fuzz: Speed up *_package_eval fuzz targets a bit #31457

1. 
   maflcko commented at 12:42 pm on December 10, 2024: member

   Each target is at least 10% faster for me when running over the current set of qa-assets, which seems nice.
2. fuzz: Speed up *_package_eval fuzz targets a bit fa33818978
3. fuzz: [refactor] Avoid confusing c-style cast fafaad5edc
4. 
   DrahtBot commented at 12:43 pm on December 10, 2024: contributor

   The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.

   Code Coverage & Benchmarks

   For details see: https://corecheck.dev/bitcoin/bitcoin/pulls/31457.

   Reviews

   See the guideline for information on the review process. A summary of reviews will appear here.

5. DrahtBot added the label Tests on Dec 10, 2024
6. in src/test/fuzz/package_eval.cpp:198 in fafaad5edc

   194@@ -195,8 +195,8 @@ FUZZ_TARGET(ephemeral_package_eval, .init = initialize_tx_pool)
   195     MockTime(fuzzed_data_provider, chainstate);
   196 
   197     // All RBF-spendable outpoints outside of the unsubmitted package
   198-    std::set<COutPoint> mempool_outpoints;
   199-    std::map<COutPoint, CAmount> outpoints_value;
   200+    std::unordered_set<COutPoint, SaltedOutpointHasher> mempool_outpoints;
   

   ---

   ---

   

   dergoegge commented at 3:24 pm on December 10, 2024:

   Using a SaltedOutpointHasher probably requires to call SeedRandomStateForTest(SeedRand::ZEROS) each iteration. Otherwise non-determinism might happen due to e.g. different order when iterating over the sets.

   ---

   

   maflcko commented at 3:47 pm on December 10, 2024:

   Good catch. This means:

   • Other tests may also need a re-seed on every iteration, if they interact with data, sorted by a salted hasher
   • The speed-up may actually come from invalidating the fuzz inputs, making a change like this harder to test

   ---

   

   maflcko commented at 10:57 am on December 12, 2024:

   Other tests may also need a re-seed on every iteration, if they interact with data, sorted by a salted hasher

   Fixed in #31481, possibly

7. maflcko marked this as a draft on Dec 10, 2024

Contributors
maflcko DrahtBot dergoegge

Labels
Tests