Friday, 2016-06-24

*** tpb has joined #timvideos00:00
JoelwCarlFK: apt-get source linux-image-$(uname -r)00:04
CarlFKJoelw: thanks.. kinda.. Package linux-source-3.14.65-70 is not available, but is referred to by another package.00:05
CarlFKso I am trying to install 4.4 ... weee!!!!00:05
CarlFKrun-parts: executing /etc/kernel/header_postinst.d/dkms 4.4.0-27-generic /boot/vmlinuz-4.4.0-27-generic00:23
CarlFKlets see if it boots :)00:23
CarlFKI think I killed it...00:26
shenkiNeuron1k: morning00:43
shenkiNeuron1k: how did your hacking go?00:44
shenkiNeuron1k: i can't see any commits in your repository00:44
shenkiNeuron1k: i thought you were going to schedule a meeting with myself, mithro and ajit this morning? i don't have anyting in my calendar00:44
Neuron1kshenki: I am still working on it00:44
Neuron1kshenki: ok, I thought we will catch up on hangouts, but I will prepare meeting00:45
shenkiNeuron1k: okay00:48
shenkiNeuron1k: regarding the code, as we've discussed in the past, you want to be pushing a commit every day at least00:48
Neuron1kshenki: I will, now after exams I can truly focus, hope that my productivity will boost now00:50
shenkiNeuron1k: okay. the unfortunate part here is the summer of code is half through00:55
shenkiNeuron1k: but we can chat about that later. good luck with the hacking00:55
Neuron1kshenki: ok, thank you00:56
*** aom_ has joined #timvideos01:57
mithrohello andi-m03:04
mithrohello aom_03:04
aom_thank so much for comment of my blog. I cannot reply for your comment, because the comment system denied  my access.03:10
mithroaom_: After you have finished the work for the midterms you should look at fixing that03:24
*** pitball has left #timvideos03:35
aom_mithro: I got it.03:49
*** ssk1328 has joined #timvideos04:29
*** sb0 has joined #timvideos04:42
CarlFKmithro: do you remember saying that mplayer didn't play the uvc stream if the serial port was in use?04:51
mithroCarlFK: only straight after loading04:52
CarlFKmithro: I think I see that with gstreamer too.    but I haven't had time to really investigate04:54
CarlFKmithro: should I open an issue somewhere so we can collect clues ?04:54
*** sb0 has quit IRC05:05
mithroCarlFK: once you have started streaming, it seems to work fine05:10
mithroCarlFK: not really, the solution is to finish the fx2-refactor that I started mid last year05:11
CarlFKmithro: does this sound right:  a serial connection blocks the first streaming connection05:16
mithroCarlFK: yes05:19
*** sab_123 has joined #timvideos05:23
mithrosab_123: getting pretty close with that API pull request05:45
sab_123mithro, cool05:45
mithrosab_123: See how much faster things go when pull requests are small and well defined?05:45
sab_123mithro, yes you are right.05:45
sab_123mithro, maybe I should have started this exercise much much eariler05:46
sab_123mithro, it is cutting it too close now05:46
mithrosab_123: yes it is05:47
mithroBe back in 15 minutes05:49
sab_123mithro, ok05:49
mithrosab_123: When I say something like "let's remove the pin configuration stuff", you should remove all of it - not just the line I commented on06:19
sab_123mithro, okay so no pin configurations06:20
sab_123mithro,  understood06:20
mithrosab_123: for now06:20
mithroWe can come back to it later06:20
*** sb0 has joined #timvideos06:20
mithrosab_123: You don't need to send an "updated again" email. I get an email when you push new commits to the pull request.06:24
sab_123mithro, okay, I send that email in after pushing in all commits06:24
mithrosab_123: "It doesn't compile" isn't a helpful message06:28
sab_123mithro, ok06:29
mithrosab_123: please include the *full* output06:31
mithrosab_123: you should also wrap it in ``` so that github doesn't try and process it06:31
mithrosab_123: I can't see the code you are trying to compile, can you push it?06:36
mithrosab_123: what version of sdcc are you using?06:37
sab_1233.6 is just released06:38
sab_123it needs atleast one arg06:39
mithrosab_123: okay, make it take the baud rate?06:40
sab_123mithro, remove the set function?06:40
sab_123mithro, or keep it06:40
mithrosab_123: no06:40
