Monday, 2016-08-15

*** tpb has joined #timvideos00:00
*** cr1901_modern has quit IRC03:05
*** cr1901_modern has joined #timvideos03:16
*** Bertl_oO is now known as Bertl_zZ04:27
mithrocr1901_modern: ping?05:13
mithrocr1901_modern: Have you use libusb under Windows?05:13
cr1901_modernmithro: Yes, you need to install libusbk for it to work b/c Windows is broken in this respect. Google Zadig for an installer.05:26
*** Evan has joined #timvideos06:09
*** Evan is now known as Guest2205406:10
Guest22054@cr1901_modern I attempted to get libusb running on windows but had problems. This is what I tried06:10
tpbTitle: Attempts at running on windows · Issue #37 · timvideos/HDMI2USB-mode-switch · GitHub (at
CarlFKmithro: do you have some time to help me figure out why modeswitch is throwing assert errors ?06:11
cr1901_modernGuest22054: I've never seen anything of the sort, believe it or not06:11
Guest22054What version of windows are you on?06:15
Guest22054or am I just installing it the wrong way cr1901_modern06:15
cr1901_modernI'm using 7, and not sure what the issue is; I've never used pyusb06:17
Guest22054pyusb requires libusb which there is a windows port of called libusb-win32 which won't install properlly on my system06:25
Guest22054supposedly its broken for windows 1006:25
*** Flandre has joined #timvideos06:51
*** Flandre is now known as evilRails06:53
*** Guest22054 has quit IRC07:05
CarlFKmithro: never mind. modeswitch is broken.08:51
CarlFKer, maybe not.  you haven't merged your pat08:53
CarlFKyes, it is so.08:56
CarlFKAssertionError: b'f' (b'f') != b'\x00' (b'\x00')08:56
CarlFK0900 f00f f00f f00f f00f 0000 6101 370008:56
CarlFK0009 0FF0 0FF0 0FF0 0FF0 0000 0108:56
evilRailsGood ole foof09:15
evilRailsHows the sprints goin?09:22
*** CarlFK has quit IRC10:06
*** puck has quit IRC10:10
*** puck has joined #timvideos10:10
*** Bertl_zZ is now known as Bertl11:13
*** Bertl is now known as Bertl_oO13:01
*** cr1901_modern1 has joined #timvideos14:11
*** cr1901_modern has quit IRC14:13
*** cr1901_modern1 is now known as cr1901_modern14:20
*** CarlFK has joined #timvideos15:24
*** ChanServ sets mode: +v CarlFK15:24
*** rohitksingh has joined #timvideos15:46
CarlFKmithro: fix modeswitch please15:56
*** evilRails has quit IRC16:14
*** rohitksingh has quit IRC17:07
*** rohitksingh has joined #timvideos17:22
*** Sajuuk has joined #timvideos19:59
Sajuukgood evening, I am wondering if there is anyone about who wouldn't mind having their mind picked regarding a revision 2.2 Numato Opsis board. Having trouble, i think, with talking to it via serial20:00
CarlFKSajuuk: hi.20:01
CarlFKyou are welcome to pick but I may not have much to offer.20:01
Sajuukso, i think i've been following the instructions on the HDMI2USB-misoc-firmware / scripts readme, and have gotten to the point were i'd like to be able to send commands to the board, but, the board doesn't present any serial devices to the OS20:02
Sajuukwould you happen to know any common things to check, or do, in this case420:02
CarlFKSajuuk: URL of readme?20:02
tpbTitle: GitHub - timvideos/HDMI2USB-misoc-firmware: A version of the HDMI2USB firmware based around the misoc+migen tools produced by m-labs. (at
CarlFKSajuuk: well.. that doesn't really have instructions20:04
Sajuukmy bad20:04
tpbTitle: HDMI2USB-misoc-firmware/scripts at master · timvideos/HDMI2USB-misoc-firmware · GitHub (at
CarlFKthats better.  this is what I call the developer path.20:05
Sajuuk... is this the path i should be taking? ;)20:06
Sajuukthere's a group of us pondering over this board atm20:06
CarlFKdo you want to hack on the code or use the device ?20:07
Sajuukuse the device20:07
CarlFKme too.20:07
CarlFKso I don't do that.20:08
Sajuukah. so where would i find the non-developer path....20:08
*** rohitksingh has quit IRC20:08
CarlFKhowever, in the last few days we have changed some things.. and now thats broken (I think)20:08
Sajuukoh dear20:08
CarlFKI just tracked down the problem 12 hours ago....20:09
CarlFKyou have an Opsis board?20:10
CarlFK(not Atlys prototype)20:10
Sajuukno, not an Atlys20:11
CarlFKok, it should power up and be functional without updating the firmware20:11
