Add fuzz inputs #252

1. 
   maflcko commented at 2:24 pm on January 5, 2026: contributor

   C.f. last one: #234

   0mv fuzz_corpora fuzz_inputs_moved
   1../b-c/bld-cmake/test/fuzz/test_runner.py -l DEBUG --par $( nproc ) --m_dir ./fuzz_inputs_moved/  --m_dir ./results_from_working_farm --m_dir ../etc/... ./fuzz_corpora
   2git restore -- ./fuzz_corpora
   3git add ./fuzz_corpora
   4git commit -m "Add fuzz inputs"
   

2. Add fuzz inputs 6be1610ae3
3. 
   maflcko commented at 2:33 pm on January 5, 2026: contributor

   main: https://drahtbot.space/host_reports/DrahtBot/reports/coverage_fuzz/monotree/bd4f4782f23b59f8/c87ae4dd4daf320f/fuzz.coverage/index.html this pull: https://drahtbot.space/host_reports/DrahtBot/reports/coverage_fuzz/monotree/bd4f4782f23b59f8/6be1610ae3cd6507/fuzz.coverage/index.html
4. 
   maflcko commented at 3:42 pm on January 5, 2026: contributor

   Only 25 new lines of code covered. My fuzzing server was down for a good chunk of time, so maybe this is expected.

   Also, Murch submitted recently in December: #246

   Not sure what to do. Seems fine to merge, or to close, and they’ll be submitted again later.

5. Add failing psbt input

   From https://github.com/bitcoin-core/qa-assets/pull/246#issuecomment-3606863084

   00c335ca2a
6. 
   maflcko commented at 10:06 am on January 6, 2026: contributor

   This includes a few regression tests for psbt stuff, so maybe it can be merged just to include those?
7. 
   murchandamus commented at 10:51 pm on January 6, 2026: contributor

   Only 25 new lines of code covered. My fuzzing server was down for a good chunk of time, so maybe this is expected.

   Also, Murch submitted recently in December: #246

   Not sure what to do. Seems fine to merge, or to close, and they’ll be submitted again later.

   It seems like we might be getting close to full coverage on many of the established fuzz targets. What would be the downsides of merging? Are you worried about the additional time to run the full harness while there is little coverage improvement? If that were the worry, should we perhaps prune more often instead of not merging?

8. 
   maflcko commented at 4:29 pm on January 7, 2026: contributor

   It seems like we might be getting close to full coverage on many of the established fuzz targets.

   The Bitcoin Core code changes, so over time the coverage will degrade and only catch up with fresh inputs. But maybe you are right and the Bitcoin core changes didn’t invalidate prior coverage, so there was not a lot new to discover.

   What would be the downsides of merging?

   No downside, just eating a bit of space and CI time.

   I think I’ll go ahead and merge this now.

9. maflcko merged this on Jan 7, 2026
10. maflcko closed this on Jan 7, 2026

    ---

Contributors
maflcko murchandamus