mithrosab_123: keep it06:40
mithrosab_123: you also haven't said anything more about
tpbTitle: Created a UART API definition for allowing proper abstraction by RacingTornado · Pull Request #17 · mithro/fx2lib · GitHub (at
mithrosab_123: so, when designing an API, consistency is one of the biggest things you should aim for07:09
sab_123mithro, I agree. But lets say fast_uart has no receive. What would be the advantage of having someone call a function in an API which has a receive declared but nothing implemented? Would it not confuse the end user?07:10
mithrosab_123: You are thinking about just one use case, there are multiple different UART implementations, right?07:11
sab_123mithro, okay07:11
sab_123mithro, I will implement these functions in fast_uart , but return something which is not valid07:13
sab_123mithro, this way we can check. Or we should have a function like enable receive , which should return a value only if receive has been enabled. This way the end user wont call uartX_receive()07:14
mithrosab_123: well, for the fast uart, the "check anything available for receiving" is always going to return false..07:15
sab_123thats true07:15
sab_123mithro, can I start the fast_uart now?07:15
mithrosab_123: Yes, the API just needs one or two naming tweaks, but the actually function structure looks okay07:18
sab_123mithro, Let me start then.07:18
mithrosab_123: you can do the timer base uart at the same time too07:19
mithrosab_123: as they both end implement the same API07:19
mithrosab_123: but they should end up in seperate pull requests07:19
sab_123Let me get started then07:19
*** Bertl is now known as Bertl_zZ07:39
*** sb0 has quit IRC07:59
ssk1328mithro: My network just came back, ping me when you are free08:23
mithrossk1328: Have you pushed everything?08:25
ssk1328mithro: Not the csr multiply thing, it will do it now.08:27
*** sb0 has joined #timvideos08:28
*** sab_123 has quit IRC08:41
*** panther has joined #timvideos09:43
*** panther is now known as sab_12309:43
mithrossk1328: so....11:07
ssk1328mithro: I have pushed all my code.11:08
ssk1328mithro: Can have our weekly meeting?11:08
mithroHold on a minute11:15
mithrossk1328: First thing, looking at the heartbeat pull request, it doesn't look like you have responded to any of _florent_'s comments?11:28
ssk1328mithro: I tried to make those changes, had some issues so I though I might first discuss those things with you in the meeting11:30
mithrossk1328: you should get in the habit of discussing the changes in the channel here rather then waiting until our meeting11:31
ssk1328mithro: I wil make sure of this next time11:32
ssk1328mithro: I gots busy working on the other stuff this week11:32
mithrossk1328: specially since _florent_ made the comments, he is likely to be the person who can answer them.11:32
mithrossk1328: so were are we at with the RLE mask stuff?11:34
ssk1328mithro: So the python file I updated is in correct direction?11:35
mithrossk1328: you mean the pull request?11:36
ssk1328Yes, the most recent file11:36
mithrossk1328: Yes, did you see my comments?11:38
ssk1328Yes, will make the required changes11:39
ssk1328But the overall class and function definitions are correct now right?11:39
mithrossk1328: What do you mean?11:39
ssk1328My previous implementations, were implemented without defining Repeat class or Pixel class11:40
mithrossk1328: you seem to be heading down a better path now, but there is still a lot for you to finish here11:43
ssk1328mithro: Yes. Hopefully now I can complete this at a much better pace11:44
mithrossk1328: I assume you haven't done any of the C side of things yet?11:45
ssk1328mithro: No. I wrote some code before, but thats pretty different from what was required.11:46
mithrossk1328: okay11:48
mithrossk1328: were are we at with the float16 document?11:48
ssk1328mithro: The FPGA mapping is not really complete, I am not sure how exactly to describe that11:49
ssk1328mithro: For example for a float multiply, will have a 11bit multiplier and 5 bit adder as the major hardware components11:50
mithrossk1328: Does the Spartan 6 have anything like a 11bit multiplier?11:50
