fix: pass prepareSlot to fork choice head for Gloas FULL vs EMPTY tie-breaker

[twoeths]

Motivation

Starting from Gloas (ePBS), fork choice has FULL vs EMPTY block variants with tie-breaker logic that depends on currentSlot (see get_payload_status_tiebreaker and modified get_head). When prepareNextSlot runs at ~67% of slot N to prepare for slot N+1, it should pass prepareSlot (N+1) to recomputeForkChoiceHead() so the tie-breaker correctly evaluates for the next slot rather than the current wall-clock slot.

This is backward compatible: pre-Gloas blocks are always PayloadStatus.FULL, so the tie-breaker is a no-op. The slot param defaults to fcStore.currentSlot when not provided, preserving existing behavior for all other callers.

Description

Thread an optional slot parameter through recomputeForkChoiceHead() → updateAndGetHead() → updateHead(), used as a fallback for fcStore.currentSlot in applyScoreChanges. Only prepareNextSlot passes it (as prepareSlot).

AI Assistance Disclosure

Used Claude Code for research (other client implementations) and code generation.

[gemini-code-assist]

Code Review

This pull request introduces an optional slot parameter to the fork-choice head recomputation logic, allowing the prepareNextSlot scheduler to override the current slot for tie-breaker logic. While this enables correct proposer head prediction, feedback suggests that updating the global canonical head state with a future slot value may cause inconsistencies for other node functions, such as attestation production, which rely on the head corresponding to the current wall-clock slot.

[github-actions]

⚠️ Performance Alert ⚠️

Possible performance regression was detected for some benchmarks.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold.

Benchmark suite	Current: bda0662	Previous: d4d0b21	Ratio
Full columns - reconstruct all 20 blobs	1.8517 ms/op	560.51 us/op	3.30
phase0 getAttestationDeltas - 250000 worstcase	19.386 ms/op	5.4304 ms/op	3.57

