Friday, 2018-08-31

*** tpb has joined #timvideos00:00
*** Guest73014 has joined #timvideos00:12
*** HeinzBoettjer has joined #timvideos00:30
*** CarlFK has quit IRC00:36
*** futarisIRCcloud has joined #timvideos00:41
*** bobe29 has joined #timvideos01:21
*** bobe29 has quit IRC01:23
*** CarlFK has joined #timvideos01:37
*** ChanServ sets mode: +v CarlFK01:37
*** mniip9 has joined #timvideos02:01
*** mniip9 has quit IRC02:05
cr1901_moderntinyfpga: Is tinyprog compatible with B2s running the old bootloader? tinyprog seems to find it just fine02:22
cr1901_modernNot about to test tho :)02:22
cr1901_modern(And to reiterate: Is it possible to distinguish BX from B2 if you have both plugged in using a command line switch? Their VID:PID seems to be the same.)02:25
tinyfpgacr1901_modern: they have different PIDs....even though it looks like they are the same02:26
tinyfpgacr1901_modern: it was a workaround I had to put into tinyprog02:26
tinyfpgacr1901_modern: because some BX boards had the wrong PID programmed on them02:26
tinyfpgacr1901_modern: but yeah, tinyprog is not compatible with the B202:27
cr1901_moderntinyfpga: http://ix.io/1lD0 Context is the following02:28
cr1901_modern(ignore the comment on tinyprog, it's wrong :)...)02:28
cr1901_modern... actually, nevermind02:29
cr1901_modernI think "having two tinyfpgas attached at the same time" is rare enough that this isn't needed02:29
cr1901_modernjust want "autodetection" to be foolproof02:30
cr1901_modernSo tinyprog doesn't try accidentally programming a B2 board :P02:30
tinyfpgacr1901_modern: I agree... at some point I will disable the workaround on tinyprog and it will no longer detect B2 boards02:51
tinyfpgacr1901_modern: : for now you can just assume it’s correct, or you could get the JSON metadata and make sure it’s a BX02:52
cr1901_modernCool... in the interim I added "if not address" to the TinyProgProgrammer class. So you can put address=None and tinyprog will do the work for you :)02:53
*** CarlFK has quit IRC03:15
*** rohitksingh_work has joined #timvideos03:52
*** sc00bz1 has joined #timvideos04:16
*** sc00bz has quit IRC04:21
*** thaytan has quit IRC05:33
*** thaytan has joined #timvideos05:33
*** ChanServ sets mode: +v thaytan05:33
*** exponent has joined #timvideos05:53
*** exponent has quit IRC05:54
*** Zimmedon25 has joined #timvideos05:57
*** Zimmedon25 has quit IRC06:01
*** Sove has joined #timvideos06:24
*** Sove has quit IRC06:28
*** CarlFK has joined #timvideos06:57
*** ChanServ sets mode: +v CarlFK06:57
*** Guest3579 has joined #timvideos07:08
*** rohitksingh_wor1 has joined #timvideos07:48
*** rohitksingh_work has quit IRC07:50
*** rohitksingh_wor1 has quit IRC08:01
*** rohitksingh_work has joined #timvideos08:03
*** patate6 has joined #timvideos08:03
*** patate6 has quit IRC08:08
*** Sigyn has quit IRC09:12
*** Sigyn has joined #timvideos09:13
*** Brace16 has joined #timvideos09:15
*** Brace16 has quit IRC09:16
*** futarisIRCcloud has quit IRC09:21
*** micolous[m] has quit IRC10:57
*** felix[m] has quit IRC10:57
*** synaption[m] has quit IRC10:57
*** CarlFK[m] has quit IRC10:57
*** jfng has quit IRC10:57
*** ducky[m] has quit IRC10:57
*** jea[m] has quit IRC10:57
*** xobs has quit IRC10:57
*** nancy[m] has quit IRC10:57
*** DSNTravellerbot[ has quit IRC10:57
*** futaris[m] has quit IRC10:57
*** shivm28[m] has quit IRC10:57
*** nbags[m] has quit IRC10:57
*** leachim619 has joined #timvideos11:03
*** nbags[m] has joined #timvideos11:07
*** leachim619 has quit IRC11:07
*** twoolie has joined #timvideos11:17
*** CarlFK[m] has