Sajuukindeeds, the owner believed it was, but we can't seem to figure it out now, and were attempting to use that documentation as our guide20:11
CarlFKboth of the hdmi outputs should send a 720p test pattern (color bars)20:12
CarlFKcan you confirm that works?20:13
Sajuukwe are setting that now :)20:13
CarlFKand again.. I am involved, but more as a user.  so things may come to an abrupt halt and I say "um.. wait for someone else to come on line"20:14
Sajuukah yes, we get the test bars from the hdmi outputs20:15
Sajuukthats fine :D20:15
Sajuukits a start at least20:15
CarlFKusb - what OS?20:16
CarlFKare you plugging it into20:16
Sajuukkubuntu...... something20:16
CarlFKif I say pastebin lsusb, you know what to do?20:17
Sajuuksure, one moment20:17
tpbTitle: USB IDs · timvideos/HDMI2USB Wiki · GitHub (at
* Sajuuk one will look at link20:20
Sajuukit would appear we have a hit on ixo-usb-jtag20:24
CarlFKoh right!20:25
* CarlFK crys20:25
CarlFKunplug usb20:25
Sajuukoh dear...20:25
Sajuukwhich one?20:25
CarlFKpower cycle20:25
CarlFKeither side20:25
CarlFKcomputer side so you aren't poking the Opsis board as much20:26
CarlFKpower off the opsis, wait 10 seconds (or some ridiculous amount of time in computer years)20:26
CarlFKpower on the the opsis, wait 30 seconds, yes 30...20:26
CarlFKthen plug in the usb cable.20:27
Sajuukwill recycle and wait that time :)20:27
CarlFKlsusb.  you should get different IDs20:27
CarlFKwhen the Opsis first powers up, it (some verb) a usb id  that somehow says "hey, if you need to fix me, now is the time!"20:28
CarlFKand on the linux box, a udev rule kicks in "ok, here we go!" and uploads some boot strappy whatever blob20:28
CarlFKand the Opsis says "thanks!  now you can fix me!" and waits for more firmwar20:29
CarlFKat least this is what I have concluded.20:29
Sajuukwell, we've done the cycle, and have gotten a different ID, although it doesn't fully match a list item20:30
Sajuuk2A19 - 544220:31
Sajuukoh! video0 has appeared in /dev!20:31
CarlFKyou should have a /dev/tty something... looking it up20:32
Sajuukawesome. At least thats a start, and we can skip faffing about with compiling firmwares :D20:34
Sajuukwe're being kicked out of our workspace atm, 2130 local time, so i'd liked to thank you for the help. If you happen to know of any documentation we should take a look at for the 'usage' route rather then the 'dev' route, that'd be a great help20:35
CarlFKSajuuk: well, I am the one that has written it ;)20:38
Sajuukokay, we'll take a look for that appearing20:38
CarlFKand really it is just my notes here and there that I try to put somewhere so I can find them later20:38
CarlFKand there are too many places to put such things, so .. chaos20:39
Sajuukyep, ttyACM0 has appeare20:39
CarlFKbest bet is to  hop in here20:39
CarlFKcat ~/.minirc.dfl20:39
CarlFKpu port             /dev/ttyACM020:39
CarlFKpu addcarreturn     Yes20:39
CarlFKcreate that 2 line file20:39
CarlFKapt install minicom20:40
CarlFKthat should get you an hdmi2usb> prompt (may need to hit enter)20:40
Sajuukah, so we can serial minicom over that.... 115200?20:40
CarlFKhelp will get you a screen full of help20:40
CarlFKsomething like that.  those are my 2 lines, everything else default20:41
Sajuukneat. Thank you very much!20:42
Sajuukthat's at least given us a great ending to the last 4hrs :D20:42
*** Bertl_oO is now known as Bertl_zZ22:29
mithroCarlFK: that issue with mode switch above looks like an endian issue. Can you confirm it happens on x86 to?22:47
CarlFKmithro: it does.  on my x86, it was working, git pull, got updates, now broken22:47
mithroCarlFK: well, tell me which commit broke it and it'll be much faster for me to fix22:48
CarlFKmithro: what is the git command to back up a rev?22:50
mithroGit reset --hard HEAD~123:03
CarlFKmithro: HEAD is now at fa2e997 Merge pull request #18 from mithro/master23:22
mithroCarlFK: hrm?23:23
CarlFKI think that is the last rev tat works23:23
mithroCarlFK: you can also just open gitk and do a reset23:23
mithroCarlFK: Okay, if it works on the commit before and not that one23:34
CarlFKmithro: no - it does work on that one, does not work on the one after it23:35
mithroCarlFK: okay - will fix shortly23:35
mithro_florent_: ping?23:46

Generated by 2.13.1 by Marius Gedminas - find it at!