Full benchmark results

Benchmark suite	Current: bda0662	Previous: d4d0b21	Ratio
getPubkeys - index2pubkey - req 1000 vs - 250000 vc	890.67 us/op	912.39 us/op	0.98
getPubkeys - validatorsArr - req 1000 vs - 250000 vc	39.591 us/op	39.413 us/op	1.00
BLS verify - blst	714.92 us/op	713.32 us/op	1.00
BLS verifyMultipleSignatures 3 - blst	1.3745 ms/op	1.3743 ms/op	1.00
BLS verifyMultipleSignatures 8 - blst	2.1827 ms/op	2.1981 ms/op	0.99
BLS verifyMultipleSignatures 32 - blst	7.0090 ms/op	6.9886 ms/op	1.00
BLS verifyMultipleSignatures 64 - blst	13.367 ms/op	13.545 ms/op	0.99
BLS verifyMultipleSignatures 128 - blst	26.064 ms/op	25.974 ms/op	1.00
BLS deserializing 10000 signatures	639.52 ms/op	641.47 ms/op	1.00
BLS deserializing 100000 signatures	6.3864 s/op	6.5157 s/op	0.98
BLS verifyMultipleSignatures - same message - 3 - blst	712.66 us/op	793.18 us/op	0.90
BLS verifyMultipleSignatures - same message - 8 - blst	950.09 us/op	906.26 us/op	1.05
BLS verifyMultipleSignatures - same message - 32 - blst	1.5691 ms/op	1.5957 ms/op	0.98
BLS verifyMultipleSignatures - same message - 64 - blst	2.3867 ms/op	2.4785 ms/op	0.96
BLS verifyMultipleSignatures - same message - 128 - blst	4.0825 ms/op	4.0819 ms/op	1.00
BLS aggregatePubkeys 32 - blst	17.734 us/op	17.763 us/op	1.00
BLS aggregatePubkeys 128 - blst	63.533 us/op	64.648 us/op	0.98
getSlashingsAndExits - default max	47.226 us/op	55.527 us/op	0.85
getSlashingsAndExits - 2k	329.27 us/op	331.73 us/op	0.99
proposeBlockBody type=full, size=empty	3.2757 ms/op	2.8211 ms/op	1.16
isKnown best case - 1 super set check	170.00 ns/op	612.00 ns/op	0.28
isKnown normal case - 2 super set checks	175.00 ns/op	184.00 ns/op	0.95
isKnown worse case - 16 super set checks	171.00 ns/op	172.00 ns/op	0.99
validate api signedAggregateAndProof - struct	1.5337 ms/op	1.5555 ms/op	0.99
validate gossip signedAggregateAndProof - struct	1.5332 ms/op	1.5494 ms/op	0.99
batch validate gossip attestation - vc 640000 - chunk 32	107.89 us/op	112.22 us/op	0.96
batch validate gossip attestation - vc 640000 - chunk 64	93.401 us/op	97.708 us/op	0.96
batch validate gossip attestation - vc 640000 - chunk 128	87.916 us/op	91.002 us/op	0.97
batch validate gossip attestation - vc 640000 - chunk 256	101.48 us/op	90.585 us/op	1.12
bytes32 toHexString	308.00 ns/op	293.00 ns/op	1.05
bytes32 Buffer.toString(hex)	181.00 ns/op	189.00 ns/op	0.96
bytes32 Buffer.toString(hex) from Uint8Array	267.00 ns/op	270.00 ns/op	0.99
bytes32 Buffer.toString(hex) + 0x	183.00 ns/op	188.00 ns/op	0.97
Return object 10000 times	0.21530 ns/op	0.21630 ns/op	1.00
Throw Error 10000 times	3.5074 us/op	3.3362 us/op	1.05
toHex	103.84 ns/op	100.56 ns/op	1.03
Buffer.from	93.902 ns/op	94.303 ns/op	1.00
shared Buffer	67.006 ns/op	65.692 ns/op	1.02
fastMsgIdFn sha256 / 200 bytes	1.5170 us/op	1.5180 us/op	1.00
fastMsgIdFn h32 xxhash / 200 bytes	162.00 ns/op	166.00 ns/op	0.98
fastMsgIdFn h64 xxhash / 200 bytes	207.00 ns/op	217.00 ns/op	0.95
fastMsgIdFn sha256 / 1000 bytes	4.8450 us/op	4.8600 us/op	1.00
fastMsgIdFn h32 xxhash / 1000 bytes	245.00 ns/op	256.00 ns/op	0.96
fastMsgIdFn h64 xxhash / 1000 bytes	259.00 ns/op	270.00 ns/op	0.96
fastMsgIdFn sha256 / 10000 bytes	42.575 us/op	42.851 us/op	0.99
fastMsgIdFn h32 xxhash / 10000 bytes	1.2810 us/op	1.2910 us/op	0.99
fastMsgIdFn h64 xxhash / 10000 bytes	826.00 ns/op	837.00 ns/op	0.99
send data - 1000 256B messages	4.2790 ms/op	4.2459 ms/op	1.01
send data - 1000 512B messages	4.3986 ms/op	4.4148 ms/op	1.00
send data - 1000 1024B messages	4.7349 ms/op	4.5355 ms/op	1.04
send data - 1000 1200B messages	4.5393 ms/op	5.0842 ms/op	0.89
