| *** tpb has joined #symbiflow | 00:00 | |
| FFY00 | so we would just maintain packages for our projects | 00:01 |
|---|---|---|
| FFY00 | so we wouldn't be maintaining an ecosystem | 00:02 |
| FFY00 | does that make sense? | 00:02 |
| FFY00 | but yeah, if conda works fine then this is not needed | 00:08 |
| *** Degi has quit IRC | 00:23 | |
| *** Degi has joined #symbiflow | 00:24 | |
| *** acomodi has quit IRC | 00:31 | |
| daniellimws | mithro: Which branch do I rebase my branch to? To get the readthedocs build | 01:29 |
| mithro | @daniellimws Looks like it is already working | 01:30 |
| daniellimws | mithro: The build failed :( And I should force push or something to trigger it on the docs-clock branch? | 01:31 |
| -_whitenotifier-3- [python-symbiflow-v2x] mithro opened issue #43: readthedocs run fails with "TypeError: connect() got an unexpected keyword argument 'priority'" - https://git.io/JfvpJ | 01:31 | |
| mithro | @daniellimws Seems to be happening on head to https://github.com/SymbiFlow/python-symbiflow-v2x/issues/43 | 01:32 |
| mithro | daniellimws: Can you see https://readthedocs.org/projects/python-symbiflow-v2x/builds/10845965/ ? | 01:34 |
| daniellimws | mithro: Yes I see "build failed". Is that correct? | 01:35 |
| mithro | daniellimws: Can you see the logs? | 01:35 |
| daniellimws | mithro: Yep I can | 01:35 |
| mithro | daniellimws: I think it's a sphinx version issue? | 01:38 |
| *** citypw has joined #symbiflow | 01:44 | |
| daniellimws | mithro: This looks weird, I didn't even change anything related to sphinx | 01:48 |
| mithro | daniellimws: Its also happening in master | 01:48 |
| mithro | daniellimws: Testing to see if https://github.com/SymbiFlow/python-symbiflow-v2x/pull/44 has any affect | 01:50 |
| tpb | Title: Require latest version of sphinx. by mithro · Pull Request #44 · SymbiFlow/python-symbiflow-v2x · GitHub (at github.com) | 01:50 |
| mithro | @daniellimws Appears not :-/ | 01:53 |
| mithro | daniellimws: Was a shot in the dark, but thought it was low effort to try | 01:53 |
| mithro | daniellimws: I've repoed locally | 01:58 |
| daniellimws | mithro: Oh nice, is it a problem with the version? | 02:05 |
| mithro | https://github.com/readthedocs/readthedocs.org/issues/6870 | 02:05 |
| daniellimws | Are you going to try version 2.2.2? | 02:08 |
| *** _whitelogger has quit IRC | 02:15 | |
| mithro | daniellimws: BTW Regarding your comment "Doesn't a D-flipflop only have input D and output Q? Should we rename this to just flipflop instead of D-flipflop?" -- I'm actually not sure :-P | 02:16 |
| daniellimws | mithro: Haha ok let's see what the other reviewers say about it | 02:16 |
| *** _whitelogger has joined #symbiflow | 02:17 | |
| mithro | daniellimws: I think I just used the name the upstream used.... | 02:17 |
| daniellimws | Yeah, the verilog modules also start with dff | 02:21 |
| mithro | daniellimws: I'm going to add you as a collaborator to the python-symbiflow-v2x repo so you can take issues and assign them to yourself if you start working on them | 02:31 |
| mithro | daniellimws: Do you have a Fomu? | 02:31 |
| daniellimws | mithro: Nope :p Looks really cool though | 02:31 |
| mithro | daniellimws: Would you like one? | 02:31 |
| daniellimws | mithro: Oh definitely :D | 02:32 |
| mithro | daniellimws: Send me an email with a postal address and phone number and I'll hook you up | 02:32 |
| daniellimws | mithro: Especially since I'm returning my FPGA board to the school now | 02:32 |
| mithro | daniellimws: Generally, people who contribute and need hardware tend to find it turning up :-) | 02:33 |
