Profiling indicated that significant time was being spent validating
coinbase outputs, ensuring that they paid to the development
organization's P2SH tax address. This check was more inefficient than
necessary for a couple of reasons:
* The tax address was always decoded from a string to a dcrutil.Address.
* The actual script being validated was always parsed for addresses to
check if they matched the tax address.
Neither of these are needed. To optimize the algorithm, only the
equality of the output script and script version are checked.
A bug when updating the chain state during forced reorganizations
was fixed. An assertion was added into the mining code so that it
now errors out if the blockchain and chainState of the block
manager are inconsistent.
The blockchain function to fetch the latest block header was
renamed for consistency and now returns a copy of the block header
instead of the pointer itself.
The legacy ticket database, which was GOB serialized and stored on
shut down, has been removed. Ticket state information is now held in
a stake node, which acts as a modularized "black box" to contain all
information about the state of the stake system. Stake nodes are now
a component of the blockchain blockNode struct, and are updated with
them.
Stake nodes, like their internal treap primitives, are immutable
objects that are created with their connect and disconnect node
functions. The blockchain database now stores all information about
the stake state of the best node in the block database. The blockchain
makes the assumption that the stake state of the best node is known at
any given time. If the states of former blocks or sidechains must be
evaluated, this can be achieved by iterating backwards along the
blockchain from the best node, and then connecting stake nodes
iteratively if necessary.
Performance improvements with this new module are dramatic. The long
delays on start up and shut down are removed. Blockchain
synchronization time is improved approximately 5-10x on the mainnet
chain. The state of the database is atomic, so unexpected shut downs
should no longer have the ability to disrupt the chain state.
An upgrade path has been added for version 1 blockchain databases.
Users with this blockchain database will automatically update when
they start their clients.
This commit is the first stage of several that are planned to convert
the blockchain package into a concurrent safe package that will
ultimately allow support for multi-peer download and concurrent chain
processing. The goal is to update btcd proper after each step so it can
take advantage of the enhancements as they are developed.
In addition to the aforementioned benefit, this staged approach has been
chosen since it is absolutely critical to maintain consensus.
Separating the changes into several stages makes it easier for reviewers
to logically follow what is happening and therefore helps prevent
consensus bugs. Naturally there are significant automated tests to help
prevent consensus issues as well.
The main focus of this stage is to convert the blockchain package to use
the new database interface and implement the chain-related functionality
which it no longer handles. It also aims to improve efficiency in
various areas by making use of the new database and chain capabilities.
The following is an overview of the chain changes:
- Update to use the new database interface
- Add chain-related functionality that the old database used to handle
- Main chain structure and state
- Transaction spend tracking
- Implement a new pruned unspent transaction output (utxo) set
- Provides efficient direct access to the unspent transaction outputs
- Uses a domain specific compression algorithm that understands the
standard transaction scripts in order to significantly compress them
- Removes reliance on the transaction index and paves the way toward
eventually enabling block pruning
- Modify the New function to accept a Config struct instead of
inidividual parameters
- Replace the old TxStore type with a new UtxoViewpoint type that makes
use of the new pruned utxo set
- Convert code to treat the new UtxoViewpoint as a rolling view that is
used between connects and disconnects to improve efficiency
- Make best chain state always set when the chain instance is created
- Remove now unnecessary logic for dealing with unset best state
- Make all exported functions concurrent safe
- Currently using a single chain state lock as it provides a straight
forward and easy to review path forward however this can be improved
with more fine grained locking
- Optimize various cases where full blocks were being loaded when only
the header is needed to help reduce the I/O load
- Add the ability for callers to get a snapshot of the current best
chain stats in a concurrent safe fashion
- Does not block callers while new blocks are being processed
- Make error messages that reference transaction outputs consistently
use <transaction hash>:<output index>
- Introduce a new AssertError type an convert internal consistency
checks to use it
- Update tests and examples to reflect the changes
- Add a full suite of tests to ensure correct functionality of the new
code
The following is an overview of the btcd changes:
- Update to use the new database and chain interfaces
- Temporarily remove all code related to the transaction index
- Temporarily remove all code related to the address index
- Convert all code that uses transaction stores to use the new utxo
view
- Rework several calls that required the block manager for safe
concurrency to use the chain package directly now that it is
concurrent safe
- Change all calls to obtain the best hash to use the new best state
snapshot capability from the chain package
- Remove workaround for limits on fetching height ranges since the new
database interface no longer imposes them
- Correct the gettxout RPC handler to return the best chain hash as
opposed the hash the txout was found in
- Optimize various RPC handlers:
- Change several of the RPC handlers to use the new chain snapshot
capability to avoid needlessly loading data
- Update several handlers to use new functionality to avoid accessing
the block manager so they are able to return the data without
blocking when the server is busy processing blocks
- Update non-verbose getblock to avoid deserialization and
serialization overhead
- Update getblockheader to request the block height directly from
chain and only load the header
- Update getdifficulty to use the new cached data from chain
- Update getmininginfo to use the new cached data from chain
- Update non-verbose getrawtransaction to avoid deserialization and
serialization overhead
- Update gettxout to use the new utxo store versus loading
full transactions using the transaction index
The following is an overview of the utility changes:
- Update addblock to use the new database and chain interfaces
- Update findcheckpoint to use the new database and chain interfaces
- Remove the dropafter utility which is no longer supported
NOTE: The transaction index and address index will be reimplemented in
another commit.
