*** tpb has joined #timvideos | 00:00 | |
mithro | wcmckee: So? Is that you :-P | 00:02 |
---|---|---|
*** williammckee has joined #timvideos | 00:02 | |
*** ewen has joined #timvideos | 00:02 | |
*** wcmckee has quit IRC | 00:02 | |
xfxf | CarlFK[m]: he's one of the new guys we're training up. make him feel welcome | 00:03 |
mithro | williammckee: Cool | 00:03 |
xfxf | I do need help with moving storage tho, seriously | 00:03 |
*** williammckee_ has joined #timvideos | 00:05 | |
*** ewen has joined #timvideos | 00:05 | |
*** ewen has joined #timvideos | 00:06 | |
*** williammckee has quit IRC | 00:07 | |
*** williammckee_ is now known as williammckee | 00:07 | |
mithro | xfxf: Dunno who you are expecting to respond? Are you after people still at the accommodation or people here at Google? | 00:09 |
xfxf | both ends. I'm omw there now and will figure it out when there | 00:09 |
mithro | williammckee: So we have a git repo here -> https://github.com/timvideos/HDMI2USB-firmware-prebuilt | 00:10 |
tpb | Title: GitHub - timvideos/HDMI2USB-firmware-prebuilt: Prebuilt firmware for the HDMI2USB devices (such as the Numato Opsis and the Digilent Atlys board) and OS drivers. (at github.com) | 00:10 |
mithro | williammckee: In the opsis/firmware directory of that github repo are three symlinks (see https://github.com/timvideos/HDMI2USB-firmware-prebuilt/tree/master/opsis/firmware) | 00:11 |
tpb | Title: HDMI2USB-firmware-prebuilt/opsis/firmware at master · timvideos/HDMI2USB-firmware-prebuilt · GitHub (at github.com) | 00:11 |
mithro | williammckee: The unstable symlink is automatically updated | 00:11 |
mithro | williammckee: We would like the stable/testing symlinks to be updated from the this spreadsheet -> https://docs.google.com/spreadsheets/d/14BENNLHrGddCI5ZVyB87Tmwt3fnEcA19S776xIYSpVo/edit#gid=0 | 00:12 |
tpb | Title: HDMI2USB Firmware Versions - Google Sheets (at docs.google.com) | 00:12 |
mithro | williammckee: The complication is that the repo ends up being like 18gigabytes when checked out on your disk | 00:13 |
ewen | mithro: Having opened FGPA kit I'm assuming the difference between the USB-A and USB-C kits is the hub that's included? And the FGPA board inside is the same? (Only cable in bag that fits the FGPA board seems to be USB-A, presumably to go into the hub) | 00:13 |
mithro | ewen: correct | 00:13 |
mithro | williammckee: There is a script here -> https://github.com/timvideos/HDMI2USB-litex-firmware/blob/master/.travis/download-prebuilt.sh -- which does a sparse, shallow clone of the repo which makes it much smaller | 00:14 |
tpb | Title: HDMI2USB-litex-firmware/download-prebuilt.sh at master · timvideos/HDMI2USB-litex-firmware · GitHub (at github.com) | 00:14 |
*** williammckee has quit IRC | 00:17 | |
*** williammckee has joined #timvideos | 00:18 | |
williammckee | mithro sorry, connection dropped out last message i got from you was from 11:11, 18gigs. have the chat log. | 00:21 |
mithro | https://logs.timvideos.us/%23timvideos/latest.log.html | 00:21 |
*** williammckee has quit IRC | 00:41 | |
*** rohitksingh has joined #timvideos | 00:43 | |
*** williammckee has joined #timvideos | 00:45 | |
ewen | mithro: Between the preparation instructions and the "now with hardware connected" instructions the prompt changes from "(LX P=arty C=or1k.linux F=linux)" to "(LX P=arty C=or1k R=litex-linux)". Is that indicative of missing steps? Or older example in the documentation? | 00:57 |
mithro | Older example I think | 00:58 |
ewen | Cool, I'll tweak document for newer example. | 00:58 |
ewen | mithro: Yes, memory initialisation failed when I did "make firmware-load" | 00:59 |
mithro | rohitksingh: How are you going with fixing the memory issue? | 00:59 |
ewen | FTR, memory failures on my attempt | 01:00 |
rohitksingh | mithro: So, yesterday's fix didn't seem to work :/ I was synthesizing gateware for the second time when everyone left for the cabs I killed the synthesis process | 01:00 |
ewen | Read delays: 1:32-33 0:32-33 completed | 01:00 |
ewen | Memtest bus failed: 64/256 errors | 01:00 |
