lnd.xprv/input/size.go

590 lines
20 KiB
Go
Raw Normal View History

package input
import (
2020-03-10 15:23:17 +03:00
"math/big"
"github.com/btcsuite/btcd/blockchain"
2020-03-10 15:23:17 +03:00
"github.com/btcsuite/btcd/btcec"
"github.com/btcsuite/btcd/wire"
2020-03-10 15:23:17 +03:00
"github.com/lightningnetwork/lnd/keychain"
)
const (
// witnessScaleFactor determines the level of "discount" witness data
// receives compared to "base" data. A scale factor of 4, denotes that
// witness data is 1/4 as cheap as regular non-witness data. Value copied
// here for convenience.
witnessScaleFactor = blockchain.WitnessScaleFactor
// The weight(weight), which is different from the !size! (see BIP-141),
// is calculated as:
// Weight = 4 * BaseSize + WitnessSize (weight).
// BaseSize - size of the transaction without witness data (bytes).
// WitnessSize - witness size (bytes).
// Weight - the metric for determining the weight of the transaction.
// P2WPKHSize 22 bytes
// - OP_0: 1 byte
// - OP_DATA: 1 byte (PublicKeyHASH160 length)
// - PublicKeyHASH160: 20 bytes
P2WPKHSize = 1 + 1 + 20
// NestedP2WPKHSize 23 bytes
// - OP_DATA: 1 byte (P2WPKHSize)
// - P2WPKHWitnessProgram: 22 bytes
NestedP2WPKHSize = 1 + P2WPKHSize
2017-02-23 22:56:47 +03:00
// P2WSHSize 34 bytes
// - OP_0: 1 byte
// - OP_DATA: 1 byte (WitnessScriptSHA256 length)
// - WitnessScriptSHA256: 32 bytes
P2WSHSize = 1 + 1 + 32
// NestedP2WSHSize 35 bytes
// - OP_DATA: 1 byte (P2WSHSize)
// - P2WSHWitnessProgram: 34 bytes
NestedP2WSHSize = 1 + P2WSHSize
// P2PKHOutputSize 34 bytes
// - value: 8 bytes
// - var_int: 1 byte (pkscript_length)
// - pkscript (p2pkh): 25 bytes
P2PKHOutputSize = 8 + 1 + 25
// P2WKHOutputSize 31 bytes
// - value: 8 bytes
// - var_int: 1 byte (pkscript_length)
// - pkscript (p2wpkh): 22 bytes
P2WKHOutputSize = 8 + 1 + P2WPKHSize
// P2WSHOutputSize 43 bytes
// - value: 8 bytes
// - var_int: 1 byte (pkscript_length)
// - pkscript (p2wsh): 34 bytes
P2WSHOutputSize = 8 + 1 + P2WSHSize
// P2SHOutputSize 32 bytes
// - value: 8 bytes
// - var_int: 1 byte (pkscript_length)
// - pkscript (p2sh): 23 bytes
P2SHOutputSize = 8 + 1 + 23
// P2PKHScriptSigSize 108 bytes
// - OP_DATA: 1 byte (signature length)
// - signature
// - OP_DATA: 1 byte (pubkey length)
// - pubkey
P2PKHScriptSigSize = 1 + 73 + 1 + 33
// P2WKHWitnessSize 109 bytes
// - number_of_witness_elements: 1 byte
// - signature_length: 1 byte
// - signature
// - pubkey_length: 1 byte
// - pubkey
P2WKHWitnessSize = 1 + 1 + 73 + 1 + 33
2017-02-23 22:56:47 +03:00
// MultiSigSize 71 bytes
// - OP_2: 1 byte
// - OP_DATA: 1 byte (pubKeyAlice length)
// - pubKeyAlice: 33 bytes
// - OP_DATA: 1 byte (pubKeyBob length)
// - pubKeyBob: 33 bytes
// - OP_2: 1 byte
// - OP_CHECKMULTISIG: 1 byte
MultiSigSize = 1 + 1 + 33 + 1 + 33 + 1 + 1
2017-02-23 22:56:47 +03:00
// WitnessSize 222 bytes
// - NumberOfWitnessElements: 1 byte
// - NilLength: 1 byte
// - sigAliceLength: 1 byte
// - sigAlice: 73 bytes
// - sigBobLength: 1 byte
// - sigBob: 73 bytes
// - WitnessScriptLength: 1 byte
// - WitnessScript (MultiSig)
WitnessSize = 1 + 1 + 1 + 73 + 1 + 73 + 1 + MultiSigSize
// InputSize 41 bytes
// - PreviousOutPoint:
// - Hash: 32 bytes
// - Index: 4 bytes
// - OP_DATA: 1 byte (ScriptSigLength)
// - ScriptSig: 0 bytes
// - Witness <---- we use "Witness" instead of "ScriptSig" for
// transaction validation, but "Witness" is stored
// separately and weight for it size is smaller. So
// we separate the calculation of ordinary data
// from witness data.
