Monday, 2017-07-24

*** tpb has joined #timvideos00:00
*** paradisaeidae_ has joined #timvideos00:19
CarlFK[m]ok, unboxed another Atlys - bars.    Now to figure out what version of firmware it is running00:21
*** paradisaeidae_ has quit IRC01:31
*** SamSagaZ has quit IRC01:34
*** SamSagaZ has joined #timvideos01:35
*** paradisaeidae_ has joined #timvideos01:49
*** paradisaeidae__ has joined #timvideos02:47
*** paradisaeidae_ has quit IRC02:49
*** paradisaeidae__ is now known as paradisaeidae02:49
*** rohitksingh_work has joined #timvideos03:33
mithroCarlFK[m]: So, I need to know the last firmware version to work on the Atlys and also more details about what is going on04:05
CarlFK[m]mithro (IRC): I can't figure out how to get a serial connection to the term to see version04:06
mithroCarlFK[m]: You connect to the exart / vizzini serial port?04:07
mithrominicom / flterm / etc?04:07
CarlFK[m]I am pretty sure I need to see something in dmesg that isn't there, so there is nothing to connect to04:07
CarlFK[m]Bus 003 Device 004: ID 04e2:1410 Exar Corp. XR21V1410 USB-UART IC04:08
CarlFK[m]Bus 003 Device 003: ID 1443:0007 Digilent Development board JTAG04:08
CarlFK[m]that may also be a problem04:08
CarlFK[m]minicom: cannot open /dev/hdmi2usb/by-num/all0/tty0: No such file or directory04:08
CarlFK[m]mithro (IRC): also, how hard would it be for you to start a new build where pattern is the default output?04:10
CarlFK[m]that might help my testing04:10
mithroBus 003 Device 004: ID 04e2:1410 Exar Corp. XR21V1410 USB-UART IC <- Exar / Vizzini USB UART04:11
mithroApparently there isn't a T in Exar04:11
*** paradisaeidae has quit IRC04:16
CarlFK[m]mithro (IRC): to get the serial port, do I need to load ./v0.0.2-81-gd0d3aea/atlys/hdmi2usb/hdmi2usb.hex04:16
CarlFK[m](or some version04:16
mithroCarlFK[m]: Not on the Atlys04:16
CarlFK[m]mithro (IRC): to get the console?04:17
mithroCarlFK[m]: Give me 3 things you attempted to do before I tell you.04:17
CarlFK[m]lusub, dmesg |grep vix, minicom04:18
*** paradisaeidae_ has joined #timvideos04:19
mithroCarlFK[m]: Did you try unplugging the USB UART, run dmesg -wH and then plugging it back in?04:19
CarlFK[m]do you want to know the 400 things I have done for the last 14 hours? ;)04:20
CarlFK[m]most of that time was spent expecting to see color bars04:20
mithroCarlFK[m]: Can you do what I suggest above?04:21
CarlFK[m]3 hours ago I figured out the back screen might be because no input04:21
CarlFK[m]I don't know which is witch, ok to un/plug both?04:21
mithroCarlFK[m]: One connector is labeled something like PROG04:22
mithroCarlFK[m]: The other is labeled something like UART04:22
* CarlFK[m] sent a long message: CarlFK[m]_2017-07-24_04:22:48.txt <>04:23
mithroHrm... cdc_xr_usb_serial -- I'm not sure what module that is - shenki is that your module?04:24
CarlFK[m][email protected]:~$ lsmod | grep viz04:26
CarlFK[m]vizzini                36864  004:26
CarlFK[m][    8.459876] vizzini: module verification failed: signature and/or required key missing - tainting kernel04:26
CarlFK[m][    8.460090] usbcore: registered new interface driver vizzini04:26
CarlFK[m][    8.460091] vizzini: Exar USB UART Driver for XR21V141x04:26
mithroCarlFK[m]: That is the right driver....04:26
mithroIt should echo some dev name under /dev/ ?04:26
mithroCarlFK[m]: Can you see if cdc_xr_usb_serial is in lsmod?04:26
CarlFK[m]I was expecting to see viz in dmesg when I plugged it in04:27
CarlFK[m]lsmod |grep cdc_xr_usb_serial - nothing04:27