ewen | Memtest data failed: 524288/524288 errors | 01:00 |
ewen | Memtest addr failed: 8191/8192 errors | 01:00 |
ewen | Memory initialization failed | 01:00 |
mithro | rohitksingh: Okay, lets look at it together shortly | 01:01 |
mithro | Need to finish a presentation for the office first | 01:01 |
mithro | williammckee: btw you all good? | 01:01 |
rohitksingh | mithro: sure. meanwhile I'll restart the build and try to get it working | 01:02 |
ewen | mithro: rohitksingh: I'd also like to sit in on looking at that to get an idea what you're changing. | 01:02 |
rohitksingh | ewen: sure! :) | 01:02 |
*** futarisIRCcloud has joined #timvideos | 01:06 | |
rohitksingh | ewen: https://github.com/enjoy-digital/litex/commit/9eb1beea041498af7bf99ed2142d72283ce1fd48 | 01:16 |
tpb | Title: fix DDR3 on arty · enjoy-digital/litex@9eb1bee · GitHub (at github.com) | 01:16 |
mithro | rohitksingh: if felix is alive, he has a working arty firmware... | 01:21 |
rohitksingh | mithro: I was gonna ask you :p Where is he? | 01:21 |
mithro | rohitksingh: I think he is unwell | 01:22 |
rohitksingh | mithro: oh :( | 01:22 |
mithro | Been working too hard | 01:22 |
rohitksingh | mithro: yup, he is right behind you in score board | 01:24 |
ewen | mithro: My github username is ewen-naos-nz | 01:33 |
mithro | ewen: invites to https://github.com/orgs/upy-fpga and https://github.com/orgs/fupy sent | 01:35 |
tpb | Title: MicroPython on FPGAs · GitHub (at github.com) | 01:35 |
mithro | ewen: Need to do the IRC channel and stuff too | 01:35 |
mithro | ewen: https://docs.google.com/document/d/1_YY2fW8DZHYG1pU1qAjK99PDjwdmdVL5DAf7bjZqEY8/edit | 01:37 |
tpb | Title: fupy - Things to rename - Google Docs (at docs.google.com) | 01:37 |
ewen | Cool, thanks | 01:37 |
ewen | mithro: My github account isn't obviously seeing invites to those organisations? Can you try again? ewen-naos-nz | 01:43 |
mithro | https://docs.google.com/document/d/1_YY2fW8DZHYG1pU1qAjK99PDjwdmdVL5DAf7bjZqEY8/edit | 01:44 |
tpb | Title: fupy - Things to rename - Google Docs (at docs.google.com) | 01:44 |
ewen | Ah, found it. It just didn't notify me. | 01:44 |
felix_ | CarlFK[m]: found your keys and will bring them to the venue | 01:50 |
felix_ | yep, that commit rohitksingh linked fixes dram on arty | 01:53 |
felix_ | after quite a lot of sleep i'm feeling much better again today. i'll walk to the venue now; please wait for me with lunch, since i haven't eaten breakfast | 01:54 |
*** CarlFK has quit IRC | 02:51 | |
*** CarlFK has joined #timvideos | 03:10 | |
*** ChanServ sets mode: +v CarlFK | 03:10 | |
*** thaytan has quit IRC | 03:14 | |
*** thaytan has joined #timvideos | 03:17 | |
*** ChanServ sets mode: +v thaytan | 03:17 | |
*** CarlFK has quit IRC | 03:19 | |
xfxf | can we please get some help carrying up stuff. full car, I'm blocking carpark, trolley has dissapeared | 03:20 |
xfxf | CarlFK[m]: stefanor et al | 03:20 |
xfxf | Thx | 03:20 |
ewen | Having played a bit with trying to upate litex to pick up the RAM fixes, and studying "make gateware" output, it appears that these warnings might be related to DDR3 not working correctly: | 04:05 |
ewen | WARNING: [DRC REQP-1839] RAMB36 async control check: The RAMB36E1 mor1kx/mor1kx_cpu/cappuccino.mor1kx_cpu/mor1kx_fetch_cappuccino/immu_gen.mor1kx_immu/itlb[0].itlb_match_regs/mem_reg has an input control pin mor1kx/mor1kx_cpu/cappuccino.mor1kx_cpu/mor1kx_fetch_cappuccino/immu_gen.mor1kx_immu/itlb[0].itlb_match_regs/mem_reg/ADDRARDADDR[10] (net: | 04:05 |
ewen | mor1kx/mor1kx_cpu/cappuccino.mor1kx_cpu/mor1kx_fetch_cappuccino/immu_gen.mor1kx_immu/itlb[0].itlb_match_regs/ctrl_alu_result_o_reg[7][5]) which is driven by a register (FDPE_1) that has an active asychronous set or reset. This may cause corruption of the memory contents and/or read values when the set/reset is asserted and is not analyzed by the default static timing analysis. It is suggested to eliminate | 04:05 |
ewen | the use of a set/reset to registers driving this RAMB pin or else use a synchronous reset in which the assertion of the reset is timed by default. | 04:05 |