// - Sequence: 4 bytes
InputSize = 32 + 4 + 1 + 4
// FundingInputSize represents the size of an input to a funding
// transaction, and is equivalent to the size of a standard segwit input
// as calculated above.
FundingInputSize = InputSize
2017-02-23 22:56:47 +03:00
// CommitmentDelayOutput 43 bytes
// - Value: 8 bytes
// - VarInt: 1 byte (PkScript length)
// - PkScript (P2WSH)
CommitmentDelayOutput = 8 + 1 + P2WSHSize
2017-02-23 22:56:47 +03:00
// CommitmentKeyHashOutput 31 bytes
// - Value: 8 bytes
// - VarInt: 1 byte (PkScript length)
// - PkScript (P2WPKH)
CommitmentKeyHashOutput = 8 + 1 + P2WPKHSize
// CommitmentAnchorOutput 43 bytes
// - Value: 8 bytes
// - VarInt: 1 byte (PkScript length)
// - PkScript (P2WSH)
CommitmentAnchorOutput = 8 + 1 + P2WSHSize
2017-02-23 22:56:47 +03:00
// HTLCSize 43 bytes
// - Value: 8 bytes
// - VarInt: 1 byte (PkScript length)
// - PkScript (PW2SH)
HTLCSize = 8 + 1 + P2WSHSize
2017-02-23 22:56:47 +03:00
// WitnessHeaderSize 2 bytes
// - Flag: 1 byte
// - Marker: 1 byte
WitnessHeaderSize = 1 + 1
// BaseTxSize 8 bytes
lnwallet/size: correct commit to-local and 2nd stage script sizes In this commit, we correct our size estimates for to-local scripts, which are used on the commitment transaction and the htlc success/timeout transactions. There have been observed cases of transactions getting stuck because our estimates were too low, and cause the transactions to not be relayed. Our previous estimate for the commitment to-local script was derived from an older version of the script. Though the estimate is greater than the actual size, this has been updated with the current estimate of 79 bytes. This estimates makes the assumption that CSV delays will be at most 4 bytes when serialized. Since this value is expressed in relative block heights, this should be more than sufficient for our needs, even though the maximum possible size for the little-endian int64 is 9 bytes (plus an OP_DATA). The other correction is to use the ToLocalScriptSize as our estimate for htlc timeout/success scripts, as they are the same script. Previously, our estimate was derived from the proper script, though we were 6 bytes shy of the new to-local estimate, since we counted the csv_delay as 1 byte, and missed some other OP_DATAs. All derived estimates have been updating depending on the new and improved ToLocalScriptSize estimate, and fix some estimates that did not include the witness length in the estimate. Finally, we correct some weight miscalculations in: - AcceptedHtlcTimeoutWitnessSize: missing data push lengths - OfferedHtlcSuccessWitnessSize: extra 73 byte sig, missing data push lengths - OfferedHtlcPenaltyWitnessSize: missing 33 byte pubkey
2018-07-25 06:30:48 +03:00
// - Version: 4 bytes
// - LockTime: 4 bytes
BaseTxSize = 4 + 4
// BaseCommitmentTxSize 125 + 43 * num-htlc-outputs bytes
// - Version: 4 bytes
// - WitnessHeader <---- part of the witness data
// - CountTxIn: 1 byte
// - TxIn: 41 bytes
// FundingInput
// - CountTxOut: 1 byte
// - TxOut: 74 + 43 * num-htlc-outputs bytes
// OutputPayingToThem,
// OutputPayingToUs,
// ....HTLCOutputs...
// - LockTime: 4 bytes
BaseCommitmentTxSize = 4 + 1 + FundingInputSize + 1 +
CommitmentDelayOutput + CommitmentKeyHashOutput + 4
// BaseCommitmentTxWeight 500 weight
BaseCommitmentTxWeight = witnessScaleFactor * BaseCommitmentTxSize
// WitnessCommitmentTxWeight 224 weight
WitnessCommitmentTxWeight = WitnessHeaderSize + WitnessSize
// BaseAnchorCommitmentTxSize 225 + 43 * num-htlc-outputs bytes
// - Version: 4 bytes
// - WitnessHeader <---- part of the witness data
// - CountTxIn: 1 byte
// - TxIn: 41 bytes
// FundingInput
// - CountTxOut: 3 byte
// - TxOut: 4*43 + 43 * num-htlc-outputs bytes
// OutputPayingToThem,
// OutputPayingToUs,
// AnchorPayingToThem,
// AnchorPayingToUs,
// ....HTLCOutputs...
// - LockTime: 4 bytes
BaseAnchorCommitmentTxSize = 4 + 1 + FundingInputSize + 3 +
2*CommitmentDelayOutput + 2*CommitmentAnchorOutput + 4
// BaseAnchorCommitmentTxWeight 900 weight
BaseAnchorCommitmentTxWeight = witnessScaleFactor * BaseAnchorCommitmentTxSize
// CommitWeight 724 weight
CommitWeight = BaseCommitmentTxWeight + WitnessCommitmentTxWeight
// AnchorCommitWeight 1124 weight
AnchorCommitWeight = BaseAnchorCommitmentTxWeight + WitnessCommitmentTxWeight
// HTLCWeight 172 weight
HTLCWeight = witnessScaleFactor * HTLCSize
// HtlcTimeoutWeight is the weight of the HTLC timeout transaction
// which will transition an outgoing HTLC to the delay-and-claim state.