mithroCarlFK[m]: That is what I am expecting too....04:27
mithroCarlFK[m]: But maybe the vizzini module outputs messages with other stupid names - it is very crap04:27
Ishan_Bansalmithro : ping04:36
mithroIshan_Bansal: pong04:36
mithroCarlFK[m]: So, any messages about something in /dev being created when you plugged in?04:37
mithroCarlFK[m]: Also what machine are you doing this on?04:37
CarlFK[m]mithro (IRC): ubuntu 16.4 - minimal install - about the same as the base install for lca04:41
mithroCarlFK[m]: We didn't use any Atlys' at LCA IIRC?04:41
mithroCarlFK[m]: but not your debian thing?04:41
CarlFK[m]the two are very similar04:42
CarlFK[m]all the same repo/set of files, some if/then where needed04:43
mithroCarlFK[m]: I'm more suspect about udev rules / modules that your debian / ansible setup might be installing04:43
Ishan_Bansalmithro, I am currently working over the huffman and the test bench and I am getting the following error :04:44
Ishan_Bansalbut the sink is been defined in the Huffman as follows:04:44
tpbTitle: Traceback (most recent call last): File "", line 62, in (at
Ishan_BansalI am not able to get the problem. For reference the code for the test bench is as follows:04:44
tpbTitle: [Python] class Huffman(PipelinedActor,Module): def __init__(self): - (at
tpbTitle: [Python] from litex.gen import * from import * from lit - (at
mithroIshan_Bansal: Can you give me links to things on github rather than pastebins?04:45
CarlFK[m]mithro (IRC): I am worried about that too04:45
CarlFK[m]mithro (IRC): I didnt install the udev package we made, but something like openocd may have installed its own04:45
mithroCarlFK[m]: So, a clean ubuntu 16.04 install with just HDMI2USB-mode-switch installed would be the best test set up04:46
mithroCarlFK[m]: If you had the development environment setup it would be 30-40 seconds to generate your own firmware which had the test pattern as the default setup (you wouldn't need the Xilinx toolchain or to regenerate the gateware)04:47
CarlFK[m]mithro (IRC): and vizzini-dkms , and minicom.. right ?04:47
mithroCarlFK[m]: don't even bother with minicom, use flterm04:47
CarlFK[m]build-essential posh fxload libftdi-dev04:48
mithroYou can install flterm via "conda install flterm" inside the HDMI2USB-mode-switch environment04:48
CarlFK[m]  that shows the things I have done to a clean install04:48
tpbTitle: Home · CarlFK/HDMI2USB-mode-switch Wiki · GitHub (at
mithroCarlFK[m]: Which part?04:49
mithroCarlFK[m]: I'm working on making it so that you can use the prebuilt gateware while hacking on the C code firmware -- which should mean you can do lot of HDMI2USB development / customisation without needing the Xilinx toolchain04:49
mithroI have to run now04:49
CarlFK[m]the 2 july 2017 bit - stop at June 30 201704:49
mithrowill be back in an hour04:49
mithroCarlFK[m]: That can't be true as you can't "apt-get install vizzini-dkms" unless you added the timvideos FPGA repo and that isn't in the list of things you have done?04:51
CarlFK[m]good catch - I added the sudo apt install -y software-properties-common ... but not the ppa bit04:52
Ishan_BansalCode for the :
Ishan_Bansaland the testBench is :05:00
tpbTitle: litejpeg/ at huffman · ishan98/litejpeg · GitHub (at
tpbTitle: litejpeg/ at huffman · ishan98/litejpeg · GitHub (at
* CarlFK[m] uploaded an image: file1500872660400.jpg (140KB) <>05:05
CarlFKjea - did you open an issue about "video_mode list" hanging?05:39
