*** tpb has joined #litex | 00:00 | |
*** st-gourichon-f has joined #litex | 01:47 | |
*** st-gourichon-fid has quit IRC | 01:47 | |
*** benh has quit IRC | 02:10 | |
*** benh has joined #litex | 02:11 | |
*** jaseg has quit IRC | 02:49 | |
*** jaseg has joined #litex | 02:52 | |
*** _florent_ has quit IRC | 03:22 | |
*** _florent_ has joined #litex | 03:23 | |
*** Degi has quit IRC | 03:30 | |
*** Degi has joined #litex | 03:30 | |
*** CarlFK has joined #litex | 04:16 | |
*** proteusguy has quit IRC | 05:54 | |
*** proteusguy has joined #litex | 06:07 | |
*** mescobar has joined #litex | 06:45 | |
*** mescobar has quit IRC | 06:54 | |
*** _whitelogger has quit IRC | 08:09 | |
*** _whitelogger has joined #litex | 08:11 | |
*** Dolu has joined #litex | 08:13 | |
*** st-gourichon-f has quit IRC | 08:47 | |
*** st-gourichon-fid has joined #litex | 08:47 | |
st-gourichon-fid | Hi! Trying to build a simple design for fomu pvt. https://github.com/litex-hub/litex-boards says "./target.py --build --load". "fomu.py" rejects "--load". | 10:23 |
---|---|---|
tpb | Title: GitHub - litex-hub/litex-boards: LiteX boards files (at github.com) | 10:23 |
st-gourichon-fid | "./fomu.py --build --board pvt" yields "litex.build.generic_platform.ConstraintError: Resource not found: serial:None" complete log on https://pastebin.ubuntu.com/p/HNNVnMhQdr/ | 10:24 |
tpb | Title: Ubuntu Pastebin (at pastebin.ubuntu.com) | 10:24 |
*** st-gourichon-f has joined #litex | 11:41 | |
*** st-gourichon-f has quit IRC | 11:58 | |
pepijndevos | Thanks florent | 12:49 |
st-gourichon-fid | Hi again. My X-not-Y problem is to build something that runs on a FOMU with a SERV CPU. Any hint? | 16:03 |
st-gourichon-fid | To run a FOMU with a SERV CPU, I tried starting with https://github.com/im-tomu/foboot and "cd hw ; python3 foboot-bitstream.py --revision pvt" but changing CPU type to "serv", but the litex used to build fomu seems too old and does not know SERV. | 16:03 |
tpb | Title: GitHub - im-tomu/foboot: Bootloader for Fomu (at github.com) | 16:03 |
st-gourichon-fid | I tried then updating the litex used to build foboot, but then the build breaks, even with vanilla foboot-bistream.py. Looks like breaking changes have made newer litex incompatible with this foboot "447c5da (HEAD, tag: v2.0.3) valentyusb: fix usb reset irq storm" | 16:03 |
*** a1k0n_ has joined #litex | 16:35 | |
*** Dolu has quit IRC | 18:37 | |
leons | Hi, I'm still kind of a noob at LiteX. Currenly playing around with FOMU. For a research project I'm trying to get a RISC-V platform with Ethernet MAC up and running. Is it possible (with reasonable effort) to get a LiteX VexRisc core with LiteEth running on an Arty A7 35T? | 19:00 |
leons | Of course willing to put in the effort of learning how to use LiteX, but it'd be good to know whether this combination would work prior to buying an expensive shiny FPGA board :) | 19:01 |
daveshah | I think it _should_ be as simple as passing --with-ethernet to the target Python script | 19:04 |
daveshah | I've definitely used that combo in the past and know it works but maybe LiteX has changed since then | 19:04 |
leons | daveshah: That'd be insanely cool. The fact that this combination worked for you makes me confident that one should at least be able to somehow make it work again. :) | 19:06 |
daveshah | The Arty A7 is one of the better supported boards, so you definitely made a good choice :) | 19:08 |
daveshah | I would expect any issues you do find will be fixed quickly as a lot of us have one | 19:08 |
leons | Just to clarify (I didn't find this in the wikis on GitHub). LiteX targets the Arty A7 35T - not the 100T, right? | 19:11 |
*** dogisfat has joined #litex | 19:11 | |
daveshah | Yes | 19:15 |