HtlcTimeoutWeight = 663
// HtlcSuccessWeight is the weight of the HTLC success transaction
// which will transition an incoming HTLC to the delay-and-claim state.
HtlcSuccessWeight = 703
// HtlcConfirmedScriptOverhead is the extra length of an HTLC script
// that requires confirmation before it can be spent. These extra bytes
// is a result of the extra CSV check.
HtlcConfirmedScriptOverhead = 3
// HtlcTimeoutWeightConfirmed is the weight of the HTLC timeout
// transaction which will transition an outgoing HTLC to the
// delay-and-claim state, for the confirmed HTLC outputs. It is 3 bytes
// larger because of the additional CSV check in the input script.
HtlcTimeoutWeightConfirmed = HtlcTimeoutWeight + HtlcConfirmedScriptOverhead
// HtlcSuccessWeightCOnfirmed is the weight of the HTLC success
// transaction which will transition an incoming HTLC to the
// delay-and-claim state, for the confirmed HTLC outputs. It is 3 bytes
// larger because of the cdditional CSV check in the input script.
HtlcSuccessWeightConfirmed = HtlcSuccessWeight + HtlcConfirmedScriptOverhead
// MaxHTLCNumber is the maximum number HTLCs which can be included in a
// commitment transaction. This limit was chosen such that, in the case
// of a contract breach, the punishment transaction is able to sweep
// all the HTLC's yet still remain below the widely used standard
// weight limits.
MaxHTLCNumber = 966
lnwallet/size: correct commit to-local and 2nd stage script sizes In this commit, we correct our size estimates for to-local scripts, which are used on the commitment transaction and the htlc success/timeout transactions. There have been observed cases of transactions getting stuck because our estimates were too low, and cause the transactions to not be relayed. Our previous estimate for the commitment to-local script was derived from an older version of the script. Though the estimate is greater than the actual size, this has been updated with the current estimate of 79 bytes. This estimates makes the assumption that CSV delays will be at most 4 bytes when serialized. Since this value is expressed in relative block heights, this should be more than sufficient for our needs, even though the maximum possible size for the little-endian int64 is 9 bytes (plus an OP_DATA). The other correction is to use the ToLocalScriptSize as our estimate for htlc timeout/success scripts, as they are the same script. Previously, our estimate was derived from the proper script, though we were 6 bytes shy of the new to-local estimate, since we counted the csv_delay as 1 byte, and missed some other OP_DATAs. All derived estimates have been updating depending on the new and improved ToLocalScriptSize estimate, and fix some estimates that did not include the witness length in the estimate. Finally, we correct some weight miscalculations in: - AcceptedHtlcTimeoutWitnessSize: missing data push lengths - OfferedHtlcSuccessWitnessSize: extra 73 byte sig, missing data push lengths - OfferedHtlcPenaltyWitnessSize: missing 33 byte pubkey
2018-07-25 06:30:48 +03:00
// ToLocalScriptSize 79 bytes
// - OP_IF: 1 byte
lnwallet/size: correct commit to-local and 2nd stage script sizes In this commit, we correct our size estimates for to-local scripts, which are used on the commitment transaction and the htlc success/timeout transactions. There have been observed cases of transactions getting stuck because our estimates were too low, and cause the transactions to not be relayed. Our previous estimate for the commitment to-local script was derived from an older version of the script. Though the estimate is greater than the actual size, this has been updated with the current estimate of 79 bytes. This estimates makes the assumption that CSV delays will be at most 4 bytes when serialized. Since this value is expressed in relative block heights, this should be more than sufficient for our needs, even though the maximum possible size for the little-endian int64 is 9 bytes (plus an OP_DATA). The other correction is to use the ToLocalScriptSize as our estimate for htlc timeout/success scripts, as they are the same script. Previously, our estimate was derived from the proper script, though we were 6 bytes shy of the new to-local estimate, since we counted the csv_delay as 1 byte, and missed some other OP_DATAs. All derived estimates have been updating depending on the new and improved ToLocalScriptSize estimate, and fix some estimates that did not include the witness length in the estimate. Finally, we correct some weight miscalculations in: - AcceptedHtlcTimeoutWitnessSize: missing data push lengths - OfferedHtlcSuccessWitnessSize: extra 73 byte sig, missing data push lengths - OfferedHtlcPenaltyWitnessSize: missing 33 byte pubkey
