Tuesday, 2018-01-16

*** tpb has joined #timvideos00:00
*** ewen has joined #timvideos00:03
*** rohitksingh has joined #timvideos00:06
rohitksinghstefanor: would like to check out that issue where we need to press enter twice?00:13
rohitksinghstefanor: And Good Morning! :)00:14
tumbleweedrohitksingh: yes please00:23
*** rohitksingh1 has joined #timvideos00:30
*** rohitksingh has quit IRC00:31
xfxfLCA AV people: folder here with all AV stuff in it, recommend starring00:32
tpbTitle: Meet Google Drive – One place for all your files (at drive.google.com)00:32
xfxfrails, williammckee: https://docs.google.com/spreadsheets/d/1Jt7R5oaXLjHoTGEqCQQbrDFclXdpKUPhu4zpevWTBtw/edit#gid=000:32
tpbTitle: Google Sheets - create and edit spreadsheets online, for free. (at docs.google.com)00:32
xfxfthey're the kits we need to build00:32
tpbTitle: LCA2018: AV Room Inventory - Google Sheets (at drive.google.com)00:33
tumbleweedrohitksingh1: it's up00:38
rohitksingh1_florent_: hi! got 5GT/s but only x1 lanes in netv2mvp00:46
CarlFKjea: http://veyepar.ndv.cdn.nextdayvideo.com/veyepar/lca/lca2018/titles/Sketching_in_hardware_examples_of_getting_from_concept_to_installation_in_the_Internet_of_Things.png00:48
*** rohitksingh has joined #timvideos01:03
*** rohitksingh1 has quit IRC01:04
rohitksinghfelix_: https://www.dropbox.com/s/zshmrftdp5o35zt/JESD204B-01_Annotated.pdf?dl=001:09
tpbTitle: Dropbox - JESD204B-01_Annotated.pdf (at www.dropbox.com)01:09
*** sb0 has quit IRC01:15
ewenmithro: You mentioned that you had a list of Xilinx Vivado modules that needed to be installed to make batch builds work?  Do you have that in some text-file form?01:22
mithroewen: Hrm?01:31
mithroewen: We have a very cut back Xilinx package which we use on CI01:31
ewenmithro: Batch install of Vivado has options like:01:31
ewenModules=DocNav:0,Kintex UltraScale:0,Virtex UltraScale+ HBM ES:0,Spartan-7:1,Artix-7:1,Kintex UltraScale+ ES:0,ARM Cortex-A53:1,Zynq UltraScale+ MPSoC:1,Zynq-7000:1,SDK Core Tools:1,ARM Cortex-A9:1,ARM Cortex R5:1,Virtex UltraScale+ ES:0,Zynq UltraScale+ MPSoC ES:0,System Generator for DSP:0,Kintex-7:1,Kintex UltraScale+:1,MicroBlaze:101:31
ewenTrying to figure out which ones of those can be safely turned off...01:31
mithroWe only care about Artix-7 at the moment, possibly Spartan-7 and Kintex-7 will be in the near future -- ultrascale+ARM+Zynq stuff in the distant future01:32
ewenOkay.  I'll randomly try turning off everything else and see if that creates a config file it'll accept to install.01:34
*** futarisIRCcloud has joined #timvideos01:35
*** samsagaz has joined #timvideos01:46
mithrofelix_: Try taking a look at https://github.com/timvideos/HDMI2USB-litex-firmware/blob/master/make.py -- https://github.com/timvideos/HDMI2USB-litex-firmware/blob/master/make.py#L126 -- builder.add_software_package("firmware", "{}/firmware".format(os.getcwd()))01:48
tpbTitle: HDMI2USB-litex-firmware/make.py at master · timvideos/HDMI2USB-litex-firmware · GitHub (at github.com)01:48
futarisIRCcloudmithro: What did you want a RPi for (yesterday) ? Did you get one?01:51
mithroTo test with the NeTV2v2 stuff01:51
mithrofutarisIRCcloud: If you could bring one, that would be awesome01:51
mithrofutarisIRCcloud: Apparently, past Tim sent you an Arty?01:52
futarisIRCcloudI've got the Arty.01:52