Contains the following upstream commits:
- ef9c50be57
- eb882f39f8
In addition to merging the fixes in the commits, this also fixes a few
more misspellings that were introduced in the new Decred code.
Contains the following upstream commits:
- 83bcfea271
- 2f6aeacfab
- Cherry-picked because it fixes an issue originally introduced by the
commit being merged
In addition to the normal required changes for syncing, the following
changes have been made in order to facilitate integration into Decred:
- Configure the NewestSha function to the new mempool config as a
closure over the block manager state instead of using the database
func directly
- Add a new NextStakeDifficulty callback to the new mempool config for
obtaining the next stake difficulty from the block manager and
configure it as a closure over the block manager state.
Upstream commit 2b6a9a56e5.
NOTE: This is only merging in the upstream changes, so while they are
useful for Decred as well, this merge commit does fully separate the
mempool as upstream does due to the new functionality provided by
Decred.
The mempool would allow low stake difficulty tickets and then remove
them after the next block was added. This prevents too low difficulty
tickets from being added in the first place. It also refactors some of
the old code in mempool.go, blockmanager.go, and mining.go for
efficiency.
A continue statement was missing which could cause an issue when
attempting to iterate through different eligible blocks on the chain
tip to mine off of. The mining code was also amended to make it return
newly generated work later on in the code if for some reason not
enough voters were found, instead of just returning a nil template.
The mining transaction selection algorithm failed to sort the stake
transactions effectively, as well as failed to sorted by fee for
these transactions. The codebase now defaults to a smaller priority
size spacing in the block, correctly sorts transactions by their stake
importance, and sub-sorts tickets by fees in all cases.
This simply exports and adds some comments to the fields of the
BlockTemplate struct.
This is primarily being done as a step toward being able to separate the
mining code into its own package, but also it makes sense on its own
because code that requests new block template necessarily examines the
returned fields which implies they should be exported.
Many of the checks for the type of stake transaction in the main sorting
loop for mempool transactions of NewBlockTemplate were redundant. Now the
cached data for the transaction type is used instead. An expensive lookup
for a transaction cache was also moved for efficiency.
The checkBlockForHiddenVotes function would corrupt block templates
by incorrectly inserting votes from other blocks. The function has
been rewritten and now correctly regenerates the block template with
the correct number of voters.
This creates a skeleton mining package that simply contains a few of the
definitions used by the mining and mempool code.
This is a step towards decoupling the mining code from the internals of
btcd and ultimately will house all of the code related to creating block
templates and CPU mining.
The main reason a skeleton package is being created before the full
blown package is ready is to avoid blocking mempool separation which
relies on these type definitions.
This introduces the concept of a new interface named TxSource which aims
to generically provide a concurrent safe source of transactions to be
considered for inclusion in a new block. This is a step towards
decoupling the mining code from the internals of btcd. Ultimately the
intent is to create a separate mining package.
The new TxSource interface relies on a new struct named miningTxDesc,
which describes each entry in the transaction source. Once this code is
refactored into a separate mining package, the mining prefix can simply
be dropped leaving the type exported as mining.TxDesc.
To go along with this, the existing TxDesc type in the mempool has been
renamed to mempoolTxDesc and changed to embed the new miningTxDesc type.
This allows the mempool to efficiently implement the MiningTxDescs
method needed to satisfy the TxSource interface.
This approach effectively separates the direct reliance of the mining
code on the mempool and its data structures. Even though the memory
pool will still be the default concrete implementation of the interface,
making it an interface offers much more flexibility in terms of testing
and even provides the potential to allow more than one source (perhaps
multiple independent relay networks, for example).
Finally, the memory pool and all of the mining code has been updated to
implement and use the new interface.
This does three things:
- Splits the priority calculation logic from the TxDesc type
- Modifies the calcPriority function to perform the value age
calculation instead of accepting it as a parameter
- Changes the starting priority to be calculated when the transaction is
added to the pool
The first is useful as it is a step towards decoupling the mining code
from the internals of the memory pool. Also, the concept of priority is
related to mining policy, so it makes more sense to have the
calculations separate than being defined on the memory pool tx
descriptor.
The second change has been made because everywhere that uses the
calcPriority function first has to calculate the value age anyways and
by making it part of the function it can be avoided altogether in
certain circumstances thereby provided a bit of optimization.
