I've gotten the gnetlist scheme backend from Mike Jarabek, gnet-verilog.scm, changed some so statements come out in one line for readability, and have read up on verilog-ams. It seems to put the usual values we have on a schematic, value=1K, for instance into a simulation, you use a feature of verilog-ams that is not generated now by gnet-verilog.scm. You have a verilog statement that instantiates a verilog module, creating a level of hierarchy, where the placed module can be more wiring and placements inside, or a behavioral model (called signal flow system in verilog-ams), or a structural model, (called conservative system in verilog-ams). If the placed module is a model, you can override its parameter values, and that's where we want to put in the value=1000 data from our schematics.
Here's a placed resistor statement, (inside of the top module of a verilog netlist -- see the whole thing below) res R1 ( .n(B), .p(A)); It has no value associated. What I am thinking is to take value from schematic placements as meaning "the main parameter", like r = resistance, c = capacitance, etc. and apply that with what is called "parameter value assignment by order" in verilog-ams. Then the above statement instantiating a resistor model would look like: res #(1000) R1 ( .n(B), .p(A)); This gets the data, 1000, from an attrib on R1 of the schematic, value=1000 without knowing it replaces parameter r = 0, the default r value for the resistor model, with r = 1000. There's no need to know the first parameter name, and we can generate netlists from plain ordinary schematics or sections of schematics originally aimed at pcb production. John Griessen ========================================================== module verilog_io ( GND , C , A ); /* Port directions begin here */ inout GND ; inout C ; inout A ; /* Wires from the design */ electrical B ; electrical GND ; electrical C ; electrical A ; /* continuous assignments */ /* Package instantiations */ cap C1 ( .p(B), .n(GND)); ind L1 ( .n(C), .p(B)); res R1 ( .n(B), .p(A)); endmodule -- Ecosensory Austin TX _______________________________________________ geda-user mailing list [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
