Saturday, 2017-07-15

*** tpb has joined #timvideos00:00
*** SamSagaZ has quit IRC03:50
mithroCan anyone who is interested in getting Opsis boards please fill out this form?
tpbTitle: Numato Opsis Order (at
mithropaddatrapper: I'm hoping to get to your pull requests today, but it's already 3:30pm and I still haven't finished the stuff in the queue before hand05:39
mithrocr1901_modern: ping?06:31
paddatrappermithro: ok no worries07:03
mithropaddatrapper: Regarding your questions07:03
mithropaddatrapper: You basically want to map putc to what ever is your serial output function07:04
mithropaddatrapper: printf uses putc to send data07:05
paddatrappermithro: ah Ok. Does the current firmware's debug statements work on any of the boards? (I'm assuming they do in the opsis)07:06
mithropaddatrapper: They did at some point in the past07:07
mithropaddatrapper: Do any of the example firmware use ISO transfers?07:08
paddatrappermithro: not that I recall, but I'll check07:08
mithropaddatrapper: You might want to take a look at
paddatrappermithro: thanks, that should be very useful07:12
mithropaddatrapper: You also want to make sure that you check the define at the top of the files07:13
paddatrappermithro: which files?07:13
mithroA lot of them have things like07:14
mithro#ifdef DEBUG_FIRMWARE07:14
mithro#include <serial.h>07:14
mithro#include <stdio.h>07:14
mithro#define printf(...)07:14
mithropaddatrapper: ALL OF THEM! :-P07:14
paddatrapperAh yes. At the moment I've got something similar, but just invoking usart_send_string directly07:15
paddatrappermithro: any idea why more than 6 characters, or more than one call to usart_send_string would cause the serial not to be sent correctly?07:17
mithropaddatrapper: Two possibilities - interrupts and buffer overflows...07:17
mithropaddatrapper: If you get an interrupt in the middle of sending your serial data its going to screw it up07:17
mithropaddatrapper: The page at seems to imply that the "debugdevice" firmware might have support for iso transfers07:19
tpbTitle: USB Testing on Linux (at
paddatrappermithro: that so sounds reasonable. There was one interrupt being triggered 8000 times a second... Could well have been triggered in the middle of the serial transfer07:19
mithropaddatrapper: You should be able to see that by using your second board in logic-analyzer mode07:20
mithropaddatrapper: You'll see the byte start, then see a large gap, then see the byte continue07:20
paddatrappermithro: ok I'll look for that07:21
mithropaddatrapper: I don't think there is much difference in ISO vers BULK handling from the FX2 side07:22
paddatrappermithro: ok. I plan on working in it today, so I'll take a look and let you know07:22
mithropaddatrapper: Some info at might be useful too07:27
mithropaddatrapper: Have you tried using the logic-analyzer bit much?07:34
paddatrappermithro: yeah I use it and the LEDs a lot of debugging, since serial is currently so unreliable08:14
mithropaddatrapper: Use the logic analyzer and sigrok, you can then set "UART" decoding on the serial output and sigrok should decode it into text for you08:15
paddatrappermithro: that's cool. At the moment I just look for the binary to see when the serial is written and can't see what it actually is.08:16
paddatrapperI'll have to try that today08:16
mithropaddatrapper: I think I've just about hit my limit for doing useful work today -- will try again tomorrow08:47
*** SamSagaZ_ has joined #timvideos14:03
*** SamSagaZ_ has quit IRC14:04
paddatrappermithro: it was an interrupt issue. Removed the interrupt and now there is no problem16:47
mithropaddatrapper: You can disable interrupts for the period of sending the byte17:01
paddatrappermithro: think I shall do that then. Do they get queued or dropped?17:01
mithroThere can only be one pending interrupt of each type17:02
mithroso if a second one comes in before you have finished servicing the first one then you have problems17:02
paddatrapperok, I'll keep that in mind17:09
*** CarlFK has quit IRC21:12
*** CarlFK has joined #timvideos22:04
*** ChanServ sets mode: +v CarlFK22:04

Generated by 2.13.1 by Marius Gedminas - find it at!