Monday, 2017-10-30

*** tpb has joined #timvideos00:00
tpbTitle: GitHub Apps - Unfurl Links 路 GitHub (at
cr1901_modernprintf debugging is useless for qemu T_T00:08
mithrocr1901_modern: Why? You can send the tty output to a different pts?00:08
cr1901_modernmithro: That doesn't matter when xmodem/the serial xfer app is swallowing stdout (and doesn't release it fast enough before swallowing my debug output)00:09
mithrocr1901_modern: Hrm? make the console be a different pts and connect your xmodem to that?00:10
cr1901_modernmithro: My current qemu command line
mithrocr1901_modern: So, what is the problem?00:11
tumbleweedmithro: pushed patches to both of those branches00:12
cr1901_modernmithro: None of my debug output after the xfer finishes is being displayed00:12
* cr1901_modern tries harder00:12
mithrowhere should the debug info be going?00:12
cr1901_modernIt should be displayed on stdout after the xmodem xfer is done (so it doesn't interfere with the return codes for the transmitter)00:15
cr1901_modernBut instead what's happening is when I log back into minicom after the xfer is done, a number of printf statements don't display00:16
mithrocr1901_modern: Are you sure you don't have multiple serial connections open?00:17
cr1901_modernmithro: Ummm, well there is a "minicom" open that I can't account for ._.00:18
cr1901_modernis there a quick linux command to figure out what fds its using?00:18
cr1901_modernactually n/m00:18
cr1901_modernthat won't help here b/c I killed qemu00:18
* cr1901_modern looks for the open minicom00:19
cr1901_modernOkay, so that minicom's attached to a different pty00:20
cr1901_modernmithro: Yes I'm sure I only have one serial connection open00:20
cr1901_modernmithro: My theory is that "sx"'s stdin buffer (which is stdout of my pty) is filled with a number of messages after the xfer finishes, but sx closes stdin as it exits and the messages get lost00:25
cr1901_modernI wonder if I can "tee" it to get the missing messages00:26
cr1901_modernwtf... when I do "echo \"write spi xmodem 0x20088000 60104 0XB6848B7C\" > /dev/pts/20", the other side doesn't receive it until I open minicom00:35
*** tvCommitBot has joined #timvideos05:14
tvCommitBot[website] mithro pushed 6 new commits to add-code-of-conduct-1:
tvCommitBotwebsite/add-code-of-conduct-1 a78993b Tim 'mithro' Ansell: Adding a reporting file.05:14
tvCommitBotwebsite/add-code-of-conduct-1 3d3cc54 Tim 'mithro' Ansell: Moving into conduct/Code.md05:14
tvCommitBotwebsite/add-code-of-conduct-1 448c777 Tim 'mithro' Ansell: Adding wrapping and front matter to code of conduct.05:14
*** tvCommitBot has left #timvideos05:14
mithroshenki: So - you going to reproduce shorne's stuff? :-P05:17
shenkimithro: tonight!05:26
shenkimithro: which board should I use?05:27
shenkimithro: i was also going to review his upstream kernel patches. they are sitting in my inbox05:27
mithroshenki: I'm unsure if shorne was using hardware or just qemu...05:27
mithroshenki: Arty or Opsis05:27
mithroWill test the Arty here sometime soon I think...05:31
*** CarlFK has quit IRC05:51
*** tvCommitBot has joined #timvideos06:10
tvCommitBot[website] mithro pushed 1 new commit to add-code-of-conduct-1:
tvCommitBotwebsite/add-code-of-conduct-1 3f20dc4 Tim 'mithro' Ansell: travis: Use rvm 2.206:10
*** tvCommitBot has left #timvideos06:10
*** aps has joined #timvideos06:14
*** CarlFK has joined #timvideos06:17
*** ChanServ sets mode: +v CarlFK06:17
*** rohitksingh has joined #timvideos06:28
*** tvCommitBot has joined #timvideos06:35
tvCommitBot[website] mithro pushed 1 new commit to add-code-of-conduct-1:
tvCommitBotwebsite/add-code-of-conduct-1 63b9654 Tim 'mithro' Ansell: Rakefile: Don't checkout again?06:35
*** tvCommitBot has left #timvideos06:35
*** rohitksingh has quit IRC06:51
*** tvCommitBot has joined #timvideos07:07
tvCommitBot[website] mithro closed pull request #17: Adding Code of Conduct. (master...add-code-of-conduct-1)