daveshah | I think it might support the 100T too but I don't have one so can't vouch for that | 19:16 |
_florent_ | leons: you can use https://github.com/litex-hub/litex-boards/blob/master/litex_boards/targets/arty.py | 19:24 |
tpb | Title: litex-boards/arty.py at master · litex-hub/litex-boards · GitHub (at github.com) | 19:24 |
_florent_ | ./arty.py --with-ethernet --build --load | 19:24 |
_florent_ | and it should try to netboot from 192.168.1.50 (will look for boot.json or boot.bin) | 19:25 |
_florent_ | https://github.com/enjoy-digital/litex/wiki/Load-Application-Code-To-CPU#ethernet-boot | 19:25 |
tpb | Title: Load Application Code To CPU · enjoy-digital/litex Wiki · GitHub (at github.com) | 19:25 |
*** dogisfat has quit IRC | 19:26 | |
leons | florent: that is really cool. I will try to get my hands on an Arty board as quick as possible. Thanks! | 19:27 |
leons | In the meantime I need to port my the embedded OS I'm planning to use to LiteX (which I can probably do with Renode it seems) and get LiteEth software support. So there's plenty to do :) | 19:29 |
daveshah | It should also be possible to do ethernet debug with litex_sim but that is lower level and slower than renode | 19:31 |
leons | In general I should be able to do all of the development with Renode, right? Or can't I use LiteEth with Renode? | 19:33 |
leons | https://github.com/timvideos/litex-buildenv/wiki/Renode | 19:34 |
tpb | Title: Renode · timvideos/litex-buildenv Wiki · GitHub (at github.com) | 19:34 |
leons | Since it runs the same bitstream(?) as the Arty A7 I thought it'd also feature an MII / MDIO interface compatible with LiteEth | 19:35 |
mithro | leons: I believe that Renode has LiteEth emulation | 19:39 |
daveshah | Yes, looks like it based on that tutorial | 19:39 |
mithro | https://docs.google.com/spreadsheets/d/1XTHfdYXuvwoYdPXm4M6qDA0D2fZCPy220-9q6qZpTw4/edit#gid=1131619550 | 19:39 |
tpb | Title: LiteX BuildEnv Support - Google Sheets (at docs.google.com) | 19:39 |
leons | mithro, daveshah: thanks, that looks promising. I'll try to get it up and running on Renode in the next days. super excited about this :) | 19:58 |
mithro | leons: Renode is very useful for CI | 20:03 |
mithro | See the robot tests and https://interrupt.memfault.com/blog/test-automation-renode | 20:03 |
tpb | Title: Firmware Testing with Renode and GitHub Actions | InterruptFirmware Testing with Renode and GitHub Actions (at interrupt.memfault.com) | 20:03 |
leons | mithro: I can imagine. In the embedded OS I'm contributing to we're currently running CI on QEMU - which is better than nothing | 20:05 |
leons | But testing on "real" hardware would of course be much nicer | 20:05 |
mithro | leons: Renode is significantly better than Qemu for CI -- it can be 100% deterministic | 20:06 |
leons | QEMU has repeatedly showed very imprecise emulation, especially of chip peripherals, many of which only implement the absolute basic functionality | 20:06 |
leons | (having heard of Renode for the first time today) I'm still struggling to understand it's exact relationship to LiteX | 20:13 |
sorear | as far as I know there is absolutely no relation beyond that antmicro uses both | 20:14 |
leons | sorear: So it just has a configuration that provides an emulated HW similar to what would be synthesized by LiteX for a specific board? | 20:15 |
*** lkcl_ has joined #litex | 20:16 | |
*** lkcl has quit IRC | 20:20 | |
*** st-gourichon-fid has quit IRC | 20:30 | |
*** st-gourichon-fid has joined #litex | 20:30 | |
*** st-gourichon-fid has quit IRC | 20:30 | |
*** st-gourichon-fid has joined #litex | 20:31 | |
*** lf has quit IRC | 23:39 | |
*** lf has joined #litex | 23:40 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!