Similar to the sweeper, when we're about to make a new breach
transaction, we ask the sweeper for a new change address, if it has one.
Then when we go to publish, we notify broadcast.
In this commit, we start to use the AuxSweeper (if present) to obtain a
new extra change addr we should add to the sweeping transaction. With
this, we'll take the set of inputs and our change addr, and then maybe
gain a new change addr to add to the sweep transaction.
The extra change addr will be treated as an extra required tx out,
shared across all the relevant inputs. This'll also be used in
NeedWalletInput to make sure that we add an extra input if needed to be
able to pay for the change addr.
In this commit, we add a new AuxSweeper interface. This'll take a set of
inputs, and a change addr for the sweep transaction, then optionally
return a new sweep output to be added to the sweep transaction.
We also add a new NotifyBroadcast method. This'll be used to notify
that we're _about_ to broadcast a sweeping transaction. The set of
inputs is passed in, which allows the caller to prepare for the ultimate
broadcast of the sweeping transaction.
We also add ExtraTxOut to BumpRequest pass fees to NotifyBroadcast. This
allows the callee to know the total fee of the sweeping transaction.
In this commit, we refactor all the other constructors for the input to
use MakeBaseInput. We also add a new set of functional options as well.
This'll be useful later on to ensure that new options are properly
applied to all the input types.
This will be used by external callers to modify the way we resolve
contracts on chain. For a given contract, we'll store an extra "blob",
that will later be presented during the sweeping phase.
In this commit, we remove the global application level lock from the
postgres backend. This lock prevents multiple write transactions from
happening at the same time, and will also block a writer if a read is on
going. Since this lock was added, we know always open DB connections
with the strongest level of concurrency control available:
`LevelSerializable`. In concert with the new auto retry logic, we ensure
that if db transactions conflict (writing the same key/row in this
case), then the tx is retried automatically.
Removing this lock should increase perf for the postgres backend, as now
concurrent write transactions can proceed, being serialized as needed.
Rather then trying to handle concurrency at the application level, we'll
set postgres do its job, with the application only needing to retry as
necessary.
In this commit, we move to add the internal key to the delivery addr. This way, we give the aux chan closer the extra information it may need to properly augment the normal co-op close process.
In this commit, we start to use the new AuxSigner to obtain+verify aux sigs for all second level HTLCs. This is similar to the existing SigPool, but we'll only attempt to do this if the AuxSigner is present (won't be for most channels).
In this commit, we add a new aux signer interface that's meant to mirror the SigPool. If present, this'll be used to (maybe) obtain signatures for second level HTLCs for certain classes of custom channels.
For the initiator, once we get the signal that the PSBT has been
finalized, we'll call into the aux funder to get the funding desc. For
the responder, once we receive the funding_created message, we'll do the
same.
We now also have local+remote aux leaves for the commitment transaction.
Some old TODO comments that in retrospect aren't required anymore are
removed as well.
In this commit, we modify the aux funding work flow slightly. We won't
be able to generate the full AuxFundingDesc until both sides has
sent+received funding params. So we'll now only attempt to bind the
tapscript root as soon as we send+recv the open_channel message.
We'll now also make sure that we pass the tapscript root all the way
down into the musig2 session creation.
In this commit, we make a new `AuxFundingController` interface capable of processing messages off the wire. In addition, we can use it to abstract away details w.r.t how we obtain a `AuxFundingDesc` for a given channel.
We'll now use this whenever we get a channel funding request, to make sure we pass along the custom state that a channel may require.
This struct will house all the information we'll need to do a class of custom channels that relies primarily on adding additional items to the tapscript root of the HTLC/commitment/funding outputs.
In this commit, we bring the timeout resolver more in line with the
success resolver by using the sweeper to handle the HTLC offered remote
timeout outputs. These are outputs that we can sweep directly from the
remote party's commitment transaction when they broadcast their version
of the commitment transaction.
With this change, we slim down the scope slightly by only doing this for
anchor channels. Non-anchor channels will continue to use the
utxonursery for this output type for now.
We might be trying to send an invoice amount that's greater than the size of the channel, but once you factor in the custom channel logic, an actual HTLC can be sent over the channel to pay that larger payment.
As a result, we'll skip over this check if a have a custom HTLC.
In this commit, we expand some of the existing chan sync tests to cover
taproot channels (the others already did). Along the way, we always
assert that the `PartialSig` is populated on retransmission. In
addition, we now send the new commit sig rather than the existing
in-memory one to test the new logic that re-signs the commitment.
In this commit, we fix an existing bug with the taproot channel type
that can cause force closes if a peer disconnects while attempting to
send the commitment signature.
Before this commit, since the `PartialSig` we send is never committed to
disk, the version read wouldn't contain the musig2 partial sig. We never
write these signatures to disk, as each time we make a new session, we
need to generate fresh nonces to avoid nonce-reuse.
Due to the above interaction, if we went to re-send a signature after a
disconnection, the `CommitSig` message we sent wouldn't actually contain
a `PartialSigWithNonce`, causing a protocol error.
This may be useful for custom channel types that base everything off the index (a global value) rather than the output index (can change with each state).
In this commit, we add a TLV blob to the PaymentDescriptor struct. We also now thread through this value from the UpdateAddHTLC message to the PaymentDescriptor mapping, and the other way around.
In this commit, we add a new AuxLeafStore which can be used to dynamically fetch the latest aux leaves for a given state. This is useful for custom channel types that will store some extra information in the form of a custom blob, then will use that information to derive the new leaf tapscript leaves that may be attached to reach state.
In this commit, we also add the custom TLV blob to the internal commitment struct that we use within the in-memory commitment linked list.
This'll be useful to ensure that we're tracking the current blob for our in memory commitment for when we need to write it to disk.
In this commit, for each channel, we'll now start to store an optional custom blob. This can be used to store extra information for custom channels in an opauqe manner.
In this commit, we update all the taproot scripts to also accept an
optional aux leaf. This aux leaf can be used to add more redemption
paths for advanced channels, or just as an extra commitment space.
With this commit, the channel is now aware of if it's a musig2 channel, that also has a tapscript root. We'll need to always pass in the tapscript root each time we: make the funding output, sign a new state, and also verify a new state.
This isn't hooked up yet to the funding manager, but with this commit, we can now start to write internal unit tests that handle musig2 channels with a tapscript root.
In this commit, we consolidate the root bucket TLVs into a new struct.
This makes it easier to see all the new TLV fields at a glance. We also
convert TLV usage to use the new type param based APis.
This lets us get rid of the mutex usage there. We also shift the algo slightly to increment by 1, then use that as the next value, which plays nicer with the atomics.
In this commit, we rename the files as assembler.go houses the primary
interfaces/abstractions of the package. In the rest of the codebase,
this file is near uniformly called interface.go, so we rename the file
to make the repo more digestible at a scan.
In this commit, we fix a bug that would cause a global message router to
be stopped anytime a peer disconnected. The global msg router only
allows `Start` to be called once, so afterwards, no messages would
properly be routed.
Olaoluwa Osuntokun