Open
Conversation
closes #292 we previously didn't set the error field for the exit function (cherry picked from commit 6c7e358)
Closes paradigmxyz/revm-inspectors#294 while retaining fix for paradigmxyz/reth#13089 The idea is that in steps (aka logs) we want `contract.address` to always be the address of contract which storage is accessed (`target_address`). However, when dealing with delegatecall frames, `contract.address` should be set to the `bytecode_address` as it is marking the destination of delegatecall. This means we can't just forward `active_call().contract` to steps anymore and need to re-construct the `Contract` value manually (cherry picked from commit da1b9ee)
ref paradigmxyz/reth#13089 we should return the original bytes here (cherry picked from commit 17a7b51)
fix gas accounting for call opcodes cc @sslivkoff @rakita (cherry picked from commit 13d515a)
the vmtracer needs state diffs for https://github.com/alloy-rs/alloy/blob/c808022d571aaffb97a3429a0fdba6fc06e8f7aa/crates/rpc-types-trace/src/parity.rs#L659-L660 so that we enter here: https://github.com/paradigmxyz/revm-inspectors/blob/1c0e405ad008453cded08b44a3cfa2ca611a6aa8/src/tracing/mod.rs#L502-L502 and record the `step-storage_change` https://github.com/paradigmxyz/revm-inspectors/blob/1c0e405ad008453cded08b44a3cfa2ca611a6aa8/src/tracing/mod.rs#L507-L507 (cherry picked from commit 27289cb)
(cherry picked from commit 1303994)
(cherry picked from commit 477a770)
(cherry picked from commit 221aead)
ref paradigmxyz/reth#16289 we need to patch uneven js input strings, which could be digits, this now behaves like geth's tracer https://github.com/ethereum/go-ethereum/blob/355228b011ef9a85ebc0f21e7196f892038d49f0/common/bytes.go#L27-L37 https://github.com/ethereum/go-ethereum/blob/355228b011ef9a85ebc0f21e7196f892038d49f0/eth/tracers/js/goja.go#L91-L91 (cherry picked from commit 76c2965)
The Geth trace builder incorrectly sets the same `returnData` value for all opcode steps in a trace, using the final call output instead of the step-specific return data buffer state. ## Expected Behavior Each opcode step should display the return data buffer state at that specific point in execution. ## Actual Behavior All opcode steps show the same returnData value - the final output of the entire call. For example, in a Fibonacci contract call that returns 1, every single opcode step shows: "returnData": "0x0000000000000000000000000000000000000000000000000000000000000001" (cherry picked from commit 2a4ae16)
… (#363) Closes paradigmxyz/revm-inspectors#350 Closes paradigmxyz/revm-inspectors#349 - Fixes `isPrecompiled` built-in to work correctly. Before we were incorrectly checking `precompiles_registered` flag which caused the fn to never be injected. Added test for it as well - Adds `slice` built-in and a test for it (cherry picked from commit 6f2ecd1)
…e (#336) Currently the revm-inspectors js tracer acts differently from the geth implementation where the `getCost()` function returns the cumulative amount of cost for each step instead of the individual opcode cost. I added a simple function that will store the previous spent value and then use that in the next step to calculate the currect opcode cost. (cherry picked from commit d596fc0)
Ref paradigmxyz/reth#18904 Closes paradigmxyz/revm-inspectors#364 Changes js objects to be created via `JsObject::with_object_proto(ctx.intrinsics())` instead of `JsObject::default()` as suggested in boa-dev/boa#4460 (comment) thus making sure helpers like `hasOwnProperty` are accessible. Changes `JsValue -> f64` conversion to happen via `to_numeric_number` so that bigints can also be used as input values for `slice` functions (cherry picked from commit 7124ebd)
closes paradigmxyz/reth#18942 we should not set this if code is disabled in the config (cherry picked from commit 8736651)
The disableCode flag was not being respected in diff mode for the PreStateTracer. When disableCode was set to true, code was correctly filtered from the pre state but was still appearing in the post state. This fix ensures that when disableCode is true, the code field is set to None for both pre and post states in diff mode, matching the expected behavior from Geth. Added test coverage to verify the fix works correctly in both cases (disableCode=true and disableCode=false). Co-authored-by: Matthias Seitz <matthias.seitz@outlook.de> (cherry picked from commit 998801f)
follow-up for paradigmxyz/revm-inspectors#355 (cherry picked from commit 9c2fd43)
Geth will convert the empty bytes to null, and omit it in the response, I think we should align with it, the empty nulls seems meanless. Below is the differ of the tx: https://etherscan.io/tx/0x337749b7c19859b2d773868b60e82a06ae008db9a80d41db76d7020d976a72e7 > left is geth, right is reth <img width="1446" height="190" alt="image" src="https://github.com/user-attachments/assets/65049603-c967-4c9c-8c7f-7b4496c902e5" /> (cherry picked from commit 922eb33)
as_str could panic if the opcode is unknown (cherry picked from commit 7aafabf)
…(#377) Closes #376 - Don't insert selfdestructed accounts in post state - Added tests to verify behavior for London and Cancun (cherry picked from commit e159a65)
This aims at resolving issues mentioned in #382 , typically this one : <img width="1818" height="694" alt="image" src="https://github.com/user-attachments/assets/51b82944-c711-421b-8479-05095ae9bbce" /> To fix this , what i did was : - Keep track of addresses that are marked as Create and were truly empty in the database snapshot. Only these addresses are filtered from the prestate diff so prefunded contract deployments remain visible. - Introduce account_was_empty helper to avoid re-computing the emptiness check and add a regression test (prestate_diff_keeps_prefunded_created_accounts) that covers a prefunded creation vs. a true empty-address creation. cc @mattsse Please check it out if it resolves should close this one paradigmxyz/reth#19703 --------- Co-authored-by: Matthias Seitz <matthias.seitz@outlook.de> (cherry picked from commit fd4cf14)
eyusufatik
approved these changes
Dec 9, 2025
Member
eyusufatik
left a comment
eyusufatik
left a comment
There was a problem hiding this comment.
approved, but no need to merge this pr
|
|
||
| /// Registers the precompiles in the JS context | ||
| fn register_precompiles<CTX: ContextTr<Journal: JournalExt>>(&mut self, context: &mut CTX) { | ||
| if !self.precompiles_registered { |
eyusufatik
added a commit
to chainwayxyz/citrea
that referenced
this pull request
Dec 9, 2025
# Description Updates revm-inspectors with fixes to multiple tracers, mainly the prestate tracer. Since `revm-inspectors` couldn't be updated, we created a fork at https://github.com/chainwayxyz/revm-inspectors/ and applied all fix PRs after v0.18.1 on the v0.18.1 tag. The changes can be reviewed here: chainwayxyz/revm-inspectors#1
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
11 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.