The current `prevector` size of 28 bytes (chosen to fill the `sizeof(CScript)` aligned size) was introduced in 2015 (https://github.com/bitcoin/bitcoin/pull/6914) before SegWit and TapRoot.
However, the increasingly common `P2WSH` and `P2TR` scripts are both 34 bytes, and are forced to use heap (re)allocation rather than efficient inline storage.
The core trade-off of this change is to eliminate heap allocations for common 34-36 byte scripts at the cost of increasing the base memory footprint of all `CScript` objects by 8 bytes (while still respecting peak memory usage defined by `-dbcache`).
Increasing the `prevector` size allows these scripts to be stored on the stack, avoiding heap allocations, reducing potential memory fragmentation, and improving performance during cache flushes. Massif analysis confirms a lower stable memory usage after flushing, suggesting the elimination of heap allocations outweighs the larger base size for common workloads.
Due to memory alignment, increasing the `prevector` size to 36 bytes doesn't change the overall `sizeof(CScript)` compared to an increase to 34 bytes, allowing us to include `P2PK` scripts as well at no additional memory cost.
Performance benchmarks for AssumeUTXO load and flush show:
- Small dbcache (450MB): ~1% performance penalty due to more frequent flushes
- Large dbcache (4500-4500MB+): ~6-7% performance improvement due to fewer heap allocations
Full IBD and reindex-chainstate with larger `dbcache` values also show an overall ~3% speedup.
Co-authored-by: Ava Chow <github@achow101.com>
Co-authored-by: Andrew Toth <andrewstoth@gmail.com>
In my benchmarks, using this pool allocator for CCoinsMap gives about
20% faster `-reindex-chainstate` with -dbcache=5000 with practically the
same memory usage. The change in max RSS changed was 0.3%.
The `validation_flush_tests` tests need to be updated because
memory allocation is now done in large pools instead of one node at a
time, so the limits need to be updated accordingly.
This frees up all associated memory with the map, not only the nodes.
This is necessary in preparation for using the PoolAllocator for
CCoinsMap, which does not actually free any memory on clear().
Use {Chain,}TestingSetup's existing {Chainstate,Block}Manager and avoid
unnecessarily creating a local one.
This also helps reduce the code diff for a later commit where we change
{Chainstate,Block}Manager's constructor signature.
After this commit, there should be no explicit instantiation of
CTxMemPool in src/test other than those in fuzz/ and setup_common
-BEGIN VERIFY SCRIPT-
find_regex="CTxMemPool\s+([^;({]+)(|\(\)|\{\});" \
&& git grep -l -E "$find_regex" -- src/test \
| grep -v -e "^src/test/util/setup_common.cpp$" \
-e "^src/test/fuzz/" \
| xargs sed -i -E "s@$find_regex@CTxMemPool\& \1 = *Assert(m_node.mempool);@g"
-END VERIFY SCRIPT-
Add an upwards reference to chainstate instances to the owning
ChainstateManager. This is necessary because there are a number
of `this_chainstate == chainman.ActiveChainstate()` checks that
will happen (as a result of assumeutxo) in functions that otherwise
don't have an easily-accessible reference to the chainstate's
ChainManager.
Since we now have multiple chainstate objects, only one of them is active at any given
time. An active chainstate has a mempool, but there's no point to others having one.
This change will simplify proposed assumeutxo semantics. See the discussion here:
https://github.com/bitcoin/bitcoin/pull/15606#pullrequestreview-692965905
Co-authored-by: Russell Yanofsky <russ@yanofsky.org>
Also adds CCoinsViewCache::ReallocateCache() to attempt to free
memory that the cacheCoins's allocator may be hanging onto when
downsizing the cache.
Adds `CChainState::m_coins{tip,db}_cache_size_bytes` data members
so that we can reference cache size on a per-chainstate basis for
flushing.