2018-07-25 06:30:48 +03:00
// - OP_DATA: 1 byte
// - revoke_key: 33 bytes
// - OP_ELSE: 1 byte
lnwallet/size: correct commit to-local and 2nd stage script sizes In this commit, we correct our size estimates for to-local scripts, which are used on the commitment transaction and the htlc success/timeout transactions. There have been observed cases of transactions getting stuck because our estimates were too low, and cause the transactions to not be relayed. Our previous estimate for the commitment to-local script was derived from an older version of the script. Though the estimate is greater than the actual size, this has been updated with the current estimate of 79 bytes. This estimates makes the assumption that CSV delays will be at most 4 bytes when serialized. Since this value is expressed in relative block heights, this should be more than sufficient for our needs, even though the maximum possible size for the little-endian int64 is 9 bytes (plus an OP_DATA). The other correction is to use the ToLocalScriptSize as our estimate for htlc timeout/success scripts, as they are the same script. Previously, our estimate was derived from the proper script, though we were 6 bytes shy of the new to-local estimate, since we counted the csv_delay as 1 byte, and missed some other OP_DATAs. All derived estimates have been updating depending on the new and improved ToLocalScriptSize estimate, and fix some estimates that did not include the witness length in the estimate. Finally, we correct some weight miscalculations in: - AcceptedHtlcTimeoutWitnessSize: missing data push lengths - OfferedHtlcSuccessWitnessSize: extra 73 byte sig, missing data push lengths - OfferedHtlcPenaltyWitnessSize: missing 33 byte pubkey
2018-07-25 06:30:48 +03:00
// - OP_DATA: 1 byte
// - csv_delay: 4 bytes
// - OP_CHECKSEQUENCEVERIFY: 1 byte
// - OP_DROP: 1 byte
// - OP_DATA: 1 byte
// - delay_key: 33 bytes
// - OP_ENDIF: 1 byte
lnwallet/size: correct commit to-local and 2nd stage script sizes In this commit, we correct our size estimates for to-local scripts, which are used on the commitment transaction and the htlc success/timeout transactions. There have been observed cases of transactions getting stuck because our estimates were too low, and cause the transactions to not be relayed. Our previous estimate for the commitment to-local script was derived from an older version of the script. Though the estimate is greater than the actual size, this has been updated with the current estimate of 79 bytes. This estimates makes the assumption that CSV delays will be at most 4 bytes when serialized. Since this value is expressed in relative block heights, this should be more than sufficient for our needs, even though the maximum possible size for the little-endian int64 is 9 bytes (plus an OP_DATA). The other correction is to use the ToLocalScriptSize as our estimate for htlc timeout/success scripts, as they are the same script. Previously, our estimate was derived from the proper script, though we were 6 bytes shy of the new to-local estimate, since we counted the csv_delay as 1 byte, and missed some other OP_DATAs. All derived estimates have been updating depending on the new and improved ToLocalScriptSize estimate, and fix some estimates that did not include the witness length in the estimate. Finally, we correct some weight miscalculations in: - AcceptedHtlcTimeoutWitnessSize: missing data push lengths - OfferedHtlcSuccessWitnessSize: extra 73 byte sig, missing data push lengths - OfferedHtlcPenaltyWitnessSize: missing 33 byte pubkey
2018-07-25 06:30:48 +03:00
// - OP_CHECKSIG: 1 byte
ToLocalScriptSize = 1 + 1 + 33 + 1 + 1 + 4 + 1 + 1 + 1 + 33 + 1 + 1
// ToLocalTimeoutWitnessSize 156 bytes
// - number_of_witness_elements: 1 byte
// - local_delay_sig_length: 1 byte
// - local_delay_sig: 73 bytes
// - zero_length: 1 byte
// - witness_script_length: 1 byte
// - witness_script (to_local_script)
ToLocalTimeoutWitnessSize = 1 + 1 + 73 + 1 + 1 + ToLocalScriptSize
