Thursday, 2017-05-18

*** msvb-lab <msvb-lab!> has quit IRC00:45
*** genpaku <genpaku!~genpaku@> has quit IRC02:18
*** genpaku <genpaku!~genpaku@> has joined #minnowboard02:20
*** Abhishek_ <Abhishek_!uid26899@gateway/web/> has quit IRC04:21
*** Abhishek_ <Abhishek_!uid26899@gateway/web/> has joined #minnowboard04:25
*** black_ant <black_ant!~black_ant@unaffiliated/black-ant/x-1505394> has quit IRC04:56
*** calculus <calculus!~calculus@gentoo/user/calculus> has quit IRC05:09
*** calculus <calculus!> has joined #minnowboard05:10
*** calculus <calculus!~calculus@gentoo/user/calculus> has joined #minnowboard05:10
*** agust <agust!> has joined #minnowboard06:38
*** vaibhava__ <vaibhava__!~vaibhava@> has quit IRC08:04
*** msvb-lab <msvb-lab!> has joined #minnowboard09:46
*** wmarone|tmp <wmarone|tmp!> has quit IRC10:28
*** wmarone|tmp <wmarone|tmp!> has joined #minnowboard10:29
*** msvb-lab <msvb-lab!> has quit IRC11:28
*** msvb-lab <msvb-lab!> has joined #minnowboard11:40
*** msvb-lab <msvb-lab!> has quit IRC11:45
*** msvb-lab <msvb-lab!> has joined #minnowboard11:56
msvb-labHas anyone actually used the UARTs on the low speed header? I12:43
msvb-labI'm getting no I/O at all (as if the cable's not connected) when using either UART1 or UART2.12:44
msvb-labAnd have tested on Wind River, Debian, and Kali:12:45
msvb-labminnow$ echo Hello >/dev/ttyS112:45
msvb-lablaptop$ cat /dev/ttyUSB012:45
msvb-labWhen using UART0 all works as expected:12:46
msvb-labminnow$ echo Hithere >/dev/ttyS012:46
msvb-lablaptop$ cat /dev/ttyUSB012:46
msvb-labstty(1) reports all UARTs set to the same baud rate.12:47
msvb-labIt seems UART1 and UART2 are unusable on the Minnow Turbot latest firmware.12:47
*** lamego <lamego!~jose@> has joined #minnowboard12:54
*** bluelightning <bluelightning!~paul@pdpc/supporter/professional/bluelightning> has quit IRC13:10
msvb-labI just created
msvb-labTo try to solve the problem of unusable UARTs on the MinnowBoard Turbot.13:14
*** pidge <pidge!> has quit IRC13:45
*** pidge <pidge!~pidge@2001:41d0:a:5dc5::> has joined #minnowboard13:45
*** stephano <stephano!~stephano@> has joined #minnowboard13:59
Bitweasilmsvb-lab, which firmware?  I'm working on 0.93 based EFI and they work perfectly.14:34
BitweasilAnd how are your flow control settings configured both in EFI and on the port?  Usually, that's a flow control issue if you can't get anything at all to come caross.14:35
msvb-labBitweasil: Firmware MNW2MAX1.X64.0095.R01.1703281124. Flow is 'off' in EFI and 'off' on the port (have tried both on the port.)14:58
msvb-labBitweasil: How are you using UART1 or UART2 to confirm they work perfectly?14:59
msvb-labBitweasil: Have you connected pin-6 of the LS header to what would otherwise connect to pin-4 of UART0 and pin-8 of LS header to what would otherwise connect to pin-5 of UART0?15:05
msvb-lab...and of course connect GND to pin-1 or pin-2 of the LS header.15:06
msvb-labI'm using a FTDI cable for this, and have tried connecting the 'unused' UART1 CTS and RTS pins.15:06
msvb-labUART0 always outputs to a 115200 terminal application, UART1 never does.15:07
BitweasilTry disconnecting your grounds.15:07
BitweasilI don't have mine hooked up, I assume there's a reason.15:08
BitweasilI've got UART1 hooked up with full flow control to a FT231X based board for main interface work, UART2 hooked up to a FT232R based board for high speed log messages, both banging away at 3M baud, but I am driving them directly as PCI devices from code in weird corners of ring 0 and not Linux.  However, I know that they work with the Ubuntu 16.04 LTS 4.4 kernel because I tested with that beforehand.15:09
msvb-labBitweasil: So you have only RX and TX connected from the MinnowBoard Turbot to your computer over a FTDI cable?15:09
BitweasilTo a FTDI breakout board, yeah.15:10
BitweasilI don't see any grounds hooked up on my setup.15:10
BitweasilBut, yes, UART2 goes straight to a FTDI board and I have no problems.15:10
