This switches the Merkle tree logic for blocks to one that runs in constant (small) space.
The old code is moved to tests, and a new test is added that for various combinations of
block sizes, transaction positions to compute a branch for, and mutations:
* Verifies that the old code and new code agree for the Merkle root.
* Verifies that the old code and new code agree for the Merkle branch.
* Verifies that the computed Merkle branch is valid.
* Verifies that mutations don't change the Merkle root.
* Verifies that mutations are correctly detected.
(cherry picked from commit bitcoin/bitcoin@eece63fa72)
Assume that when a wallet transaction has a valid block hash and transaction position
in it, the transaction is actually there. We're already trusting wallet data in a
much more fundamental way anyway.
To prevent backward compatibility issues, a new record is used for storing the
block locator in the wallet. Old wallets will see a wallet file synchronized up
to the genesis block, and rescan automatically.
(cherry picked from commit bitcoin/bitcoin@391dff16fe)
CBlockHeader.hashFinalSaplingRoot has been renamed to hashLightClient.
CBlockIndex now stores:
- hashLightClient as from the block header
- hashFinalSaplingRoot, which is accurate for all blocks prior to
Heartwood activation, and all blocks from Heartwood activation onward
that are connected at some point to the main chain in ConnectBlock().
- hashChainHistoryRoot, which is null prior to Heartwood activation, and
set per ZIP 221 from Heartwood activation.
The new block index fields are only written to disk for client version
2.1.2 and above, which will be the first Heartwood-aware clients (even
if Heartwood doesn't have an activation height).
Previously we had both nVersion as a class parameter *and* a serialization
argument, and in several inherited serializers the latter was set to the former,
in order to pass the serialized object's version into underlying parsers. #3180
pulled in the upstream changes to clean this up, and in doing so these lines
became no-ops - setting the class parameter to itself. Clang throws warnings on
this, which turn into errors on the MacOS builder.
We can just remove these, because upstream already had done so in earlier PRs,
indicating that they were not being relied on by underlying parsers.
Remove the nType and nVersion as parameters to all serialization methods
and functions. There is only one place where it's read and has an impact
(in CAddress), and even there it does not impact any of the recursively
invoked serializers.
Instead, the few places that need nType or nVersion are changed to read
it directly from the stream object, through GetType() and GetVersion()
methods which are added to all stream classes.
For steady-state operation, this reduces the average time between wallet disk
writes from once per block to once per hour.
On -rescan, witness caches are only written out at the end along with the best
block, increasing speed while ensuring that on-disk state is kept consistent.
Witness caches are now never recreated during a -reindex, on the assumption that
the blocks themselves are not changing (the chain is just being reconstructed),
and so the witnesses will remain valid.
Part of #1749.
The genesis blocks and miner tests have been regenerated, because changing the
block header serialisation format changes the block hash, and thus validity.
The Equihash solutions have been removed from the bloom test inputs for
simplicity (block validity is not checked there; only a valid serialisation is
necessary).
This field has no defined semantics. While it was added as a result of
discussions about merged mining in #724, this field will not necessarily ever be
used for that purpose.
The main and test networks are configured to use parameters that are currently
low-memory but usable with the basic solver; they will be increased once the
solver is optimised. The regtest network is configured to have extremely low
memory usage for speed.
Note that Bitcoin's double-hasher is used for the difficulty check. This does
not match the paper, but is simpler than changing the block header
serialization. Single hashing is kept for the EquiHash solver because there is
no requirement on execution time there, only on memory usage.
Kris Nuttycombe
Pieter Wuille
Jack Grigg
Kris Nuttycombe
Daira Hopwood
Sean Bowe
Per Grön
Peter Todd
jtimon
Wladimir J. van der Laan
Pavel Janík
sandakersmann
Michael Ford
Matt Corallo
Luke Dashjr