send data - 1000 2048B messages	4.7424 ms/op	5.1691 ms/op	0.92
send data - 1000 4096B messages	5.8113 ms/op	6.0219 ms/op	0.97
send data - 1000 16384B messages	18.360 ms/op	39.225 ms/op	0.47
send data - 1000 65536B messages	200.00 ms/op	203.07 ms/op	0.98
enrSubnets - fastDeserialize 64 bits	776.00 ns/op	760.00 ns/op	1.02
enrSubnets - ssz BitVector 64 bits	278.00 ns/op	285.00 ns/op	0.98
enrSubnets - fastDeserialize 4 bits	105.00 ns/op	111.00 ns/op	0.95
enrSubnets - ssz BitVector 4 bits	270.00 ns/op	279.00 ns/op	0.97
prioritizePeers score -10:0 att 32-0.1 sync 2-0	209.56 us/op	207.17 us/op	1.01
prioritizePeers score 0:0 att 32-0.25 sync 2-0.25	239.64 us/op	238.14 us/op	1.01
prioritizePeers score 0:0 att 32-0.5 sync 2-0.5	347.22 us/op	351.72 us/op	0.99
prioritizePeers score 0:0 att 64-0.75 sync 4-0.75	614.25 us/op	605.29 us/op	1.01
prioritizePeers score 0:0 att 64-1 sync 4-1	711.89 us/op	708.23 us/op	1.01
array of 16000 items push then shift	1.3189 us/op	1.3208 us/op	1.00
LinkedList of 16000 items push then shift	7.0460 ns/op	6.8920 ns/op	1.02
array of 16000 items push then pop	66.108 ns/op	65.860 ns/op	1.00
LinkedList of 16000 items push then pop	6.0770 ns/op	5.9760 ns/op	1.02
array of 24000 items push then shift	1.9258 us/op	1.9477 us/op	0.99
LinkedList of 24000 items push then shift	6.5490 ns/op	6.4460 ns/op	1.02
array of 24000 items push then pop	95.580 ns/op	94.309 ns/op	1.01
LinkedList of 24000 items push then pop	6.2680 ns/op	5.9740 ns/op	1.05
intersect bitArray bitLen 8	4.7780 ns/op	4.7840 ns/op	1.00
intersect array and set length 8	29.829 ns/op	29.847 ns/op	1.00
intersect bitArray bitLen 128	23.879 ns/op	24.393 ns/op	0.98
intersect array and set length 128	503.68 ns/op	503.48 ns/op	1.00
bitArray.getTrueBitIndexes() bitLen 128	1.1470 us/op	1.0080 us/op	1.14
bitArray.getTrueBitIndexes() bitLen 248	1.8930 us/op	1.7550 us/op	1.08
bitArray.getTrueBitIndexes() bitLen 512	3.6860 us/op	3.5810 us/op	1.03
Full columns - reconstruct all 6 blobs	126.05 us/op	186.89 us/op	0.67
Full columns - reconstruct half of the blobs out of 6	97.653 us/op	105.47 us/op	0.93
Full columns - reconstruct single blob out of 6	31.984 us/op	35.001 us/op	0.91
Half columns - reconstruct all 6 blobs	397.70 ms/op	383.80 ms/op	1.04
Half columns - reconstruct half of the blobs out of 6	199.29 ms/op	191.22 ms/op	1.04
Half columns - reconstruct single blob out of 6	71.054 ms/op	67.634 ms/op	1.05
Full columns - reconstruct all 10 blobs	181.88 us/op	686.21 us/op	0.27
Full columns - reconstruct half of the blobs out of 10	101.49 us/op	106.63 us/op	0.95
Full columns - reconstruct single blob out of 10	35.151 us/op	32.184 us/op	1.09
Half columns - reconstruct all 10 blobs	660.99 ms/op	644.10 ms/op	1.03
Half columns - reconstruct half of the blobs out of 10	332.18 ms/op	323.94 ms/op	1.03
Half columns - reconstruct single blob out of 10	70.318 ms/op	67.492 ms/op	1.04
Full columns - reconstruct all 20 blobs	1.8517 ms/op	560.51 us/op	3.30
Full columns - reconstruct half of the blobs out of 20	258.02 us/op	176.01 us/op	1.47
Full columns - reconstruct single blob out of 20	32.265 us/op	32.861 us/op	0.98
Half columns - reconstruct all 20 blobs	1.3097 s/op	1.2682 s/op	1.03
Half columns - reconstruct half of the blobs out of 20	657.85 ms/op	637.62 ms/op	1.03
Half columns - reconstruct single blob out of 20	71.011 ms/op	67.081 ms/op	1.06
Set add up to 64 items then delete first	2.5906 us/op	2.1559 us/op	1.20
OrderedSet add up to 64 items then delete first	3.3959 us/op	3.3727 us/op	1.01
Set add up to 64 items then delete last	2.4441 us/op	2.4053 us/op	1.02