BitweasilYou do have the pins properly swapped, TX to RX, RX to TX, right?15:10
msvb-labOkay, I'll try to duplicate your configuration. And I'll set flow control to on in the UEFI setup.15:10
BitweasilLet me check my configs.  Just a sec.15:11
BitweasilAnything interesting in dmesg?15:11
BitweasilAnd you do see the devices in lspci, right?15:12
msvb-labI've swapped RX and TX a couple times, but I'm using the diagram:15:12
msvb-labto know which cable connector goes where. I've just replugged what was in the RX pin of UART0 to UART1 according to the diagram.15:12
msvb-labI haven't tried looking at the PCI table, thanks for suggesting that.15:13
BitweasilYou did make sure your FTDI adapters are set to 3.3V, right?15:13
BitweasilThe IO pins are not 5V tolerant.15:13
BitweasilYeah, that diagram looks correct.15:14
msvb-labThe FTDI adapter I'm using is already working on UART0, but I think the only 3V3 or 5V found is on VCC isn't it?15:15
BitweasilEFI -> DEvice manager -> System Setup -> South Cluster Configuration -> LPSS & SCC Configuration, I have the devices in PCI mode (because I hate parsing ACPI tables), HSUART #1 support and flowcontrol enable, #2 support enable, flow control disable.15:16
BitweasilUh, pretty sure that controls TX voltages as well.15:16
BitweasilI've got one funky cable that has 3.3V TX and 5V VCC.15:16
BitweasilBut my other adapters are straightforward and all set for 3.3V.15:17
msvb-labSo I can touch a multimeter to the TX/RX and as long as the USB end is plugged into 5V I should see voltage?15:17
BitweasilNo, you'd have to send data through it.15:17
BitweasilAnd a multimeter is going to average.15:17
msvb-labTo be honest, I haven't connected an oscilloscope yet either but I'm cringing at the hours.15:18
BitweasilBang away a bunch of 0xff through it, since it's TTL logic, you should see "mostly" ones.15:18
BitweasilEven a cheap DSO138 can do something useful with 115200.15:18
BitweasilIt's a good sanity check.15:18
msvb-labWhat's the 'one' variant of /dev/zero?15:19
BitweasilDunno.  Again, I don't use the ports from inside Linux.15:19
msvb-labI could do laptop$ cat /dev/<one-here> >/dev/ttyUSB0 to get all ones and hope to read 3V3.15:19
BitweasilI'd just write a quick C binary to stream it.15:19
BitweasilWell, it won't be all 1s.15:19
BitweasilBut if you do that, a voltmeter shouldn't read *more* than 3.3V.15:19
BitweasilOr just check the docs.15:20
BitweasilMy FTDI adapters have a jumper on the FT232R ones, the FT231X ones have a solderable bit on the bottom.15:20
msvb-labBitweasil: So you think it might be that UART0 accepts RX and TX streams at 5V but UART1 silently fails?15:21
BitweasilI really don't know what the issue you're having is.  But I know that those high speed ports work just fine when properly set up.15:22
BitweasilI thought the console UART was a 3.3V unit as well, actually.15:22
BitweasilPretty much everything on this board is 3.3V.15:22
msvb-labI'll try with a second (non FTDI) cable as well. I had assumed it couldn't be the cable if it was working fine on UART0.15:23
BitweasilYeah, UART0 is 3.3V too.15:23
BitweasilQuit feeding it 5V if you are.15:24
BitweasilThat'll fry things eventually.15:24
msvb-labIf my memory serves me, the FTDI cable I have came with an older MinnowBoard Max.15:25
msvb-labIt's quite unlikely to be the problem, but I like the idea of disconnecting GND and setting flow control.15:26
BitweasilGet it working without flow control first.15:26
BitweasilAlso, I think you swap CTS/RTS pins when wiring up for flow control.15:27
BitweasilNot sure if you're doing that on the adapter.15:27
BitweasilSorry, I wired this up months ago and haven't touched it.15:27
BitweasilOh, speaking of FTDI, you see that post on hackaday where some nutter replaced the fake chips on his adapters with real ones?15:27
BitweasilSeems... a silly way to get legit FTDI, but I guess it worked.15:28