*** tvCommitBot has left #timvideos07:07
*** tvCommitBot has joined #timvideos07:07
tvCommitBot[website] mithro deleted add-code-of-conduct-1 at 63b9654:
*** tvCommitBot has left #timvideos07:07
*** rohitksingh has joined #timvideos07:14
cr1901_modernmithro: I made a little bit of progress when I switched to flterm. So here's what happened when trying to do an xmodem xfer w/ flterm using qemu:07:18
cr1901_modern1. flterm errors out immediately w/ 11 transfer failures07:18
cr1901_modern2. HDMI2USB echos "Invalid chars in CRC." in the middle of a bunch of garbage. I have no idea where this string is defined in HDMI2USB.07:18
cr1901_modern3. Every 10 seconds, qemu-litex echoes a load of crap to the screen. For every newline in this crap, the "H2U time>" prompt is refreshed. It does this 11 times. The crap is identical each time.07:18
cr1901_modern4. Prompt goes back to normal after 11 times.07:18
cr1901_modernSo I suspect what is happening is that there is a nice mismatch between when the qemu serial port is actually retrieving data versus when its sent07:18
cr1901_modernAnd the "crap" that's being emitted is in fact the contents of the first packet, 11 times07:19
cr1901_modernmithro: Plz ignore everything I just wrote lmao. This is a different "oversight" in my flterm fixes. I'll make a mental note and we can discuss after I get qemu working.07:23
cr1901_modernmithro: Actually, my list still applies; the garbage still happens when I account for the flterm "oversight"07:29
mithroshenki: For the arty, you might need a crossover cable07:51
*** aps has quit IRC09:04
*** jwhitmore has joined #timvideos09:28
mithroshorne: I get 'buildroot login:' with qemu now too!09:42
*** rohitksingh has quit IRC09:44
shornemithro: I was just using qemu so far11:04
shornenow that qemu is working I will try to get serial output on arty again11:05
shornebut... I was having issues with getting any output, so I might need to integrate _florent_ 's adv_dbg_sys patches so I can debug whats going on11:06
shorneafter that we can work on linux networking11:06
shorneIs there any work on the liteth driver in linux?11:06
*** aps has joined #timvideos12:06
*** tsglove has quit IRC12:41
shenkibenh was interested in writing a driver. i'll ask him about that tomorrow12:50
shenkii'm not aware of any existing code for ethernet12:51
shorneshenki: I think now with userspace on qemu working it should not take too much time12:52
shenkishorne: cool12:53
shorneThe bios code is probably a good place to start12:55
tpbTitle: liteeth/liteeth/software/libuip at 937c2407276371d9f8c1bb19c4ae8e97e581da83 路 enjoy-digital/liteeth 路 GitHub (at
*** accesspoint has joined #timvideos12:56
*** tsglove has joined #timvideos14:07
*** tsglove2 has joined #timvideos14:28
*** tsglove has quit IRC14:31
*** tsglove3 has joined #timvideos14:51
*** tsglove2 has quit IRC14:54
*** jwhitmore has quit IRC15:12
*** tsglove3 has quit IRC15:25
*** tsglove3 has joined #timvideos15:27
*** tsglove3 has joined #timvideos15:28
*** tsglove2 has joined #timvideos15:29
*** tsglove3 has quit IRC15:33
*** tsglove3 has joined #timvideos15:34
*** tsglove2 has quit IRC15:35
*** tsglove3 has quit IRC15:35
*** tsglove3 has joined #timvideos15:37
*** tsglove2 has joined #timvideos15:38
*** tsglove3 has quit IRC15:41
*** tsglove3 has joined #timvideos16:00
*** tsglove2 has quit IRC16:03
*** rohitksingh has joined #timvideos16:21
*** tsglove2 has joined #timvideos16:36
*** tsglove3 has quit IRC16:39
*** jwhitmore has joined #timvideos16:53
*** tsglove3 has joined #timvideos17:03
*** tsglove2 has quit IRC17:06
*** sb0 has quit IRC17:08
*** tsglove2 has joined #timvideos17:12
*** sb0 has joined #timvideos17:12
*** jwhitmore has quit IRC17:14
*** tsglove3 has quit IRC17:15
*** jwhitmore has joined #timvideos17:16
*** tsglove3 has joined #timvideos17:24
*** tsglove2 has quit IRC17:27
*** jwhitmore has quit IRC17:35
*** tsglove2 has joined #timvideos17:49
*** tsglove2 has quit IRC17:51
*** tsglove2 has joined #timvideos17:51
tumbleweedmithro: that build failed, but it seemed successful?