CarlFKc2 that tumbleweed built last year: [ 3438.269161] vizzini 1-1.4.1:1.0: ttyVIZ0: XR21v14x usb uart device05:49
CarlFKthis gets testing status:   git describe: v0.0.3-710-g947321b-dirty05:51
mithroBack now06:19
mithroCarlFK[m]: So yeah, something like /dev/ttyVIZ0 is the serial port you should be connecting too06:20
mithroCarlFK[m]: I assume you should be asleep now, however?06:20
CarlFKmithro: na, i have at least an hour left in me06:20
CarlFKmithro: I have a c2 hooked up that I can connect to the console - so that verifies the atlys is working06:21
mithroCarlFK: so it was just that the default output is black?06:21
CarlFKwel, thats is one problem06:21
CarlFKdon't know why the /dev/ttyVIZ0 isn't coming up06:22
CarlFKon my fresh ubuntu box06:22
CarlFKmithro: here is my current problem: ~/HDMI2USB-firmware-prebuilt/archive/master/v0.0.3-730-gf9c031d/atlys/hdmi2usb/lm32$ hdmi2usb-mode-switch -v --flash-gateware flash.bin06:25
CarlFKmove usb to c206:25
mithroCarlFK: Well, that doesn't involved the HDMI2USB firmware at all - that is entirely the vizzini-dkms stuff06:25
CarlFK[email protected]:~$ flterm --port /dev/ttyVIZ006:25
CarlFK[FLTERM] Starting...06:25
CarlFKhitting enter, etc.. no response06:25
mithroCarlFK: Can you hit the reset button?06:26
mithroWhile keeping the flterm connected06:27
CarlFKmithro: reset - nothing.  including the leds don't go off06:27
mithroOkay, what do you think is the reset button? :-P06:27
CarlFKthe one that reboots my other atlys06:27
mithroThat doesn't tell me which - I seem to recall a little red button labelled something like MRST?06:28
* CarlFK[m] uploaded an image: file1500877724308.jpg (25KB) <>06:29
mithroCarlFK: welp, that seems rather likely to be the right one06:30
mithroCarlFK: you seem to have a working Atlys that has v0.0.3-710-g947321b-dirty ?06:32
CarlFKmithro: I haven't streamed yet, but I can get color bars and connection to it06:33
mithroCarlFK: Can you hit the same button on that board and see what you get on the serial console?06:34
*** paradisaeidae_ has quit IRC06:34
CarlFKmithro: it reboots - lots of stuff.. do you need the details?06:34
mithroCan you pastebin it?06:34
tpbTitle: Ubuntu Pastebin (at
mithroCarlFK: that is rather interesting...06:40
Ishan_Bansalmithro : I have send you the links of the huffman see once when you are free.06:40
mithroCarlFK: That doesn't have the HDMI2USB firmware on it, it has the "VideoSoC" on it -- which is all of HDMI2USB without the USB/JPEG encoder....06:42
CarlFKmithro: I think that is the result of me flashing random tings expecting to see color bars06:43
Ishan_Bansalmithro : got that, Actually I got two functions by the Huffman which is creating that problem. :)06:50
CarlFKmithro: that should have come from HDMI2USB-firmware-prebuilt/archive/master/v0.0.3-710-g947321b/atlys/video/lm32/flash.bin  right?06:54
CarlFKmithro: I am trying to flash that onto a 2nd atlys, still getting black/nothing on console06:55
CarlFKmithro: I can't see to replicate it07:00
mithroNo - that is "video"07:03
mithroCarlFK: You want the "hdmi2usb" directory07:04
mithroOh wait07:04
mithroSorry - yes it is likely that working atlys was flashed from that flash.bin file you linked (assuming the versions match exactly)07:04
mithroCarlFK: So that binary works on one Atlys but not another?07:06
mithroCarlFK: Do you have something that you have flashed on this Atlys that works?07:07
CarlFKmithro: seems that binary doesn't work.  not sure this is working07:09
mithroCarlFK: What do you mean?07:09