msvb-labSince the shape of the FTDI cable I have is fixed on USB and the TTL sides (where it is a solid row of plastic header) I haven't swapped at all on UART0 where the shape just fits.15:28
msvb-labThen using the diagram (where it says UART0/RX change to UART1/RX) I've not need to troubleshoot much there.15:29
msvb-labSame with CTS RTS but since you're only using RX/TX and all docs say that CTS/RTS are unused, I'm not focusing on that.15:29
BitweasilOh, you know, I have one of those cables, and I was having some reliability issues with the high speed UARTs at 3M baud with that - switched to the little breakout boards and short jumpers and no problems.15:29
BitweasilI didn't bother troubleshooting, but if you've got a little breakout board for the serial adapter and some jumpers, you might try that.15:30
msvb-labThere's also the FTDI cable on Sparkfun that has 5V/3V mismatch because of their 'misunderstanding' with the manufacturer and they're selling stil for full price.15:30
BitweasilI got all this working back around November and it's been rock solid since then.15:30
BitweasilYeah, I have that one.15:30
BitweasilWorked fine for the console.  Not using that on the high speed serial for some reason that probably was important.15:30
BitweasilBut it's probably fine at 115200.15:30
*** m_w <m_w!> has joined #minnowboard15:31
*** calculus <calculus!~calculus@gentoo/user/calculus> has quit IRC15:47
*** calculus <calculus!~calculus@gentoo/user/calculus> has joined #minnowboard15:49
*** stephano <stephano!~stephano@> has quit IRC15:53
*** ka6sox is now known as zz_ka6sox16:11
*** stephano <stephano!~stephano@> has joined #minnowboard16:39
*** calculus <calculus!~calculus@gentoo/user/calculus> has quit IRC17:19
*** calculus <calculus!~calculus@gentoo/user/calculus> has joined #minnowboard17:20
*** calculus <calculus!~calculus@gentoo/user/calculus> has quit IRC18:27
*** calculus <calculus!> has joined #minnowboard18:29
*** calculus <calculus!~calculus@gentoo/user/calculus> has joined #minnowboard18:29
*** dvhart <dvhart!> has joined #minnowboard18:58
*** jkridner|pd <jkridner|pd!~jkridner@pdpc/supporter/active/jkridner> has quit IRC19:18
*** m_w <m_w!> has quit IRC19:22
*** jkridner <jkridner!~jkridner@pdpc/supporter/active/jkridner> has joined #minnowboard19:24
*** bluelightning <bluelightning!~paul@pdpc/supporter/professional/bluelightning> has joined #minnowboard19:59
*** zz_ka6sox is now known as ka6sox20:12
*** bluelightning_ <bluelightning_!~paul@> has joined #minnowboard21:12
*** bluelightning_ <bluelightning_!~paul@pdpc/supporter/professional/bluelightning> has joined #minnowboard21:12
*** bluelightning <bluelightning!~paul@pdpc/supporter/professional/bluelightning> has quit IRC21:13
*** bluelightning_ is now known as bluelightning21:13
*** ccyRegeaneWolfe_ <ccyRegeaneWolfe_!~cccyRegea@> has joined #minnowboard21:17
*** cccyRegeaneWolfe <cccyRegeaneWolfe!~cccyRegea@> has quit IRC21:18
*** ccyRegeaneWolfe_ is now known as cccyRegeaneWolfe21:18
msvb-labBitweasil: By the way, after a second (closer) look at the cable it's labeled 'FTDI 3.3V' and I suspect it's a real (not fake) one.21:29
*** dvhart <dvhart!> has quit IRC21:35
msvb-labBitweasil: I've solved the UART1 problem, which was a combination of (1) poor MinnowBoard documentation wrongly stating tty device on elinux (and main pages?), (2) wild color matching and swapping of RX to TX contacts especially when moving from UART0 to UART1, and (3) default UART1 serial configuration not matching the default UART0 one.21:53
msvb-labBitweasil: Thanks for your help, much of which was useful. I really appreciate it.21:54
*** dvhart <dvhart!> has joined #minnowboard22:32
*** lamego <lamego!~jose@> has quit IRC22:32
*** msvb-lab <msvb-lab!> has quit IRC22:34
*** agust <agust!> has quit IRC22:39
*** msvb-lab <msvb-lab!> has joined #minnowboard22:47
*** dvhart <dvhart!> has quit IRC22:53
*** stephano <stephano!~stephano@> has quit IRC23:23
*** msvb-lab <msvb-lab!> has quit IRC23:47

Generated by 2.11.0 by Marius Gedminas - find it at!