| *** tpb has joined #symbiflow | 00:00 | |
| litghost | clay_1: You are using fasm2bels in a weird way. fasm2bels is designed to convert the input FASM file into a top level verilog file that matches the original so that testbenches can run on the post-place&route output | 00:23 |
|---|---|---|
| litghost | clay_1: Some of the arguments to fasm2bels were specifically for the original bitstream, before you replaced it | 00:23 |
| litghost | clay_1: Solution is to invoke fasm2bels without arguments specific to the original design | 00:24 |
| *** citypw has joined #symbiflow | 01:35 | |
| sf-slack | <aryap> hello | 01:57 |
| sf-slack | <aryap> is the symbiflow fork of vtr significantly different to upstream, currently? | 01:57 |
| litghost | aryap: Yes | 01:58 |
| sf-slack | <aryap> looks like there was a sync only a few days ago | 01:58 |
| sf-slack | <aryap> ok | 01:58 |
| litghost | aryap: We periodically bring in upstream changes, but there are significant downstream changes required for the flow | 01:58 |
| sf-slack | <aryap> is this mostly plumbing or significant work on QoR? | 01:59 |
| sf-slack | <aryap> i guess i can just run both | 01:59 |
| *** drom has joined #symbiflow | 02:00 | |
| sf-slack | <aryap> thanks :pray: | 02:01 |
| *** drom is now known as wavedrom | 02:04 | |
| hackerfoo | aryap: Upstream VtR won't work with symbiflow-arch-defs. We are working to upstream our changes, but it takes time. | 02:12 |
| sf-slack | <aryap> aha, great | 02:16 |
| *** wavedrom has quit IRC | 02:29 | |
| mithro | aryap: litghost has been working hard to get a bunch of stuff upstream lately | 02:44 |
| sf-slack | <aryap> nice work :clap: | 02:45 |
| sf-slack | <aryap> should i be able to run any of the titan benchmarks on the xc7a50t arch from symbiflow-arch-defs? | 02:59 |
| sf-slack | <aryap> i get this a lot: ```Error 1: [blah]/symbiflow-vtr/libs/libvtrutil/src/vtr_util.cpp:190 Failed to convert string '{iopath_A1_O5}' to float``` | 02:59 |
| mithro | aryap: That means you have broken the symbiflow architecture generation -- it hasn't patched in the timing information | 03:05 |
| mithro | aryap: Nobody has gotten the titan benchmarks (**not** to be confused with a benchmark containing opentitan) running on symbiflow architectures -- but that would be an excellent project | 03:06 |
| *** wavedrom has joined #symbiflow | 03:55 | |
| *** az0re has quit IRC | 06:21 | |
| *** OmniMancer has joined #symbiflow | 06:26 | |
| *** OmniMancer1 has joined #symbiflow | 06:28 | |
| *** OmniMancer has quit IRC | 06:31 | |
| *** az0re has joined #symbiflow | 06:37 | |
| *** HEGAZY has joined #symbiflow | 07:36 | |
| *** wavedrom has quit IRC | 07:38 | |
| *** HEGAZY has quit IRC | 07:40 | |
| *** _whitelogger has quit IRC | 08:06 | |
| *** _whitelogger has joined #symbiflow | 08:08 | |
| *** Vonter has quit IRC | 08:43 | |
| *** Vonter has joined #symbiflow | 08:45 | |
| *** az0re has quit IRC | 08:49 | |
| *** tux3 has quit IRC | 09:03 | |
| *** clay_1 has joined #symbiflow | 09:11 | |
| *** upcho has joined #symbiflow | 11:50 | |
| *** upcho has quit IRC | 11:58 | |
| *** clay_1 has quit IRC | 13:37 | |
| *** clay_1 has joined #symbiflow | 13:48 | |
| *** Bertl_oO is now known as Bertl | 15:16 | |
| *** HEGAZY has joined #symbiflow | 15:30 | |
| *** citypw has quit IRC | 15:51 | |
| *** tux3 has joined #symbiflow | 15:54 | |
| *** HEGAZY has quit IRC | 15:55 | |
| *** Hegazy has joined #symbiflow | 15:55 | |
| *** OmniMancer1 has quit IRC | 15:55 | |
| *** Hegazy has quit IRC | 15:56 | |
| *** HEGAZY has joined #symbiflow | 15:57 | |
| *** phire has quit IRC | 16:34 | |
| *** phire has joined #symbiflow | 16:35 | |
| *** wavedrom has joined #symbiflow | 17:04 | |
| HEGAZY | Hi all, again, remember me? | 17:10 |
| HEGAZY | so we talked yesterday about the Gsoc project for #symbiflow , I've thought it through and I am intersted in "system verilog feature detector" you told me about, | 17:10 |
| HEGAZY | I have an initial plan on how the work is going to go, but first i have couple of questions I need to discuss with you :D | 17:10 |
| HEGAZY | mithro said "For example, we would love to work on a "system verilog feature detector" which is able to give you a list of the system verilog functionality a project is using and thus which tools will are more likely to be compatible with the project" | 17:11 |
| HEGAZY | - I see that the third party tools takes .sv files directly, so are the tools already support some kinda of verilog structure format that i should return as a result? or is this for future integration? | 17:11 |
| HEGAZY | - why not use an existing parser for sv, I mean all the third party tools like yosys parse the code 1st and their parser opensource that can be ported here right? | 17:11 |
