Wednesday, 2016-06-08

*** tpb has joined #timvideos00:00
*** aom has joined #timvideos00:37
mithrocr1901_modern: so, where is my frequency counter pull request? :-P00:58
*** Bertl_oO is now known as Bertl_zZ01:00
mithrotumbleweed: when is debcamp?01:00
mithrotumbleweed: I'm assuming you have a week(?) of debcamp before debconf and there will be work on getting recording in order during that?01:02
mithroCarlFK: are you using voctomix all the time now?01:03
CarlFKmithro: yep01:04
mithroHow do you have the HDMI2USB and your cameras connected to it?01:04
CarlFKHDMI2USB  -> eathernet -> core01:06
tpbTitle: voctomix-outcasts/ at master · CarlFK/voctomix-outcasts · GitHub (at
CarlFKhmm both sources use that...01:06
CarlFKblack magic hdmi card in a desktop on the core box01:06
CarlFKyou can kinda see the command I use to bring up all the vocto parts:
tpbTitle: dvsmon/ at master · CarlFK/dvsmon · GitHub (at
mithroCan you put the dates of your conferences into the TimVideos calendar?01:09
cr1901_modernmithro: pong. Not much happened last week. I added a status command to read the frequency counter, but wanted to talk to you about where that functionality should go01:13
cr1901_modernI'll work on finishing up the pull request tomorrow01:15
mithrocr1901_modern: so the firmware has a status command01:15
mithrocr1901_modern: I'd like the pixel clock to be printed as part of that01:15
mithrocr1901_modern: see
tpbTitle: Have frequency (and hsync/vsync) information on the inputs in the status message · Issue #94 · timvideos/HDMI2USB-misoc-firmware · GitHub (at
cr1901_modernmithro: That is already what I did :)01:16
cr1901_modernmithro: At this stage though, the frequency counter info that's printed out is global, and not tied to a specific input interface01:17
*** aom has quit IRC01:19
cr1901_modernThe "infrastructure's" there already however :). (I must say, it was tough finding a good spot to insert the new command lol)01:20
*** aom has joined #timvideos01:23
mithrocr1901_modern: so, frequency counter should be coming from the HDMI clock pins?02:05
cr1901_modernmithro: Yes, it should. But since there are commands dedicated JUST to query input interfaces, I thought of putting the command to query the frequency there02:09
mithrocr1901_modern: I just want the number in the status output as shown in that bug02:18
*** travis-ci has joined #timvideos02:19
travis-ci[mithro/HDMI2USB-misoc-firmware/fx2-refactor-old#203] (08f6b99): The build has errored. (
*** travis-ci has left #timvideos02:19
CarlFKmithro: did you see  "Open Camps will be the largest mission-driven open source conference in  the world,       with 6,000+ attendees joining for 25+ conferences & events,  featuring 400+ sessions on dozens of open source technologies.      "02:19
tpbTitle: Open Camps : Jul 8-17, 2016 (at
cr1901_modernmithro: Okay, very good.02:19
mithroIts always nerve wrecking deleting old git repositories02:24
CarlFKmithro: I think "open source" is the mission02:45
CarlFKmithro: I added NodePDX June 20 to my personal calendar and did "copy to Timvideos event" .. did that show up where you want it?02:46
*** travis-ci has joined #timvideos02:46
travis-ci[mithro/HDMI2USB-misoc-firmware/cdc-opsis-working#204] (e9d9e48): The build failed. (
*** travis-ci has left #timvideos02:46
mithroCarlFK: I thought for a moment it meant "mission" in the Christian sense02:50
*** rohitksingh_work has joined #timvideos04:32
*** aom has quit IRC04:59
*** sb0 has quit IRC05:05
*** sb0 has joined #timvideos05:19
*** nueces has quit IRC05:19
*** aom has joined #timvideos05:22
*** ssk1328 has joined #timvideos05:24
*** Bertl_zZ is now known as Bertl05:52
*** se6astian|away is now known as se6astian06:07
*** Bertl is now known as Bertl_oO06:36
mithrohey ssk132806:36
ssk1328mithro: Hi06:37
ssk1328I just saw your mail06:37
mithrossk1328: great06:37
mithrossk1328: as I mentioned, figuring out the exact goals is a negotiation, so please do speak up if you think I'm being unreasonable or if you have a better idea06:37
*** panther has joined #timvideos06:42
*** panther is now known as sab_12306:42
mithrosab_123: hey06:42
sab_123mithro, hi06:42
sab_123mithro, so although I cant see the data I send back on wireshark06:43
sab_123mithro, I see the URB status flag is succesful06:44
mithrosab_123: That means you are not sending the data back06:44
sab_123yep, I know something funny06:44
sab_123I just cant figure out what06:44
sab_123The documentation clearly states06:45
sab_123if using manual mode06:45