| daniellimws | mithro: That's really awesome | 02:33 |
| mithro | daniellimws: [email protected] | 02:34 |
| mithro | daniellimws: What FPGA board did you have through Uni? My guess is an Arty? | 02:35 |
| mithro | daniellimws: Looks like this pull request fixed it -> https://github.com/SymbiFlow/python-symbiflow-v2x/pull/45 | 02:35 |
| tpb | Title: Add the readthedocs dependencies to environment.yml by mithro · Pull Request #45 · SymbiFlow/python-symbiflow-v2x · GitHub (at github.com) | 02:35 |
| daniellimws | mithro: It was a basys3 dev board, so far only had this one semester to use it, and the semester is ending | 02:36 |
| mithro | I don't recommend going to this page -> https://joelw.id.au/FPGA/CheapFPGADevelopmentBoards -- You'll end up very poor if you are anything like me :-P | 02:36 |
| tpb | Title: Cheap FPGA Development Boards | Joel's Compendium of Total Knowledge (at joelw.id.au) | 02:36 |
| daniellimws | mithro: And it was annoying that I had to download a 20GB toolchain for simple projects, and also a virtual machine since I'm on Mac | 02:37 |
| daniellimws | mithro: Hope symbiflow settles this problem | 02:37 |
| mithro | @daniellimws: That is our goal! | 02:37 |
| daniellimws | mithro: I'll merge #45, ok? | 02:41 |
| mithro | daniellimws: You should rebase your changes on top of master now #45 is merged | 02:44 |
| *** az0re has quit IRC | 02:54 | |
| *** az0re has joined #symbiflow | 02:54 | |
| *** gsmecher has joined #symbiflow | 02:55 | |
| mithro | FFY00: Did I ever share https://docs.google.com/spreadsheets/d/1B2pXefk0tOIx0h-JO_e1SP3JQsKQUqseBLpVMRH8aa4/edit#gid=1527456559 with you? | 02:57 |
| tpb | Title: LiteX / SymbiFlow / TimVideos Conda Packages Usage - Google Sheets (at docs.google.com) | 02:57 |
| FFY00 | nop | 02:58 |
| mithro | FFY00: That is kinda of a matrix of what uses what packages | 02:58 |
| mithro | (See the multiple sheets) | 02:59 |
| FFY00 | btw, you can make gcc link to a different libc than the default one | 03:00 |
| FFY00 | is there any reason this is not done? | 03:01 |
| FFY00 | it would save a lot of gcc builds | 03:01 |
| mithro | FFY00: It's more complicated than that, there are a lots of subtile dependencies when you start talking about libc, calling conventions, libgcc / soft float / etc | 03:02 |
| FFY00 | hum | 03:02 |
| FFY00 | what about linux vs elf with the same libc? | 03:03 |
| mithro | You mean baremetal verse linux? | 03:04 |
| FFY00 | yes | 03:04 |
| FFY00 | do we still need 2 gcc? | 03:04 |
| mithro | FFY00: Yes, baremetal and linux can have different calling conventions and stuff like syscall stuff | 03:04 |
| mithro | FFY00: Learned this the hard way through many lost hours of pulling hair out | 03:05 |
| FFY00 | yeah | 03:05 |
| FFY00 | what about riscv64 with multilib support? | 03:06 |
| FFY00 | do we still need riscv32> | 03:06 |
| FFY00 | you should be able to just pass -m32 | 03:07 |
| mithro | https://wiki.osdev.org/Libgcc_without_red_zone | 03:09 |
| tpb | Title: Libgcc without red zone - OSDev Wiki (at wiki.osdev.org) | 03:09 |
| mithro | FFY00: generally things /should/ work a multilib support is enabled but it is surprising how many things still cause headaches... | 03:10 |
| FFY00 | right | 03:11 |
| FFY00 | one riscv dev was asking why I put riscv64 and riscv32 in the repos | 03:11 |
| FFY00 | is there something in particular I should be getting from that wiki page? | 03:13 |
| FFY00 | I think I missed it | 03:13 |
| mithro | FFY00: An example of some complexity when dealing with various options | 03:15 |
| FFY00 | I got that part | 03:16 |