joined #timvideos11:30
*** micolous[m] has joined #timvideos11:30
*** futaris[m] has joined #timvideos11:30
*** jea[m] has joined #timvideos11:30
*** felix[m]1 has joined #timvideos11:30
*** DSNTravellerbot[ has joined #timvideos11:30
*** ducky[m] has joined #timvideos11:30
*** nancy[m] has joined #timvideos11:30
*** shivm28[m] has joined #timvideos11:30
*** sscout24 has joined #timvideos11:33
*** sscout24 has quit IRC11:38
*** ManyRaptors22 has joined #timvideos11:41
*** ManyRaptors22 has quit IRC11:42
*** shenki has quit IRC12:06
*** shenki has joined #timvideos12:24
*** ChanServ sets mode: +v shenki12:24
*** rohitksingh_work has quit IRC12:41
*** ducky[m] has quit IRC12:54
*** nancy[m] has quit IRC12:54
*** futaris[m] has quit IRC12:54
*** micolous[m] has quit IRC12:55
*** nbags[m] has quit IRC12:55
*** felix[m]1 has quit IRC12:55
*** DSNTravellerbot[ has quit IRC12:55
*** CarlFK[m] has quit IRC12:55
*** jea[m] has quit IRC12:55
*** shivm28[m] has quit IRC12:55
*** nbags[m] has joined #timvideos13:05
*** CarlFK[m] has joined #timvideos13:27
*** xobs has joined #timvideos13:27
*** DSNTravellerbot[ has joined #timvideos13:27
*** jfng has joined #timvideos13:27
*** futaris[m] has joined #timvideos13:27
*** micolous[m] has joined #timvideos13:27
*** felix[m]1 has joined #timvideos13:27
*** jea[m] has joined #timvideos13:27
*** nancy[m] has joined #timvideos13:27
*** shivm28[m] has joined #timvideos13:27
*** synaption[m] has joined #timvideos13:27
*** ducky[m] has joined #timvideos13:27
*** sc00bz1 is now known as sc00bz14:13
*** twoolie has quit IRC14:31
mithroMorning everyone - back in the US now14:42
*** mrdata14 has joined #timvideos14:46
*** mrdata14 has quit IRC14:51
felix_hi mithro! in case you haven't read the other channel: the first prototype of the axiom photonsdi board will be sent to production soon :) still need to check a few things, but the design itself should be done14:55
mithrofelix_: So, how goes the code side?14:55
felix_haven't made that much progress on that side; the hardware ddesign ate way more time than expected :/14:56
mithrofelix_: The hardware side whas the least interesting side of things14:56
mithros/whas/was/14:56
felix_i discussed in #photonsdi if i should focus on the hardware or gateware side first14:58
felix_now that the hardware is more or less done, i'll continue on the gateware side14:59
mithrofelix_: Okay cool15:02
*** Whiskey16 has joined #timvideos15:32
CarlFKmithro: back in mtv?16:34
mithroyeap16:46
CarlFKmithro: welcome back.    do you have time to figure out why CI (jenkins?) isn't building opsis/hdmi2usb?16:47
CarlFKwhich may still have: usb stream broken https://github.com/timvideos/HDMI2USB-litex-firmware/issues/43116:49
tpbTitle: usb stream broken · Issue #431 · timvideos/HDMI2USB-litex-firmware · GitHub (at github.com)16:49
mithroCarlFK: I told you why - https://travis-ci.org/timvideos/HDMI2USB-litex-firmware16:50
mithrohttps://travis-ci.org/timvideos/HDMI2USB-litex-firmware/builds16:50
*** Kripton has quit IRC16:51
CarlFKmithro: im confused.  is that the cause or fix? or.. what am I looking at?16:56
*** Kripton has joined #timvideos16:56
mithroCarlFK: If it isn't green, then it probably didn't successfully build the firmware16:56
CarlFKmithro: thanks - I see a build from 4 days ago.. il check it out.17:01
*** ensyde has joined #timvideos17:50
*** ensyde has quit IRC17:54
*** bungle29 has joined #timvideos18:33
mithroCarlFK: Hopefully later today I'm going to do a docs cleanup18:46