lnwallet/size: correct commit to-local and 2nd stage script sizes In this commit, we correct our size estimates for to-local scripts, which are used on the commitment transaction and the htlc success/timeout transactions. There have been observed cases of transactions getting stuck because our estimates were too low, and cause the transactions to not be relayed. Our previous estimate for the commitment to-local script was derived from an older version of the script. Though the estimate is greater than the actual size, this has been updated with the current estimate of 79 bytes. This estimates makes the assumption that CSV delays will be at most 4 bytes when serialized. Since this value is expressed in relative block heights, this should be more than sufficient for our needs, even though the maximum possible size for the little-endian int64 is 9 bytes (plus an OP_DATA). The other correction is to use the ToLocalScriptSize as our estimate for htlc timeout/success scripts, as they are the same script. Previously, our estimate was derived from the proper script, though we were 6 bytes shy of the new to-local estimate, since we counted the csv_delay as 1 byte, and missed some other OP_DATAs. All derived estimates have been updating depending on the new and improved ToLocalScriptSize estimate, and fix some estimates that did not include the witness length in the estimate. Finally, we correct some weight miscalculations in: - AcceptedHtlcTimeoutWitnessSize: missing data push lengths - OfferedHtlcSuccessWitnessSize: extra 73 byte sig, missing data push lengths - OfferedHtlcPenaltyWitnessSize: missing 33 byte pubkey
2018-07-25 06:30:48 +03:00
// ToLocalPenaltyWitnessSize 156 bytes
// - number_of_witness_elements: 1 byte
// - revocation_sig_length: 1 byte
// - revocation_sig: 73 bytes
lnwallet/size: correct commit to-local and 2nd stage script sizes In this commit, we correct our size estimates for to-local scripts, which are used on the commitment transaction and the htlc success/timeout transactions. There have been observed cases of transactions getting stuck because our estimates were too low, and cause the transactions to not be relayed. Our previous estimate for the commitment to-local script was derived from an older version of the script. Though the estimate is greater than the actual size, this has been updated with the current estimate of 79 bytes. This estimates makes the assumption that CSV delays will be at most 4 bytes when serialized. Since this value is expressed in relative block heights, this should be more than sufficient for our needs, even though the maximum possible size for the little-endian int64 is 9 bytes (plus an OP_DATA). The other correction is to use the ToLocalScriptSize as our estimate for htlc timeout/success scripts, as they are the same script. Previously, our estimate was derived from the proper script, though we were 6 bytes shy of the new to-local estimate, since we counted the csv_delay as 1 byte, and missed some other OP_DATAs. All derived estimates have been updating depending on the new and improved ToLocalScriptSize estimate, and fix some estimates that did not include the witness length in the estimate. Finally, we correct some weight miscalculations in: - AcceptedHtlcTimeoutWitnessSize: missing data push lengths - OfferedHtlcSuccessWitnessSize: extra 73 byte sig, missing data push lengths - OfferedHtlcPenaltyWitnessSize: missing 33 byte pubkey
2018-07-25 06:30:48 +03:00
// - OP_TRUE: 1 byte
// - witness_script_length: 1 byte
// - witness_script (to_local_script)
ToLocalPenaltyWitnessSize = 1 + 1 + 73 + 1 + 1 + ToLocalScriptSize
// ToRemoteConfirmedScriptSize 37 bytes
// - OP_DATA: 1 byte
// - to_remote_key: 33 bytes
// - OP_CHECKSIGVERIFY: 1 byte
// - OP_1: 1 byte
// - OP_CHECKSEQUENCEVERIFY: 1 byte
ToRemoteConfirmedScriptSize = 1 + 33 + 1 + 1 + 1
// ToRemoteConfirmedWitnessSize 113 bytes
// - number_of_witness_elements: 1 byte
// - sig_length: 1 byte
// - sig: 73 bytes
// - witness_script_length: 1 byte
// - witness_script (to_remote_delayed_script)
ToRemoteConfirmedWitnessSize = 1 + 1 + 73 + 1 + ToRemoteConfirmedScriptSize
// AcceptedHtlcScriptSize 142 bytes
// - OP_DUP: 1 byte
// - OP_HASH160: 1 byte
// - OP_DATA: 1 byte (RIPEMD160(SHA256(revocationkey)) length)
// - RIPEMD160(SHA256(revocationkey)): 20 bytes
// - OP_EQUAL: 1 byte
// - OP_IF: 1 byte
// - OP_CHECKSIG: 1 byte
// - OP_ELSE: 1 byte
// - OP_DATA: 1 byte (remotekey length)
// - remotekey: 33 bytes
// - OP_SWAP: 1 byte
// - OP_SIZE: 1 byte
// - 32: 1 byte
// - OP_EQUAL: 1 byte
// - OP_IF: 1 byte
// - OP_HASH160: 1 byte
// - OP_DATA: 1 byte (RIPEMD160(payment_hash) length)
// - RIPEMD160(payment_hash): 20 bytes
// - OP_EQUALVERIFY: 1 byte
// - 2: 1 byte
// - OP_SWAP: 1 byte
// - OP_DATA: 1 byte (localkey length)
// - localkey: 33 bytes
// - 2: 1 byte
// - OP_CHECKMULTISIG: 1 byte
// - OP_ELSE: 1 byte
// - OP_DROP: 1 byte
// - OP_DATA: 1 byte (cltv_expiry length)
// - cltv_expiry: 4 bytes
// - OP_CHECKLOCKTIMEVERIFY: 1 byte
// - OP_DROP: 1 byte
// - OP_CHECKSIG: 1 byte
// - OP_ENDIF: 1 byte
// - OP_1: 1 byte // These 3 extra bytes are used for both confirmed and regular
// - OP_CSV: 1 byte // HTLC script types. The size won't be correct in all cases,
// - OP_DROP: 1 byte // but it is just an upper bound used for fee estimation in any case.