mithrofutarisIRCcloud: Do you have any time to work on Linux stuff before the weekend?01:54
futarisIRCcloudFor RPi, I've got a RPi Zero (1.2), RPi Zero W and a Raspberry Pi 1 (Model B / 256 MB).01:54
futarisIRCcloudProbably only a few hours here and there. What are we looking to do?01:55
mithrofutarisIRCcloud: Trying to improve the drivers which are supported01:55
mithroI know shenki was working on Ethernet01:55
futarisIRCcloudNB, https://core-electronics.com.au/raspberry-pi.html is good, if you want any RPi related stuff in Australia (quickly).01:56
tpbTitle: Raspberry Pi Australia - Express Delivery (at core-electronics.com.au)01:56
xfxfLCA AV team: being familiar with this doc is recommended, it lists caveats and our packup/packdown schedule:02:06
tpbTitle: LCA: Schedule, Caveats/Notes, Team - Google Docs (at docs.google.com)02:06
*** sb0 has joined #timvideos02:15
mithrofutarisIRCcloud: Okay02:26
mithrofutarisIRCcloud: So, it would be good to verify the status of all the Linux drivers02:27
futarisIRCcloudOk. I'll try and test it all today. Anything beyond the default scripts / checkouts that I should be using?02:28
felix_meh, i waited for the people start walking to some place for lunch and now they're gone02:31
ewenmithro: upy-fpga GitHub organisation renamed to fupy, apparently successfully02:45
ewenLots of references to update.02:46
mithroewen: \o/02:46
mithroewen: Is any forwarding going on02:46
mithrofutarisIRCcloud: Default is probably the best to start with02:46
ewenmithro: new website at: https://fupy.github.io/02:47
tpbTitle: Welcome to FPGA MicroPython (FuPy) | fupy.github.io (at fupy.github.io)02:47
ewenmithro: In theory some things get auto-forwarded.  But, eg, GitHub Pages just went 404 immediately AFAICT.02:48
mithroI think technically the u should be a mu, right?02:48
mithroewen: Can we leave something at the old name which does the forward?02:48
ewenThe best idea I have so far is to rename the fupy-placeholder to upy-fpga (ie the old name) and put some basic redirection details there.02:48
ewenAnd yes, mu, if I can figure out how to encode that correctly for, eg, the website.02:49
ewen(I'm assuming we're not getting mu into the URLs...)02:49
rohitksinghmithro: sent you mail02:50
mithroewen: Okay02:52
ewenmithro: UTF-8 literal added to title of https://fupy.github.io/02:54
tpbTitle: Welcome to FPGA MicroPython (FμPy) | fupy.github.io (at fupy.github.io)02:54
mithroewen: \o/ :-P02:55
ewenmithro: According to https://help.github.com/articles/renaming-an-organization/ there are auto-redirects for _repositories_ from old-org to new-org, but not for the _organisation_name_ itself.  So we might want to put something back in the old organisation name as a redirector.02:56
tpbTitle: Renaming an organization - User Documentation (at help.github.com)02:56
mithroewen: Sounds good to me!03:26
CarlFK[m]mithro (IRC): gsoc app03:50
mithrofutarisIRCcloud: https://docs.google.com/spreadsheets/d/1LCu33Bi37fgG0SrRTf5dNqOujSBeWO_mXWHYyxhcWVM/edit#gid=003:53
tpbTitle: Google Sheets - create and edit spreadsheets online, for free. (at docs.google.com)03:53
ewenmithro: fupy placholder renamed to old upy-fpga name, with trivial website to redirect users: https://upy-fpga.github.io/04:25
tpbTitle: MicroPython on FPGA has moved! | upy-fpga.github.io (at upy-fpga.github.io)04:25
*** rohitksingh has quit IRC04:31
*** rohitksingh has joined #timvideos04:51
mithroewen: thank you so much!04:52
cr1901_modernmithro: Current status... debugging my example_design for mimasv2 in the litesdcard repo. For some reason doesn't want to cooperate04:58
cr1901_modernprobably something dumb on my end04:58
cr1901_modernI wonder if the PLL isn't locking, b/c not even the LED is functioning w/ the current bitstream05:06
cr1901_modern(that would be _weird_/unexpected, but it would be an explanation)05:06
mithrocr1901_modern: oh, I was also going to suggest that maybe we should get sdcard via SPI emulation with bitbanging mode going05:09
cr1901_modernmithro: Alternatively, it could be that I forgot to connect the clock to a BUFG :P05:11
cr1901_modernor at all05:11
CarlFK[m]mithro (IRC): the firmware docs are blessed!05:16
*** kmehall has joined #timvideos05:27
cr1901_modernmithro: So the litesdcard stuff works on mimasv2 using an example design. Which means now I have to progressively add stuff in to figure out why it doesn't work in litex-buildenv05:30
cr1901_modern(well at the very least, my card is detected. I get a CRC error/Timeout on SET_BLOCK_COUNT, but the BIST passes w/ 0 errors anyway. _florent_, is this normal?05:31
mithroI would first check that the mimasv2 works under the environment first...05:31
cr1901_modernwhat do you mean "under the environment?"05:32
tumbleweedjea: https://github.com/voc/voctomix/blob/master/voctocore/default-config.ini#L68-L9005:33
tpbTitle: voctomix/default-config.ini at master · voc/voctomix · GitHub (at github.com)05:33
tumbleweedolasd: do we have the background images we made for minidc cambridge anywhere?05:34
*** rohitksingh-demo has joined #timvideos05:39
*** rohitksingh-demo has quit IRC05:41
CarlFK[m]mithro (IRC): flashed  v0.0.4-72-g6e326e3 - flterm doesn't connect ... wanna come look?05:41
*** sb0 has quit IRC06:10
*** rohitksingh-demo has joined #timvideos06:15
*** rohitksingh-demo has quit IRC06:15
*** rohitksingh-demo has joined #timvideos06:15
*** rohitksingh-demo has quit IRC06:18
*** rohitksingh-demo has joined #timvideos06:20
*** rohitksingh-demo has quit IRC06:24
*** rohitksingh-demo has joined #timvideos06:24
*** rohitksingh-demo has quit IRC06:25
*** sb0 has joined #timvideos06:59
tpbTitle: Issues · timvideos/HDMI2USB-litex-firmware · GitHub (at github.com)07:17
*** futarisIRCcloud has quit IRC07:25
jearails: CB11.00.401 (tutorials) -> CB11.04.400, then CB11.00.100 (talks) -> CB11.00.40107:32
*** CarlFK has quit IRC07:33
*** futarisIRCcloud has joined #timvideos08:16
*** ewen has quit IRC08:23
mithrocr1901_modern: btw -> https://sigrok.org/gitweb/?p=libsigrokdecode.git;a=tree;f=decoders/sdcard_sd;h=1deeb7a5bb2228bb1e95b2abafb309033b3a289f;hb=HEAD and -> https://sigrok.org/gitweb/?p=libsigrokdecode.git;a=tree;f=decoders/sdcard_sd;h=1deeb7a5bb2228bb1e95b2abafb309033b3a289f;hb=HEAD08:30
tpbTitle: sigrok.org Git - libsigrokdecode.git/tree - decoders/sdcard_sd/ (at sigrok.org)08:30
cr1901_modernmithro: Yes, that was my next idea. After I figure out how to reduce bram usage08:30
cr1901_modern_florent_/mithro: Something bothers me about litesdcard. According to ISE, litesdcard on its own takes up _more_ resources than lm3208:31
cr1901_modernThat... doesn't sound right08:31
