* Standardize the DLEq serialization function naming
They mismatched from the rest of the project.
This commit is technically incomplete as it doesn't update the dkg crate.
* Rewrite DKG encryption to enable per-message decryption without side effects
This isn't technically true as I already know a break in this which I'll
correct for shortly.
Does update documentation to explain the new scheme. Required for blame.
* Add a verifiable system for blame during the FROST DKG
Previously, if sent an invalid key share, the participant would realize that
and could accuse the sender. Without further evidence, either the accuser
or the accused could be guilty. Now, the accuser has a proof the accused is
in the wrong.
Reworks KeyMachine to return BlameMachine. This explicitly acknowledges how
locally complete keys still need group acknowledgement before the protocol
can be complete and provides a way for others to verify blame, even after a
locally successful run.
If any blame is cast, the protocol is no longer considered complete-able
(instead aborting). Further accusations of blame can still be handled however.
Updates documentation on network behavior.
Also starts to remove "OnDrop". We now use Zeroizing for anything which should
be zeroized on drop. This is a lot more piece-meal and reduces clones.
* Tweak Zeroizing and Debug impls
Expands Zeroizing to be more comprehensive.
Also updates Zeroizing<CachedPreprocess([u8; 32])> to
CachedPreprocess(Zeroizing<[u8; 32]>) so zeroizing is the first thing done
and last step before exposing the copy-able [u8; 32].
Removes private keys from Debug.
* Fix a bug where adversaries could claim to be using another user's encryption keys to learn their messages
Mentioned a few commits ago, now fixed.
This wouldn't have affected Serai, which aborts on failure, nor any DKG
currently supported. It's just about ensuring the DKG encryption is robust and
proper.
* Finish moving dleq from ser/deser to write/read
* Add tests for dkg blame
* Add a FROST test for invalid signature shares
* Batch verify encrypted messages' ephemeral keys' PoP
While the previous construction achieved n/2 average detection,
this will run in log2(n). Unfortunately, the need to keep entropy
around (or take in an RNG here) remains.
Technically, non-0-amount outputs can still appear and this considered them
as part of the global 0-amount pool. Now, only outputs which are 0-amount are
counted.
* Remove the explicit included participants from FROST
Now, whoever submits preprocesses becomes the signing set. Better separates
preprocess from sign, at the cost of slightly more annoying integrations
(Monero needs to now independently lagrange/offset its key images).
* Support caching preprocesses
Closes https://github.com/serai-dex/serai/issues/40.
I *could* have added a serialization trait to Algorithm and written a ton of
data to disk, while requiring Algorithm implementors also accept such work.
Instead, I moved preprocess to a seeded RNG (Chacha20) which should be as
secure as the regular RNG. Rebuilding from cache simply loads the previously
used Chacha seed, making the Algorithm oblivious to the fact it's being
rebuilt from a cache. This removes any requirements for it to be modified
while guaranteeing equivalency.
This builds on the last commit which delayed determining the signing set till
post-preprocess acquisition. Unfortunately, that commit did force preprocess
from ThresholdView to ThresholdKeys which had visible effects on Monero.
Serai will actually need delayed set determination for #163, and overall,
it remains better, hence it's inclusion.
* Document FROST preprocess caching
* Update ethereum to new FROST
* Fix bug in Monero offset calculation and update processor
A type alias of MoneroAddress is provided to abstract away the generic.
To keep the rest of the library sane, MoneroAddress is used everywhere.
If someone wants to use this library with another coin, they *should* be
able to parse a custom address and then recreate it as a Monero address.
While that's annoying to them, better them than any person using this
lib for Monero.
Closes#152.
* Add dkg crate
* Remove F_len and G_len
They're generally no longer used.
* Replace hash_to_vec with a provided method around associated type H: Digest
Part of trying to minimize this trait so it can be moved elsewhere. Vec,
which isn't std, may have been a blocker.
* Encrypt secret shares within the FROST library
Reduces requirements on callers in order to be correct.
* Update usage of Zeroize within FROST
* Inline functions in key_gen
There was no reason to have them separated as they were. sign probably
has the same statement available, yet that isn't the focus right now.
* Add a ciphersuite package which provides hash_to_F
* Set the Ciphersuite version to something valid
* Have ed448 export Scalar/FieldElement/Point at the top level
* Move FROST over to Ciphersuite
* Correct usage of ff in ciphersuite
* Correct documentation handling
* Move Schnorr signatures to their own crate
* Remove unused feature from schnorr
* Fix Schnorr tests
* Split DKG into a separate crate
* Add serialize to Commitments and SecretShare
Helper for buf = vec![]; .write(buf).unwrap(); buf
* Move FROST over to the new dkg crate
* Update Monero lib to latest FROST
* Correct ethereum's usage of features
* Add serialize to GeneratorProof
* Add serialize helper function to FROST
* Rename AddendumSerialize to WriteAddendum
* Update processor
* Slight fix to processor
* Create message types for FROST key gen
Taking in reader borrows absolutely wasn't feasible. Now, proper types
which can be read (and then passed directly, without a mutable borrow)
exist for key_gen. sign coming next.
