*** tpb has joined #symbiflow | 00:00 | |
*** futarisIRCcloud has joined #symbiflow | 00:57 | |
*** Vonter has quit IRC | 03:16 | |
*** Vonter has joined #symbiflow | 03:38 | |
*** space_zealot has quit IRC | 03:44 | |
*** citypw has joined #symbiflow | 03:54 | |
*** _whitelogger has quit IRC | 05:38 | |
*** _whitelogger has joined #symbiflow | 05:41 | |
*** OmniMancer has joined #symbiflow | 07:26 | |
*** Bertl_zZ is now known as Bertl | 07:33 | |
sf-slack2 | <acomodi> mithro: I have commented PR https://github.com/SymbiFlow/symbiflow-arch-defs/pull/775. `<tile>` tags appear in XMLs (only xc7-related), but they are "invisible" to VtR | 08:57 |
---|---|---|
tpb | Title: Sign in to GitHub · GitHub (at github.com) | 08:57 |
sf-slack2 | <acomodi> mithro: more details are in the comment on GH | 08:57 |
*** citypw has quit IRC | 09:28 | |
*** space_zealot has joined #symbiflow | 11:21 | |
*** futarisIRCcloud has quit IRC | 12:19 | |
*** OmniMancer has quit IRC | 12:56 | |
*** OmniMancer has joined #symbiflow | 12:58 | |
*** _whitelogger has quit IRC | 13:14 | |
*** _whitelogger has joined #symbiflow | 13:17 | |
mithro | acomodi: Okay great | 14:20 |
mithro | mkurc: Where are we with the Yosys attribute + parameter stuff? | 14:20 |
sf-slack2 | <mkurc> @mithro I fixed handling of default values of floating point (aka real) parameters | 14:24 |
sf-slack2 | <mkurc> So if there are no more review issues then it is ready on my side. | 14:25 |
mithro | mkurc: Okay I'll take a look shortly | 14:26 |
sf-slack2 | <mkurc> @mithro @litghost Have you managed to successfully model an async set/reset flip-flop (eg. FDSE) with timings ? I've been trying to make such a test for V2X SDF import with no success. | 14:47 |
sf-slack2 | <mkurc> I tried to use the exact example as given in the VPR doc (https://docs.verilogtorouting.org/en/latest/tutorials/arch/timing_modeling), this time no async reset and the VPR segfaults with it. | 14:47 |
mithro | mkurc: Example? | 14:48 |
mithro | FYI it is a public holiday here in the US so litghost probably won't be around today... | 14:48 |
sf-slack2 | <mkurc> Ahh, ok. | 14:48 |
sf-slack2 | <mkurc> I can prepare a short doc with explanations / examples | 14:48 |
mithro | mkurc: That shouldn't be needed, just an example pasted here would be enough | 14:51 |
sf-slack2 | <mkurc> @mithro The example is there https://docs.verilogtorouting.org/en/latest/tutorials/arch/timing_modeling/ under the "Sequential block (no internal paths)" section. | 14:52 |
sf-slack2 | <mkurc> @mithro The first problem is that when such a model+pb_type is injected into the v2x test suite then the VPR crashes (segfaults) with the assertion failure: `vpr/src/timing/timing_graph_builder.cpp:250 add_block_to_timing_graph: Assertion 'clk_port' failed.` | 14:53 |
sf-slack2 | <mkurc> @mitrho But this is just a D flip-flop | 14:54 |
mithro | mkurc: I saw that issue recently | 14:54 |
mithro | mkurc: Let me find the fix | 14:55 |
sf-slack2 | <mkurc> @mitrho: Ok | 14:55 |
sf-slack2 | <mkurc> @mitrho And another thing: when having a FF with async set/reset like FDSE then I see no way to define timing between SR and Q. | 14:57 |
mithro | mkurc: It is caused by the eblif not having the clock pin | 14:57 |
*** lavatech_ has joined #symbiflow | 14:58 | |
mithro | mkurc: https://github.com/SymbiFlow/symbiflow-arch-defs/pull/703/commits/5b0105bdc55f5d5acb9235252cc248c597b4f6b7 that is the commit you need I think | 15:00 |
tpb | Title: mux_gen + v2x: Support generating FASM annotations for muxes. by mithro · Pull Request #703 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com) | 15:00 |
mithro | mkurc: I'll pull it out into it's own pull request | 15:00 |
litghost | Fdse is syncronous reset | 15:00 |
litghost | Fdce/fdpe is async reset | 15:01 |
*** space_zealot has quit IRC | 15:01 | |
litghost | In sync case, it is clock to q timing, and setup/hold on sr | 15:01 |
litghost | In async it recovery+removal on sr | 15:03 |
litghost | https://github.com/SymbiFlow/prjxray-db/blob/master/artix7/timings/CLBLL_L.sdf#L255 | 15:03 |
tpb | Title: prjxray-db/CLBLL_L.sdf at master · SymbiFlow/prjxray-db · GitHub (at github.com) | 15:03 |
mithro | mkurc: https://github.com/SymbiFlow/symbiflow-arch-defs/pull/783 | 15:27 |
tpb | Title: Small improvements to eblif generation by mithro · Pull Request #783 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com) | 15:27 |
sf-slack2 | <mkurc> @mitrho Thanks. And yes, I meant FDCE for async reset. | 15:27 |
*** OmniMancer has quit IRC | 16:21 | |
mithro | mkurc: I added a bunch of tests into that pull request -> https://github.com/SymbiFlow/symbiflow-arch-defs/pull/783 | 17:58 |
tpb | Title: Small improvements to EBLIF generation by mithro · Pull Request #783 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com) | 17:58 |
mithro | kgugala: What is the status of https://github.com/SymbiFlow/symbiflow-arch-defs/pull/756? | 18:45 |
tpb | Title: [DNM][WIP] Use Yosys to generate blif in v2x tests by kgugala · Pull Request #756 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com) | 18:45 |
mithro | litghost / acomodi: kmurry just landed the reformatting patch to vtr | 18:53 |
sf-slack2 | <acomodi> mithro: that's great! We need to merge it to the symbiflow fork ASAP. There already is a PR open with a merge from upstream, but currently it is failing during routing. I am looking at the issue. More information in the PR itself: https://github.com/SymbiFlow/vtr-verilog-to-routing/pull/59 | 19:20 |
tpb | Title: Merge upstream by acomodi · Pull Request #59 · SymbiFlow/vtr-verilog-to-routing · GitHub (at github.com) | 19:20 |
*** ZipCPU has quit IRC | 19:21 | |
*** ZipCPU has joined #symbiflow | 19:26 | |
*** mats has quit IRC | 19:27 | |
*** mats has joined #symbiflow | 19:53 | |
mithro | acomodi: It looks like xc7 is failing on master at the moment? | 19:57 |
mithro | acomodi: Nope, I'm work sorry... | 19:58 |
*** lavatech_ has quit IRC | 20:03 | |
*** Bertl is now known as Bertl_zZ | 23:00 |
Generated by irclog2html.py 2.13.1 by Marius Gedminas - find it at mg.pov.lt!