You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think it would be a great idea to join forces in this regard! Your list of points on what such an IR HDL would need pretty much matches up with what I have observed so far.
My ultimate goal is to capture SystemVerilog, Verilog, and VHDL in its entirety -- with an initial focus on synthesis/simulation. I also want to maintain compatibility with existing commercial tools, to avoid a "ideological" biforcation of the community into either an open-source and commercial lock-in. I would hope that we could eventually move a lot of the language frontend part into the open source domain, to allow for language innovation and have the tool vendors focus on their core business.
As a first step I think it would be great to tie moore/llhd into the SymbiFlow compliance suite and start exploring ways to support the flow with LLHD, e.g. by using moore as a frontend to feed designs.
Some time ago I opened a discussion about creating truly common middle-level HDL/representation, see the f4pga/ideas#19
It would be nice to hear your opinion, maybe even it's possible to join the forces.
The text was updated successfully, but these errors were encountered: