Sunday, 2017-08-13

*** tpb has joined #timvideos00:00
*** jea[m] has joined #timvideos00:06
mithroxfxf: ping?00:09
Kamilionhm, anyone got experience with bulk flash drives with custom branding? 100-200 units, conswag sort of thing?00:12
CarlFKKamilion: #pumpingstationone00:17
CarlFKI suspect 200 is too low00:18
CarlFKI was trying to get 100 SD cards with a custom raspberry pi image on them.  not a chance00:18
*** CarlFK[m] has quit IRC00:24
*** jea[m] has quit IRC00:24
KamilionHm, we can do mass imaging internally; but we can't do silkscreening or laser engraving internally00:28
Kamilioni suppose you're right though, just use the hackerspace's epilog00:29
Kamilionhad a look on ebay for an epilog, just for giggles; as I used to have one 6-7 years ago... click on a used one for $5500, then the "You may also like" drops six 40/60W epilog clones between $400 and $2000 new on me.00:39
Kamilionthis $1800 one's even got an ethernet port and pretends to be a Just Another Printer On The Network00:40
KamilionThey say you get what you pay for, but sheesh, it might cost me more than $340 to get a hundred USB drives lasered up...
tpbTitle: Professional 40W Laser Engraving Machine With Exhaust Fan USB Port 12"x 8" (at
*** cr1901_modern has joined #timvideos01:07
mithroCarlFK: I've got a project for you01:10
mithrocr1901_modern: ping?01:10
CarlFKmithro: what's that?01:11
cr1901_modernmithro: I just got back home, not much has occurred since we last chat :)01:12
mithroCarlFK: Could you build a little rig which lets you power cycle an Opsis via a computer? You just need like a little relay or something on the power connector connected to a USB microcontroller or GPIO on some board01:12
mithrocr1901_modern: Have you pushed your xmodem code anywhere?01:12
tpbTitle: GitHub - cr1901/HDMI2USB-litex-firmware: A version of the HDMI2USB firmware based around the misoc+migen tools produced by m-labs. (at
mithrocr1901_modern: are you just working on the xmodem side of things at the moment?01:14
cr1901_modernmithro: yes. Though I can do SFL in parallel if you wish01:14
mithrocr1901_modern: I don't see any of the "spiflash"ing side of things?01:14
cr1901_modernmithro: firmware/flash.c01:15
cr1901_modernmithro: Wait, I apologize01:15
cr1901_modernI screwed up01:15
*** paradisaeidae_ has joined #timvideos01:16
mithroforget to push something?01:16
cr1901_modernmithro: Yes. It does appear I forgot to push something... hmmm...01:17
tpbTitle: GitHub - cr1901/litex: Build your hardware, easily! (at
cr1901_modernPushed now01:22
*** paradisaeidae_ has quit IRC01:26
mithrocr1901_modern: You only seem to call "erase_flash_sector" at the moment?01:28
*** CarlFK[m] has joined #timvideos01:30
tpbTitle: HDMI2USB-litex-firmware/flash.c at master · cr1901/HDMI2USB-litex-firmware · GitHub (at
*** tangrs has quit IRC01:31
cr1901_modernReading works fine; the xmodem part is what's broken01:31
mithrocr1901_modern: Ahh01:32
cr1901_modernThe same code will work for the Opsis _if_ I design the quad SPI interface correctly01:34
cr1901_modernmithro: Does opsis have a JTAG port, or is that the fx2's responsibility?01:36
*** tangrs has joined #timvideos01:44
mithrocr1901_modern: Well it has JTAG port for an external JTAG programmer, the fx2 can also be a USB->JTAG01:44
cr1901_modernmithro: Is that what the mode-switch tool is for?01:47
mithrocr1901_modern: Yes01:47
cr1901_modernswap from serial port to "jtag/maintenance mode"?01:47
mithrocr1901_modern: yeah01:47
cr1901_modernWell for the time being if I need JTAG I'll use my external programmer. Once xmodem is implemented, reprogramming the firmware/gateware on Windoze will be less dire, and if I flash a bad gateware/firmware, I can use JTAG to revert01:48
*** jea[m] has joined #timvideos01:49
*** cr1901 has joined #timvideos02:16
*** cr1901 has quit IRC02:20
tpbTitle: USB Relay Modules With GPIOs and Analog Inputs (at
mithroCarlFK: yes, something like that02:22
CarlFKmithro: $25 - 2 relays, usb powered,  shows up as a serial port02:27
mithroCarlFK: sure02:27
CarlFKmithro: the price is about $15 + $5 per relay.    seems silly to get only one, other than size of the device...02:28
CarlFKmithro:  I figured 1 for what we want, 1 more "just in case" ...  can you think of any use for more than 2?02:29
*** paradisaeidae has joined #timvideos02:41
*** rohitksingh has joined #timvideos03:55
*** paradisaeidae has quit IRC04:01
mithroHi rohitksingh!04:02
rohitksinghmithro: Hi!04:40
mithroHow goes your long weekend?04:40
rohitksinghmithro: Its going good :) Just published Please don't laugh though at the amateur code :p04:42
tpbTitle: GitHub - rohitk-singh/usb-device: USB 2.0 Device IP core using Migen with out-of-box AXI IP Core (at
mithrorohitksingh: Would be good to add a picture of how you connected it up...04:51