sab_123I just need to write the data into Epxx06:45
mithrosab_123: the manual for the FX2 is quite poor in areas06:45
mithroand blatantly wrong in other areas06:45
sab_123and arm the transfer by writing something in SUDPTRL06:45
sab_123how would you suggest I proceed?06:46
sab_123I see something equally strange in the main loop of my FX206:46
sab_123I know the second time I get a packet06:46
mithrosab_123: you have a serial console on the FX2 you can print information out06:47
sab_123a SUDAV interrupt is being raised06:47
sab_123Serial console screws around with the timing of everything06:47
mithrosab_123: the timing isn't actually all that important06:47
sab_123Right now I seem to have 2 problems06:48
mithroLinux is *very* lenient with you violating response times because there is a huge number of broken devices out there06:48
sab_1231- Cant send data back on EP006:48
sab_1232- I see my SUDAV interrupt is being entered into. However my main while loop is calling setup_handle_data06:49
sab_123is not06:49
sab_123If you have time I can show you on team viewer06:49
mithrosab_123: I don't have the time to do that06:50
sab_123mithro, no problems06:50
sab_123I get my interrupt which sets a variable06:50
mithrosab_123: you want a reliable serial port you can print things on so you can print out debugging information06:50
sab_123I have an LA06:51
sab_123I use that for everything06:51
sab_123I find it much easier to use06:51
sab_123Logic Analyser06:51
tpbTitle: fx2lib_serial_uart/bulkloop.c at i2c_interrupts · RacingTornado/fx2lib_serial_uart · GitHub (at
sab_123This line is being executed06:53
tpbTitle: fx2lib_serial_uart/bulkloop.c at i2c_interrupts · RacingTornado/fx2lib_serial_uart · GitHub (at
sab_123This isn ot06:53
sab_123For the second interrupt06:53
sab_123the second time I get the SUDAV interrupt06:53
sab_123Thats one06:55
sab_123The second is the data itself06:55
sab_123Do you think I should continue to investigate this06:55
sab_123Or move over to something else for the time being06:56
mithrosab_123: you should look at the SUDAV interrupt first06:57
sab_123it is being executed06:57
mithrosab_123: you really need to format this code better, it is very hard for me to read and thus I'm probably not going to have time to help06:58
sab_123mithro, do you mean the spaces?06:59
mithrosab_123: Things like indenting06:59
mithrosab_123: brace positioning, etc06:59
sab_123mithro, it is indented as far as I can see07:00
sab_123but I will take a look at it again07:00
tpbTitle: fx2lib_serial_uart/bulkloop.c at i2c_interrupts · RacingTornado/fx2lib_serial_uart · GitHub (at
tpbTitle: fx2lib_serial_uart/bulkloop.c at i2c_interrupts · RacingTornado/fx2lib_serial_uart · GitHub (at
sab_123mithro, okay got it07:00
sab_123I will find a way to indent these things07:00
*** se6astian is now known as se6astian|away07:00
mithrosab_123: you might want to take a look at the linux tool called "indent" to automatically fix that07:01
sab_123mithro, okay07:01
mithrosab_123: you should output a signal every time you go around the main loop07:08
sab_123i did07:08
sab_123it toggle07:08
sab_123I will debug a little more07:09
sab_123And when I run out of ideas07:09
sab_123i will get back to you07:09
sab_123mithro, indent is pretty cool07:09
sab_123I didnt know about it07:09
mithrosab_123: yeah, some projects like gstreamer have it as a precommit hook so it always means the code is indented correctly07:10
ssk1328mithro: I had a question regarding implementing RLE in python07:10
mithrosab_123: we should set up something like that on fx2lib eventually, but it requires fixing the indenting there first07:11
mithrossk1328: shoot07:11
ssk1328mithro: Generally RLE is like this
tpbTitle: Run-length encoding - Wikipedia, the free encyclopedia (at
ssk1328mithro: But what we want is to have a repeat like function, in which we can specify what has to be repeated how many times07:11
ssk1328As given in the hardware design docs07:12
tpbTitle: Hardware Fader Design Doc for the MiSoC Firmware - Google Docs (at
mithrossk1328: in the example on wikipedia you can think of there being an "implicit" repeat instruction before each of their numbers.07:13
ssk1328mithro: Yeah, for one single level they are essentially equivalent07:14
ssk1328mithro: Since we are aiming at multiple level of RLE ours will be a bit different07:14
mithrossk1328: yeah07:15
mithrossk1328: Let's start with a simple text format and then we can figure out a binary equivalent a bit later07:15
