reth

reth copied to clipboard

This validates sidechain state roots by first unwinding, trying to insert the new payload, and then re-committing the unwinded data.

Fixes #3060 Fixes #3057

TODO:

• [x] fix latestValidHash - currently is incorrect for invalid payloads on forks
• [x] remove unwraps

Codecov Report

Merging #3074 (33fa694) into main (4de1ff0) will decrease coverage by 0.03%. Report is 11 commits behind head on main. The diff coverage is 61.66%.

Files Changed	Coverage Δ
crates/interfaces/src/executor.rs	100.00% <ø> (ø)
crates/interfaces/src/provider.rs	12.50% <0.00%> (-87.50%)	:arrow_down:
crates/payload/basic/src/lib.rs	0.00% <0.00%> (ø)
crates/payload/builder/src/error.rs	0.00% <0.00%> (ø)
crates/rpc/rpc-engine-api/src/error.rs	60.00% <0.00%> (-10.59%)	:arrow_down:
crates/blockchain-tree/src/blockchain_tree.rs	82.61% <50.00%> (-0.77%)	:arrow_down:
crates/consensus/beacon/src/engine/mod.rs	76.08% <96.29%> (+0.29%)	:arrow_up:
crates/blockchain-tree/src/chain.rs	76.81% <100.00%> (-0.25%)	:arrow_down:

... and 18 files with indirect coverage changes

Flag	Coverage Δ
integration-tests	16.84% <0.00%> (-0.03%)	:arrow_down:
unit-tests	64.05% <61.66%> (-0.02%)	:arrow_down:

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
reth binary	25.94% <ø> (ø)
blockchain tree	82.48% <57.89%> (-0.36%)	:arrow_down:
pipeline	90.07% <ø> (ø)
storage (db)	74.58% <ø> (-0.18%)	:arrow_down:
trie	94.71% <ø> (ø)
txpool	48.20% <ø> (+0.53%)	:arrow_up:
networking	77.69% <ø> (-0.04%)	:arrow_down:
rpc	58.64% <0.00%> (-0.07%)	:arrow_down:
consensus	63.95% <96.29%> (+0.19%)	:arrow_up:
revm	32.26% <ø> (ø)
payload builder	6.54% <0.00%> (-0.04%)	:arrow_down:
primitives	87.83% <0.00%> (-0.08%)	:arrow_down:

What's blocking us here?

What's blocking us here?

Need to convert some unwraps to errors, then this should be r4r

I'm split if we should go forward with this as an intermediate solution. imo we should have taken the in-mem diff approach instead of non-committing unwind

Agree that the in memory approach is strictly better, whether or not we want to merge this is there an issue for the in-memory design?

Update here, we can probably include this as an intermediate solution. Otherwise this just needs to be rebased

this needs some work because of error mapping since rebase - will update when this is working again

Ready for review now

converted to draft because this should not be merged - I'll check the hive results with this PR, post them here, and close the PR. This should give us a good idea of how far we'll get with hive tests if we implement sidechain root validation in some other way.

Hive results:

INFO[08-10|14:35:31] API: suite ended                         suite=0
INFO[08-10|14:35:32] simulation ethereum/engine finished      suites=1 tests=109 failed=7
7 tests failed

Closing the PR now