06:16 <hzeller> Mostly for own educational purposes and to play around with the parts easier, I put together a simple Makefile to build an example for the BASYS3 board out of the symbiflow tree https://github.com/hzeller/symbiflow-simple-sample

06:16 <tpb> Title: GitHub - hzeller/symbiflow-simple-sample: Using Symbiflow arch defs to get BASYS3 board entertained with logic (at github.com)

13:52 <hackerfoo> hzeller: Nice!

13:57 <sf-slack2> <acomodi> update on equivalent tiles: I have just murax with equivalent tiles and it works fine on HW

13:59 <sf-slack2> <acomodi> I have also updated the PR to get the tiles tags in the `arch.xml` file. I think it is ready to be reviewed: https://github.com/SymbiFlow/symbiflow-arch-defs/pull/653

13:59 <tpb> Title: WIP: Modified prjxray import to add tiles tags by acomodi · Pull Request #653 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com)

16:02 <mithro> Morning

16:02 <sf-slack2> <mgielda> [time_of_day]

16:08 <sf-slack2> <pgielda> Good morning.

16:34 <mithro> kgugala: So I think we should proceed with the explicit flipflops for now and defer the implicit flipflops?

16:35 <mithro> kgugala: What is the status of https://github.com/SymbiFlow/symbiflow-arch-defs/pull/639/files ? I think you just need to fix the model.xml part?

16:35 <tpb> Title: v2x dsp_in_registered test fix by kgugala · Pull Request #639 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com)

16:36 <mithro> kgugala: Can you reply to https://github.com/SymbiFlow/symbiflow-arch-defs/pull/639/files#r278361302 ?

16:36 <tpb> Title: v2x dsp_in_registered test fix by kgugala · Pull Request #639 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com)

16:41 <sf-slack2> <kgugala> @mithro sure we can go with explicit dffs for now, as for #639 fixing model.xml is what we need to have the current code to work fine (dsp_in_registered with explicit dffs). To have the rest of the tests there working we'd need to switch them to explicit dffs or fix implicit inference in v2x

16:55 <sf-slack2> <kgugala> @mithro I have replied to your question in #639

16:58 <mithro> kgugala: So - model.xml files should *only* be top level models and includes to sub model.xml files?

16:59 <mithro> kgugala: Do you understand what I mean?

17:10 <sf-slack2> <kgugala> @mithro isn't it like that now? currently v2x generates model.xml which is a top level, and may include some sub models

17:22 <sf-slack2> <kgugala> but you wrote `FYI - A model never includes another model. Models are low level primitives.` - so I thought you did not want there to be any includes. But the above suggests you do want the model to be able to include other models. Which is it?

20:31 <litghost> kgugala: A model can include other models

20:31 <litghost> kgugala: A primative v2x should not include other models (e.g. leaf pbtype)

20:32 <litghost> kgugala: But intermediate v2x's which include other intermediate pbtypes and primative pbtypes, those should include the models of the children of the node

20:42 <litghost> mithro: If I read kgugala's question, he is talking about the XML model including another model, e.g. intermediate XML include another intermediate or primative XML?

20:43 <mithro> Just doing a quick example

20:43 <litghost> https://github.com/SymbiFlow/symbiflow-arch-defs/blob/master/xc7/primitives/slicel/ntemplate.slicelN.model.xml

20:43 <tpb> Title: symbiflow-arch-defs/ntemplate.slicelN.model.xml at master · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com)

20:43 <litghost> Like I said, we do this all the time by hand

20:44 <mithro> litghost / kgugala: https://gist.github.com/mithro/94ff11701379637990e5b3be16b7f46c

20:44 <tpb> Title: model.xml.md · GitHub (at gist.github.com)

20:45 <litghost> mithro: Of course

20:45 <litghost> mithro: The model tag cannot really include another mode, without some weird XPath trickry

20:45 <litghost> mithro: It is probably possible, but seems unnessary and confusing

20:46 <mithro> litghost: We are talking about https://github.com/SymbiFlow/symbiflow-arch-defs/pull/639/files/a8e33ac14c63d5d0268ca9ee39d81819b4dbc33f#r278249196

20:46 <tpb> Title: v2x dsp_in_registered test fix by kgugala · Pull Request #639 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com)

20:47 <mithro> litghost: All properties have to be propagated out to the top level ports...

20:50 <litghost> mithro: That particular thread is confusing, because we already have an unambiguous way to declare a blackbox (e.g. leaf)

20:50 <litghost> "(* blackbox *)"

20:51 <litghost> VPR doesn't support packing intermediate pbtype nodes, so the entire discussion is moot

20:51 <litghost> mithro: If someone wants to instance the equivilant of an intermediate node, Yosys techmapping needs to burn it down to the primative (leaf) pbtypes

20:52 <litghost> mithro: I think the "all at once" mode is never correct

20:53 <mithro> litghost: So, only pb_types which have a blif=`.subckt <model_name>` should end up in the model.xml

20:53 <litghost> mithro: Yes

20:54 <mithro> litghost: I would like to discuss the "all at once" mode but lets not block on that for now

20:54 <litghost> mithro: And intermediate nodes should always include their included children wholesale

20:54 <mithro> litghost: Yes

20:54 <litghost> mithro: So leaf's define models, and intermediate's define includes to other XMLs

20:54 <mithro> litghost: Yeap

21:27 <mithro> kgugala: ^

21:27 <mithro> kerel: Did you understand that?

21:31 <mithro> opps

21:31 <mithro> kgugala: Did you understand that?

21:49 <mithro> kgugala / litghost: https://github.com/SymbiFlow/prjxray-db/commit/e29750043d678016526d238da864748dd8c822dc

21:49 <tpb> Title: Updating artix7 based on "Merge pull request #810 from mithro/import-… · SymbiFlow/prjxray-db@e297500 · GitHub (at github.com)

23:42 <hackerfoo> ^ That's what I have so far for a schematic of the RAM blocks in an M slice.