Outbound queue v2 #4

yrong · 2024-10-18T01:35:21Z

Resolves: https://linear.app/snowfork/issue/SNO-1205

Remove Channel & Pricing params & Estimation logic on chain
Global nonce
Locked Fee
abi encode new message format for V2(respect Contracts V2 Snowfork/snowbridge#1300)
submit delivery proof
Seperate message V1 & V2 with different routes/exporters
Reward flow(design in https://linear.app/snowfork/project/polkadot-sdk-unified-rewards-718300ef867d/overview)
XCMV5 integration(require a dev branch from Parity with XCMV5 support)

yrong · 2024-10-21T05:36:08Z

Workflow

on AH we add a custom exporter returns a lower fee(without the cost on Ethereum) charged from the user when exporting to Ethereum, essentially predict the route by content of XCM, more details in Predicate route table by Xcm paritytech/polkadot-sdk#6074, since there is no XCMV5 branch ready for tests yet, as a workaround we can execute raw xcm with some custom instruction on AH to differ V2 with V1.
on BH we route the new XCM(with V2 specific instructions) to the new pallet OutboundQueueV2, in which we convert the XCM to Message structure as following:

pub struct Message {
    /// Origin
    pub origin: H256,
    /// ID
    pub id: H256,
    /// Fee
    pub fee: u128,
    /// Commands
    pub commands: BoundedVec<Command, ConstU32<5>>,
}

which means we can execute multiple commands on Ethereum in one message.

Then we abi encode the message and store it into a merkle tree as before, meanwhile add the pending fee into LockedFee storage and increase the nonce. more detais here
add a new extrinsic submit_delivery_proof in which we verify the event log is truely happened on Ethereum side, and add the fee to RewardLeger.

vgeddes · 2024-10-22T17:44:51Z

bridges/snowbridge/primitives/core/src/merkle.rs

Hey I'd prefer to move the crate to primitives/merkle-tree rather than adding more code to core. In fact I think the core crate has too much code and potentially needs be split up further.

vgeddes · 2024-10-22T17:51:38Z

bridges/snowbridge/primitives/core/src/outbound_v2.rs

We shouldn't just add V2 code and leave v1 code as is. The two versions need to co-exist in a coherent and neatly ordered manner.

I think we should move to this layout:

* core/src/outbound/v1/mod.rs * core/src/outbound/v2/mod.rs

vgeddes · 2024-10-22T18:08:15Z

bridges/snowbridge/primitives/core/src/lib.rs

@@ -172,3 +174,13 @@ impl Default for AssetMetadata {

 /// Maximum length of a string field in ERC20 token metada
 const METADATA_FIELD_MAX_LEN: u32 = 32;
+
+// Origin for high-priority governance commands
+pub fn primary_governance_origin() -> H256 {


This is now a V1-only type, so should go in core/src/v1/mod.rs.

850fe96
Actually it's V2 specific.

The usage in

polkadot-sdk/bridges/snowbridge/pallets/outbound-queue-v2/src/send_message_impl.rs

Lines 45 to 47 in ba03ae6

if ticket.origin != primary_governance_origin() {

ensure!(!Self::operating_mode().is_halted(), SendError::Halted);

}

is to allow upgrading even when bridge is halted.

vgeddes · 2024-10-22T18:08:26Z

bridges/snowbridge/primitives/core/src/lib.rs

+}
+
+// Origin for lower-priority governance commands
+pub fn second_governance_origin() -> H256 {


#4 (comment)

...rachains/integration-tests/emulated/tests/bridges/bridge-hub-westend/src/tests/snowbridge.rs

vgeddes · 2024-10-22T18:48:57Z

bridges/snowbridge/pallets/outbound-queue-v2/src/types.rs

 	pub id: H256,
+	/// Commands to execute in Ethereum
+	pub commands: BoundedVec<CommandWrapper, ConstU32<5>>,


Should probably limit to 8 (I like constants being a factor of 2). And we should probably use a constant.

Suggested change

pub commands: BoundedVec<CommandWrapper, ConstU32<5>>,

pub commands: BoundedVec<CommandWrapper, ConstU32<MAX_COMMANDS>>,

vgeddes · 2024-10-22T18:54:05Z

bridges/snowbridge/pallets/outbound-queue-v2/src/types.rs

-		Token::Tuple(vec![
-			Token::FixedBytes(Vec::from(x.channel_id.as_ref())),
+		let header = vec![
+			Token::FixedBytes(x.origin.as_bytes().to_owned()),


Can we use the the sol! macro to auto-generate encoders for InboundMessage? Its from this library which is already used by outbound queue: https://alloy.rs/

Token::FixedBytes is from the older ethabi-decode crate which we should stop using and migrate to alloy crates.

vgeddes · 2024-10-22T18:54:42Z

bridges/snowbridge/primitives/core/src/outbound_v2.rs

+			mode: OperatingMode,
+		},
+		/// Transfer ether from an agent contract to a recipient account
+		TransferNativeFromAgent {


This command doesn't exist anymore, I believe

vgeddes · 2024-10-22T18:55:47Z

bridges/snowbridge/primitives/core/src/outbound_v2.rs

+		}
+
+		/// ABI-encode the Command.
+		pub fn abi_encode(&self) -> Vec<u8> {


Can we use the sol! macro from the alloy-rs crates to generate the encoders automatically? We should deprecate use of ethabi::encode, as its an unmaintained library.

