Add pallet_revive to Polkadot AssetHub by athei · Pull Request #1050 · polkadot-fellows/runtimes

athei · 2026-01-06T20:17:06Z

In addition to adding the pallet I also:

KSM: Removed the old WeightToFee definition and replaced all occurrences with the new one
KSM: Removed magic numbers from the WeightToFee definition
DOT: Replaced WeightToFee with new definition required for pallet_revive
Fees: Moved the fee definition to the constants crate so they are shared between all system chains

The only difference to the Kusama config is:

type NativeToEthRatio = ConstU32<100_000_000>; //  vs. 1_000_000 on KSM due to different decimals
type GasScale = ConstU32<1_000>; //  vs 100_000 on KSM due to different decimals

github-actions

Found fee-calculation semantic risks (notably proof-size charging), an inconsistency in migration call filtering for the new Revive pallet, and a fragile extrinsic construction workaround introduced by the new UncheckedExtrinsic type; also a couple of test/formatting regressions and Revive config choices that warrant tightening/justification.

Suggestions that couldn't be attached to a specific line

system-parachains/asset-hubs/asset-hub-kusama/src/lib.rs:2779, 2813-2814

Two issues in the updated tests: (1) there is a stray blank line/whitespace-only line in the imports/aliases section ([Line 2779]) which should be removed to avoid churn/noise; (2) the proof-size fee assertion was weakened from assert_eq!(fee, CENTS, "10kb maps to CENT") to only checking non-zero ([Line 2813-2814]). This reduces the test’s ability to detect accidental fee-regression in proof-size pricing. Action: remove the whitespace-only line and replace the weakened assertion with an explicit expected mapping for proof_size under the new WeightToFee (even if the expected value changes, assert that value/range).

system-parachains/asset-hubs/asset-hub-polkadot/src/ah_migration/call_filter.rs:141, 210

Revive is set to OFF in call_allowed_status before migration ([Line 141]) but is included as ON in call_allowed_before_migration ([Line 210]). This is internally inconsistent and makes it unclear whether Revive calls are intended to be permitted pre-migration. Action: decide the intended policy and make both functions agree (and add a short comment for Revive specifically, since this pallet is newly introduced and likely sensitive).

system-parachains/asset-hubs/asset-hub-polkadot/src/staking/mod.rs:559, 568

Extrinsic construction is changed to use <UncheckedExtrinsic as TypeInfo>::Identity::new_transaction/new_bare(...).into() ([Line 559], [Line 568]). This is a fragile coupling to SCALE-info metadata (TypeInfo::Identity) rather than the actual runtime extrinsic API, and it may break if the TypeInfo derive/Identity changes or if UncheckedExtrinsic’s wrapper semantics evolve. Action: introduce an explicit base extrinsic type alias and construct that instead, e.g. type BaseUx = sp_runtime::generic::UncheckedExtrinsic<Address, RuntimeCall, Signature, TxExtension>; BaseUx::new_transaction(...).into(), or add/consume a dedicated constructor provided by the new pallet_revive::evm::runtime::UncheckedExtrinsic wrapper.

system-parachains/constants/src/kusama.rs

system-parachains/constants/src/polkadot.rs

system-parachains/asset-hubs/asset-hub-polkadot/src/lib.rs

athei · 2026-01-06T20:24:14Z

/cmd fmt

github-actions · 2026-01-06T20:32:29Z

Command "fmt" has started 🚀 See logs here

github-actions · 2026-01-06T20:33:39Z

Command "fmt" has finished ✅ See logs here

In polkadot-fellows/runtimes#1050 I am unifying the used `WeightToFee` formula to use `BlockRatioFee` everywhere. This type is exported by `pallet_revive` and is independent of any functionality here. However, so far it required the runtime to use `pallet_revive` in order to discover the `Balance` type. This PR removes this constraint so it can be used by all runtimes. Even the ones which do not use `pallet_revive`. --------- Co-authored-by: cmd[bot] <41898282+github-actions[bot]@users.noreply.github.com>

In polkadot-fellows/runtimes#1050 I am unifying the used `WeightToFee` formula to use `BlockRatioFee` everywhere. This type is exported by `pallet_revive` and is independent of any functionality here. However, so far it required the runtime to use `pallet_revive` in order to discover the `Balance` type. This PR removes this constraint so it can be used by all runtimes. Even the ones which do not use `pallet_revive`. --------- Co-authored-by: cmd[bot] <41898282+github-actions[bot]@users.noreply.github.com> (cherry picked from commit 8d6ccfd)

athei · 2026-01-08T15:50:46Z

@claravanstaden This PR changes the weight formula for all system chains. Including bridge hub. This integration/runtime test are now failing:
https://github.com/polkadot-fellows/runtimes/actions/runs/20818991053/job/59803475344?pr=1050
https://github.com/polkadot-fellows/runtimes/actions/runs/20818991053/job/59806987895?pr=1050