ssk1328So essentially we won't be storing some encoded 1 and 0 in our memory but high level C functions that do the equivalent job, right?07:16
mithrossk1328: hrm? Eventually we will need to store an encoded version in the memory, as the hardware mask block will need to understand it07:18
ssk1328mithro: That clears things a bit. So I dont have to focus on getting something like repeat function working, but I should be able to encode in my way of using 1s and 0s07:23
mithrossk1328: hrm?07:23
ssk1328mithro: I am not sure what hrm stands for!07:24
mithrossk1328: It's the sound you make when you don't understand "hrmmmmmm"07:25
ssk1328mithro: So what I was trying to go was to write a repeat function, which can be used in the way mention in hardware fader design doc07:27
ssk1328So what I was trying to say, that earlier I was trying to write a repeat function, which can be used in the way mention in hardware fader design doc07:27
mithrossk1328: Well, for converting a mask into a full size image in Python you will have to do something like that07:27
ssk1328mithro: But what got me confused was how can i generate something like this from image. ie image to mask. Now things are clear07:29
*** aom has quit IRC07:36
*** sab_123 has quit IRC07:51
mithroWell, that took fricken forever, but I think I have working udev rules now08:07
mithroCarlFK / tumbleweed:
tpbTitle: HDMI2USB-mode-switch/udev at master · timvideos/HDMI2USB-mode-switch · GitHub (at
mithrossk1328: ping?08:27
mithrossk1328: When you are happy you have fixed all the issues I mentioned in the pull request you should respond to each comment saying what you have done (Fixed, Change to XYZ instead, Removed, etc)08:28
tumbleweedmithro: yes
tpbTitle: DebConf16 (at
mithrotumbleweed: In my calendar now10:00
*** panther has joined #timvideos11:24
*** panther is now known as sab_12311:24
sab_123rohitksingh_work, hi11:53
rohitksingh_worksab_123: hi!12:15
*** se6astian|away is now known as se6astian12:23
*** ssk1328 has quit IRC12:43
sab_123Do you know how to relate the opcode with the USB field12:45
sab_123Do you know in which packet (field) the data will be sent12:45
sab_123section 3.312:45
rohitksingh_worksab_123: would have to grep through libmpsse and libftdi codes to find them out...must be in there...they both use libusb12:48
rohitksingh_worklet me start my VM and see if I can find them12:49
sab_123rohitksingh_work, I am looking at the USB traces and I cant relate them12:51
*** sb0 has quit IRC13:06
rohitksingh_worksab_123: this might be of some help jtag directly using libusb13:15
tpbTitle: jtag-mpsse/jtag-mpsse.c at master · swetland/jtag-mpsse · GitHub (at
sab_123rohitksingh_work, that looks interesting, but the problem is I am looking at the wireshark packets and I cant seem to relate13:16
rohitksingh_worksab_123: you'll have to check out those libusb functions and see how are they formatting the usb packets13:17
*** sb0 has joined #timvideos13:18
*** panther_ has joined #timvideos13:19
*** panther_ has joined #timvideos13:20
*** panther_ has joined #timvideos13:20
*** sab_123 has quit IRC13:22
*** panther_ is now known as sab_12313:23
sab_123rohitksingh_work, ok13:23
*** rohitksingh_work has quit IRC13:39
*** sab_123 has quit IRC13:53
*** CarlFK has quit IRC14:14
*** rohitksingh has joined #timvideos14:17
*** CarlFK has joined #timvideos14:19
*** ChanServ sets mode: +v CarlFK14:19
*** rohitksingh1 has joined #timvideos15:11
*** rohitksingh has quit IRC15:12
*** sb0 has quit IRC15:22
*** CarlFK has quit IRC16:20
*** CarlFK has joined #timvideos16:41
*** ChanServ sets mode: +v CarlFK16:41
*** se6astian is now known as se6astian|away16:47
*** nueces has joined #timvideos16:58
*** CarlFK has quit IRC17:00
*** CarlFK has joined #timvideos17:07
*** ChanServ sets mode: +v CarlFK17:07
*** se6astian|away is now known as se6astian17:58
*** rohitksingh1 has quit IRC18:12
*** rohitksingh has joined #timvideos18:15
*** rohitksingh has quit IRC18:45
*** rohitksingh has joined #timvideos18:46
*** rohitksingh has quit IRC18:59
*** rohitksingh has joined #timvideos19:00
*** rohitksingh has quit IRC19:20
*** nueces has quit IRC19:22
*** Bertl_oO is now known as Bertl_zZ19:42
*** CarlFK is now known as bri20:12
*** bri is now known as CarlFK20:13
*** f15h has joined #timvideos20:18
*** f15h has quit IRC20:28
*** se6astian has left #timvideos20:41
*** ssk1328 has joined #timvideos22:20
ssk1328mithro: Just added comments to the pull requests, will discuss the changes tomorrow22:21

Generated by 2.13.1 by Marius Gedminas - find it at!