// - OP_ENDIF: 1 byte
AcceptedHtlcScriptSize = 3*1 + 20 + 5*1 + 33 + 7*1 + 20 + 4*1 +
33 + 5*1 + 4 + 8*1
// AcceptedHtlcTimeoutWitnessSize 219
lnwallet/size: correct commit to-local and 2nd stage script sizes In this commit, we correct our size estimates for to-local scripts, which are used on the commitment transaction and the htlc success/timeout transactions. There have been observed cases of transactions getting stuck because our estimates were too low, and cause the transactions to not be relayed. Our previous estimate for the commitment to-local script was derived from an older version of the script. Though the estimate is greater than the actual size, this has been updated with the current estimate of 79 bytes. This estimates makes the assumption that CSV delays will be at most 4 bytes when serialized. Since this value is expressed in relative block heights, this should be more than sufficient for our needs, even though the maximum possible size for the little-endian int64 is 9 bytes (plus an OP_DATA). The other correction is to use the ToLocalScriptSize as our estimate for htlc timeout/success scripts, as they are the same script. Previously, our estimate was derived from the proper script, though we were 6 bytes shy of the new to-local estimate, since we counted the csv_delay as 1 byte, and missed some other OP_DATAs. All derived estimates have been updating depending on the new and improved ToLocalScriptSize estimate, and fix some estimates that did not include the witness length in the estimate. Finally, we correct some weight miscalculations in: - AcceptedHtlcTimeoutWitnessSize: missing data push lengths - OfferedHtlcSuccessWitnessSize: extra 73 byte sig, missing data push lengths - OfferedHtlcPenaltyWitnessSize: missing 33 byte pubkey
2018-07-25 06:30:48 +03:00
// - number_of_witness_elements: 1 byte
// - sender_sig_length: 1 byte
// - sender_sig: 73 bytes
// - nil_length: 1 byte
// - witness_script_length: 1 byte
// - witness_script: (accepted_htlc_script)
AcceptedHtlcTimeoutWitnessSize = 1 + 1 + 73 + 1 + 1 + AcceptedHtlcScriptSize
// AcceptedHtlcPenaltyWitnessSize 252 bytes
lnwallet/size: correct commit to-local and 2nd stage script sizes In this commit, we correct our size estimates for to-local scripts, which are used on the commitment transaction and the htlc success/timeout transactions. There have been observed cases of transactions getting stuck because our estimates were too low, and cause the transactions to not be relayed. Our previous estimate for the commitment to-local script was derived from an older version of the script. Though the estimate is greater than the actual size, this has been updated with the current estimate of 79 bytes. This estimates makes the assumption that CSV delays will be at most 4 bytes when serialized. Since this value is expressed in relative block heights, this should be more than sufficient for our needs, even though the maximum possible size for the little-endian int64 is 9 bytes (plus an OP_DATA). The other correction is to use the ToLocalScriptSize as our estimate for htlc timeout/success scripts, as they are the same script. Previously, our estimate was derived from the proper script, though we were 6 bytes shy of the new to-local estimate, since we counted the csv_delay as 1 byte, and missed some other OP_DATAs. All derived estimates have been updating depending on the new and improved ToLocalScriptSize estimate, and fix some estimates that did not include the witness length in the estimate. Finally, we correct some weight miscalculations in: - AcceptedHtlcTimeoutWitnessSize: missing data push lengths - OfferedHtlcSuccessWitnessSize: extra 73 byte sig, missing data push lengths - OfferedHtlcPenaltyWitnessSize: missing 33 byte pubkey
2018-07-25 06:30:48 +03:00
// - number_of_witness_elements: 1 byte
// - revocation_sig_length: 1 byte
// - revocation_sig: 73 bytes
// - revocation_key_length: 1 byte
// - revocation_key: 33 bytes
// - witness_script_length: 1 byte
// - witness_script (accepted_htlc_script)
AcceptedHtlcPenaltyWitnessSize = 1 + 1 + 73 + 1 + 33 + 1 + AcceptedHtlcScriptSize
// OfferedHtlcScriptSize 136 bytes
// - OP_DUP: 1 byte
// - OP_HASH160: 1 byte
// - OP_DATA: 1 byte (RIPEMD160(SHA256(revocationkey)) length)
// - RIPEMD160(SHA256(revocationkey)): 20 bytes
// - OP_EQUAL: 1 byte
// - OP_IF: 1 byte
// - OP_CHECKSIG: 1 byte
// - OP_ELSE: 1 byte
// - OP_DATA: 1 byte (remotekey length)
// - remotekey: 33 bytes
// - OP_SWAP: 1 byte
// - OP_SIZE: 1 byte
// - OP_DATA: 1 byte (32 length)
// - 32: 1 byte
// - OP_EQUAL: 1 byte
// - OP_NOTIF: 1 byte
// - OP_DROP: 1 byte
// - 2: 1 byte
// - OP_SWAP: 1 byte
// - OP_DATA: 1 byte (localkey length)
// - localkey: 33 bytes
// - 2: 1 byte
// - OP_CHECKMULTISIG: 1 byte
// - OP_ELSE: 1 byte
// - OP_HASH160: 1 byte
// - OP_DATA: 1 byte (RIPEMD160(payment_hash) length)
// - RIPEMD160(payment_hash): 20 bytes
// - OP_EQUALVERIFY: 1 byte
// - OP_CHECKSIG: 1 byte
// - OP_ENDIF: 1 byte
// - OP_1: 1 byte
// - OP_CSV: 1 byte
// - OP_DROP: 1 byte
