eth/gasestimator: check ErrGasLimitTooHigh conditions #32268

mininny · 2025-07-24T07:48:05Z

This PR makes 2 changes to how EIP-7825 behaves.

When eth_estimateGas or eth_createAccessList is called without any gas limit in the payload, geth will choose the block's gas limit or the RPCGasCap, which can be larger than the maxTxGas.

When this happens for estimateGas, the gas estimation just errors out and ends, when it should continue doing binary search to find the lowest possible gas limit.

This PR will:

Add a check to see if hi is larger than maxTxGas and cap it to maxTxGas if it's larger. And add a special case handling for gas estimation execute when it errs with ErrGasLimitTooHigh
Decrease the RPCGasCap to 30M (== maxTxGas) to make createAccessList choose 30M as the highest gas cap for the tx. This can also be fixed by adding a check like this to use maxTxGas when the specified RPCGasCap is larger than maxTxGas. I'd be glad to fix it either way!

gasCap :=b.RPCGasCap()  
if gasCap > params.MaxTxGas { 
    gasCap = params.MaxTxGas
}

I've created this PR for fusaka-devnet-3 first, and will open another pr to master (or there is some periodic merges to master) if needed.

Replace manual byte-by-byte XOR implementation with the optimized bitutil.XORBytes function. This improves performance by using word-sized operations on supported architectures while maintaining the same functionality. The optimized version processes data in bulk rather than one byte at a time --------- Co-authored-by: Felix Lange <fjl@twurst.com>

Improve binary search, preventing the potential overflow in certain L2 cases

Seems the `signal.result` was not sent back in shorten case, this will cause a deadlock. --------- Signed-off-by: jsvisa <delweng@gmail.com> Co-authored-by: Gary Rong <garyrong0905@gmail.com>

Adds the heal time and snap sync time to grafana --------- Co-authored-by: Gary Rong <garyrong0905@gmail.com>

eth/ethconfig/config.go

rjl493456442 · 2025-07-25T06:36:59Z

eth/gasestimator/gasestimator.go

@@ -62,6 +62,9 @@ func Estimate(ctx context.Context, call *core.Message, opts *Options, gasCap uin
 	if call.GasLimit >= params.TxGas {
 		hi = call.GasLimit
 	}
+	if hi >= params.MaxTxGas {


This might break L2 projects whose transactions potentially consume more gas in order to amortize L1 costs.

I might have misunderstood the code, but even if the L2s specified gas higher than MaxTxGas, wouldn't it err with ErrGasLimitTooHigh and return error for the estimateGas call anyway?

Yeah, you are right!

I will ask L2 forks how they plan to integrate the EIP-7825 first, but generally your changes look good to me.

The gas capping should only be applied for Osaka

// Cap the maximum gas allowance according to EIP-7825 if the estimation targets Osaka if hi > params.MaxTxGas { blockNumber, blockTime := opts.Header.Number, opts.Header.Time if opts.BlockOverrides != nil { if opts.BlockOverrides.Number != nil { blockNumber = opts.BlockOverrides.Number.ToInt() } if opts.BlockOverrides.Time != nil { blockTime = uint64(*opts.BlockOverrides.Time) } } if opts.Config.IsOsaka(blockNumber, blockTime) { hi = params.MaxTxGas } }

understood, thank you! I've updated the code to your suggestion 🙏

The `errors.new` function does not require string formatting, so its performance is better than that of `fmt.Errorf`.

…m#32269) Correct the error message in the ExecuteStatelessPayloadV4 function to reference newPayloadV4 and the Prague fork, instead of incorrectly referencing newPayloadV3 and Cancun. This improves clarity during debugging and aligns the error message with the actual function and fork being validated. No logic is changed. --------- Co-authored-by: rjl493456442 <garyrong0905@gmail.com>

Alternative implementation of ethereum#32159

The errors.new function does not require string formatting, so its performance is better than that of fmt.Errorf.