* Move FROST signing to messages, not Readers/Writers/Vec<u8>
Also takes the nonce handling code and makes a dedicated file for it,
aiming to resolve complex types and make the code more legible by
replacing its previously inlined state.
* clippy
* Update FROST tests
* read_signature_share
* Update the Monero library to the new FROST packages
* Update processor to latest FROST
* Tweaks to terminology and documentation
Unbeknowst to me, height doesn't have a universal definition of the
chain length.
Bitcoin defines height as the block number, with getblockcount existing
for the chain length.
Ethereum uses the unambiguous term "block number".
Monero defines height as both the block number and the chain length.
Instead of arguing about who's right, it's agreed it referring to both
isn't productive. While we could provide our own definition, taking a
side, moving to the unambiguous block number prevents future hiccups.
height is now only a term in the Monero code, where it takes its
Monero-specific definition, as documented in the processor.
* Label the version as an alpha
* Add versions to Cargo.tomls
* Update to Zeroize 1.5
* Drop patch versions from monero-serai Cargo.toml
* Add a repository field
* Move generators to OUT_DIR
IIRC, I didn't do this originally as it constantly re-generated them.
Unfortunately, since cargo is complaining about .generators, we have to.
* Remove Timelock::fee_weight
Transaction::fee_weight's has a comment, "Assumes Timelock::None since
this library won't let you create a TX with a timelock". Accordingly,
this is dead code.
Despite being slower and only used for blinding values, its still
extremely performant. 20 is far more standard and will avoid an eye
raise from reviewers.
There is a slight note we only implement u64 varint's, while Monero does
it for arbitrary uints, yet that's not relevant at this time. It is
documented in #25.
Creates a new monero-generators crate so the monero crate can run the
code in question at build time.
Saves several seconds from running the tests.
Closes https://github.com/serai-dex/serai/issues/101.
* Apply Zeroize to nonces used in Bulletproofs
Also makes bit decomposition constant time for a given amount of
outputs.
* Fix nonce reuse for single-signer CLSAG
* Attach Zeroize to most structures in Monero, and ZOnDrop to anything with private data
* Zeroize private keys and nonces
* Merge prepare_outputs and prepare_transactions
* Ensure CLSAG is constant time
* Pass by borrow where needed, bug fixes
The past few commitments have been one in-progress chunk which I've
broken up as best read.
* Add Zeroize to FROST structs
Still needs to zeroize internally, yet next step. Not quite as
aggressive as Monero, partially due to the limitations of HashMaps,
partially due to less concern about metadata, yet does still delete a
few smaller items of metadata (group key, context string...).
* Remove Zeroize from most Monero multisig structs
These structs largely didn't have private data, just fields with private
data, yet those fields implemented ZeroizeOnDrop making them already
covered. While there is still traces of the transaction left in RAM,
fully purging that was never the intent.
* Use Zeroize within dleq
bitvec doesn't offer Zeroize, so a manual zeroing has been implemented.
* Use Zeroize for random_nonce
It isn't perfect, due to the inability to zeroize the digest, and due to
kp256 requiring a few transformations. It does the best it can though.
Does move the per-curve random_nonce to a provided one, which is allowed
as of https://github.com/cfrg/draft-irtf-cfrg-frost/pull/231.
* Use Zeroize on FROST keygen/signing
* Zeroize constant time multiexp.
* Correct when FROST keygen zeroizes
* Move the FROST keys Arc into FrostKeys
Reduces amount of instances in memory.
* Manually implement Debug for FrostCore to not leak the secret share
* Misc bug fixes
* clippy + multiexp test bug fixes
* Correct FROST key gen share summation
It leaked our own share for ourself.
* Fix cross-group DLEq tests
Considering they take 7 seconds to generate, thanks to #68, the ability
to generate them at the start instead of on first BP is greatly
appreciated.
Also performs minor cleanups regarding BPs.
* Use a struct in an enum for Bulletproofs
* verification bp working for just one proof
* add some more assert tests
* Clean BP verification
* Implement batch verification
* Add a debug assertion w_cache isn't 0
It's initially set to 0 and if not updated, this would be broken.
* Correct Monero workflow yaml
* Again try to corrent Monero workflow yaml
* Again
* Finally
* Re-apply weights as required by Bulletproofs
Removing these was insecure and my fault.