rohitksinghmithro: Yeah definitely.04:53
CarlFKmithro:  can you look at my atlys setup soon - I am guessing you are going to want me to move a jumper around04:58
mithroCarlFK: At the moment I have other higher priority things to look at...04:58
CarlFKno prob - I have an Opsis hooked up too.  I forget what started this ;)05:00
CarlFKI think it was the 50/60 hz thing that DC maybe bumped into05:01
mithroCarlFK: Oh, testing two outputs, 1 input + encoder with 60Hz verse 50Hz outputs05:03
CarlFKmithro: sounds like it05:04
CarlFKoutput0: [email protected] from input1 (underflows: 1662)05:12
mithroCarlFK: And do they go away at 50Hz?05:20
CarlFKmithro: looks like it.    I did "s on" and have been watching05:23
mithroCarlFK: If you add a second input on 50Hz, do you start getting underflows there too?05:23
CarlFKmithro: I need to run a cable, right?  just enabling it isnt' enough?05:24
mithroI think so, for inputs if no data is coming it won't use up bandwidth IIRC05:25
CarlFKmithro:  yes, 2nd input and getting: output0: [email protected] from input1 (underflows: 497)05:41
CarlFKmithro:  switched to 60 - output0: [email protected] from input1 (underflows: 187946)05:41
CarlFKmithro: screen saver kicked in, no more underruns06:02
CarlFKim heading to bed06:03
mithrorohitksingh: So what does your USB core do so far?07:06
rohitksinghmithro: So far it just logs all activity on USB lines. For now, you can just issue multiple USB resets which forces Linux to try to re-enumerate the device. You can check the USB Get Descriptor packets sent by Linux which tries to know the maximum packet size supported by the device. This is followed by Set Address request. No requests have been implemented so far.07:14
mithroAhh cool07:14
*** twoolie has joined #timvideos11:21
*** sb0 has joined #timvideos11:33
*** twoolie has quit IRC12:56
*** rohitksingh1 has joined #timvideos13:11
*** rohitksingh has quit IRC13:14
*** medicalwei has joined #timvideos13:37
*** rohitksingh1 has quit IRC18:06
*** nikivi has quit IRC20:04
*** nikivi has joined #timvideos21:19
*** techman83 has quit IRC22:01
*** techman83 has joined #timvideos22:06
*** ChanServ sets mode: +v techman8322:06
mithrocr1901_modern: any progress? I was thinking that you should test the xmodem code outside the hardware itself22:59
cr1901_modernmithro: I have been. I've been using qemu-litex23:01
cr1901_modernmithro: And no progress; ask me in an hour23:01
mithrocr1901_modern: and
tpbTitle: HDMI2USB-litex-firmware/Makefile at master · timvideos/HDMI2USB-litex-firmware · GitHub (at
cr1901_modernmithro: You want me to add a spiflash check?23:04
mithrocr1901_modern: I was thinking more xmodem as it doesn't have any actual hardware dependency...23:05
cr1901_modernmithro: Fair. That's a good idea23:06
mithrocr1901_modern: Actually it would be super easy to just supply fake "spiflash_write" / "spiflash_read" which just store the info in a buffer somewhere that you can check later..23:06
cr1901_modernmithro: Well, the tests run on the development machine, right (amd64 env)?23:08
mithrocr1901_modern: Yes23:08
cr1901_modernI'll need to make a wrapper for serial port access on linux (termios.h)/windows, but that won't be difficult23:09
cr1901_modernI need to test against minicom, other clients23:09
mithrocr1901_modern: I would connect the serial stuff to just stdin/stdout?23:09
mithrocr1901_modern: minicom and other programs shell out to another tool if I understand correctly23:12
mithrocr1901_modern: lrzsz IIRC23:12
cr1901_modernOh... I totally forgot23:13
cr1901_modernErm yea, that should work. Main idea is that it doesn't make sense for me to write my own transmit program, which will probably have a bug or two, and test against my own receiver, which already doesn't work23:14
mithrocr1901_modern: Yeah23:14
cr1901_modern(and I'm sure it's some stupid mistake :/...)23:14
mithromuch easier to debug / understand on your host computer23:20
felix_cr1901_modern: for the serial port wraper have a look at libserialport23:27
cr1901_modernfelix_: Thanks, but as mithro correctly pointed out, doesn't make sense in this context.23:28
cr1901_modernI just wasn't thinking :P23:28
mithrocr1901_modern: it would be nice to be able to reuse some of this stuff on the pic and FX2 in the future23:37
cr1901_modernmithro: I told you I had an old xmodem library that worked :P23:38
cr1901_modernit fits all your needs :D23:38
mithroSo, why aren't we using it now?23:39
cr1901_modernmithro: Because you said you didn't want to have an external dep23:39
cr1901_modernAnd additionally it does need to be cleaned up23:39
mithroI'm okay with an external dep?23:41
tpbTitle: GitHub - cr1901/libmodem at refactor (at

Generated by 2.13.1 by Marius Gedminas - find it at!