Could it be related to changing weight fees? Can you have a look?

yrong · 2026-01-09T06:06:46Z

@claravanstaden This PR changes the weight formula for all system chains. Including bridge hub. This integration/runtime test are now failing: https://github.com/polkadot-fellows/runtimes/actions/runs/20818991053/job/59803475344?pr=1050 https://github.com/polkadot-fellows/runtimes/actions/runs/20818991053/job/59806987895?pr=1050

Could it be related to changing weight fees? Can you have a look?

@athei I've made a fix in #1051

seunlanlege · 2026-01-09T12:09:54Z

Amazing, will this include: paritytech/polkadot-sdk#10476 ?

system-parachains/asset-hubs/asset-hub-polkadot/src/lib.rs

athei · 2026-01-12T13:50:21Z

Amazing, will this include: paritytech/polkadot-sdk#10476 ?

Yes. Both kusama and polkadot will get this fix as part of this PR.

athei

I needed to adjust hard coded weight values due to the weight formula changes. I pinged the people I think should review those changes.

system-parachains/encointer/src/treasuries_xcm_payout.rs

athei · 2026-01-12T15:28:53Z

system-parachains/bridge-hubs/bridge-hub-kusama/primitives/src/lib.rs

-	pub const BridgeHubKusamaBaseXcmFeeInKsms: u128 = 757_124_666;
+	pub const BridgeHubKusamaBaseXcmFeeInKsms: u128 = 36_453_538;

 	/// Transaction fee that is paid at the Kusama BridgeHub for delivering single inbound message.
 	/// (initially was calculated by test `BridgeHubKusama::can_calculate_fee_for_complex_message_delivery_transaction` + `33%`)
-	pub const BridgeHubKusamaBaseDeliveryFeeInKsms: u128 = 3_142_112_953;
+	pub const BridgeHubKusamaBaseDeliveryFeeInKsms: u128 = 709_970_588;

 	/// Transaction fee that is paid at the Kusama BridgeHub for delivering single outbound message confirmation.
 	/// (initially was calculated by test `BridgeHubKusama::can_calculate_fee_for_complex_message_confirmation_transaction` + `33%`)
-	pub const BridgeHubKusamaBaseConfirmationFeeInKsms: u128 = 575_036_072;
+	pub const BridgeHubKusamaBaseConfirmationFeeInKsms: u128 = 181_274_402;


@acatangiu @franciscoaguirre The weight formula changed. I adjusted those values in order to make the tests succeed. Please review if it is okay to make this change.

Why are the number that much lower? There shouldn't be this much of a difference, no?

If we just made things cheaper then I'm fine with it

Why are the number that much lower?

I think because of in-memory weights: #1025

athei · 2026-01-12T15:29:07Z

system-parachains/bridge-hubs/bridge-hub-polkadot/primitives/src/lib.rs

-	pub const BridgeHubPolkadotBaseXcmFeeInDots: Balance = 114_080_750;
+	pub const BridgeHubPolkadotBaseXcmFeeInDots: Balance = 5_638_281;

 	/// Transaction fee that is paid at the Polkadot BridgeHub for delivering single inbound message.
 	/// (initially was calculated by test `BridgeHubPolkadot::can_calculate_fee_for_standalone_message_delivery_transaction` + `33%`)
-	pub const BridgeHubPolkadotBaseDeliveryFeeInDots: Balance = 471_317_032;
+	pub const BridgeHubPolkadotBaseDeliveryFeeInDots: Balance = 106_495_676;

 	/// Transaction fee that is paid at the Polkadot BridgeHub for delivering single outbound message confirmation.
 	/// (initially was calculated by test `BridgeHubPolkadot::can_calculate_fee_for_standalone_message_confirmation_transaction` + `33%`)
-	pub const BridgeHubPolkadotBaseConfirmationFeeInDots: Balance = 86_267_609;
+	pub const BridgeHubPolkadotBaseConfirmationFeeInDots: Balance = 27_027_016;


@acatangiu @franciscoaguirre The weight formula changed. I adjusted those values in order to make the tests succeed. Please review if it is okay to make this change.

acatangiu · 2026-01-12T15:41:30Z

I needed to adjust hard coded weight values due to the weight formula changes. I pinged the people I think should review those changes.