OrderedSet add up to 64 items then delete last	3.4861 us/op	3.2602 us/op	1.07
Set add up to 64 items then delete middle	2.2592 us/op	2.1581 us/op	1.05
OrderedSet add up to 64 items then delete middle	5.0096 us/op	4.8092 us/op	1.04
Set add up to 128 items then delete first	4.3122 us/op	4.3912 us/op	0.98
OrderedSet add up to 128 items then delete first	6.4300 us/op	6.7935 us/op	0.95
Set add up to 128 items then delete last	4.3087 us/op	4.2812 us/op	1.01
OrderedSet add up to 128 items then delete last	6.1934 us/op	6.1508 us/op	1.01
Set add up to 128 items then delete middle	4.0490 us/op	4.0470 us/op	1.00
OrderedSet add up to 128 items then delete middle	12.488 us/op	11.773 us/op	1.06
Set add up to 256 items then delete first	8.0051 us/op	8.2040 us/op	0.98
OrderedSet add up to 256 items then delete first	11.851 us/op	12.691 us/op	0.93
Set add up to 256 items then delete last	8.0902 us/op	8.2317 us/op	0.98
OrderedSet add up to 256 items then delete last	12.713 us/op	12.374 us/op	1.03
Set add up to 256 items then delete middle	8.1359 us/op	8.0575 us/op	1.01
OrderedSet add up to 256 items then delete middle	37.949 us/op	36.482 us/op	1.04
pass gossip attestations to forkchoice per slot	2.8562 ms/op	2.5782 ms/op	1.11
forkChoice updateHead vc 100000 bc 64 eq 0	459.12 us/op	426.24 us/op	1.08
forkChoice updateHead vc 600000 bc 64 eq 0	2.7322 ms/op	2.5981 ms/op	1.05
forkChoice updateHead vc 1000000 bc 64 eq 0	5.0060 ms/op	4.4096 ms/op	1.14
forkChoice updateHead vc 600000 bc 320 eq 0	2.8698 ms/op	2.6140 ms/op	1.10
forkChoice updateHead vc 600000 bc 1200 eq 0	2.6997 ms/op	2.6021 ms/op	1.04
forkChoice updateHead vc 600000 bc 7200 eq 0	3.1113 ms/op	2.9427 ms/op	1.06
forkChoice updateHead vc 600000 bc 64 eq 1000	3.3152 ms/op	3.1515 ms/op	1.05
forkChoice updateHead vc 600000 bc 64 eq 10000	3.3465 ms/op	3.1909 ms/op	1.05
forkChoice updateHead vc 600000 bc 64 eq 300000	7.3327 ms/op	7.1417 ms/op	1.03
computeDeltas 1400000 validators 0% inactive	13.517 ms/op	13.033 ms/op	1.04
computeDeltas 1400000 validators 10% inactive	12.956 ms/op	12.466 ms/op	1.04
computeDeltas 1400000 validators 20% inactive	12.263 ms/op	11.527 ms/op	1.06
computeDeltas 1400000 validators 50% inactive	8.9418 ms/op	8.7072 ms/op	1.03
computeDeltas 2100000 validators 0% inactive	20.435 ms/op	20.061 ms/op	1.02
computeDeltas 2100000 validators 10% inactive	22.153 ms/op	18.978 ms/op	1.17
computeDeltas 2100000 validators 20% inactive	18.106 ms/op	17.674 ms/op	1.02
computeDeltas 2100000 validators 50% inactive	13.746 ms/op	10.451 ms/op	1.32
altair processAttestation - 250000 vs - 7PWei normalcase	1.8949 ms/op	1.6932 ms/op	1.12
altair processAttestation - 250000 vs - 7PWei worstcase	2.8655 ms/op	2.4525 ms/op	1.17
altair processAttestation - setStatus - 1/6 committees join	103.86 us/op	105.92 us/op	0.98
altair processAttestation - setStatus - 1/3 committees join	212.75 us/op	198.72 us/op	1.07
altair processAttestation - setStatus - 1/2 committees join	290.76 us/op	281.57 us/op	1.03
altair processAttestation - setStatus - 2/3 committees join	376.41 us/op	374.96 us/op	1.00
altair processAttestation - setStatus - 4/5 committees join	516.19 us/op	507.68 us/op	1.02
altair processAttestation - setStatus - 100% committees join	607.80 us/op	599.06 us/op	1.01
altair processBlock - 250000 vs - 7PWei normalcase	4.3957 ms/op	2.9508 ms/op	1.49
altair processBlock - 250000 vs - 7PWei normalcase hashState	17.894 ms/op	14.191 ms/op	1.26
altair processBlock - 250000 vs - 7PWei worstcase	23.910 ms/op	20.076 ms/op	1.19
altair processBlock - 250000 vs - 7PWei worstcase hashState	43.711 ms/op	39.607 ms/op	1.10