*** Sigyn has left #timvideos18:53
mithrocr1901_modern: you get your replacement tinyfpga?19:19
mithroCarlFK: when I get this docs updated, can you give them a look over?19:20
CarlFKmithro: sure thing19:20
cr1901_modernmithro: Yes, BX has the same problem when booting from SPI flash19:23
mithrocr1901_modern: were you able to boot lm32 using ROM?19:24
mithroWe had that working at PyCon Au19:25
cr1901_modernmithro: Yes19:34
cr1901_modernBut micropython will not work on that19:34
mithroSo, the issue is the memory mapped spiflash?19:37
cr1901_modernCorrect, but it's not the SPI flash core portion that's failing19:38
mithroHave you put a scope in the flash to see the fetches are working okay?19:38
cr1901_modernThis is what I'm doing now19:38
mithroWe have the lm32 configuration with spiflash booting on the MimasV219:39
mithroIt's very slow without cache :-P19:39
cr1901_modernwhich is why I said at least I-cache should be enabled19:39
mithroBut the behaviour should be identical to that of the BX19:40
cr1901_modernDifferent FPGA :P19:40
cr1901_modernthe ice40lp doesn't like pipelined softcores for some reason19:40
cr1901_modernpicorv32 seems to work fine19:41
mithroWithout i-cache you can atleast see every instruction19:41
mithroPicorv32 is using a different spiflash controller?19:41
cr1901_modernLast time I disabled icache, I had the same problem19:42
cr1901_modernNo, it's not that19:42
cr1901_modernI think picorv32 just doesn't tickle the bug that lm32 does19:42
mithroIt's the lm32 with spiflash from LiteX that is the problem19:43
mithroAs lm32 from rom works19:43
mithroYou got a trace of the spiflash accesses?19:44
mithroWith the cache disabled you should see each fetch...19:45
cr1901_modernI'll do that and get back to you19:45
cr1901_modernBtw, I still want to add an "lm32-lite" which has icache enabled19:45
mithroSure19:47
mithroYou have a MimasV2 right?19:48
cr1901_modernyes19:48
mithroYou should be able to compare the SPI accesses for the same configuration on the MimasV2 and the TinyFPGA BX19:48
mithroSee how they differ...19:48
cr1901_modernTrue19:48
tinyfpgacr1901_modern: make sure the SPI core is issuing a “resume” command to the SPI flash to wake it up first19:49
mithrotinyfpga: that was my guess to :-)19:50
cr1901_moderntinyfpga: I have the proper switch to disable the final power down on icepack19:50
tinyfpgacr1901_modern: the ICE40 will put the SPI flash to sleep after waking it up... the SPI flash will ignore all commands until it receives a “resume” command19:50
tinyfpgacr1901_modern: ok19:50
mithroAlthough, who knows if it is working correctly?19:50
cr1901_moderntinyfpga: What you described WAS the original incarnation of my tinyfpga woes :)19:50
tinyfpgacr1901_modern: yeah, next step is comparing the SPI commands being sent with the commands in the Adesto SPI flash datasheet19:51
cr1901_modernmithro: I know it works correctly b/c I've seen traces of the data returned19:51
mithroMaybe someone broke out since then :-P19:52
mithroS/out/it/19:52
cr1901_modernI saw traces of the data returned since a few days ago :)19:52
cr1901_modernI'm pretty confident I've eliminated all the easy sources of error :/19:52
mithroOkay19:52
cr1901_modernWhat I want now is to figure out which nets in the FPGA are the culprit and a "rip out nets that don't matter" mode to create a minimal offending bitstream in nextpnr19:53
mithroLunchtime for me19:54
CarlFKmithro: did you say I shouldn't see: Bus 003 Device 043: ID 16c0:06ad Van Ooijen Technische Informatica20:04
mithroCarlFK: Depends on when you are seeing it20:05