@bkontur can you please look?

pgherveou · 2026-01-12T15:41:41Z

system-parachains/asset-hubs/asset-hub-polkadot/src/lib.rs

+	type InstantiateOrigin = EnsureSigned<Self::AccountId>;
+	type RuntimeHoldReason = RuntimeHoldReason;
+	type CodeHashLockupDepositPercent = CodeHashLockupDepositPercent;
+	type ChainId = ConstU64<420_420_418>;


that's kusama value here isn' it?

Good catch! I'll need to adjust.

kianenigma · 2026-01-12T16:36:11Z

system-parachains/asset-hubs/asset-hub-polkadot/src/staking/mod.rs


 	fn create_transaction(call: RuntimeCall, extension: TxExtension) -> UncheckedExtrinsic {
-		UncheckedExtrinsic::new_transaction(call, extension)
+		<UncheckedExtrinsic as TypeInfo>::Identity::new_transaction(call, extension).into()


Can't guess why but I trust you :D

Maybe you shouldn't. I just took this from the kusama runtime. I will have a look at this and maybe they both need to be changed so something more elegant.

Okay I switched to explicitly naming the correct type. Also fixed up the runtime where I copied the code from.

kianenigma · 2026-01-12T16:42:34Z

system-parachains/asset-hubs/asset-hub-polkadot/src/lib.rs

+	type MaxEthExtrinsicWeight = MaxEthExtrinsicWeight;
+	// Must be set to `false` in a live chain
+	type DebugEnabled = ConstBool<false>;
+	type GasScale = ConstU32<80_000>;


The PRdoc where this is introduced recommends 50k and 100k. What is the rationale here?

This value depends on the deposit fees, weight fees and decimals of the chain. We need to choose it in a way that the numerical gas values are not higher than on Ethereum. This is because wallets use hard coded values for transfers. See: #1050 (comment)

So the tuning is different for different chains.

system-parachains/constants/src/polkadot.rs

franciscoaguirre · 2026-01-12T18:48:40Z

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

 ) {
-	// On AH the xcm fee is 26_789_690 and the ED is 3_300_000
-	send_token_from_ethereum_to_asset_hub_with_fee([1; 32], 30_000_000);
+	// On AH, ED is 0.1 DOT. Make both the transfer amount (in WETH) and the XCM fee below the ED.


Isn't the ED 0.01 DOT?

Yes should be 0.01.

franciscoaguirre · 2026-01-12T19:01:03Z

system-parachains/asset-hubs/asset-hub-polkadot/src/lib.rs

 		Staking: pallet_staking_async = 89,

+		// Contracts
+		Revive: pallet_revive = 90,


franciscoaguirre · 2026-01-12T19:07:31Z

system-parachains/bridge-hubs/bridge-hub-kusama/primitives/src/lib.rs

-	pub const BridgeHubKusamaBaseXcmFeeInKsms: u128 = 757_124_666;
+	pub const BridgeHubKusamaBaseXcmFeeInKsms: u128 = 36_453_538;

 	/// Transaction fee that is paid at the Kusama BridgeHub for delivering single inbound message.
 	/// (initially was calculated by test `BridgeHubKusama::can_calculate_fee_for_complex_message_delivery_transaction` + `33%`)
-	pub const BridgeHubKusamaBaseDeliveryFeeInKsms: u128 = 3_142_112_953;
+	pub const BridgeHubKusamaBaseDeliveryFeeInKsms: u128 = 709_970_588;

 	/// Transaction fee that is paid at the Kusama BridgeHub for delivering single outbound message confirmation.
 	/// (initially was calculated by test `BridgeHubKusama::can_calculate_fee_for_complex_message_confirmation_transaction` + `33%`)
-	pub const BridgeHubKusamaBaseConfirmationFeeInKsms: u128 = 575_036_072;
+	pub const BridgeHubKusamaBaseConfirmationFeeInKsms: u128 = 181_274_402;


Why are the number that much lower? There shouldn't be this much of a difference, no?

If we just made things cheaper then I'm fine with it

athei · 2026-01-12T19:47:40Z

/merge

fellowship-merge-bot · 2026-01-12T19:49:27Z

Enabled auto-merge in Pull Request

Available commands

/merge: Enables auto-merge for Pull Request
/merge cancel: Cancels auto-merge for Pull Request
/merge help: Shows this menu

For more information see the documentation

bkontur · 2026-01-13T07:58:47Z

system-parachains/asset-hubs/asset-hub-kusama/src/xcm_config.rs

 	type Trader = (
 		UsingComponents<
-			WeightToFee,
+			WeightToFee<Runtime>,


Very ultra nit or just question, why not to use directly KsmWeightToFee?

