Open rswarbrick opened 6 months ago
@rswarbrick: This has been closed by mistake instead of https://github.com/lowRISC/opentitan/issues/21716 (which I now closed), right? Reopening
Oh, nuts. You're completely right, sorry!
Looking at the task, I think there are two possible approaches.
I'm reasonably convinced that (2) would be a silly thing to do: we'd be waiving a warning by applying an on-the-fly edit to the RTL. This comes with some risk and I can't see a benefit over a lint waiver. So I'm assuming we're going with option (1). Will assign Andreas (as he has much more knowledge of the pulp project than me) and make time estimates accordingly.
Moving to M5 because Verilator is not critical for tapeout signoff
I realise that this is just bumping things down the road(!) but I don't believe this is critical for the M5 milestone. Marking as worthy of discussion in the triage meeting.
Description
The pulp code that we're vendoring for
rv_dm
contains some code that triggers Verilator warnings about address truncation. We're currently waiving the warning inpulp_riscv_dbg.vlt
.When reviewing the waiver in #22486, @andreaskurth pointed out that a better approach would be to fix the underlying RTL. This issue is to remind us to do so.