Aren’t we missing out on a lot of reductions? #150

1. 
   murchandamus commented at 1:44 pm on October 6, 2023: contributor

   Hey,

   I noticed that I get a bunch of new seeds when I fuzz, but by merging my results into the existing corpus, it only takes the ones that add new features and coverage. All the seeds that reduce the size but achieve the same coverage are not adopted into the corpus. Obviously, if I merged both the existing and my new seeds into an empty third directory, they’d get traversed by increasing size and I would get keep the shortest seeds that achieve the same coverage.

   Should I:

   1. merge my new seeds into the existing corpus and only upstream new features/coverage?
   2. merge everything into an empty directory, add everything that ends up in that direct to the corpus and upstream it?

   Currently, I’m following approach 1. which keeps the corpus smaller, but misses out on the reductions. Following 2. would grow the corpus more quickly, but then when we squash it at the branch off point we’d keep the best reduced seeds.

2. 
   sipa commented at 1:45 pm on October 6, 2023: contributor

   See also #130 (comment) and the discussion afterwards.
3. murchandamus closed this on Oct 6, 2023

   ---

4. 
   maflcko commented at 4:40 pm on October 13, 2023: contributor

   To implement this idea in a hacky way, one could do: (untested, don’t try this at home)

   0mv fuzz_seed_corpus fuzz_inputs_upstream
   1../bitcoin-core/test/fuzz/test_runner.py -l DEBUG --par $( nproc ) --m_dir ./fuzz_inputs_upstream/  --m_dir ../results_from_working_farm/fuzz_seed_corpus/ -x wallet_notifications ./fuzz_seed_corpus
   2git restore -- ./fuzz_seed_corpus
   3git add ./fuzz_seed_corpus
   4git commit -m "Add my fuzz inputs (append-only)"
   

5. 
   murchandamus commented at 7:18 pm on October 13, 2023: contributor

   Yeah, something like that would work. The question is whether we want to do that?
6. murchandamus reopened this on Oct 13, 2023

   ---

7. 
   maflcko commented at 8:05 am on October 14, 2023: contributor

   I don’t think there is an easy answer. I think anyone can decide for themselves for now?
8. 
   murchandamus commented at 2:10 pm on October 16, 2023: contributor

   Okay, I would like to do that at least right before we resquash the qa-assets repository.
9. 
   murchandamus commented at 2:05 pm on October 17, 2023: contributor

   I’ve created two variants as potential contributions to compressing the qa-assets repo with #153 and #154
10. 
    maflcko commented at 10:14 am on October 19, 2023: contributor

    Edited my comment to replace rm -rf with mv: #150 (comment)
11. 
    maflcko commented at 10:15 am on October 19, 2023: contributor

    Anything left to be done here?
12. 
    murchandamus commented at 1:09 pm on October 19, 2023: contributor

    Edited my comment to replace rm -rf with mv: #150 (comment)

    Yeah, I had just merged into a /tmp/qa_assets_merge and copied the result back. I think we’re good.

13. murchandamus closed this on Oct 19, 2023

    ---

Contributors
murchandamus sipa maflcko