| FFY00 | but as long as support is there it should work | 03:16 |
| FFY00 | the question is | 03:16 |
| FFY00 | are you expecting a riscv32 instead of riscv64 when compiling things? | 03:17 |
| mithro | FFY00: Most of our riscv targets are 32bit rather than 64bit | 03:18 |
| FFY00 | right, but does the build system support using a riscv64 compiler? | 03:19 |
| FFY00 | because I need to provide some sort of justification to have it in the repos | 03:19 |
| mithro | FFY00: yes, I believe it does | 03:20 |
| FFY00 | okay, so I can remove it | 03:21 |
| FFY00 | if needed in the future I still have the build scripts so I can just build it again | 03:21 |
| mithro | https://wiki.osdev.org/OS_Specific_Toolchain and https://wiki.osdev.org/Creating_a_C_Library have a lot of interesting information about things in this area too | 03:22 |
| tpb | Title: OS Specific Toolchain - OSDev Wiki (at wiki.osdev.org) | 03:22 |
| FFY00 | thanks | 03:23 |
| FFY00 | I totally forgot this was in osdev | 03:23 |
| FFY00 | what was the usb implementation you were telling me about? | 03:23 |
| FFY00 | I think I know it but I can't remember the name | 03:24 |
| mithro | FFY00: ValentyUSB? daisho? | 03:29 |
| mithro | Luna? | 03:29 |
| *** space_zealot has quit IRC | 03:29 | |
| FFY00 | yes | 03:30 |
| FFY00 | valentyusb | 03:30 |
| FFY00 | is this the one you told me about in the meeting? | 03:30 |
| mithro | https://github.com/im-tomu/valentyusb | 03:30 |
| tpb | Title: GitHub - im-tomu/valentyusb: FPGA USB stack written in LiteX (at github.com) | 03:30 |
| *** space_zealot has joined #symbiflow | 03:30 | |
| mithro | FFY00: ValentyUSB is USB1.1 -- for USB3.0 you want the daisho (https://github.com/enjoy-digital/usb3_pipe) and in the future LUNA stuff that ktemkin is working on | 03:31 |
| tpb | Title: GitHub - enjoy-digital/usb3_pipe: USB3 PIPE interface for Xilinx 7-Series / Lattice ECP5 (at github.com) | 03:31 |
| mithro | FFY00: xobs did a lot of the work on ValentyUSB and foboot -> https://github.com/im-tomu/foboot | 03:32 |
| tpb | Title: GitHub - im-tomu/foboot: Bootloader for Fomu (at github.com) | 03:32 |
| mithro | Going to have some dinner, might be back later, might not | 03:34 |
| FFY00 | I might be able to help with HID stuff if you are interested | 03:35 |
| FFY00 | I'll probably leave soon | 03:35 |
| *** space_zealot has quit IRC | 03:35 | |
| *** gsmecher has quit IRC | 03:48 | |
| sf-slack | <timo.callahan> @kgugala please disregard my earlier message, I was trying to use the git flow from my previous job :) | 04:01 |
| *** Bertl_oO is now known as Bertl_zZ | 04:13 | |
| *** titanbiscuit has quit IRC | 05:35 | |
| *** titanbiscuit has joined #symbiflow | 05:36 | |
| *** _whitelogger has quit IRC | 05:48 | |
| *** _whitelogger has joined #symbiflow | 05:50 | |
| *** OmniMancer has quit IRC | 06:37 | |
| *** OmniMancer has joined #symbiflow | 06:39 | |
| *** kraiskil has joined #symbiflow | 07:57 | |
| *** kraiskil has quit IRC | 08:02 | |
| *** kraiskil has joined #symbiflow | 08:14 | |
| *** kraiskil has quit IRC | 09:15 | |
| *** kraiskil has joined #symbiflow | 09:28 | |
| *** kraiskil has quit IRC | 10:28 | |
| *** space_zealot has joined #symbiflow | 10:31 | |
| *** kraiskil has joined #symbiflow | 10:41 | |
| *** kraiskil has quit IRC | 10:49 | |
| *** Bertl_zZ is now known as Bertl | 11:00 | |
| *** kraiskil has joined #symbiflow | 11:04 | |
| *** kraiskil has quit IRC | 11:33 | |
| -_whitenotifier-3- [symbiflow-arch-defs] rakeshm75 opened issue #1424: Branch : Quicklogic : Error during generation of top_bit.v file - https://git.io/Jffsy | 11:45 | |
