-
Notifications
You must be signed in to change notification settings - Fork 2.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
lnwire+netann: update structure of g175 messages to be pure TLV #9175
base: master
Are you sure you want to change the base?
lnwire+netann: update structure of g175 messages to be pure TLV #9175
Conversation
ee6ff8c
to
4160fd5
Compare
4160fd5
to
5d3c965
Compare
Ok i've also now updated this PR to include validating the signature of a |
PureTLVMessage describes an LN message that is a pure TLV stream. If the message includes a signature, it will sign all the TLV records in the inclusive ranges: 0 to 159 and 1000000000 to 2999999999. A comprehensive test is added that shows how two versions of the same message remain forward compatible.
So that we can use it as a TLV record type.
Such that all fields are now TLV (including the signature).
such that all fields are now TLV including the signature.
This commit expands the ChannelAnnouncement2 validation for the case where it is announcing a P2WSH channel.
5d3c965
to
6f0ce3b
Compare
@@ -210,3 +210,5 @@ replace google.golang.org/protobuf => github.com/lightninglabs/protobuf-go-hex-d | |||
go 1.22.6 | |||
|
|||
retract v0.0.2 | |||
|
|||
replace github.com/lightningnetwork/lnd/tlv => github.com/ellemouton/lnd/tlv v0.0.0-20241012094556-298ff9eaed58 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think we can remove this now?
) | ||
|
||
const ( | ||
pureTLVUnsignedRangeOneStart = 160 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should have godoc
comments.
const ( | ||
pureTLVUnsignedRangeOneStart = 160 | ||
pureTLVSignedSecondRangeStart = 1000000000 | ||
pureTLVUnsignedRangeTwoStart = 3000000000 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
How did we arrive at these values?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Also I think we should try to capture this in the spec, independent of G 1.75.
|
||
// SerialiseFieldsToSign serialises all the records from the given | ||
// PureTLVMessage that fall within the signed TLV range. | ||
func SerialiseFieldsToSign(msg PureTLVMessage) ([]byte, error) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We might want to adopt a similar structure for the signed portion of the dyn commit messages @ProofOfKeags. Basically we copied over from the way BOLT 12 works, and we can use the TLV type values, then let this handle how serialization works (does require a message to be pure TLV).
// Any extra fields in the signed range that we do not yet know about, | ||
// but we need to keep them for signature validation and to produce a | ||
// valid message. | ||
ExtraFieldsInSignedRange map[uint64][]byte |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should something like this be split out into an embeddable struct?
// Any extra fields in the signed range that we do not yet know about, | ||
// but we need to keep them for signature validation and to produce a | ||
// valid message. | ||
ExtraFieldsInSignedRange map[uint64][]byte |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Also why can't we just continue to store it in the serialized form?
// AuthenticatedGossiper. Any announcements that don't match this | ||
// chain hash will be ignored. This is an internal config value obtained | ||
// from ChainParams. | ||
chainHash *chainhash.Hash |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why do we need it here if we have ChainParams
?
ChainHash chainhash.Hash | ||
// ChainParams holds the chain parameters for the active network this | ||
// node is participating on. | ||
ChainParams *chaincfg.Params |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think this can be a normal value rather than a pointer.
[]byte, error) { | ||
txscript.ScriptClass, btcutil.Address, error) { | ||
|
||
pkScript, err := lnwallet.FetchPKScriptWithQuit( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What about the merkle proof case?
@@ -108,7 +110,8 @@ func CreateChanAnnouncement(chanProof *models.ChannelAuthProof, | |||
|
|||
// FetchPkScript defines a function that can be used to fetch the output script | |||
// for the transaction with the given SCID. | |||
type FetchPkScript func(*lnwire.ShortChannelID) ([]byte, error) | |||
type FetchPkScript func(*lnwire.ShortChannelID) (txscript.ScriptClass, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Does the arg here need to be a pointer?
@ellemouton, remember to re-request review from reviewers when ready |
Update the structure of all gossip 1.75 messages to be purely TLV.
With this update, the signature is also a TLV record. There is a un-signed
and signed range of TLVs. The signature along with things like SPV proofs
will fall in the un-signed range.
I've used the same un-signed range used by Bolt12 messages.
This PR introduces a new
PureTLVMessage
interface for messages like thisand adds a comprehensive test to ensure that old clients (who are aware of
fewer fields of a message than new clients) and new clients still generate the
same correct digest of a message regardless of if they are aware of all the
records or not.
Bonus Commit
Ok i've also now updated this PR to include validating the signature of a
ChannelAnnouncement2 message for the case where it is announcing a P2WSH channel.