// - OP_ENDIF: 1 byte
OfferedHtlcScriptSize = 3*1 + 20 + 5*1 + 33 + 10*1 + 33 + 5*1 + 20 + 7*1
// OfferedHtlcSuccessWitnessSize 320 bytes
lnwallet/size: correct commit to-local and 2nd stage script sizes In this commit, we correct our size estimates for to-local scripts, which are used on the commitment transaction and the htlc success/timeout transactions. There have been observed cases of transactions getting stuck because our estimates were too low, and cause the transactions to not be relayed. Our previous estimate for the commitment to-local script was derived from an older version of the script. Though the estimate is greater than the actual size, this has been updated with the current estimate of 79 bytes. This estimates makes the assumption that CSV delays will be at most 4 bytes when serialized. Since this value is expressed in relative block heights, this should be more than sufficient for our needs, even though the maximum possible size for the little-endian int64 is 9 bytes (plus an OP_DATA). The other correction is to use the ToLocalScriptSize as our estimate for htlc timeout/success scripts, as they are the same script. Previously, our estimate was derived from the proper script, though we were 6 bytes shy of the new to-local estimate, since we counted the csv_delay as 1 byte, and missed some other OP_DATAs. All derived estimates have been updating depending on the new and improved ToLocalScriptSize estimate, and fix some estimates that did not include the witness length in the estimate. Finally, we correct some weight miscalculations in: - AcceptedHtlcTimeoutWitnessSize: missing data push lengths - OfferedHtlcSuccessWitnessSize: extra 73 byte sig, missing data push lengths - OfferedHtlcPenaltyWitnessSize: missing 33 byte pubkey
2018-07-25 06:30:48 +03:00
// - number_of_witness_elements: 1 byte
// - nil_length: 1 byte
// - receiver_sig_length: 1 byte
// - receiver_sig: 73 bytes
// - sender_sig_length: 1 byte
// - sender_sig: 73 bytes
// - payment_preimage_length: 1 byte
// - payment_preimage: 32 bytes
// - witness_script_length: 1 byte
// - witness_script (offered_htlc_script)
OfferedHtlcSuccessWitnessSize = 1 + 1 + 1 + 73 + 1 + 73 + 1 + 32 + 1 + OfferedHtlcScriptSize
// OfferedHtlcPenaltyWitnessSize 246 bytes
// - number_of_witness_elements: 1 byte
// - revocation_sig_length: 1 byte
// - revocation_sig: 73 bytes
// - revocation_key_length: 1 byte
// - revocation_key: 33 bytes
// - witness_script_length: 1 byte
// - witness_script (offered_htlc_script)
lnwallet/size: correct commit to-local and 2nd stage script sizes In this commit, we correct our size estimates for to-local scripts, which are used on the commitment transaction and the htlc success/timeout transactions. There have been observed cases of transactions getting stuck because our estimates were too low, and cause the transactions to not be relayed. Our previous estimate for the commitment to-local script was derived from an older version of the script. Though the estimate is greater than the actual size, this has been updated with the current estimate of 79 bytes. This estimates makes the assumption that CSV delays will be at most 4 bytes when serialized. Since this value is expressed in relative block heights, this should be more than sufficient for our needs, even though the maximum possible size for the little-endian int64 is 9 bytes (plus an OP_DATA). The other correction is to use the ToLocalScriptSize as our estimate for htlc timeout/success scripts, as they are the same script. Previously, our estimate was derived from the proper script, though we were 6 bytes shy of the new to-local estimate, since we counted the csv_delay as 1 byte, and missed some other OP_DATAs. All derived estimates have been updating depending on the new and improved ToLocalScriptSize estimate, and fix some estimates that did not include the witness length in the estimate. Finally, we correct some weight miscalculations in: - AcceptedHtlcTimeoutWitnessSize: missing data push lengths - OfferedHtlcSuccessWitnessSize: extra 73 byte sig, missing data push lengths - OfferedHtlcPenaltyWitnessSize: missing 33 byte pubkey
2018-07-25 06:30:48 +03:00
OfferedHtlcPenaltyWitnessSize = 1 + 1 + 73 + 1 + 33 + 1 + OfferedHtlcScriptSize
)
2020-03-10 15:23:17 +03:00
// dummySigner is a fake signer used for size (upper bound) calculations.
type dummySigner struct {
Signer
}
// SignOutputRaw generates a signature for the passed transaction according to
// the data within the passed SignDescriptor.
func (s *dummySigner) SignOutputRaw(tx *wire.MsgTx, signDesc *SignDescriptor) (
[]byte, error) {
// Always return worst-case signature length, excluding the one byte
// sighash flag.
return make([]byte, 73-1), nil
}
var (
// dummyPubKey is a pubkey used in script size calculation.
dummyPubKey = btcec.PublicKey{
X: &big.Int{},
Y: &big.Int{},
}
// dummyAnchorScript is a script used for size calculation.
dummyAnchorScript, _ = CommitScriptAnchor(&dummyPubKey)
// dummyAnchorWitness is a witness used for size calculation.