Co-authored-by: DangerousFreedom <dangfreed@tutanota.com>
* Consolidate GitHub CI actions, split out Monero
build now includes the specified Rust toolchain/components.
Added a test dependencies action which grabs Foundry and Monero.
Split the Monero v14 job into a matrixed job in its own workflow flow.
It's now only run when Monero has changes.
* Correct Monero unit/integration tests run timing
Additionally tests a feature-less Monero build.
Also removes a pointless Monero file, which already should have been
removed, causing this workflow to be triggered.
* Correct exclusion and paths
Updates to FROST should re-run the Monero tests to ensure it didn't
introduce API incompatibilities.
* Initial stab at Bulletproofs+
Does move around the existing Bulletproofs code, does still work as
expected.
* Make the Clsag RCTPrunable type work with BP and BP+
* Initial set of BP+ bug fixes
* Further bug fixes
* Remove RING_LEN as a constant
* Monero v16 TX support
Doesn't implement view tags, nor going back to v14, nor the updated BP
clawback logic.
* Support v14 and v16 at the same time
Introduces missing CLSAG checks. The only difference now should be the
additional rejection of torsioned points, which is relevant to
https://github.com/serai-dex/serai/issues/25. Considering this is only
currently used for FROST verification, this should be fine.
Closes https://github.com/serai-dex/serai/issues/19 by making it
irrelevant.
Increases priority of https://github.com/serai-dex/serai/issues/68, as
now it's used for the BP generators which are done at first-proof.
Also merges BP's stricter hash_to_point with the library's, since CLSAG
has the same bound.
* Initial attempt at Bulletproofs
I don't know why this doesn't work. The generators and hash_cache lines
up without issue. AFAICT, the inner product proof is valid as well, as
are all included formulas.
* Add yinvpow asserts
* Clean code
* Correct bad imports
* Fix the definition of TWO_N
Bulletproofs work now :D
* Tidy up a bit
* fmt + clippy
* Compile a variety of XMR dependencies with optimizations, even under dev
The Rust bulletproof implementation is 8% slower than C right now, under
release. This is acceptable, even if suboptimal. Under debug, they take
a quarter of a second to two seconds though, depending on the amount of
outputs, which justifies this move.
* Remove unnecessary deref in BPs
* Remove the Monero CMake and make
* Download the Monero daemon instead of building it
* Cache the Monero daemon
Prevents hammering the Monero servers, should reduce CI time.
* Correct YAML
* Add back sodium-dev
* Create an independent job for downloading the Monero daemon
Improves parallelism while decreasing the amount of work re-done if
build fails. Also increases modularity.
* Correct Monero job definition
* Correct skipping the Monero download on cache hit
* begin to setup ci
* attempt to fix build
* fix paths in build script
* fix
* satisfy clippy
* update fmt check to use nightly
* use nightly for build
* fmt
* fix fmt install
* update test script
* try to fix fmt
* merge w develop
* maybe fix build script
* install wasm toolchain
* install solc-select, use stable rust to build
* Correct clippy warnings
Currently intended to be done with:
cargo clippy --features "recommended merlin batch serialize experimental
ed25519 ristretto p256 secp256k1 multisig" -- -A clippy::type_complexity
-A dead_code
* Remove try-runtime
I tried to get this to work for an hour. I have no idea why it doesn't,
yet it doesn't.
* Rewrite workflow
Splits tasks into a more modular structure. Also uses
actions-rs/toolchain.
* Add a cache
* Immediately try building ETH/Monero while this is fixed
Adds solc-select use.
* Revert selective advance building of ETH/XMR
ETH builds now, so it hopefully should work now.
Also moves from on push to on push to develop.
* Install Monero runtime dependencies
Specify missing Rust toolchain setting.
* Correct multi-line commands
* Fix multi-line commands again
Cache Ethereum artifacts.
* Add Foundry
* Move Clippy under build
* Minimal rustup
Adds wasm Clippy. Puts Clippy before build.
* Use nightly clippy
* Remove old clippy call from under build
* Have the Monero build script support ARCH specification
Requirement for CI.
* Add WASM toolchain to tests
* Remove Ethereum cache which did not work as needed
* Remove extraneous quotes which broke builds on Arch
Co-authored-by: Luke Parker <lukeparker5132@gmail.com>
Currently intended to be done with:
cargo clippy --features "recommended merlin batch serialize experimental
ed25519 ristretto p256 secp256k1 multisig" -- -A clippy::type_complexity
-A dead_code
The two-generator limit wasn't required nor beneficial. This does
theoretically optimize FROST, yet not for any current constructions. A
follow up proof which would optimize current constructions has been
noted in #38.
