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
While dealing with #108 I have bumped into one issue which has many ways of being solved IMO.
The problem is that the XDC constraints that add attributes do not reflect outside of nextpnr, as the XDC parser is embedded within. This means that the logical netlist won't have information, for instance, on the IOSTANDARD attribute of a pad, unless this is specified in the design.
The attributes are quite important when, for instance, writing the FASM features.
Some of the possible solutions:
enhance the physical netlist schema to include information on additional attributes assigned to the cells during p&r (e.g. when reading the XDCs)
read the XDC file once again alongside with logical and physical netlists
add the attributes when generating the logical netlist and read the XDC at that step
The text was updated successfully, but these errors were encountered:
Another option would be to write out a new logical netlist from nextpnr.
In the long-long term, this would be needed if a place and route tool wanted to perform optimisations that manipulate the netlist, such as retiming, resynthesis, or control set remapping.
While dealing with #108 I have bumped into one issue which has many ways of being solved IMO.
The problem is that the XDC constraints that add attributes do not reflect outside of nextpnr, as the XDC parser is embedded within. This means that the logical netlist won't have information, for instance, on the IOSTANDARD attribute of a pad, unless this is specified in the design.
The attributes are quite important when, for instance, writing the FASM features.
Some of the possible solutions:
The text was updated successfully, but these errors were encountered: