Closed algofoogle closed 6 months ago
With latest OpenROAD head, the crash not reproducible. So you can update OpenROAD commit and run again. @algofoogle
@algofoogle OpenLane head (which doesn't necessarily follows OpenROAD head) doesn't reproduce the crash. Can you update OpenLane and try again ?
I will give it a try! I don’t have a failing case right now but I will try the update after the next time I hit the error :)
@algofoogle Great. Please open a new issue if something surfaces.
Description
Hi there! This may be exactly the same as #2035 and already fixed (but awaiting deployment?). In any case I'm including this bug report in case it helps provide another test case.
I'm using caravel_user_project tag
gfmpw-1c
to harden a user project macro, and I have frequently had it crash on step 20 (Global Routing Resizer Timing Optimizations). This seems to be unpredictable, but can be worked around for any given iteration of my design if I fiddle marginally with things likeFP_CORE_UTIL
.Reported error:
Expected Behavior
Flow should complete normally.
Environment report
Reproduction material
issue_reproducible.tar.gz
Relevant log output