ewen | There's perhaps 30-ish of them (presumably different pins). | 04:05 |
*** CarlFK has joined #timvideos | 04:31 | |
*** ChanServ sets mode: +v CarlFK | 04:31 | |
ewen | Of note, synthesis has 100 (exactly) "Unused sequential element [...] was removed" | 04:34 |
ewen | Which is possibly related to later warnings? | 04:34 |
ewen | Overall synthesis of gateware has about 680 warnings (mostly in various groups of warnings repeated 10s/100s of times), not counting warnings about Verilog / SystemVerilog not specified | 04:38 |
jea | xfxf: if we have people email us (team@lca) about things involving AV, such as 'my talk has audio in it', should we forward that on to you? or would you like it handled somewhere else? | 04:57 |
jea | like, we could probably give you access to an RT queue if you want. (not that I think you would :P) | 04:57 |
mithro | https://theamphour.com/375-an-interview-with-tim-mithro-ansell/ | 05:08 |
mithro | I can't bare to listen to it.... Can someone check if it sounds okay :-P | 05:08 |
mithro | rohitksingh: ^ | 05:09 |
ewen | mithro: listening now. Both you and host say "umm"/"uh" a bunch, which in some recordings would get cleaned up. But otherwise sounds great interview so far, lots of interesting information, on same level with rest of the Amp Hour interviews (I've heard a bunch of Amp Hour podcasts) | 05:24 |
*** futarisIRCcloud has quit IRC | 05:46 | |
xfxf | jea: forward would be good to ryan@nextdayvideo | 06:17 |
jea | xfxf: cool, will do | 06:17 |
ewen | mithro: per pull request comment, RAM issues were fixed by building with rohitksingh/your changes; have been able to boot to H2U> prompt, and also telnet to CPU running on FPGA. Moving on to getting it to boot Linux | 06:46 |
rohitksingh | ewen: awesome! | 06:49 |
ewen | Booted linux via TFTP on FPGA: | 07:00 |
ewen | # uname -a | 07:00 |
ewen | Linux buildroot 4.14.0-rc2-131008-ga7a47e6 #2 Mon Jan 15 17:47:17 AEDT 2018 openrisc GNU/Linux | 07:00 |
ewen | mithro: Listened to whole Amp Hour podcast. Great interview, with heaps of useful pointers for people. Should come across well. | 07:04 |
* ewen disappears off for the evening; back 10:30ish tomorrow | 07:14 | |
*** ewen has quit IRC | 07:15 | |
*** xobs has joined #timvideos | 07:57 | |
*** rqou has quit IRC | 08:02 | |
mithro | I'm heading back to the office now | 08:17 |
mithro | https://pypi.python.org/pypi/pyverilog | 08:22 |
tpb | Title: pyverilog 1.1.1 : Python Package Index (at pypi.python.org) | 08:22 |
mithro | Gah my badge is not working | 08:28 |
mithro | xfxf: Xobs arrives at about 10am tomorrow - his plan is to go to the hotel and drop stuff off then come to Google | 08:30 |
mithro | xfxf: does that sound like the best plan? | 08:30 |
xfxf | ya he already emailed me. i'm gonna leave his key at front desk | 08:30 |
xfxf | (of hotel) | 08:30 |
*** sb0 has quit IRC | 08:37 | |
_florent_ | hi, i should have updated all ddr3 parameters in the HDMI2USB-litex-firmware, i should have done that before, sorry for the troubles | 08:43 |
mithro | Hey _florent_ | 08:44 |
_florent_ | hi mithro | 08:44 |
mithro | _florent_: I'm in the process of testing https://github.com/enjoy-digital/litex/pull/50 with the HDMI2USB designs | 08:44 |
tpb | Title: Import migen+misoc changes by mithro · Pull Request #50 · enjoy-digital/litex · GitHub (at github.com) | 08:44 |
_florent_ | mithro: ok good, not sure how we'll do the merge, i'll probably pick up the commits of your pull request manually | 08:49 |
mithro | _florent_: It would make future merging easier if you just merged the pull request | 08:52 |
mithro | _florent_: What hardware / designs do you need to test with? | 08:52 |
_florent_ | mithro: ok then i'll comment on the pull request | 08:53 |
mithro | _florent_: I'm trying to get things into a state were I can do automated merges much more easily | 08:54 |
mithro | _florent_: Also trying to get some more automated CI stuff going | 08:56 |