mithrocr1901_modern: I don't think it has be optimised at all?08:31
cr1901_modernEven an unoptimized design I am very surprised by this (but I haven't deeply checked either)08:32
cr1901_modernmithro: I'm making progress though. On the example design with _just_ the CPU attached, we get past sdcard_init(). Whereas in litex_buildenv we crash there08:33
mithrocr1901_modern: It's easy to do silly things which ends up consuming a lot of resources :-P08:33
cr1901_modernmithro: Fair enough. I recently had a large amount of BRAM be allocated as distributed b/c I forgot a signal.08:33
cr1901_modern_florent_: https://hastebin.com/esehevuhak.go I have an interesting failure... just thought I'd post it here while I look through litesdcard firmware for clues08:46
cr1901_modern(this is on mimasv2)08:46
*** rohitksingh-demo has joined #timvideos08:51
cr1901_modernThe C firmware and the litex_server firmware are not semantically equivalent08:56
rohitksingh-demomithro, _florent_: Can't get kernel module to compile.09:27
rohitksingh-democomplains about DMA_WRITER_INTERRUPT undefined'09:28
rohitksingh-demoshould have been more clearer. I meant the kernel module in netv2-soc repp09:30
rohitksingh-demomithro: could test the test/test_identifier.py either...complains about missing csr.csv which indeed isn't being generated09:33
*** rohitksingh-demo has quit IRC09:37
*** rohitksingh-demo has joined #timvideos10:01
*** rohitksingh-demo has quit IRC10:05
*** rohitksingh has quit IRC10:10
*** rohitksingh-demo has joined #timvideos10:16
*** rohitksingh-demo has joined #timvideos10:19
*** rohitksingh-demo has quit IRC10:20
*** rohitksingh-demo has joined #timvideos10:21
*** paddatrapper has quit IRC10:24
*** futarisIRCcloud has quit IRC10:25
*** paddatrapper has joined #timvideos10:27
*** futarisIRCcloud has joined #timvideos10:50
*** rohitksingh-demo has quit IRC10:54
CarlFK[m]mithro: we are at the front door, let us in please11:04
xfxfstefanor: https://github.com/xfxf/video-scripts/blob/master/udev/40-bmweb-udev.rules11:22
tpbTitle: video-scripts/40-bmweb-udev.rules at master · xfxf/video-scripts · GitHub (at github.com)11:22
mithro_florent_: I'm pretty sure that I've borked the interrupt stuff in your netv2mvp thingy11:50
*** rohitksingh-demo has joined #timvideos11:52
rohitksingh-demofelix_: http://hamsterworks.co.nz/mediawiki/index.php/MMCM_reset12:50
tpbTitle: MMCM reset - Hamsterworks Wiki! (at hamsterworks.co.nz)12:50
*** PeterMortensen has joined #timvideos13:16
*** andi-m has quit IRC13:30
*** paddatrapper has quit IRC13:30
*** andi-m has joined #timvideos13:32
*** rohitksingh-demo has quit IRC13:38
*** CarlFK has joined #timvideos13:57
*** ChanServ sets mode: +v CarlFK13:57
*** paddatrapper has joined #timvideos14:10
*** PeterMortensen has quit IRC14:49
*** PeterMortensen__ has joined #timvideos15:06
*** PeterMortensen__ has quit IRC15:17
*** PeterMortensen_ has joined #timvideos17:04
*** PeterMortensen_ has quit IRC17:43
_florent_cr1901_modern: litesdcard ressource usage is probably not optimal, is hasn't been optimized on that point yet17:46
_florent_cr1901_modern: i think cdc is using way too much registers, i'll have a look17:47
_florent_rohitksingh-demo: strange that you only get x1 and not x2, is your motherboard only x1?18:07
_florent_rohitksingh-demo: this should fix the csr.csv issue you had:18:07