Suggested change

WeightToFee<Runtime>,

KsmWeightToFee<Runtime>,

When somebody will serach for WeightToFee, it could get little bit confusing :)

KsmWeightToFee as WeightToFee fee::WeightToFee as KsmWeightToFee, type WeightToFee = KsmWeightToFee<Runtime>;

Basically, we are doing some aliasing, everywhere we import WeightToFee from constants:

use system_parachains_constants::kusama::{currency::*, fee::WeightToFee as KsmWeightToFee}; ... type WeightToFee = KsmWeightToFee<Runtime>;

Maybe, if we rename WeightToFee to KsmWeightToFee in the constants, we won't need to do any aliasing and just use directly KsmWeightToFee.

I am completely open to whatever people think is the most elegant. I used so much aliasing to keep the code churn down.

athei added 3 commits January 6, 2026 16:26

Kusama: Fully replace old WeightToFee type

86bdf87

polkadot: Add pallet_revive

6faa9b8

Update CHANGELOG

3d6c5d2

github-actions bot reviewed Jan 6, 2026

View reviewed changes

Propagate features

d256194

Update from athei running command 'fmt'

caaf8b4

athei mentioned this pull request Jan 7, 2026

Make BlockRatioFee no longer require pallet_revive paritytech/polkadot-sdk#10740

Merged

athei added 7 commits January 7, 2026 16:13

Update to latest pallet-revive

061ecf5

Fix remaining system chains to use new weight formula

5ca20f4

Remove stray white lines

088a220

Remove useless .into()

3ace928

Merge branch 'main' into at/scc-polkadot

7b2664c

forward try-runtime feature

5cf454a

map testnet accounts in genesis

049b935

Fix broken snowbridge tests by sending proper amount (#1051)

a666392

TorstenStueber reviewed Jan 12, 2026

View reviewed changes

system-parachains/asset-hubs/asset-hub-polkadot/src/lib.rs Outdated Show resolved Hide resolved

athei added 4 commits January 12, 2026 11:09

Remove already applied migration

39fddd7

Adjust expected XCM fees to new weight formula

5e7156d

Tune GasScale so that ERC20 transfers match the gas values on Eth

0cfb0a5

Adjust REMOTE_XCM_TRANSFER_REMOTE_EXECUTION_FEE to new fee formula

fb99985

Merge branch 'main' into at/scc-polkadot

1a723d0

athei commented Jan 12, 2026

View reviewed changes

pgherveou reviewed Jan 12, 2026

View reviewed changes

kianenigma reviewed Jan 12, 2026

View reviewed changes

polkadot-fellows deleted a comment from github-actions bot Jan 12, 2026

kianenigma approved these changes Jan 12, 2026

View reviewed changes

athei added 3 commits January 12, 2026 14:29

Fix chain id for polkadot asset hub

951c3be

Don't use type info to construct types

5b73b0d

Merge branch 'main' into at/scc-polkadot

7f691a3

ggwpez reviewed Jan 12, 2026

View reviewed changes

system-parachains/constants/src/polkadot.rs Show resolved Hide resolved

ggwpez approved these changes Jan 12, 2026

View reviewed changes

franciscoaguirre approved these changes Jan 12, 2026

View reviewed changes

fellowship-merge-bot bot enabled auto-merge (squash) January 12, 2026 19:49

fellowship-merge-bot bot merged commit 4d9eca7 into main Jan 12, 2026
140 of 157 checks passed

bkontur reviewed Jan 13, 2026

View reviewed changes

bkchr deleted the at/scc-polkadot branch February 4, 2026 13:51

Conversation

athei commented Jan 6, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot left a comment

Choose a reason for hiding this comment

Suggestions that couldn't be attached to a specific line

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

athei commented Jan 6, 2026

Uh oh!

github-actions bot commented Jan 6, 2026

Uh oh!

github-actions bot commented Jan 6, 2026

Uh oh!

athei commented Jan 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

yrong commented Jan 9, 2026

Uh oh!

seunlanlege commented Jan 9, 2026

Uh oh!

Uh oh!

athei commented Jan 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

athei left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

franciscoaguirre Jan 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

acatangiu commented Jan 12, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

franciscoaguirre Jan 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

athei commented Jan 12, 2026

Uh oh!

fellowship-merge-bot bot commented Jan 12, 2026

Uh oh!

Uh oh!

Choose a reason for hiding this comment

athei commented Jan 6, 2026 •

edited

Loading

athei commented Jan 8, 2026 •

edited

Loading

athei commented Jan 12, 2026 •

edited

Loading

franciscoaguirre Jan 12, 2026 •

edited

Loading

franciscoaguirre Jan 12, 2026 •

edited

Loading