@User: It would not be my intention to make the pad input unreadable when the pad is configured for output. It is too late for an easy fix (the GPIO control block has been hardened and added to layout which was hand-routed) but I'm sure it can be done. I'll take a look at it. The example user project had only very simple I/O operations on the user side, so the issue didn't come up in the testbenches.
@Matt Venn: Looking at the verilog for the GPIO pad control logic, it should always have a valid signal on the input line to the user project area whenever the management SoC cedes control of the GPIO to the user project. I looked at your code but didn't see anything obviously wrong, so I would have to clone your repo and debug.
Tim Edwards
11/29/2020, 2:25 AM
Feel free to share a screenshot of gtkwave if you still can't figure out what is going on.
Linen is a search-engine friendly community platform. We offer integrations with existing Slack/Discord communities and make those conversations Google-searchable.