When an `std::future::Future` is `poll()`ed, we're only supposed to
use the latest `Waker` provided. However, we currently push an
`StdWaker` onto our callback list every time `poll` is called,
waking every `Waker` but also using more and more memory until the
`Future` itself is woken.
Here we take a step towards fixing this by giving each `Future` a
unique index and storing which `Future` an `StdWaker` came from in
the callback list. This sets us up to deduplicate `StdWaker`s by
`Future`s in the next commit.
In the next commit we'll fix a memory leak due to keeping too many
`std::task::Waker` callbacks in `FutureState` from redundant `poll`
calls, but first we need to split handling of `StdWaker`-based
future wake callbacks from normal ones, which we do here.
As part of the ongoing async signer work, our holder signatures must
also be capable of being obtained asynchronously. We expose a new
`ChannelMonitor::signer_unblocked` method to retry pending onchain
claims by re-signing and rebroadcasting transactions. Unfortunately, we
cannot retry said claims without them being registered first, so if
we're not able to obtain the signature synchronously, we must return the
transaction as unsigned and ensure it is not broadcast.
This method is meant to be used as a last resort when a user is forced
to broadcast the current state, even if it is stale, in an attempt to
claim their funds in the channel. Previously, we'd return the commitment
and HTLC transactions such that they broadcast them themselves. Doing so
required a different code path, one which was not tested, to obtain
these transactions than our usual path when force closing. It's not
worth maintaining both, and it's much simpler for us to broadcast
instead.
A client node might choose not to handle `Event::BumptTransaction`
events and leave bumping / Anchor output spending to a trusted
counterparty.
However, `Event::BumptTransaction` currently doesn't offer any clear
indication what channel and/or counterparty it is referring to. In order
to allow filtering these events, we here expose the `channel_id` and
`counterparty_node_id` fields.
This exposes details around pending HTLCs in ChannelDetails. The state
of the HTLC in the state machine is also included, so it can be
determined which protocol message the HTLC is waiting for to advance.
While this isn't expected to materially improve performance, it
does get us ahash 0.8, which allows us to reduce fuzzing
randomness, making our fuzzers much happier.
Sadly, by default `ahash` no longer tries to autodetect a
randomness source, so we cannot simply rely on `hashbrown` to do
randomization for us, but rather have to also explicitly depend on
`ahash`.
In the next commit we'll bump the `hashbrown` version, which no
longer randomizes its hasher by default. Thus, we'll need to call
a different constructor in no-std builds from std builds.
Here we do a quick prefactor to use wrappers for `FromIterator`
constructors instead of calling the tables directly to make the
version bump changeset smaller.
In the next commit we'll bump the `hashbrown` version, which no
longer randomizes its hasher by default. Thus, we'll need to call
a different constructor in no-std builds from std builds.
Here we do a quick prefactor to use wrappers for constructors
instead of calling the tables directly to make the version bump
changeset smaller.
As we've added more and more configuration parameters which change
our behavior, we're increasingly missing coverage in the general
`full_stack_target` fuzzer. Sadly, a reachable `unwrap` slipped in
uncaught that should have been stopped by the `full_stack_target`.
Here, we update the `full_stack_target` fuzzer to read a full
config object to substantially increase coverage.
We use these params for our HubPreferentialScorer and without these
utility functions then we have to manually do this and calculate which
is the source and which is the target node.
We generally allow routing nodes to forward less than the expected HTLC
amount, if the receiver knowingly accepts this and claims the
underpaying HTLC (see `ChannelConfig::accept_underpaying_htlcs`). This
use case is in particular useful for the LSPS2/JIT channel setting where
the intial underpaying HTLC pays for the channel open.
While we previously exposed the withheld amount as
`PaymentClaimable::counterparty_skimmed_fee_msat` on the receiver side,
we did not individually provide it on the forwarding node's side.
Here, we therefore expose this additionally withheld amount via
`PaymentForwarded::skimmed_fee_msat`.
- The first test make sure that the OutboundV1Channel is not
immediately removed when peers disconnect, but is removed after N timer
ticks.
- The second test makes sure that the SendOpenChannel is rebroadcasted
for the OutboundV1Channel if peer reconnects within time.
- Do not remove channel immediately when peer_disconnect, instead
removed it after some time if peer doesn't reconnect soon (handled in
previous commit).
- Do not mark per ok_to_remove if we have some OutboundV1Channels too.
- Rebroadcast SendOpenChannel for outboundV1Channel when peer
reconnects.
- Update the relevant tests to account for the behavior change.
- Repurpose the test_disconnect_in_funding_batch to test that all
channels in the batch close when one them closes.
