Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Simulation of module with undriven signal causes generated VCD to not contain the signal, but contain an individual signal for each character of the undriven signal's name #345

Closed
the6p4c opened this issue Apr 1, 2020 · 2 comments

Comments

@the6p4c
Copy link

the6p4c commented Apr 1, 2020

This snippet illustrates the issue:

from nmigen import *
from nmigen.back import pysim

class Bug(Elaboratable):
    def __init__(self):
        self.reg = Signal()
        self.out = Signal()

    def elaborate(self, platform):
        m = Module()
        m.d.sync += self.out.eq(~self.out)

        # driving the signal fixes the issue
        # uncomment: vcd will be correct
        # m.d.comb += self.reg.eq(0)

        return m

bug = Bug()
fragment = Fragment.get(bug, None)
sim = pysim.Simulator(fragment)
sim.add_clock(1e-6)
with sim.write_vcd(vcd_file='bug.vcd', traces=[bug.out, bug.reg]):
    sim.run_until(1e-6 * 10, run_passive=True)

With the line driving reg inside elaborate commented out, the generated VCD starts:

$comment Generated by nMigen $end
$date 2020-04-01 22:07:22.464464 $end
$timescale 100 ps $end
$var wire 1 3 r $end
$var wire 1 4 e $end
$var wire 1 5 g $end
$scope module top $end
$var wire 1 0 clk $end
$var wire 1 1 rst $end
$var wire 1 2 out $end
$upscope $end
$enddefinitions $end

i.e. reg has somehow become three signals, r, e and g.

Driving reg combinatorically or with registered logic fixes the signal's name and value in the VCD.

@the6p4c
Copy link
Author

the6p4c commented Apr 1, 2020

Experimenting, it appears this happens if the signal is unused in any way - either undriven, or not part of the module's logic.

@whitequark
Copy link
Member

whitequark commented Apr 1, 2020

Ouch, I think I made a mistake fixing #280.

@whitequark whitequark added this to the 0.3 milestone Apr 3, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Development

No branches or pull requests

2 participants