See example from https://github.com/paritytech/polkadot-sdk/blob/aeebf2f383390f2f86527d70212162d5dbea8b93/bridges/snowbridge/pallets/inbound-queue/src/envelope.rs#L11

vgeddes · 2024-10-22T19:06:16Z

bridges/snowbridge/primitives/core/src/outbound_v2.rs

+	// The base transaction cost, which includes:
+	// 21_000 transaction cost, roughly worst case 64_000 for calldata, and 100_000
+	// for message verification
+	const MAXIMUM_BASE_GAS: u64 = 185_000;


This won't be needed in V2.

yrong · 2024-10-29T12:38:51Z

Check details in emulated tests how to construct transfer message for both ENA&PNA with XCMV5 instructions.

cargo test -p bridge-hub-westend-integration-tests --lib tests::snowbridge_v2  -- --nocapture

vgeddes · 2024-10-30T21:26:53Z

bridges/snowbridge/pallets/outbound-queue-v2/src/send_message_impl.rs

+			BoundedVec::try_from(ticket.encode()).map_err(|_| SendError::MessageTooLarge)?;
+
+		T::MessageQueue::enqueue_message(message.as_bounded_slice(), origin);
+		Self::deposit_event(Event::MessageQueued { message: ticket.clone() });


This isn't sufficient for dry-running, because the message at this stage in the processing pipeline isn't the same that is forwarded to the Ethereum gateway. Its missing information such as the gas limit. Without knowing the gas limits, we can't dry run properly on Ethereum.

In other words, we need the dry-run API to output the final message, the SCALE-encoded equivalent of this type: https://github.com/Snowfork/snowbridge/blob/fa9b8a7b3551cc9df8ee3239dc2830be5d84728e/contracts/src/v2/Types.sol#L8

This makes it much more easy for the UX to dry-run the resulting message on Ethereum.

We can ignore the message nonce for dry-run purposes on Ethereum.

So I you think you should explore adding a custom runtime API that takes the ExportMessage XCM, converts it to the final ethereum command.

Meanwhile to remove the fee when validate ExportMessage so that we can decrease the delivery fee config on AH and make it as an user input.

More context in paritytech#4826 (comment)

yrong mentioned this pull request Oct 18, 2024

Initialize bridge V2 paritytech/polkadot-sdk#6113

Draft

yrong mentioned this pull request Oct 22, 2024

Parachain Relayer V2 Snowfork/snowbridge#1321

Open

vgeddes reviewed Oct 22, 2024

View reviewed changes

claravanstaden mentioned this pull request Oct 23, 2024

Snowbridge V2 - Unified Rewards claravanstaden/polkadot-sdk#1

Open

yrong changed the base branch from v2 to master October 24, 2024 16:55

yrong changed the base branch from master to xcm-v5 October 29, 2024 02:03

yrong marked this pull request as ready for review October 29, 2024 12:44

vgeddes reviewed Oct 30, 2024

View reviewed changes

yrong force-pushed the outbound-queue-v2 branch 2 times, most recently from 80dad67 to 74d23af Compare November 4, 2024 02:06

Merge from xcm-v5

df137c0

yrong force-pushed the outbound-queue-v2 branch from fffd895 to df137c0 Compare November 5, 2024 17:03

yrong added 7 commits November 6, 2024 09:16

V2 message from message queue

c3e8ff1

Cleanup

9e4a427

Update sol binding

a1de9ef

No fee for governance order

d5c1115

Merge branch 'xcm-v5' into outbound-queue-v2

8533707

Cleanup

32657dc

Fix tests

5272515

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Outbound queue v2 #4

Outbound queue v2 #4

yrong commented Oct 18, 2024 •

edited

Loading

yrong commented Oct 21, 2024 •

edited

Loading

vgeddes Oct 22, 2024

yrong Oct 23, 2024

vgeddes Oct 22, 2024

yrong Oct 23, 2024

vgeddes Oct 22, 2024

yrong Oct 23, 2024 •

edited

Loading

vgeddes Oct 22, 2024

yrong Oct 24, 2024

vgeddes Oct 22, 2024

yrong Oct 24, 2024

vgeddes Oct 22, 2024

yrong Oct 24, 2024

vgeddes Oct 22, 2024

yrong Oct 23, 2024

vgeddes Oct 22, 2024

yrong Oct 24, 2024

vgeddes Oct 22, 2024

yrong Oct 23, 2024 •

edited

Loading

yrong commented Oct 29, 2024 •

edited

Loading

vgeddes Oct 30, 2024

yrong Oct 31, 2024

yrong Nov 1, 2024

	if ticket.origin != primary_governance_origin() {
	ensure!(!Self::operating_mode().is_halted(), SendError::Halted);
	}

	pub commands: BoundedVec<CommandWrapper, ConstU32<5>>,
	pub commands: BoundedVec<CommandWrapper, ConstU32<MAX_COMMANDS>>,

Outbound queue v2 #4

Are you sure you want to change the base?

Outbound queue v2 #4

Conversation

yrong commented Oct 18, 2024 • edited Loading

yrong commented Oct 21, 2024 • edited Loading

Workflow

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

yrong Oct 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

yrong Oct 23, 2024 • edited Loading

Choose a reason for hiding this comment

yrong commented Oct 29, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

yrong commented Oct 18, 2024 •

edited

Loading

yrong commented Oct 21, 2024 •

edited

Loading

yrong Oct 23, 2024 •

edited

Loading

yrong Oct 23, 2024 •

edited

Loading

yrong commented Oct 29, 2024 •

edited

Loading