_florent_rohitksingh-demo: https://github.com/enjoy-digital/netv2-soc/commit/76caeebf0e99d4d32a27da54c023d6de38e3038418:08
tpbTitle: targets: generate csr.csv · enjoy-digital/[email protected] · GitHub (at github.com)18:08
*** PeterMortensen_ has joined #timvideos18:23
_florent_rohitksingh-demo: this should fix your issue when compiling pcie kernel code:18:40
_florent_rohitksingh-demo: https://github.com/enjoy-digital/netv2-soc/commit/38bdfe5de57694d66fddce7fbb444092811e8d3118:42
tpbTitle: targets: fix irq constant generation for pcie target (we have cpu + p… · enjoy-digital/[email protected] · GitHub (at github.com)18:42
*** PeterMortensen_ has quit IRC18:54
*** samsagaz has quit IRC19:14
*** daveshah has joined #timvideos19:25
*** CarlFK has quit IRC21:11
*** CarlFK has joined #timvideos21:23
*** ChanServ sets mode: +v CarlFK21:23
* CarlFK[m] eats breakfast21:24
CarlFKanyone awake?21:31
CarlFKmore important: who wants to go get coffee?21:31
*** CarlFK has quit IRC21:42
jea[m]CarlFK: Did you head to google?22:48
micolous[m]I watched the 34c3 infrastructure review, in it they had level monitoring and video tiles for all their rooms. Any chance someone knows where the code is for that? It's probably better than what I had for monitoring22:50
CarlFK[m]jea: arrived and got into the hackfest come on over22:51
*** CarlFK has joined #timvideos23:02
*** ChanServ sets mode: +v CarlFK23:02
mithroJust heading in now23:18
xfxfmorning.  i'm at guthrie theatre testing AV - we have problems23:24
xfxfHDMI2USB doesn't want to work, i get a sideways skewed picture.  would upload a photo but no phone reception here23:24
xfxfthis board is running a firmware from pycon - is there a prebuilt gateware/firmware i can try flashing onto this?23:25
xfxfping CarlFK[m] stefanor23:25
xfxf(have tried 720p50, 720p60, 60 works better than 50, but neither are right)23:25
tumbleweedxfxf: yeah, try current firmware23:26
tumbleweedalso try using a magic cable between the opsis and their equipment23:26
tumbleweedwhat is the setup?23:28
xfxftumbleweed: can you link me to current firmware?  I don't.have time.to.compile.it up while AV are waiting23:28
xfxfIt's a crestron system. Have seen same issue before with them23:28
tumbleweedI mean, what is your side of the setup?23:29
xfxfsorry, phone reception in this room non-existant - had to wander outside to send pic above23:29
xfxfjoel's hdmi2usb23:30
xfxfnot much 'setup' on our end23:30
tumbleweedso, magic cable to hdmi2usb to what to crestron?23:30
xfxfit's a proprietry uni av system with lots of gear, i question why you want to know specifically what crestron gear it is23:31
tumbleweedI'm asking how you're connecting to it23:31
tumbleweedsee my suggestion above?23:31
tumbleweedhttps://github.com/timvideos/HDMI2USB-firmware-prebuilt/blob/master/archive/master/v0.0.4-72-g6e326e3/opsis/hdmi2usb/lm32/image-gateware%2Bbios%2Bfirmware.bin is the latest build23:31
tpbTitle: HDMI2USB-firmware-prebuilt/image-gateware+bios+firmware.bin at master · timvideos/HDMI2USB-firmware-prebuilt · GitHub (at github.com)23:31
xfxfre redmere directly to their kit?  just done that, zero difference23:31
xfxfta, i'll flash that23:31
xfxfwill I need an updated mode-switch to flash it?23:32
tumbleweedupdated from what?23:32
tumbleweedyou want >= 0.0.23:32