| ZirconiumX | HEGAZY: essentially "because nobody's done it yet". | 17:12 |
| ZirconiumX | VHDL support is more or less third-party | 17:13 |
| HEGAZY | done what exactly? there is this project for example https://github.com/ben-marshall/verilog-parser | 17:13 |
| tpb | Title: GitHub - ben-marshall/verilog-parser: A Flex/Bison Parser for the IEEE 1364-2001 Verilog Standard. (at github.com) | 17:13 |
| ZirconiumX | That's Verilog not SystemVerilog | 17:13 |
| ZirconiumX | Yosys has a native Verilog parser | 17:14 |
| ZirconiumX | With some SV extensions | 17:14 |
| HEGAZY | great, didn't know that nobody's done it yet :D | 17:16 |
| ZirconiumX | There's talk of trying to integrate the Google SV parser | 17:19 |
| ZirconiumX | But it would essentially need a build system transplant | 17:19 |
| ZirconiumX | Like, I'm pretty sure ghdlsynth couldn't be mainlined because it drags in an Ada codebase | 17:24 |
| *** _whitelogger has quit IRC | 17:54 | |
| mithro | HEGAZY: If you already had good system verilog support you wouldn't need the system verilog feature tester | 17:54 |
| mithro | HEGAZY: did you see the sv-test project? | 17:54 |
| *** lambda has quit IRC | 17:55 | |
| *** _whitelogger has joined #symbiflow | 17:56 | |
| *** lambda has joined #symbiflow | 17:56 | |
| HEGAZY | mithro: yup saw it, i was wondering about the existance of such a project not how effective it is, i got you now | 17:58 |
| mithro | HEGAZY: SystemVerilog and Verilog are about as similar as C++ and C | 17:59 |
| mithro | HEGAZY: If you look at https://symbiflow.github.io/sv-tests/ you'll see lots of red | 17:59 |
| tpb | Title: SystemVerilog Report (at symbiflow.github.io) | 17:59 |
| HEGAZY | dalance sv-parser is doing well right? | 18:05 |
| mithro | HEGAZY: yeah - but that is just a parser and isn't integrated into tools like Yosys or Verilator | 18:10 |
| mithro | litghost: Does this look okay? https://github.com/SymbiFlow/prjxray-db/compare/master...mithro:master -- It's missing zynq as that hasn't finished yet | 18:10 |
| tpb | Title: Comparing SymbiFlow:master...mithro:master · SymbiFlow/prjxray-db · GitHub (at github.com) | 18:10 |
| *** HEGAZY has quit IRC | 19:45 | |
| *** HEGAZY has joined #symbiflow | 20:22 | |
| *** HEGAZY has quit IRC | 20:28 | |
| *** Bertl is now known as Bertl_oO | 20:57 | |
| *** az0re has joined #symbiflow | 21:04 | |
| *** Vicky35 has joined #symbiflow | 21:06 | |
| *** Vicky35 is now known as vickykiller | 21:06 | |
| *** vickykiller has quit IRC | 21:10 | |
| sf-slack | <aryap> uh oh. i ran the instructions to generate the arch from HEAD. unmodified | 21:34 |
| sf-slack | <aryap> ok interesting | 21:34 |
| sf-slack | <aryap> oh i see. i have to use the arch.timing.xml file | 21:37 |
| *** az0re has quit IRC | 21:52 | |
| *** clay_1 has quit IRC | 22:31 | |
| *** pdp7 has quit IRC | 22:31 | |
| *** diamondman has quit IRC | 22:31 | |
| *** futarisIRCcloud has quit IRC | 22:31 | |
| *** digshadow has quit IRC | 22:31 | |
| *** daveshah has quit IRC | 22:31 | |
| *** daveshah has joined #symbiflow | 22:35 | |
| *** digshadow has joined #symbiflow | 22:35 | |
| *** diamondman has joined #symbiflow | 22:36 | |
| *** benreynwar has quit IRC | 22:36 | |
| *** flammit has quit IRC | 22:36 | |
| *** litghost has quit IRC | 22:36 | |
| *** brent has quit IRC | 22:36 | |
| *** futarisIRCcloud has joined #symbiflow | 22:39 | |
| *** pdp7 has joined #symbiflow | 22:39 | |
| *** benreynwar has joined #symbiflow | 22:41 | |
| *** flammit has joined #symbiflow | 22:41 | |
| *** litghost has joined #symbiflow | 22:44 | |
| *** brent has joined #symbiflow | 22:45 | |
| *** _florent_ has quit IRC | 22:46 | |
| *** sorear has quit IRC | 22:46 | |
| *** _florent_ has joined #symbiflow | 22:47 | |
| *** sorear has joined #symbiflow | 22:48 | |
| *** nickray has quit IRC | 22:50 | |
| *** emilazy has quit IRC | 22:50 | |
| *** mithro has quit IRC | 22:50 | |
| *** bubble_buster has quit IRC | 22:50 | |
| *** bubble_buster has joined #symbiflow | 22:51 | |
| *** mithro has joined #symbiflow | 22:52 | |
| *** emilazy has joined #symbiflow | 22:55 | |
| *** nickray has joined #symbiflow | 22:55 | |
| *** perillamint has quit IRC | 22:56 | |
| *** elms has quit IRC | 22:56 | |
| *** swetland has quit IRC | 22:56 | |
| *** ovf has quit IRC | 22:56 | |
| *** ovf has joined #symbiflow | 22:59 | |
| *** elms has joined #symbiflow | 23:01 | |
| *** perillamint has joined #symbiflow | 23:02 | |
| *** swetland has joined #symbiflow | 23:02 | |
| litghost | mithro: Diff looks good | 23:07 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!