The third change ensure the starting priority is safe for reentrancy
which will be important once the mempool is split into a separate
package.
This introduces the concept of a mining policy struct which is used to
control block template generation instead of directly accessing the
config struct. This is a step toward decoupling the mining code from
the internals of btcd. Ultimately the intent is to create a separate
mining package.
Now that the memory pool minimum fee calculation code is also
calculating a more precise value instead of rounding up to the nearest
kilobyte boundary, the comment in NewBlockTemplate regarding this
behavior is no longer accurate. Thus, this removes the comment.
Also, while here, change the calculation to use an int64 instead of
float since it matches the precision of the new calculation code used by
the memory pool and can also avoid the need for the slower floating
point math.
Introduce an ECDSA signature verification into btcd in order to
mitigate a certain DoS attack and as a performance optimization.
The benefits of SigCache are two fold. Firstly, usage of SigCache
mitigates a DoS attack wherein an attacker causes a victim's client to
hang due to worst-case behavior triggered while processing attacker
crafted invalid transactions. A detailed description of the mitigated
DoS attack can be found here: https://bitslog.wordpress.com/2013/01/23/fixed-bitcoin-vulnerability-explanation-why-the-signature-cache-is-a-dos-protection/
Secondly, usage of the SigCache introduces a signature verification
optimization which speeds up the validation of transactions within a
block, if they've already been seen and verified within the mempool.
The server itself manages the sigCache instance. The blockManager and
txMempool respectively now receive pointers to the created sigCache
instance. All read (sig triplet existence) operations on the sigCache
will not block unless a separate goroutine is adding an entry (writing)
to the sigCache. GetBlockTemplate generation now also utilizes the
sigCache in order to avoid unnecessarily double checking signatures
when generating a template after previously accepting a txn to the
mempool. Consequently, the CPU miner now also employs the same
optimization.
The maximum number of entries for the sigCache has been introduced as a
config parameter in order to allow users to configure the amount of
memory consumed by this new additional caching.
This commit converts all block height references to int32 instead of
int64. The current target block production rate is 10 mins per block
which means it will take roughly 40,800 years to reach the maximum
height an int32 affords. Even if the target rate were lowered to one
block per minute, it would still take roughly another 4,080 years to
reach the maximum.
In the mean time, there is no reason to use a larger type which results
in higher memory and disk space usage. However, for now, in order to
avoid having to reserialize a bunch of database information, the heights
are still serialized to the database as 8-byte uint64s.
This is being mainly being done in preparation for further upcoming
infrastructure changes which will use the smaller and more efficient
4-byte serialization in the database as well.
This change moves IsFinalizedTransaction to txscript and also changes
the first argument to take a wire.MsgTx instead of btcutil.Tx. This
is needed for an upcoming diff in which txscript will require
IsFinalizedTransaction and we do not want to import the btcd/blockchain.
This commit implements a new type, named scriptNum, for handling all
numeric values used in scripts and converts the code over to make use of
it. This is being done for a few of reasons.
First, the consensus rules for handling numeric values in the scripts
require special handling with subtle semantics. By encapsulating those
details into a type specifically dedicated to that purpose, it
simplifies the code and generally helps prevent improper usage.
Second, the new type is quite a bit more efficient than big.Ints which
are designed to be arbitrarily large and thus involve a lot of heap
allocations and additional multi-precision bookkeeping. Because this
new type is based on an int64, it allows the numbers to be stack
allocated thereby eliminating a lot of GC and also eliminates the extra
multi-precision arithmetic bookkeeping.
The use of an int64 is possible because the consensus rules dictate that
when data is interpreted as a number, it is limited to an int32 even
though results outside of this range are allowed so long as they are not
interpreted as integers again themselves. Thus, the maximum possible
result comes from multiplying a max int32 by itself which safely fits
into an int64 and can then still appropriately provide the serialization
of the larger number as required by consensus.
Finally, it more closely resembles the implementation used by Bitcoin
Core and thus makes is easier to compare the behavior between the two
implementations.
This commit also includes a full suite of tests with 100% coverage of
the semantics of the new type.
Because FetchTransactionStore in GetBlockTemplate occasionally accesses
the internal blockchain memory structure while it is being read or modified,
a race can occur. To prevent this, FetchTransactionStore is instead
routed through the internal channel for blockchain requests.
This commit modifies finalized transaction check used by the memory pool
and block templates to use the network adjusted time instead of the
unadjusted local time. This helps keep the transactions accepted to the
memory pool, and hence allowed to relay, more consistent across nodes.
By exporting StandardVerifyFlags, clients can ensure they create
transactions that btcd will accept into its mempool.
This flag doesn't belong in txscript. It belongs in a
policy package. However, this is currently the least worse place.
Remove ScriptCanonicalSignatures and use the new
ScriptVerifyDERSignatures flag. The ScriptVerifyDERSignatures
flag accomplishes the same functionality.