| *** kraiskil has joined #symbiflow | 11:45 | |
| *** space_zealot has quit IRC | 12:43 | |
| *** fengling has quit IRC | 12:44 | |
| *** space_zealot has joined #symbiflow | 14:39 | |
| *** kraiskil has quit IRC | 14:41 | |
| *** kraiskil has joined #symbiflow | 14:53 | |
| *** space_zealot has quit IRC | 14:53 | |
| *** OmniMancer has quit IRC | 14:56 | |
| *** proteus-guy has quit IRC | 14:57 | |
| *** proteus-guy has joined #symbiflow | 15:01 | |
| *** nonlinear has joined #symbiflow | 15:18 | |
| *** nonlinear1 has quit IRC | 15:18 | |
| *** gsmecher has joined #symbiflow | 15:36 | |
| *** citypw has quit IRC | 15:39 | |
| *** Bertl is now known as Bertl_oO | 15:48 | |
| *** FFY00 has quit IRC | 16:03 | |
| *** FFY00 has joined #symbiflow | 16:04 | |
| *** futarisIRCcloud has quit IRC | 16:34 | |
| -_whitenotifier-3- [symbiflow-arch-defs] acomodi opened issue #1425: Cannot pack MUXF8 with new yosys master+wip - https://git.io/Jffay | 16:50 | |
| *** lopsided98 has quit IRC | 17:40 | |
| *** lopsided98 has joined #symbiflow | 17:41 | |
| -_whitenotifier-3- [python-symbiflow-v2x] mithro opened issue #47: Figure out how to skip copyright header in verilog output in rendered Sphinx docs - https://git.io/JffwA | 17:43 | |
| -_whitenotifier-3- [python-symbiflow-v2x] mithro opened issue #48: Figure out how to use netlistsvg with single whitebox modules - https://git.io/Jffrv | 17:46 | |
| *** space_zealot has joined #symbiflow | 17:48 | |
| -_whitenotifier-3- [python-symbiflow-v2x] mithro opened issue #49: Add symbolator rendered module blocks to the documentation - https://git.io/JffrT | 17:50 | |
| *** space_zealot has quit IRC | 18:17 | |
| -_whitenotifier-3- [python-symbiflow-v2x] mithro opened issue #50: Come up with a good "template" for documentation of the tests - https://git.io/Jffom | 18:29 | |
| *** space_zealot has joined #symbiflow | 19:24 | |
| *** space_zealot has quit IRC | 19:41 | |
| -_whitenotifier-3- [fpga-tool-perf] mithro opened issue #67: Make sure kokoro is using the symbiflow-kokoro user on this repo - https://git.io/Jff61 | 19:42 | |
| mithro | @acomodi: Great work on getting the tar of the automatic generation of the tarball pull request merged! https://github.com/SymbiFlow/symbiflow-arch-defs/pull/1406#event-3238518867 | 19:44 |
| tpb | Title: install: create tar of the installed toolchain by acomodi · Pull Request #1406 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com) | 19:44 |
| mithro | @acomodi: How big are the resulting output? | 19:44 |
| mithro | Looks like 800mb -> https://console.cloud.google.com/storage/browser/symbiflow-arch-defs/artifacts/prod/foss-fpga-tools/symbiflow-arch-defs/continuous/install/4/20200416-002215/ | 19:48 |
| tpb | Title: Google Cloud Platform (at console.cloud.google.com) | 19:48 |
| *** ayazar has joined #symbiflow | 20:02 | |
| *** space_zealot has joined #symbiflow | 20:05 | |
| *** adjtm has quit IRC | 20:19 | |
| -_whitenotifier-3- [python-symbiflow-v2x] mithro opened issue #51: Add a `make format` and `make format-check` to the repository. - https://git.io/JffP1 | 20:31 | |
| -_whitenotifier-3- [python-symbiflow-v2x] mithro opened issue #52: Add verible's Verilog auto-formatter to the repository - https://git.io/JffPy | 20:33 | |
| -_whitenotifier-3- [python-symbiflow-v2x] mithro opened issue #53: Make sure that a `v2x` binary is installed using setup.py's script entry points - https://git.io/JffPx | 20:37 | |
| *** space_zealot has quit IRC | 20:37 | |
| -_whitenotifier-3- [python-symbiflow-v2x] mithro opened issue #54: Make sure the contents of this `README.md` appear in the readthedocs output - https://git.io/JffPp | 20:38 | |
