lbcd/blockchain/process.go
Dave Collins 491acd4ca6 blockchain: Rework to use new db interface.
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.
2016-04-11 16:47:27 -05:00

240 lines
8.3 KiB
Go

// Copyright (c) 2013-2016 The btcsuite developers
// Use of this source code is governed by an ISC
// license that can be found in the LICENSE file.
package blockchain
import (
"fmt"
database "github.com/btcsuite/btcd/database2"
"github.com/btcsuite/btcd/wire"
"github.com/btcsuite/btcutil"
)
// BehaviorFlags is a bitmask defining tweaks to the normal behavior when
// performing chain processing and consensus rules checks.
type BehaviorFlags uint32
const (
// BFFastAdd may be set to indicate that several checks can be avoided
// for the block since it is already known to fit into the chain due to
// already proving it correct links into the chain up to a known
// checkpoint. This is primarily used for headers-first mode.
BFFastAdd BehaviorFlags = 1 << iota
// BFNoPoWCheck may be set to indicate the proof of work check which
// ensures a block hashes to a value less than the required target will
// not be performed.
BFNoPoWCheck
// BFDryRun may be set to indicate the block should not modify the chain
// or memory chain index. This is useful to test that a block is valid
// without modifying the current state.
BFDryRun
// BFNone is a convenience value to specifically indicate no flags.
BFNone BehaviorFlags = 0
)
// blockExists determines whether a block with the given hash exists either in
// the main chain or any side chains.
//
// This function MUST be called with the chain state lock held (for reads).
func (b *BlockChain) blockExists(hash *wire.ShaHash) (bool, error) {
// Check memory chain first (could be main chain or side chain blocks).
if _, ok := b.index[*hash]; ok {
return true, nil
}
// Check in the database.
var exists bool
err := b.db.View(func(dbTx database.Tx) error {
var err error
exists, err = dbTx.HasBlock(hash)
return err
})
return exists, err
}
// processOrphans determines if there are any orphans which depend on the passed
// block hash (they are no longer orphans if true) and potentially accepts them.
// It repeats the process for the newly accepted blocks (to detect further
// orphans which may no longer be orphans) until there are no more.
//
// The flags do not modify the behavior of this function directly, however they
// are needed to pass along to maybeAcceptBlock.
//
// This function MUST be called with the chain state lock held (for writes).
func (b *BlockChain) processOrphans(hash *wire.ShaHash, flags BehaviorFlags) error {
// Start with processing at least the passed hash. Leave a little room
// for additional orphan blocks that need to be processed without
// needing to grow the array in the common case.
processHashes := make([]*wire.ShaHash, 0, 10)
processHashes = append(processHashes, hash)
for len(processHashes) > 0 {
// Pop the first hash to process from the slice.
processHash := processHashes[0]
processHashes[0] = nil // Prevent GC leak.
processHashes = processHashes[1:]
// Look up all orphans that are parented by the block we just
// accepted. This will typically only be one, but it could
// be multiple if multiple blocks are mined and broadcast
// around the same time. The one with the most proof of work
// will eventually win out. An indexing for loop is
// intentionally used over a range here as range does not
// reevaluate the slice on each iteration nor does it adjust the
// index for the modified slice.
for i := 0; i < len(b.prevOrphans[*processHash]); i++ {
orphan := b.prevOrphans[*processHash][i]
if orphan == nil {
log.Warnf("Found a nil entry at index %d in the "+
"orphan dependency list for block %v", i,
processHash)
continue
}
// Remove the orphan from the orphan pool.
orphanHash := orphan.block.Sha()
b.removeOrphanBlock(orphan)
i--
// Potentially accept the block into the block chain.
err := b.maybeAcceptBlock(orphan.block, flags)
if err != nil {
return err
}
// Add this block to the list of blocks to process so
// any orphan blocks that depend on this block are
// handled too.
processHashes = append(processHashes, orphanHash)
}
}
return nil
}
// ProcessBlock is the main workhorse for handling insertion of new blocks into
// the block chain. It includes functionality such as rejecting duplicate
// blocks, ensuring blocks follow all rules, orphan handling, and insertion into
// the block chain along with best chain selection and reorganization.