Adds explicit no_std support to the core DLEq proof.
Closes#34.
While Serai only needs the simple DLEq which was already present under
monero, this migrates the implementation of the cross-group DLEq I
maintain into Serai. This was to have full access to the ecosystem of
libraries built under Serai while also ensuring support for it.
The cross_group curve, which is extremely experimental, is feature
flagged off. So is the built in serialization functionality, as this
should be possible to make nostd once const generics are full featured,
yet the implemented serialization adds the additional barrier of
std::io.
Increases usage of standardization while expanding dalek_ff_group.
Closes https://github.com/serai-dex/serai/issues/26 by moving
dfg::EdwardsPoint to only be for the prime subgroup.
Updates how change outputs are handled, with a far more logical
construction offering greater flexibility.
prepare_outputs can not longer error. SignaableTransaction::new will.
Modifies FROST behavior so group_key has the offset applied regardless
of if view was called. The unaltered secret_share and
verification_shares (as they have differing values depending on the
signing set) are no longer publicly accessible.
Doesn't fully utilize ec's hash2curve module as k256 Scalar doesn't have
FromOkm for some reason. The previously present bigint reduction is
preserved.
Updates ff/group to 0.12.
Premised on https://github.com/cfrg/draft-irtf-cfrg-frost/pull/205 being
merged, as while this Ed25519 is vector compliant, it's technically not
spec compliant due to that conflict.
While it was fine as-is, as it only had one variable length property,
this is a bit more robust. Also binds the Curve ID, which should declare
differently even for just different basepoints, and therefore adds two
variable length properties (justifying the transcript).
No functional changes have been made to signing, with solely slight API
changes being made.
Technically not actually FROST v5 compatible, due to differing on zero
checks and randomness, yet the vectors do confirm the core algorithm.
For any valid FROST implementation, this will be interoperable if they
can successfully communicate. For any devious FROST implementation, this
will be fingerprintable, yet should still be valid.
Relevant to https://github.com/serai-dex/serai/issues/9 as any curve can
now specify vectors for itself and be tested against them.
Moves the FROST testing curve from k256 to p256. Does not expose p256
despite being compliant. It's not at a point I'm happy with it, notably
regarding hash to curve, and I'm not sure I care to support p256. If it
has value to the larger FROST ecosystem...
It was never used as we derive entropy via the other fields in the
transcript, and explicitly add fields directly as needed for entropy.
Also drops an unused crate and corrects a bug in FROST's Schnorr
implementation which used the Group's generator, instead of the Curve's.
Also updates the Monero crate's description.
Also updates Bulletproofs from C to not be length prefixed, yet rather
have Rust calculate their length.
Corrects an error in key_gen where self was blamed, instead of the
faulty participant.
Saves roughly 0.8s when running the tests, which took 16.6s and now take
15.8 (5%).
Removes the larger sample size, which replaced the closest selected
decoy with the real spend, per advice of Rucknium.
Saves ~8% during FROST key gen, even with dropping a vartime for a
constant time (as needed to be secure), as the new batch verifier is
used where batch verification previously wasn't. The new multiexp API
itself also offered a very slight performance boost, which may solely be
a measurement error.
Handles most of https://github.com/serai-dex/serai/issues/10. The blame
function isn't binary searched nor randomly sorted yet.
Changes the output index to a u8. While it may expand to a u16 at some
point, this can remain canonical using little endian serialization while
dropping the latter byte if it's 0 (or simply only using u16 when it's
actually possible).
Honestly, the borrowed keys are frustrating, and this probably reduces
performance while no longer offering an order when iterating. That said,
they enable full u16 indexing and should mildly improve the API.
Cleans the Proof of Knowledge handling present in key gen.
Currently solely used for single signer change outputs, intended to be
used for funds into Serai and multisig change outputs (dependent on #2).
Also cleans the file layout, makes scanning a bit more robust, doesn't
return outputs of amount 0, and shuffles outputs.
Remove's CLSAG's msg Rc for the msg available through AlgorithmMachine.
Potentially slightly more inefficient, as it needs to be converted from
a slice to a [u8; 32], yet removes a re-impl.
Also removes a match for an if.
Updates decoy selection with an explicit panic, the removal of a divide
by 0 (causing tests to fail on new chains), and a minor optimization
when dealing with a large quantity of locked outputs.
Also increases documentation, acknowledging infinite loops and breakage
from Monero more.
While all the transcript/extension code works as expected, which means,
they don't cause any conflicts, n was still capped at u64::MAX at
creation when it needs to be u16. Furthermore, participant index and
scalars/points were little endian instead of big endian/curve dependent.