*** Sigyn has joined #timvideos | 08:57 | |
*** sb0 has joined #timvideos | 09:08 | |
mithro | _florent_: How are you programming your NeTV2v2 when using as a PCIe board? | 09:16 |
_florent_ | mithro: i'm programming in sram, then rebooting the computer | 09:19 |
mithro | _florent_: I mean - your using a jtag programming cable or something? | 09:20 |
_florent_ | mithro: yes with a jtag hs2 | 09:20 |
mithro | _florent_: Rohit is about to try and get my netv2v2 to work with your soc | 09:21 |
_florent_ | mithro: ok, first thing to check is lspci --vv | 09:21 |
_florent_ | mithro: try to see if the board is seen and recognized as gen2 x2 | 09:22 |
mithro | _florent_: I think we want to see if we can get the uart working first :-P | 09:23 |
mithro | rohitksingh: https://github.com/bunnie/netv2mvp-hardware | 09:23 |
tpb | Title: GitHub - bunnie/netv2mvp-hardware: Hardware for NeTV2MVP (at github.com) | 09:23 |
_florent_ | mithro: which targert are you testing? video or pcie? | 09:24 |
mithro | _florent_: I think he is starting with base :-P | 09:25 |
mithro | _florent_: Give us a while, still getting computers in the right place and stuff :-P | 09:30 |
felix_ | rohitksingh: 'openocd -f board/kc705.cfg -c init -c "pld load 0 bitstream.bit" -c exit' is what i use to load the bitstream in the sram of the fpga on an ac701; doing that and then rebooting the machine works well for testing pcie stuff | 09:37 |
felix_ | if the netv2 uses a different programmer, you have to duplicate the kc705.cfg file and edit that accordingly | 09:38 |
jea | is everyone still going down at google? | 09:52 |
rohitksingh | felix_: thanks! let me try | 09:52 |
mithro | rohitksingh: https://github.com/timvideos/HDMI2USB/wiki/Xilinx-Platform-Cable-USB-under-Linux | 10:21 |
tpb | Title: Xilinx Platform Cable USB under Linux · timvideos/HDMI2USB Wiki · GitHub (at github.com) | 10:21 |
mithro | jea: I'm still here | 10:21 |
mithro | But everyone else went to get food | 10:21 |
mithro | they are bringing food back hopefully | 10:21 |
CarlFK[m] | Praying now | 11:05 |
mithro | _florent_: We are having trouble programming the netv2v2 - what pin are you using as the vref? | 11:08 |
*** rqou has joined #timvideos | 11:08 | |
mithro | Does anyone at the hackfest have an RPi? | 11:11 |
*** sb0 has quit IRC | 11:14 | |
*** rqou has quit IRC | 11:16 | |
jea | I left mine at home this year | 11:17 |
jea | if really needed i could get people to bring it down on the weekend | 11:18 |
*** rqou has joined #timvideos | 11:19 | |
_florent_ | mithro: ah yes, you have to remove r24c and keep the mcu in reset (connect sreset/pin17 iof p11c to gnd) | 11:22 |
rohitksingh | _florent_: yeah we found that out after much pain | 11:23 |
_florent_ | rohitksingh: so it's working now? | 11:23 |
rohitksingh | _florent_ : just got to get it detected in JTAG...searching for a soldering iron now | 11:24 |
_florent_ | rohitksingh: ok | 11:25 |
*** olasd has quit IRC | 11:35 | |
*** sb0 has joined #timvideos | 11:38 | |
*** olasd has joined #timvideos | 11:52 | |
shorne | mithro: you pinged? | 12:13 |
*** sb0 has quit IRC | 12:19 | |
CarlFK | shorne: he went next door to do some soldering | 12:33 |
*** sb0 has joined #timvideos | 13:01 | |
*** CarlFK has quit IRC | 13:06 | |
*** rohitksingh has quit IRC | 13:10 | |
*** CarlFK has joined #timvideos | 13:39 | |
*** ChanServ sets mode: +v CarlFK | 13:39 | |
*** TimGremalm has joined #timvideos | 17:06 | |
*** hozer has joined #timvideos | 17:13 | |
*** CarlFK has quit IRC | 21:28 | |
CarlFK[m] | coming down for breakfast at Song bird hotel | 21:28 |
CarlFK[m] | jea: we are across the street Getting Coffee go up across the street come back down a block | 22:33 |
CarlFK[m] | Got coffee, walking to venue | 22:35 |
jea[m] | CarlFK: I’ve just left, will catch up with you soon | 22:38 |
CarlFK[m] | mithro: gsoc | 22:51 |
jea[m] | CarlFK: which way did you walk? | 22:54 |
CarlFK[m] | I am on the path with the bin chicken | 22:57 |
*** CarlFK has joined #timvideos | 23:38 | |
*** ChanServ sets mode: +v CarlFK | 23:38 |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!