//
// It returns a bool which indicates whether or not the block is an orphan and
// any errors that occurred during processing. The returned bool is only valid
// when the error is nil.
//
// This function is safe for concurrent access.
func (b *BlockChain) ProcessBlock(block *btcutil.Block, timeSource MedianTimeSource, flags BehaviorFlags) (bool, error) {
b.chainLock.Lock()
defer b.chainLock.Unlock()
fastAdd := flags&BFFastAdd == BFFastAdd
dryRun := flags&BFDryRun == BFDryRun
blockHash := block.Sha()
log.Tracef("Processing block %v", blockHash)
// The block must not already exist in the main chain or side chains.
exists, err := b.blockExists(blockHash)
if err != nil {
return false, err
}
if exists {
str := fmt.Sprintf("already have block %v", blockHash)
return false, ruleError(ErrDuplicateBlock, str)
}
// The block must not already exist as an orphan.
if _, exists := b.orphans[*blockHash]; exists {
str := fmt.Sprintf("already have block (orphan) %v", blockHash)
return false, ruleError(ErrDuplicateBlock, str)
}
// Perform preliminary sanity checks on the block and its transactions.
err = checkBlockSanity(block, b.chainParams.PowLimit, timeSource, flags)
if err != nil {
return false, err
}
// Find the previous checkpoint and perform some additional checks based
// on the checkpoint. This provides a few nice properties such as
// preventing old side chain blocks before the last checkpoint,
// rejecting easy to mine, but otherwise bogus, blocks that could be
// used to eat memory, and ensuring expected (versus claimed) proof of
// work requirements since the previous checkpoint are met.
blockHeader := &block.MsgBlock().Header
checkpointBlock, err := b.findPreviousCheckpoint()
if err != nil {
return false, err
}
if checkpointBlock != nil {
// Ensure the block timestamp is after the checkpoint timestamp.
checkpointHeader := &checkpointBlock.MsgBlock().Header
checkpointTime := checkpointHeader.Timestamp
if blockHeader.Timestamp.Before(checkpointTime) {
str := fmt.Sprintf("block %v has timestamp %v before "+
"last checkpoint timestamp %v", blockHash,
blockHeader.Timestamp, checkpointTime)
return false, ruleError(ErrCheckpointTimeTooOld, str)
}
if !fastAdd {
// Even though the checks prior to now have already ensured the
// proof of work exceeds the claimed amount, the claimed amount
// is a field in the block header which could be forged. This
// check ensures the proof of work is at least the minimum
// expected based on elapsed time since the last checkpoint and
// maximum adjustment allowed by the retarget rules.
duration := blockHeader.Timestamp.Sub(checkpointTime)
requiredTarget := CompactToBig(b.calcEasiestDifficulty(
checkpointHeader.Bits, duration))
currentTarget := CompactToBig(blockHeader.Bits)
if currentTarget.Cmp(requiredTarget) > 0 {
str := fmt.Sprintf("block target difficulty of %064x "+
"is too low when compared to the previous "+
"checkpoint", currentTarget)
return false, ruleError(ErrDifficultyTooLow, str)
}
}
}
// Handle orphan blocks.
prevHash := &blockHeader.PrevBlock
if !prevHash.IsEqual(zeroHash) {
prevHashExists, err := b.blockExists(prevHash)
if err != nil {
return false, err
}
if !prevHashExists {
if !dryRun {
log.Infof("Adding orphan block %v with parent %v",
blockHash, prevHash)
b.addOrphanBlock(block)
}
return true, nil
}
}
// The block has passed all context independent checks and appears sane
// enough to potentially accept it into the block chain.
err = b.maybeAcceptBlock(block, flags)
if err != nil {
return false, err
}
// Don't process any orphans or log when the dry run flag is set.
if !dryRun {
// Accept any orphan blocks that depend on this block (they are
// no longer orphans) and repeat for those accepted blocks until
// there are no more.
err := b.processOrphans(blockHash, flags)
if err != nil {
return false, err
}
log.Debugf("Accepted block %v", blockHash)
}
return false, nil
}