tpbTitle: debian Pastezone (at
*** tsglove3 has quit IRC17:53
*** tsglove3 has joined #timvideos17:55
*** tsglove2 has quit IRC17:58
*** jwhitmore has joined #timvideos18:05
*** rohitksingh has quit IRC18:19
*** rohitksingh has joined #timvideos18:33
*** tsglove2 has joined #timvideos18:34
*** tsglove3 has quit IRC18:37
*** tsglove3 has joined #timvideos18:40
*** tsglove2 has quit IRC18:43
*** rohitksingh has quit IRC19:07
*** rohitksingh has joined #timvideos19:13
*** tsglove3 has quit IRC19:22
*** tsglove3 has joined #timvideos19:22
*** rohitksingh has quit IRC19:54
*** tsglove2 has joined #timvideos20:01
*** tsglove3 has quit IRC20:04
*** aps has quit IRC20:24
*** tsglove3 has joined #timvideos20:35
*** tsglove2 has quit IRC20:38
tpbTitle: GitHub - 0xAX/0bt: 馃捒 馃惂 some experiments with loading of the Linux kernel (at
*** tsglove2 has joined #timvideos20:47
*** accesspoint has left #timvideos20:47
*** tsglove2 has quit IRC20:49
*** tsglove3 has quit IRC20:51
mithroMerge all the changes!20:57
tumbleweedthanks :)21:02
mithrotumbleweed: Need to take another look at
tpbTitle: Use CRLF line-endings by stefanor 路 Pull Request #341 路 timvideos/HDMI2USB-litex-firmware 路 GitHub (at
CarlFK[m]5:19 - pxe booted installer.. lets see how fast this new ssd thing works22:21
CarlFK[m]it's been 5 min... im not impressed :p22:26
mithroCarlFK[m]: So you going to do some testing?22:35
mithroCarlFK[m]: btw what relay boards did you buy?22:35
CarlFK[m]mithro : trying to. figured I should have a clean/known/up to date box22:36
tumbleweedmithro: any ideas about that build failure?22:36
tumbleweedI assume the CI failures on the CR-LF branch are due to you locking down the prebuilt repo?22:37
mithrotumbleweed: link?22:37
tpbTitle: debian Pastezone (at
CarlFK[m]SainSmart USB Eight Channel Relay Board for Automation - 12 V   Link:
CarlFK[m]somewhere is a comment describing how it flutters when you connect/reset the usb connection22:41
CarlFK[m]if you plug in usb, then power, all is well22:41
CarlFK[m]power, then usb and 3 of the relays click a few times22:41
CarlFK[m]also "sudo reboot" the pc that the usb is plugged int, and the relays click (which powercycles the opsis)22:42
CarlFK[m]so I would say lets look for another version and maybe it will be better22:44
mithroFYI - I think we are going to rename the "HDMI2USB-litex-firmware" to something like "litebuild" because we now have 2 projects using it for setup22:44
mithros/2 projects/3 projects/22:44
mithroThere will still be a HDMI2USB-litex-firmware for the HDMI2USB specific stuff22:44
mithroBut it will be much smaller22:44
CarlFK[m]heh 'Had no way to find ip address'22:48
*** tsglove2 has joined #timvideos22:50
*** jwhitmore has quit IRC22:57
CarlFK[m]"The relays DO switch all on/off very quickly when your computer boots up, but if you're fine with this it works well."22:58
CarlFK[m]" on computer startup, the board goes through some kind of initialization and trips the relays rapidly a few times." "Now I feel like I need a power control for the 12V for the relays so they are unpowered during reboot cycles ..."23:00
mithroCarlFK[m]: Is there a reason you picked that relay board?23:00
CarlFK[m]a ps1er said "I got something from that company"23:00
mithroThe current users of "litebuild" would be - HDMI2USB, MicroPython on FPGAs and Linux on LiteX23:01
CarlFK[m]I liked the usb-cb connector23:01
CarlFK[m]usb-b (no c)23:01
CarlFK[m]8 relays for $30 was the cheepest per relay cost. and I made up reasons why I might need 5.23:02
CarlFK[m]so yes, lots of reasons.  No good reasons.23:02
cr1901_modernmithro: I'm not 100% sure what you have in mind for litebuild, but throw ideas out there and I will tell you all the ways Windows makes it difficult :)23:05
cr1901_modern(this is hopefully a joke)23:05
mithrocr1901_modern: I'm thinking of trying to use meson stuff23:06
cr1901_modernmithro: Cool! I use meson for libmodem already without issue23:07
cr1901_modernNeed to still add sdcc support, but that shouldn't be too hard (priorities lol)23:08

Generated by 2.13.1 by Marius Gedminas - find it at!