VecReadWrapper is only used in TLVs so there is no need to prepend
a length before writing/reading the objects - we can instead simply
read until we reach the end of the TLV stream.
This substantially improves deserialization performance when LLVM
decides not to inline many short methods, eg when not building
with LTO/codegen-units=1.
Even with the default bench params of LTO/codegen-units=1, the
serialization benchmarks on an Intel 2687W v3 take:
test routing::network_graph::benches::read_network_graph ... bench: 1,955,616,225 ns/iter (+/- 4,135,777)
test routing::network_graph::benches::write_network_graph ... bench: 165,905,275 ns/iter (+/- 118,798)
With the new `serialized_length()` method potentially being
significantly more efficient than `LengthCalculatingWriter`, this
commit ensures we call `serialized_length()` when calculating
length of a larger struct.
Specifically, prior to this commit a call to
`serialized_length()` on a large object serialized with
`impl_writeable`, `impl_writeable_len_match`, or
`encode_varint_length_prefixed_tlv` (and
`impl_writeable_tlv_based`) would always serialize all inner fields
of that object using `LengthCalculatingWriter`. This would ignore
any `serialized_length()` overrides by inner fields. Instead, we
override `serialized_length()` on all of the above by calculating
the serialized size using calls to `serialized_length()` on inner
fields.
Further, writes to `LengthCalculatingWriter` should never fail as
its `write` method never returns an error. Thus, any write failures
indicate a bug in an object's write method or in our
object-creation sanity checking. We `.expect()` such write calls
here.
As of this commit, on an Intel 2687W v3, the serialization
benchmarks take:
test routing::network_graph::benches::read_network_graph ... bench: 2,039,451,296 ns/iter (+/- 4,329,821)
test routing::network_graph::benches::write_network_graph ... bench: 166,685,412 ns/iter (+/- 352,537)
When writing out libsecp256k1 objects during serialization in a
TLV, we potentially calculate the TLV length twice before
performing the actual serialization (once when calculating the
total TLV-stream length and once when calculating the length of the
secp256k1-object-containing TLV). Because the lengths of secp256k1
objects is a constant, we'd ideally like LLVM to entirely optimize
out those calls and simply know the expected length. However,
without cross-language LTO, there is no way for LLVM to verify that
there are no side-effects of the calls to libsecp256k1, leaving
LLVM with no way to optimize them out.
This commit adds a new method to `Writeable` which returns the
length of an object once serialized. It is implemented by default
using `LengthCalculatingWriter` (which LLVM generally optimizes out
for Rust objects) and overrides it for libsecp256k1 objects.
As of this commit, on an Intel 2687W v3, the serialization
benchmarks take:
test routing::network_graph::benches::read_network_graph ... bench: 2,035,402,164 ns/iter (+/- 1,855,357)
test routing::network_graph::benches::write_network_graph ... bench: 308,235,267 ns/iter (+/- 140,202)
Now that our MSRV supports the native methods, we have no need
for the helpers anymore. Because LLVM was already matching our
byte_utils methods as byteswap functions, this should have no
impact on generated (optimzied) code.
This removes most of the byte_utils usage, though some remains to
keep the patch size reasonable.
This makes a small difference for NetworkGraph deserialization
as it enables more inlining across different files, hopefully
better matching user performance as well.
As of this commit, on an Intel 2687W v3, the serialization
benchmarks take:
test routing::network_graph::benches::read_network_graph ... bench: 2,037,875,071 ns/iter (+/- 760,370)
test routing::network_graph::benches::write_network_graph ... bench: 320,561,557 ns/iter (+/- 176,343)
NetworkGraph is one of the largest structures we generally
deserialize, so it makes for a good benchmark, even if it isn't the
most complicated one.
As of this commit, on an Intel 2687W v3, these benchmarks take:
test routing::network_graph::benches::read_network_graph ... bench: 2,101,420,078 ns/iter (+/- 6,649,020)
test routing::network_graph::benches::write_network_graph ... bench: 344,696,835 ns/iter (+/- 229,061)
Channel serialization should happen "as if
remove_uncommitted_htlcs_and_mark_paused had just been called".
This is true for the most part, but outbound RemoteRemoved HTLCs
were being serialized as normal, even though
`remote_uncommitted_htlcs_and_mark_paused` resets them to
`Committed`.
This led to a bug identified by the `chanmon_consistency_target`
fuzzer wherein, if we receive a update_*_htlc message bug not the
corresponding commitment_signed prior to a serialization roundtrip,
we'd force-close the channel due to the peer "attempting to
fail/claim an HTLC which was already failed/claimed".
While trying to debug the issue ultimately tracked down to a
`PeerHandler` locking bug in #891, the ability to deliver only
individual messages at a time in chanmon_consistency looked
important. Specifically, it initially appeared there may be a race
when an update_add_htlc was delivered, then a node sent a payment,
and only after that, the corresponding commitment-signed was
delivered.
This commit adds such an ability, greatly expanding the potential
for chanmon_consistency to identify channel state machine bugs.
This makes it so that users cannot usefully implement their own
`EventsProvider`, which would require substantial new logic in the
bindings generator (for generic methods). In the case of
`EventsProvider`, because there are no Rust methods which accept an
`EventsProvider` as an argument, this is perfectly OK as the
generated code would be entirely unused anyway.
This stores transaction templates temporarily until their locktime
is reached, avoiding broadcasting (or RBF bumping) transactions
prior to their locktime. For those broadcasting transactions
(potentially indirectly) via Bitcoin Core RPC, this ensures no
automated rebroadcast of transactions on the client side is
required to get transactions confirmed.
This somewhat cleans up the public API of PackageSolvingData to
make it harder to get an invalid amount and use it, adding further
debug assertion to check it at test-time.
