<!-- 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>mchehab</u>: ping hverkuil: <u>pinchartl</u>: I have very little time, your best bet is Monday morning your time (mail me at what time you want to discuss it) <br> Today I'm leaving at 4pm my time, this weekend I am gone both evenings, on Tuesday I am gone in the evening and from Wednesday onwards I am on vacation. <br> Monday evening is the only evening I am free. posciak: do we by any chance have any details about the summit arrangements yet? <br> I was hoping to start reserving things hverkuil: <u>pinchartl</u>: actually, I might have time Tuesday morning your time. I don't have to leave until 7pm or so my time. <br> <u>posciak</u>: Mauro made an announcement on linux-media on September 1st. <br> Thursday morning the plan is to visit Samsung engineers. <br> workshop is on Monday. <br> I arrive on Sunday, depart on Friday. posciak: oh I must've missed that hverkuil: <u>pinchartl</u>: I mailed you with times (I realized that early morning on Monday/Tuesday works for me too). Please reply to the mail when we can discuss it. posciak: <u>hverkuil</u>: I guess we don't know any logistics apart from what you mentioned? <br> invitations/reservations/etc? javier__: <u>pinchartl</u>: yes, I know that branch. That's the one I was talking about when I refered to your patches <br> <u>pinchartl</u>: but that branch does not even build so I guess you used an older kernel to test those <br> <u>pinchartl</u>: I fixed the build issues, added async probe support to tvp5150 and also cherry-picked some DTS patches from the board vendor tree to add DT support for tvp5151 since platform data support has been removed from omap3isp <br> the tvp5150 is detected and the pad links created and I can enable the pad links and set the format but got the mismatch I mentioned on VIDIOC_STREAMON <br> but I'll dig deeper to see why that's happening hverkuil: <u>posciak</u>: Mauro is in charge of the invitations, and you're obviously invited (besides, I noticed that you have a topic that you want to discuss anyway) <br> Not sure what you mean with reservations. The hotel is the one the kernel summit is in. posciak: <u>hverkuil</u>: usually there were some invitations to register to the summit sent out for participants <br> on LF page <br> in any case, we have Monday for workshop, do we plan to do anything else? <br> like have more discussions on Tue, etc. <br> did you already reserve your hotel? this is via the LF page specia rate in Conrad? hverkuil: <u>posciak</u>: you're confusing two things: invites to the kernel summit and invites to the workshop. I'm talking about the second. <br> I'm not invited to the kernel summit (again :-( ), so I can't attend the Tuesday session. So Monday is the workshop, Wednesday is the open session of the kernel summit and Thursday is a meeting with Samsung. I don't know how many people can attend the Samsung meeting. <br> Mauro said it was a 1.5 hour drive, and I don't know how many cars/space there is. posciak: <u>hverkuil</u>: I thought Wed was kernel summit invitees only. hverkuil: Anyway, Tuesday is open for me, but I may have to work that day (very busy at work). <br> <u>posciak</u>: no, that's Tue. posciak: <u>hverkuil</u>: the "Register" page on kernel summit website says Wed, but I don't know <br> in any case <br> I'm not confusing workshop with core, my point was previously there were also invites to register for workshop attendees <br> and there was a shared day core+workshop attendees <br> maybe not this year <br> in any case, I wanted to coordinate with media people :) <br> to have more time to talk etc. <br> so was trying to figure out what everyone's plans were <br> and how to plan my hotel etc. <br> so looks like we have Mon, you are probably busy on Tue, and then Thu seems also taken by Samsung visit hverkuil: http://events.linuxfoundation.org/events/linux-kernel-summit/program/summit-agenda <br> That's what I am going with. <br> According to Mauro's email for the workshop all you need to do is to get your name on the google doc link, and your name is on it. posciak: <u>hverkuil</u>: this is what I was looking at: http://events.linuxfoundation.org/events/linux-kernel-summit/attend/register, but your link wins ;) hverkuil: Mauro will arrange the invite. I think that happened last year as well. <br> Ah, yes. There is some confusion there. Since I won't leave until Friday I'm good either way (Tue or Wed). <br> Send an email to the linuxfoundation that they should sync up those two links. <br> I didn't use the linux foundation website for the hotel reservation, I used a Cisco system (same price, and actually easier billing for me). posciak: thanks <br> I'm basically trying to figure out how long to stay, if people are availlable/want to talk on Tue/Wed, I'd stay longer, otherwise seems like I'd only be there for Mon <br> (and we had topics to discuss) hverkuil: On the day off I can always make time to talk to you. It's rare that we see each other in person, so it's more important to take that opportunity. posciak: ok so sounds like I'd probably prefer to stay Mon-Tue <br> yeah I have plenty of things to talk about :) hverkuil: If you want to take half a day (or more) to discuss topics, then I would suggest staying until Wed. Then we can talk Tue or Wed (depending on the actual summit schedule) posciak: I'm definitely very busy as well, but should be able to work from there too <br> ok <br> maybe I should leave Wed evening then hverkuil: This says that the invite only is on Wed: https://lwn.net/Articles/650226/ <br> I think it is more likely to be on Wed rather than Tue. <br> It makes no sense to have non-invites wait for a day before the open session starts. posciak: exactly <br> which also suggests Tue could have interesting sessions for us to attend as well potentially hverkuil: right <br> So leave on Thu, that way we'll have time to talk on Wed. posciak: good point <br> alright <br> that settles it <br> time to make reservations <br> oh there seems to be a reception on Sun hverkuil: I think I arrive around 3pm on Sunday, so I'll probably make the reception. <br> I've contacted the LF asking them to fix the webpage (whichever is wrong) posciak: thanks <br> I guess another question would be if workshop attendees are invited to the reception ;) <br> I'm gonna assume yes ;) hverkuil: I'm pretty sure they are. ***: benjiG has left <br> awalls has left <br> benjiG has left Nikhil_D: quick question: on 4.1 kernel, writing 9 to /sys/class/v4l/video0/dev_debug <br> doesn't print the fields of the structures passed to the ioctls <br> Got it, dev_debug is not a level, it has flags to be set <br> nvr mind pinchartl: <u>javier__</u>: ok, sorry about the misunderstanding <br> <u>javier__</u>: I think you need to configure the CCDC input to height / 2 when using bt.656 Nikhil_D: ping pinchartl <br> vb2_set_plane_payload is to be used by the driver to say how much memory it would use ***: prabhakarlad has left Nikhil_D: So should that include the data_offset also pinchartl: <u>Nikhil_D</u>: no, it's used by the driver to tell many bytes have been written in the plane <br> that's not clearly defined <br> I've posted a patch set a while ago to clarify (and repurpose) the data_offset field <br> I need to dig that up Nikhil_D: meaning internally, the V4L2 fw should set the bytesused = payload_size + data_offset pinchartl: we reached a consensus in the end but I haven't resent the patch <br> let me check Nikhil_D: Also, the verify_length should be called after the plane parameters are populated by qbuf_dmabuf, right?? pinchartl: <u>Nikhil_D</u>: http://www.spinics.net/lists/linux-api/msg09564.html ***: awalls has left <br> debris` has quit IRC (Ping timeout: 256 seconds) Reventlov: pp/2