CarlFK(02:06:17 AM) mithro: CarlFK: So that binary works on one Atlys but not another?07:09
CarlFK(02:07:03 AM) mithro: CarlFK: Do you have something that you have flashed on this Atlys that works?07:09
mithroLet me start again07:09
mithroYou have two Atlys boards.07:10
mithro (a) One Atlys board has something like master/v0.0.3-710-g947321b/atlys/video/lm32/flash.bin on it (according to the version info you can get from serial) but you are unsure how you got that Atlys into that state.07:10
mithro (b) One Atlys board you tried to flash what we think (a) has on it and but you still get nothing via the serial port.07:11
mithroWith (b), was it working / doing anything before you start flashing it with things?07:12
CarlFKI don't know.  I haven't used it at a show (no mounting hardware attached to it's feet)07:13
CarlFKoh!!! month or so ago (April maybe?) I setup a box for you? and someone else to try somethhing ?07:14
mithroCarlFK: No idea?07:15
CarlFKI remember asking if you wanted me to leave it in place - that might have been the board that is flashed07:15
mithroCarlFK: I think maybe flashing the Digilent firmware that the Atlys boards ship with and seeing if that works is a potential idea....07:15
mithroCarlFK: can you give me the output of your attempt to flash the video/lm32 to (b) above?07:16
mithroCarlFK: And did you power cycle the Atlys after "flashing" that?07:16
CarlFKpower cycle - no.07:18
CarlFKdid now.. and I have version   git describe: v0.0.3-730-gf9c031d-dirty07:18
CarlFKthats the latest in ~/HDMI2USB-firmware-prebuilt/archive/master07:19
CarlFKH2U 00:00:49>x c p 007:20
CarlFKConnecting pattern to output0   - patter on my lcd07:20
mithroI'm confused....07:21
CarlFKmithro: I have not been power cycling07:21
mithroCarlFK: You saying that things are now working after power cycling?07:21
CarlFKbut just target: video07:22
CarlFKlet me try hdmi2usb07:22
mithroI was sure I had added "power cycle" to the end of my instructions - but it appears I also failed at doing that...07:25
CarlFK        target: hdmi2usb07:28
CarlFK  git describe: v0.0.3-730-gf9c031d-dirty07:28
mithroCarlFK: Getting console output is a good initial starting point07:30
mithroCarlFK: Do you get video output is the next one07:30
mithroCarlFK: Then do you get video input07:30
mithroI'm going to disappear in a couple of minutes again07:31
CarlFKout yes07:32
CarlFKchecing in now..07:32
CarlFKmithro: in and out work07:33
mithroCarlFK: so it works?07:34
CarlFKmithro: seem so - should I try the encoder over usb or do I need the .hex for that07:35
CarlFK[m]mithro (IRC):  hdmi2usb-mode-switch -v --load-fx2-firmware ...     assert len(exart_uarts) == len(atlys_boards)08:14
tpbTitle: Ubuntu Pastebin (at
Ishan_Bansalmithro : is their any way to use a signal as an index of another signal ?10:10
*** rohitksingh_work has quit IRC12:41
*** rohitksingh has joined #timvideos13:31
*** rohitksingh has quit IRC13:57
*** CarlFK has quit IRC14:14
*** rohitksingh has joined #timvideos14:52
*** rohitksingh has quit IRC16:59
*** CarlFK has joined #timvideos17:48
*** ChanServ sets mode: +v CarlFK17:48
*** CarlFK has quit IRC19:23
*** CarlFK has joined #timvideos19:33
*** ChanServ sets mode: +v CarlFK19:33
*** CarlFK has quit IRC19:58
*** CarlFK has joined #timvideos20:13
*** ChanServ sets mode: +v CarlFK20:13
paddatrappermithro: can we push our meeting this week out 30 minutes? I have to work until 10:20 on Thursday21:23

Generated by 2.13.1 by Marius Gedminas - find it at!