dummyAnchorWitness, _ = CommitSpendAnchor(
&dummySigner{},
&SignDescriptor{
KeyDesc: keychain.KeyDescriptor{
PubKey: &dummyPubKey,
},
WitnessScript: dummyAnchorScript,
},
nil,
)
// AnchorWitnessSize 116 bytes
AnchorWitnessSize = dummyAnchorWitness.SerializeSize()
)
// EstimateCommitTxWeight estimate commitment transaction weight depending on
// the precalculated weight of base transaction, witness data, which is needed
// for paying for funding tx, and htlc weight multiplied by their count.
func EstimateCommitTxWeight(count int, prediction bool) int64 {
// Make prediction about the size of commitment transaction with
// additional HTLC.
if prediction {
count++
}
htlcWeight := int64(count * HTLCWeight)
baseWeight := int64(BaseCommitmentTxWeight)
witnessWeight := int64(WitnessCommitmentTxWeight)
return htlcWeight + baseWeight + witnessWeight
}
// TxWeightEstimator is able to calculate weight estimates for transactions
// based on the input and output types. For purposes of estimation, all
// signatures are assumed to be of the maximum possible size, 73 bytes. Each
// method of the estimator returns an instance with the estimate applied. This
// allows callers to chain each of the methods
type TxWeightEstimator struct {
hasWitness bool
inputCount uint32
outputCount uint32
inputSize int
inputWitnessSize int
outputSize int
}
// AddP2PKHInput updates the weight estimate to account for an additional input
// spending a P2PKH output.
func (twe *TxWeightEstimator) AddP2PKHInput() *TxWeightEstimator {
twe.inputSize += InputSize + P2PKHScriptSigSize
twe.inputWitnessSize++
twe.inputCount++
return twe
}
// AddP2WKHInput updates the weight estimate to account for an additional input
// spending a native P2PWKH output.
func (twe *TxWeightEstimator) AddP2WKHInput() *TxWeightEstimator {
twe.AddWitnessInput(P2WKHWitnessSize)
return twe
}
// AddWitnessInput updates the weight estimate to account for an additional
// input spending a native pay-to-witness output. This accepts the total size
// of the witness as a parameter.
func (twe *TxWeightEstimator) AddWitnessInput(witnessSize int) *TxWeightEstimator {
twe.inputSize += InputSize
twe.inputWitnessSize += witnessSize
twe.inputCount++
twe.hasWitness = true
return twe
}
// AddNestedP2WKHInput updates the weight estimate to account for an additional
// input spending a P2SH output with a nested P2WKH redeem script.
func (twe *TxWeightEstimator) AddNestedP2WKHInput() *TxWeightEstimator {
twe.inputSize += InputSize + NestedP2WPKHSize
twe.inputWitnessSize += P2WKHWitnessSize
twe.inputCount++
twe.hasWitness = true
return twe
}
// AddNestedP2WSHInput updates the weight estimate to account for an additional
// input spending a P2SH output with a nested P2WSH redeem script.
func (twe *TxWeightEstimator) AddNestedP2WSHInput(witnessSize int) *TxWeightEstimator {
twe.inputSize += InputSize + NestedP2WSHSize
twe.inputWitnessSize += witnessSize
twe.inputCount++
twe.hasWitness = true
return twe
}
// AddP2PKHOutput updates the weight estimate to account for an additional P2PKH
// output.
func (twe *TxWeightEstimator) AddP2PKHOutput() *TxWeightEstimator {
twe.outputSize += P2PKHOutputSize
twe.outputCount++
return twe
}
// AddP2WKHOutput updates the weight estimate to account for an additional
// native P2WKH output.
func (twe *TxWeightEstimator) AddP2WKHOutput() *TxWeightEstimator {
twe.outputSize += P2WKHOutputSize
twe.outputCount++
return twe
}
// AddP2WSHOutput updates the weight estimate to account for an additional
// native P2WSH output.
func (twe *TxWeightEstimator) AddP2WSHOutput() *TxWeightEstimator {
twe.outputSize += P2WSHOutputSize
twe.outputCount++
return twe
}
// AddP2SHOutput updates the weight estimate to account for an additional P2SH
// output.
func (twe *TxWeightEstimator) AddP2SHOutput() *TxWeightEstimator {
twe.outputSize += P2SHOutputSize
twe.outputCount++
return twe
}
// Weight gets the estimated weight of the transaction.
func (twe *TxWeightEstimator) Weight() int {
txSizeStripped := BaseTxSize +
wire.VarIntSerializeSize(uint64(twe.inputCount)) + twe.inputSize +
wire.VarIntSerializeSize(uint64(twe.outputCount)) + twe.outputSize
weight := txSizeStripped * witnessScaleFactor
if twe.hasWitness {
weight += WitnessHeaderSize + twe.inputWitnessSize
}
return weight
}
// VSize gets the estimated virtual size of the transactions, in vbytes.
func (twe *TxWeightEstimator) VSize() int {
// A tx's vsize is 1/4 of the weight, rounded up.
return (twe.Weight() + witnessScaleFactor - 1) / witnessScaleFactor
}