ssk1328mithro: It probably implements this using DSP blocks.11:51
mithrossk1328: The idea was to find out and document it11:51
ssk1328mithro: Understood now.11:52
ssk1328mithro: I think this shouldn't take much time to find out11:53
mithrossk1328: okay11:54
mithrossk1328: where are we at with the RGB conversion migen/misoc code?11:54
ssk1328mithro: I have completed the code and added required model function sin the test bench.11:55
mithrossk1328: and they work?11:55
ssk1328mithro: The code complies but the output is not correct, need to debug11:55
mithrossk1328: Code is not "complete" until it works correctly11:57
ssk1328mithro: Okay, my complete was in the sense that I had added all the required hardware components11:58
mithrossk1328: how do you know that if it doesn't work correctly?11:58
mithrossk1328: IE The reason it might not be working correctly is because you have forgotten  a required hardware component or something11:59
ssk1328mithro: The test setup is same as the other csc stuff11:59
ssk1328mithro: Curently the output is perfectly black image11:59
mithrossk1328: Can you send a pull request to litevideo with the conversion code in it for myself and _florent_ to review12:06
mithrossk1328: so what about the operation code?12:06
ssk1328Yeah. Will do that.12:06
ssk1328mithro: Still in very basic stage. I attempted doing the float multiply thing, I have to integrate it with CSR registers and access it from firmware12:08
mithrossk1328: okay12:08
mithrossk1328: so, still got quite a bit to do - I think you are going to be working quite hard over the next couple of days?12:09
ssk1328mithro: There are too many open things in the project right now. I think I should aim at completing one at a time perfectly12:09
ssk1328mithro: Yeah. I have been since the past 3-4 days as well.12:09
mithrosab_123: Did you see the comments on your second pull request12:10
sab_123mithro, yes12:10
sab_123mithro, Looking through it12:10
sab_123mithro, small question12:10
mithrosab_123: It looks like you are starting to head down the right path now12:10
sab_123mithro, SDCC implements printf() as small or large (depending on what you select)12:11
sab_123mithro, but for gets() , the manual seems to be absent.12:11
sab_123Can we write our own gets()?12:11
sab_123Or you want the one in SDCC12:11
mithrosab_123: you should be using the one in sdcc12:11
mithrosab_123: gets is part of stdio.h12:12
sab_123mithro, there are conflicts I guess12:13
ssk1328mithro: So are we doing the weekly video meeting? We have discussed most of the stuff already.12:13
mithrossk1328: It looks like you haven't gotten around to the weekly report this week either?12:13
ssk1328mithro: Yeah, I thought I should try finishing before that.12:14
ssk1328*finishing midterm tasks12:14
mithrossk1328: okay, I'll let you get on with stuff12:15
ssk1328mithro: I am adding some comments to the RLE repo, regarding doubts I have.12:16
mithroI have to go home now12:34
mithroI'll be around tomorrow12:37
*** Bertl_zZ is now known as Bertl_oO14:09
*** ssk1328 has quit IRC14:54
*** Kripton has quit IRC17:34
*** Kripton has joined #timvideos17:37
*** sb0 has quit IRC17:42
*** sab_123 has quit IRC17:52
*** Bertl_oO is now known as Bertl_zZ18:14
CarlFKtumbleweed: what is the chance of using the vizzini-dkms with a 3.14 kernel?19:07
CarlFKI am trying to get my odroid c2 to boot a kernel that has the source handy.  4.4 wont boot, and 3.14 I am having trouble finding the kernel source package.19:08
CarlFKif zzini-dkms with a 3.14 kernel is hard, then I'll give up on using the odroid this week.19:08
tumbleweedCarlFK: I have no idea19:10
tumbleweedwhy doesn't it work?19:10
CarlFKtumbleweed: cpu is an arm19:10
CarlFKPackage linux-source-3.14.65-70 is not available, but is referred to by another package.19:10
CarlFKpretty sure I need that in order to talk to the Atlys (I have 4 of them)19:11
CarlFKUnpacking linux-headers-3.14.65-70 (20160621) ...19:38
CarlFKmaybe that's enough...19:38
CarlFKgot it!! (with the help of a local friend)23:24

Generated by 2.13.1 by Marius Gedminas - find it at!