This PR addresses a flakiness in the rollback test discussed in ethereum#32252 I found `nonce` collision caused transactions occasionally fail to send. I tried to change error message in the failed test like: ``` if err = client.SendTransaction(ctx, signedTx); err != nil { t.Fatalf("failed to send transaction: %v, nonce: %d", err, signedTx.Nonce()) } ``` and I occasionally got test failure with this message: ``` === CONT TestFlakyFunction/Run_#100 rollback_test.go:44: failed to send transaction: already known, nonce: 0 --- FAIL: TestFlakyFunction/Run_#100 (0.07s) ``` Although `nonces` are obtained via `PendingNonceAt`, we observed that, in rare cases (approximately 1 in 1000), two transactions from the same sender end up with the same nonce. This likely happens because `tx0` has not yet propagated to the transaction pool before `tx1` requests its nonce. When the test succeeds, `tx0` and `tx1` have nonces `0` and `1`, respectively. However, in rare failures, both transactions end up with nonce `0`. We modified the test to explicitly assign nonces to each transaction. By controlling the nonce values manually, we eliminated the race condition and ensured consistent behavior. After several thousand runs, the flakiness was no longer reproducible in my local environment. Reduced internal polling interval in `pendingStateHasTx()` to speed up test execution without impacting stability. It reduces test time for `TestTransactionRollbackBehavior` from about 7 seconds to 2 seconds.

Improvement: preallocate capacity for `logs` at first to avoid reallocating multi times.

…#32298)

…eum#32304)

rjl493456442 · 2025-07-31T06:19:00Z

Oops, the compilation is failing

: # github.com/ethereum/go-ethereum/internal/ethapi/override [github.com/ethereum/go-ethereum/internal/ethapi/override.test]
internal/ethapi/override/override_test.go:45:39: cannot use &precompileContract{} (value of type *precompileContract) as vm.PrecompiledContract value in map literal: *precompileContract does not implement vm.PrecompiledContract (missing method Name)
internal/ethapi/override/override_test.go:46:39: cannot use &precompileContract{} (value of type *precompileContract) as vm.PrecompiledContract value in map literal: *precompileContract does not implement vm.PrecompiledContract (missing method Name)

This adds a cross-client protocol test for a recently discovered bug in Nethermind.

This pull request optimizes trie hashing by reducing memory allocation overhead. Specifically: - define a fullNodeEncoder pool to reuse encoders and avoid memory allocations. - simplify the encoding logic for shortNode and fullNode by getting rid of the Go interfaces.

This adds a method on vm.EVM to set the jumpdest cache implementation. It can be used to maintain an analysis cache across VM invocations, to improve performance by skipping the analysis for already known contracts. --------- Co-authored-by: lmittmann <lmittmann@users.noreply.github.com> Co-authored-by: Felix Lange <fjl@twurst.com>

…imit estimation to continue estimating with lower caps

rjl493456442 · 2025-08-01T12:40:20Z

Just realized you are opening the PR against [ethereum:fusaka-devnet-3](https://github.com/ethereum/go-ethereum/tree/fusaka-devnet-3). Any particular reason?

Sorry I messed your branch, I thought it's against master

mininny · 2025-08-05T06:26:04Z

@rjl493456442 Hi, sorry for taking a while to check back on this.
I originally created this against fusaka-devnet-3 because I was working on testing the latest fusaka spec.

I just created another PR to master with clean commit history since this isn't necessarily for fusaka specific. #32348

rjl493456442 · 2025-08-05T11:57:20Z

Implemented in the #32348

reallesee and others added 3 commits July 22, 2025 23:06

eth/gasestimator: fix potential overflow (ethereum#32255)

3b67602

Improve binary search, preventing the potential overflow in certain L2 cases

triedb/pathdb: fix an deadlock in history indexer (ethereum#32260)

16117eb

Seems the `signal.result` was not sent back in shorten case, this will cause a deadlock. --------- Signed-off-by: jsvisa <delweng@gmail.com> Co-authored-by: Gary Rong <garyrong0905@gmail.com>

mininny requested a review from rjl493456442 as a code owner July 24, 2025 07:48

eth/protocols/snap: add healing and syncing metrics (ethereum#32258)

b369a85

Adds the heal time and snap sync time to grafana --------- Co-authored-by: Gary Rong <garyrong0905@gmail.com>

rjl493456442 reviewed Jul 25, 2025

View reviewed changes

nthumann and others added 2 commits July 28, 2025 09:13

core: replace the empty fmt.Errorf with errors.New (ethereum#32274)

29eebb5

The `errors.new` function does not require string formatting, so its performance is better than that of `fmt.Errorf`.

rjl493456442 added the osaka label Jul 28, 2025

rjl493456442 and others added 5 commits July 28, 2025 14:57

cmd, eth, internal: introduce debug_sync (ethereum#32177)

a7aed7b

Alternative implementation of ethereum#32159

all: replace fmt.Errorf with errors.New (ethereum#32286)