| -_whitenotifier-3- [python-symbiflow-v2x] mithro opened issue #55: Make sure info in @kgugala's talk is in the v2x documentation - https://git.io/JffXe | 20:40 | |
| *** OmniMancer has joined #symbiflow | 21:15 | |
| *** OmniMancer1 has joined #symbiflow | 21:16 | |
| *** OmniMancer has quit IRC | 21:19 | |
| sf-slack | <acomodi> mithro: yep, around 850M, will check tomorrow if everything is all right with the package by using that in tool-perf | 21:25 |
| *** space_zealot has joined #symbiflow | 21:34 | |
| sf-slack | <pgielda> decompressed its 12G | 21:43 |
| sf-slack | <pgielda> most of it /share | 21:43 |
| sf-slack | <pgielda> and most of share is | 21:43 |
| sf-slack | <pgielda> 2674296./arch/xc7a50t-bottom_test 3274000./arch/xc7z010_test 5878852./arch/xc7a50t_test | 21:44 |
| sf-slack | <pgielda> is this needed in the install package? | 21:46 |
| sf-slack | <pgielda> we also have things like | 21:46 |
| sf-slack | <pgielda> ./prjxray/prjxray-db/.travis | 21:46 |
| sf-slack | <pgielda> probably not needed as well | 21:46 |
| sf-slack | <acomodi> so, the bottom_test is actually not needed, and definitely neither the travis file. I think also the xray-db files are not needed as they can be downloaded separately | 21:49 |
| litghost | pgielda: The only thing that are needed are the final arch.xml, the final rrgraph.capnp (and maybe the .rrgraph.xml), the lookahead.capnp and place_delay.capnp, plus a handful of other small files | 21:50 |
| litghost | pgielda: E.g. pinmap files | 21:50 |
| *** adjtm has joined #symbiflow | 21:51 | |
| sf-slack | <acomodi> I'll get ready a follow-up PR to have only what is needed | 21:51 |
| litghost | acomodi: If we want to keep the rrgraph.xml in the tarball, I suggest compressing it before going into the tarball, and the user can decompress it if they need it | 21:54 |
| *** kraiskil has quit IRC | 21:54 | |
| *** space_zealot has quit IRC | 21:56 | |
| sf-slack | <acomodi> litghost: Ok, makes sense as it does not get used in VPR and we can save additional space, but still can be useful for debug purposes | 22:00 |
| litghost | Bingo | 22:00 |
| litghost | It is actually possible to generate the xml from the capnp using VPR directly, but it is convient to include if it compresses nicely, which I believe it will | 22:00 |
| sf-slack | <acomodi> Sure | 22:02 |
| litghost | Other option would be to provide a shell script to invoke VPR to generate the xml from the capnp | 22:04 |
| sf-slack | <acomodi> I'll check what is the best way to go. If regenerating the xml from capnp does not take too much time maybe that would be better, as for bigger parts, maybe even the compressed xml could occupy too much space | 22:07 |
| litghost | The conversion command is roughly "vpr -read <capnp> -write <xml> -pack wire.eblif" | 22:09 |
| litghost | All you need VPR to do is get far enough to attempt to read the RR graph | 22:09 |
| litghost | Which is will do for the post-pack timing analysis | 22:09 |
| duck2 | capnp could also compress well, it's created in the "unpacked" format and contains a lot of \0's. also metadata is text | 22:19 |
| duck2 | sorry, forgot that it's going to end up in a tarball | 22:20 |
| *** ayazar has quit IRC | 22:34 | |
| *** space_zealot has joined #symbiflow | 22:50 | |
| *** adjtm has quit IRC | 23:13 | |
| *** futarisIRCcloud has joined #symbiflow | 23:18 | |
| *** adjtm has joined #symbiflow | 23:27 | |
| *** space_zealot has quit IRC | 23:32 | |
| *** space_zealot has joined #symbiflow | 23:39 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!