patchwork.
[13:20] <mchehab> e. g. the e-mails that go through linuxtv-commits@linuxtv.org
[13:21] <mchehab> gah
[13:21] <mchehab> it means either 2 things:
[13:21] <mchehab> 1) patchwork got unsubscribed from vger;
[13:21] <mchehab> 2) vger rejected the e-mails
[13:21] <mchehab> (it may also be due to disk full)
[13:22] <hverkuil> They arrived fine in linux-media.
[13:22] <mchehab> no, it is not disk full
[13:22] <hverkuil> So not 2.
[13:29] <mchehab> it seems that it is working properly
[13:32] <mchehab> this one is from today https://patchwork.linuxtv.org/patch/65372/
[13:32] <hverkuil> I don't see '[PATCH v3] media: videobuf2: Print videobuf2 buffer state by name', nor do I see my PR: '[GIT PULL FOR v5.9] ti-vpe cleanups'
[13:33] <hverkuil> '[GIT PULL FOR v5.9] Various fixes' is the last thing I see in patchwork.
[13:33] <hverkuil> https://patchwork.linuxtv.org/project/linux-media/list/
[13:34] <mchehab> I don't see:  videobuf2: Print videobuf2 buffer state by name
[13:34] <mchehab> on my inbox
[13:34] <mchehab> can you find it on lore?
[13:35] <mchehab> https://lore.kernel.org/linux-media/
[13:36] <mchehab> I can see it there
[13:36] <mchehab> but didn't arrived here yet
[13:36] <mchehab> forget it... local problem
[13:37] <mchehab> let me try to download it from lore and try to run the script which imports it
[13:38] <hverkuil> thanks!
[13:55] <mchehab> somehow, it is rejecting the e-mail from lore:
[13:55] <mchehab> cat raw |./manage.py parsemail | echo "Rejected."
[13:55] <mchehab> Rejected.
[13:56] <mchehab> knowing why patchwork is rejececting something is hard
[13:57] <mchehab> btw, I even upgraded to the latest stable version, just in case
[13:57] <hverkuil> Do you want me to redirect the posts to you? See if that works?
[13:58] <mchehab> v2.1.4 -> v2.2.2
[13:59] <mchehab> yeah, if you could send me a mbox file with all e-mails since the last merged patch (including it), it would be very helpful
[13:59] <mchehab> (or a maildir)
[13:59] <mchehab> the e-mails should be exactly identical to what patchwork is supposed to receive
[14:00] <mchehab> it works via .forward
[14:00] <mchehab> I'll add some prints at the python file which parses it
[14:00] <mchehab> (I used to have something like that in the past)
[14:00] <hverkuil> redirected 5 patches. Fingers crossed.
[14:01] <mchehab> just 5 files?
[14:01] <mchehab> I need not only the patches, but also their replies
[14:02] <mchehab> as it will ignore everything that it is not patches, it is usually easy to just place everything starting from a timestamp on a file
[14:02] <mchehab> at the mbox format
[14:02] <hverkuil> I'll add a bunch of replies too...
[14:03] <hverkuil> redirected everything that came after "[GIT PULL FOR v5.9] Various fixes"
[14:24] <mchehab> did you redirected them to patchwork@linuxtv.org?
[14:27] <mchehab> $ cat raw |./manage.py parsemail
[14:27] <mchehab> Parsing mail loaded from stdin
[14:27] <mchehab> Duplicate mail for message ID %s <20200709174336.79112-1-ezequiel@collabora.com>
[14:28] <mchehab> well, patch is there already:
[14:28] <mchehab> https://patchwork.linuxtv.org/project/linux-media/patch/20200709174336.79112-1-ezequiel@collabora.com/
[14:29] <mchehab> State   Changes Requested
[14:29] <mchehab> your comment e-mails were also caught by patchwork
[14:33] <mchehab> ok, just manually imported one patch
[14:33] <mchehab> with:
[14:33] <mchehab> wget https://lore.kernel.org/linux-media/20200710131813.452513-2-paul.elder@ideasonboard.com/raw && cat raw |./manage.py parsemail
[14:34] <mchehab> patch went fine on patchwork: https://patchwork.linuxtv.org/project/linux-media/patch/20200710131813.452513-2-paul.elder@ideasonboard.com/
[14:37] <mchehab> bouncing a patch to patchwork also worked fine
[14:37] <mchehab> https://patchwork.linuxtv.org/project/linux-media/patch/20200710131813.452513-1-paul.elder@ideasonboard.com/
[14:37] <mchehab> hverkuil: the problem seems to be between vger and patchwork@linuxtv.org
[14:38] <mchehab> btw, not sure what happened, but my mailbox received lots of duplicated e-mail on my media ML inbox
[14:38] <mchehab> maybe some issue on vger?
[14:42] <hverkuil> Eh, that might be my mistake. I redirected the posts to you, not to patchwork.linuxtv.org.
[14:56] <mchehab> **** Address already subscribed to linux-media
[14:57] <mchehab> vger has already it subscribed
[14:57] <mchehab> that's said, I found 2 issues on patchwork (besides the version upgrade):
[14:58] <mchehab> 1. the job that were supposed to answer e-mail were broken
[14:58] <mchehab> due to a bug at patchwork
[14:58] <mchehab> 2. there were some e-mails that were on hold at the server
[14:58] <mchehab> the server is set to take some time to handle mails
[14:58] <mchehab> in order to avoid some kinds of attacks
[15:00] <mchehab> anyway, both got fixed and the .forward rule restored
[15:00] <mchehab> It would be worth to rebounce those e-mails to patchwork@vger.kernel.org
[15:00] <mchehab> in case it lost something
[15:00] <mchehab> sending duplicate stuff won't hurt patchwork
[15:01] <mchehab> as it uses the message ID to avoid duplicaton
[15:03] <hverkuil> mchehab: you mean: rebounce to patchwork@linuxtv.org ? (instead of patchwork@vget.kernel.org)
[15:03] <mchehab> yes
[15:04] <mchehab> that's what I meant ;-)
[15:04] <mchehab> just noticed I wrote the wrong address....
[15:04] <mchehab> well, it is Friday ;-)
[15:04] * mchehab needs to rest
[15:04] <mchehab> this was a stressful week
[15:04] <hverkuil> Done.
[15:04] <mchehab> thank you!