32d537c

The errors.new function does not require string formatting, so its performance is better than that of fmt.Errorf.

downloader: fix typos, grammar and formatting (ethereum#32288)

b64a500

core/state: preallocate capacity for logs list (ethereum#32291)

a56558d

Improvement: preallocate capacity for `logs` at first to avoid reallocating multi times.

rjl493456442 changed the title ~~fix: ErrGasLimitTooHigh conditions for estimateGas and createAccessList~~ eth/gasestimator: check ErrGasLimitTooHigh conditions Jul 29, 2025

ericxtheodore and others added 4 commits July 30, 2025 10:39

core/state: improve PrettyPrint function (ethereum#32293)

d14d4d2

core/types: expose sigHash as Hash for SetCodeAuthorization (ethereum…

2d95ba7

…#32298)

common/hexutil: replace customized bit sizer with bit.Uintsize (ether…

0814d99

…eum#32304)

accounts/abi: precompile regex (ethereum#32301)

4d9d728

rjl493456442 approved these changes Jul 31, 2025

View reviewed changes

fjl and others added 4 commits July 31, 2025 12:13

cmd/devp2p/internal/v4test: add test for ENRRequest (ethereum#32303)

d4a3bf1

This adds a cross-client protocol test for a recently discovered bug in Nethermind.

eth: fix typos and outdated comments (ethereum#32324)

9c58810

rjl493456442 force-pushed the feat/mininny/allow-gaslimit-errors-in-estimateGas branch from 8121be4 to b58e83f Compare August 1, 2025 12:31

rjl493456442 requested review from fjl, gballet, zsfelfoldi and s1na as code owners August 1, 2025 12:31

rjl493456442 requested review from lightclient, MariusVanDerWijden and jwasinger as code owners August 1, 2025 12:31

mininny and others added 4 commits August 1, 2025 20:32

Allow ErrGasLimitTooHigh errors when executing transactions for gas l…

4d1264f

…imit estimation to continue estimating with lower caps

Use params.MaxTxGas for RPCGasCap

db12f63

Cap maximum gas allowance to EIP-7825 in gas estimates only for osaka

a7857cd

core/vm, internal/ethapi: fix lint

5ef01bb

rjl493456442 force-pushed the feat/mininny/allow-gaslimit-errors-in-estimateGas branch from b58e83f to 5ef01bb Compare August 1, 2025 12:37

mininny mentioned this pull request Aug 5, 2025

eth/gasestimator: check ErrGasLimitTooHigh conditions #32348

Merged

rjl493456442 closed this Aug 5, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

eth/gasestimator: check ErrGasLimitTooHigh conditions #32268

eth/gasestimator: check ErrGasLimitTooHigh conditions #32268

Uh oh!

mininny commented Jul 24, 2025

Uh oh!

Uh oh!

rjl493456442 Jul 25, 2025

Uh oh!

mininny Jul 25, 2025

Uh oh!

rjl493456442 Jul 28, 2025

Uh oh!

rjl493456442 Jul 31, 2025 •

edited

Loading

Uh oh!

mininny Jul 31, 2025

Uh oh!

rjl493456442 commented Jul 31, 2025

Uh oh!

rjl493456442 commented Aug 1, 2025

Uh oh!

mininny commented Aug 5, 2025

Uh oh!

rjl493456442 commented Aug 5, 2025

Uh oh!

Uh oh!

eth/gasestimator: check ErrGasLimitTooHigh conditions #32268

eth/gasestimator: check ErrGasLimitTooHigh conditions #32268

Uh oh!

Conversation

mininny commented Jul 24, 2025

Uh oh!

Uh oh!

rjl493456442 Jul 25, 2025

Choose a reason for hiding this comment

Uh oh!

mininny Jul 25, 2025

Choose a reason for hiding this comment

Uh oh!

rjl493456442 Jul 28, 2025

Choose a reason for hiding this comment

Uh oh!

rjl493456442 Jul 31, 2025 • 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!

mininny Jul 31, 2025

Choose a reason for hiding this comment

Uh oh!

rjl493456442 commented Jul 31, 2025

Uh oh!

rjl493456442 commented Aug 1, 2025

Uh oh!

mininny commented Aug 5, 2025

Uh oh!

rjl493456442 commented Aug 5, 2025

Uh oh!

Uh oh!

rjl493456442 Jul 31, 2025 •

edited

Loading