xfxfblah, i should have grabbed one of the later laptops23:33
xfxfi don't have recent tooling on this23:33
xfxf(this is my personal laptop)23:33
tumbleweedit's in the PPA23:33
tumbleweedor debian unstable23:33
xfxfthis is debian 16.04 IIRC23:33
tpbTitle: Support packages for FPGA development : “TimVideos.Us - Live event streaming” team (at launchpad.net)23:34
xfxfer, ubuntu 16.0423:34
xfxfsorry, lacking coffee :)23:34
tumbleweedwe can tell23:34
xfxfokay, that's updated; i throw the blob above at hdmi2usb-mode-switch's --flash-gateware option?23:37
xfxfherm, --help doesn't have that.  old version must still be installed23:37
tumbleweeddpkg -l hdmi2usb-mode-switch23:39
xfxfya sorted23:39
xfxfthis had uber old stuff compiled & installed manually23:39
xfxfpackages work well, thx23:39
tpbTitle: Snippet | IRCCloud (at www.irccloud.com)23:40
tumbleweed-v to see what's going on23:41
tumbleweedbut I'm guessing you need to sudo hdmi2usb-mode-switch --mode serial, and then try again23:41
tumbleweed(the 818181 nonsense)23:41
xfxfstill not working23:43
tpbTitle: Snippet | IRCCloud (at www.irccloud.com)23:43
tumbleweedthat's still 81 messyness23:43
tumbleweedtry again?23:44
tumbleweedsometimes I have to go between serial and jtag a couple of times23:44
tumbleweedsometimes a power cycle helps23:44
xfxfdoesn't seem to be doing much. changing modes is fine, flashing always same error23:45
tumbleweedis your ixo-usb-jtag package up to date too?23:46
xfxfno idea, will apt get install now23:46
xfxfixo-usb-jtag is already the newest version (0.0.1-1~ubuntu16.04~ppa1).23:46
tumbleweedand hdmi2usb-fx2-firmware23:47
tumbleweedand hdmi2usb-udev23:47
xfxfhdmi2usb-udev is already the newest version (0.0.1-1~ubuntu16.04~ppa1).23:47
xfxfhdmi2usb-udev set to manually installed.23:47
xfxfhdmi2usb-fx2-firmware is already the newest version (0.0.0~git20151225-1~ubuntu16.04~ppa1).23:47
xfxfhdmi2usb-fx2-firmware set to manually installed.23:47
tumbleweedthat all seems fine23:47
tumbleweedlet's try with an opsis here...23:47
tumbleweedoh, you didn't sudo with the mode serial23:49
tumbleweeddid it actually do anything?23:49
xfxfi assume it did23:49
xfxfi'll try with sudo23:49
tumbleweedit appears to have23:49
tumbleweedbecause it was in serial afterwards23:49
tumbleweedah, you were root duh23:49
tumbleweedanyway, worksforme :(23:50
xfxfi can sudo if you want :P23:50
xfxfherm.  kinda stuck then, do we just wing it and figure it out saturday?23:50
xfxfwe just have an opportunity to get something fixed here if something's broken23:50
tumbleweedwe could bring another opsis and some random toys23:51
tumbleweedif we knew where to go23:51
tumbleweedI feel like I've seen the problem you're having before23:52
tumbleweedbut can't recall exactly what it was23:52
xfxfi've organised that we can come back later23:54
tumbleweedcan we bypass the crestron system?23:56
tumbleweedthat's my usual strategy23:56
tumbleweedrun hdmi extenders up to the projectors23:56
tumbleweedor go hdmi -> vga -> crestron :P23:56
xfxflikely not, unfortunately23:56
tumbleweedcan we get into the cabinet where it feeds their hdmi extender?23:57
tumbleweedxfxf: let me ask again. Can we come help right now?23:58
tumbleweedor is later better?23:58
*** ewen has joined #timvideos23:59

Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!