CarlFKflashed image-gateware+bios+firmware.v0.0.4-255-g4a97842.opsis.hdmi2usb.lm32.bin,  power cycled opsis, 5 min later20:06
CarlFKnot sure I have ever seen this before [738275.293613] usb 3-1: SerialNumber: hw_opsis20:07
CarlFKbut this is the fist time I have flashed something after April20:07
CarlFKmithro: syslog http://paste.ubuntu.com/p/zXcfYZ2p86/20:09
tpbTitle: Ubuntu Pastebin (at paste.ubuntu.com)20:09
mithroCarlFK: This seems like you didn't get the hdmi2usb firmware...20:10
CarlFKmithro:  hash checks out: sha256sum image-gateware+bios+firmware.v0.0.4-255-g4a97842.opsis.hdmi2usb.lm32.bin20:14
CarlFKfcb4315d4aa9ca38202ff6e503f548e0af849daeaa2cbbbd05aeb6ea5aa9a7bf20:14
CarlFKhttps://github.com/timvideos/HDMI2USB-firmware-prebuilt/blob/master/archive/master/v0.0.4-255-g4a97842/opsis/hdmi2usb/lm32/sha256sum.txt20:14
tpbTitle: HDMI2USB-firmware-prebuilt/sha256sum.txt at master · timvideos/HDMI2USB-firmware-prebuilt · GitHub (at github.com)20:14
*** shorne has quit IRC20:16
*** shorne has joined #timvideos20:18
mithroCarlFK: Don't know.,..20:21
CarlFKmithro: any guesses what I can do ?20:37
mithroTry a different version?20:38
mithroI could use help filling out https://docs.google.com/spreadsheets/d/1XTHfdYXuvwoYdPXm4M6qDA0D2fZCPy220-9q6qZpTw4/edit#gid=022:14
tpbTitle: LiteX BuildEnv Support - Google Sheets (at docs.google.com)22:14
mithrocr1901_modern: https://docs.google.com/spreadsheets/d/1XTHfdYXuvwoYdPXm4M6qDA0D2fZCPy220-9q6qZpTw4/edit#gid=132611137922:30
tpbTitle: LiteX BuildEnv Support - Google Sheets (at docs.google.com)22:30
CarlFKmithro: is nexys_video different from just nexys?22:47
mithroYes22:49
mithroAdding nexys support should be pretty easy though22:50
mithroIt should mostly be data entry22:50
*** tac-tics has joined #timvideos22:55
tac-ticsCarlFK, sure I am22:55
CarlFKtac-tics: do you have nexys_video or nexys ? (seems they are 2 different things)22:56
tac-ticsThis is what I have: https://reference.digilentinc.com/reference/programmable-logic/nexys-4/start?redirect=122:57
tpbTitle: Nexys 4 [Reference.Digilentinc] (at reference.digilentinc.com)22:57
CarlFKtac-tics: did you dl the 25 gig of Xilinx Vivado ?22:58
tac-ticsno22:59
tac-ticsI downloaded 0gigs of no videos.22:59
CarlFKhttps://github.com/timvideos/litex-buildenv/wiki/Xilinx-Vivado23:00
tpbTitle: Xilinx Vivado · timvideos/litex-buildenv Wiki · GitHub (at github.com)23:00
CarlFKhmm, those docs are a little wonky.  I updated the file names for the parts I used.23:02
CarlFKhttps://www.xilinx.com/member/forms/download/xef.html?filename=Xilinx_Vivado_SDK_2018.2_0614_1954.tar.gz23:02
tpbTitle: OAM Login Page Redirect (at www.xilinx.com)23:02
CarlFKthat's current as of about 2 weeks ago23:02
tac-ticsI'll hae to play with it when I get home23:03
tac-ticsCan't seem to create an account on their site >_>23:06
tac-ticsHow dumb23:07
*** keithn13 has joined #timvideos23:07
*** keithn13 has quit IRC23:08
*** liori27 has joined #timvideos23:18
CarlFKit may be annoying. but somehow I've done it.23:22
mithroCarlFK: https://github.com/timvideos/litex-buildenv/wiki23:30
tpbTitle: Home · timvideos/litex-buildenv Wiki · GitHub (at github.com)23:30
*** twoolie has joined #timvideos23:40
cr1901_modernmithro: Internally, HX4k == HX8K :)23:42
CarlFKmithro: has the qemu for or1k been .. patched or pulled or whtaever to work with current linux?23:45
*** tac-tics has quit IRC23:45
mithroCarlFK: mangled?23:45
CarlFKlol - yes, that.23:45
mithroDunno23:52
CarlFKim guessing it would have been you23:55
mithro?23:55

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