Øyvind Harboe
Øyvind Harboe
> @oharboe this seems to have stalled out on the size of your LEF macros. Is there something here to pursue as this doesn't seem to be a pdn issue....
I tried to to whittle it down, but I did not get a faster failure. ``` NUM_CORES=16 openroad -python ~/OpenROAD-flow-scripts/tools/OpenROAD/etc/deltaDebug.py --persistence 4 --use_stdout --error_string "MPL-2067" --base_db_path bazel-out/k8-fastbuild/bin/build/results/asap7/BoomTile/base/2_3_floorplan_tdms.odb --step "./run-me-BoomTile-asap7-base.sh" ```
> Does the fix for #4242 work here? This does not crash anymore, so for that part yes, but mpl2 does not find a solution, whereas classic placement does find...
@maliberty Please confirm that you can repeat the problem. deltaDebug did not produce a faster failure, so the reproduction case is the above.
> Is that the whole maker? Are all shapes/nets visible? On second try, the snapshot looks like it has everything. I replaced the image and uploaded the `make detail_route_issue` file.
@maliberty Please confirm that this issue has what is needed to reproduce. Thanks!
@osamahammad21 You should have access to https://drive.google.com/file/d/10ECM456C94DEikW51WINpRCTr4MvYN14/view?usp=sharing now
Same crash with: ``` $ firtool --version LLVM (http://llvm.org/): LLVM version 18.0.0git Optimized build with assertions. CIRCT firtool-1.61.0 ```
Without `--extract-test-code`, it doesn't crash.
> What command are you running? I believe we should have isolated the SFC flow so the standard verilog generation step doesn't invoke SFC at all, but there could be...