litghost
litghost
I thought we had discussed managing this as a plugin? Am I misremembering?
> @litghost , @kgugala : Still awaiting response, if this is merged then we can proceed with merging k4n8 device support in Symbiflow-arch-defs. I already gave my response. All of...
This change is part of the equivalent tiles change, which has a PR open here: https://github.com/SymbiFlow/vtr-verilog-to-routing/pull/296
> Can the lookahead be generated once and then saved / loaded? That's already being done. However, consider that GH actions only has 4 threads, so if it takes ~6...
> How big is the file, could it just be shipped with the interchange files in some way? So there is additional complexity in your suggestion. The lookahead is generated...
> To sum up, the tests performed show a **reduce in run-time** equal to **27.36%** and an **increase in CPD** equal to **3.10%** w.r.t. the baseline. I think this is...
> To sum up, the tests performed show a **reduce in run-time** equal to **27.36%** and an **increase in CPD** equal to **3.10%** w.r.t. the baseline. To be clear, the...
Okay, so the sampling method and Dijkstra expansion (for place delay) should be both go upstream, yes? The penalty cost stuff is something we need to handle separately.
For my own clarity, all the changes you have that are not upstream only affect the generation of the lookahead files, not in their use?
> @litghost I think that, with the new information I got and summarized above, there are still issues to be solved with the lookahead if we want to stop using...