phase0 processBlock - 250000 vs - 7PWei normalcase	1.3913 ms/op	1.3156 ms/op	1.06
phase0 processBlock - 250000 vs - 7PWei worstcase	17.954 ms/op	17.424 ms/op	1.03
altair processEth1Data - 250000 vs - 7PWei normalcase	293.02 us/op	299.84 us/op	0.98
getExpectedWithdrawals 250000 eb:1,eth1:1,we:0,wn:0,smpl:16	3.2610 us/op	3.4010 us/op	0.96
getExpectedWithdrawals 250000 eb:0.95,eth1:0.1,we:0.05,wn:0,smpl:220	20.347 us/op	21.603 us/op	0.94
getExpectedWithdrawals 250000 eb:0.95,eth1:0.3,we:0.05,wn:0,smpl:43	5.6150 us/op	5.8660 us/op	0.96
getExpectedWithdrawals 250000 eb:0.95,eth1:0.7,we:0.05,wn:0,smpl:19	3.5000 us/op	3.8070 us/op	0.92
getExpectedWithdrawals 250000 eb:0.1,eth1:0.1,we:0,wn:0,smpl:1021	91.027 us/op	98.067 us/op	0.93
getExpectedWithdrawals 250000 eb:0.03,eth1:0.03,we:0,wn:0,smpl:11778	1.4238 ms/op	1.3909 ms/op	1.02
getExpectedWithdrawals 250000 eb:0.01,eth1:0.01,we:0,wn:0,smpl:16384	1.8674 ms/op	1.8241 ms/op	1.02
getExpectedWithdrawals 250000 eb:0,eth1:0,we:0,wn:0,smpl:16384	1.8503 ms/op	1.8228 ms/op	1.02
getExpectedWithdrawals 250000 eb:0,eth1:0,we:0,wn:0,nocache,smpl:16384	3.7834 ms/op	3.8248 ms/op	0.99
getExpectedWithdrawals 250000 eb:0,eth1:1,we:0,wn:0,smpl:16384	2.0849 ms/op	2.0672 ms/op	1.01
getExpectedWithdrawals 250000 eb:0,eth1:1,we:0,wn:0,nocache,smpl:16384	4.0627 ms/op	4.1838 ms/op	0.97
Tree 40 250000 create	319.27 ms/op	314.07 ms/op	1.02
Tree 40 250000 get(125000)	100.34 ns/op	97.612 ns/op	1.03
Tree 40 250000 set(125000)	1.0651 us/op	1.0328 us/op	1.03
Tree 40 250000 toArray()	10.546 ms/op	9.3883 ms/op	1.12
Tree 40 250000 iterate all - toArray() + loop	15.754 ms/op	9.6914 ms/op	1.63
Tree 40 250000 iterate all - get(i)	40.276 ms/op	35.182 ms/op	1.14
Array 250000 create	2.1196 ms/op	2.0867 ms/op	1.02
Array 250000 clone - spread	673.38 us/op	674.19 us/op	1.00
Array 250000 get(125000)	0.29600 ns/op	0.30200 ns/op	0.98
Array 250000 set(125000)	0.29800 ns/op	0.30600 ns/op	0.97
Array 250000 iterate all - loop	57.528 us/op	57.491 us/op	1.00
phase0 afterProcessEpoch - 250000 vs - 7PWei	39.596 ms/op	39.730 ms/op	1.00
Array.fill - length 1000000	2.0588 ms/op	2.2538 ms/op	0.91
Array push - length 1000000	9.0964 ms/op	7.7584 ms/op	1.17
Array.get	0.20708 ns/op	0.20894 ns/op	0.99
Uint8Array.get	0.26307 ns/op	0.24656 ns/op	1.07
phase0 beforeProcessEpoch - 250000 vs - 7PWei	19.037 ms/op	13.121 ms/op	1.45
altair processEpoch - mainnet_e81889	251.89 ms/op	250.39 ms/op	1.01
mainnet_e81889 - altair beforeProcessEpoch	20.719 ms/op	13.979 ms/op	1.48
mainnet_e81889 - altair processJustificationAndFinalization	5.8310 us/op	5.3810 us/op	1.08
mainnet_e81889 - altair processInactivityUpdates	3.6870 ms/op	3.5392 ms/op	1.04
mainnet_e81889 - altair processRewardsAndPenalties	21.287 ms/op	16.581 ms/op	1.28
mainnet_e81889 - altair processRegistryUpdates	546.00 ns/op	565.00 ns/op	0.97
mainnet_e81889 - altair processSlashings	143.00 ns/op	145.00 ns/op	0.99
mainnet_e81889 - altair processEth1DataReset	133.00 ns/op	141.00 ns/op	0.94
mainnet_e81889 - altair processEffectiveBalanceUpdates	1.7652 ms/op	1.5245 ms/op	1.16
mainnet_e81889 - altair processSlashingsReset	707.00 ns/op	711.00 ns/op	0.99
mainnet_e81889 - altair processRandaoMixesReset	1.2280 us/op	1.1980 us/op	1.03
mainnet_e81889 - altair processHistoricalRootsUpdate	137.00 ns/op	151.00 ns/op	0.91
mainnet_e81889 - altair processParticipationFlagUpdates	437.00 ns/op	457.00 ns/op	0.96
mainnet_e81889 - altair processSyncCommitteeUpdates	111.00 ns/op	125.00 ns/op	0.89
mainnet_e81889 - altair afterProcessEpoch	41.761 ms/op	41.648 ms/op	1.00
capella processEpoch - mainnet_e217614	863.93 ms/op	789.28 ms/op	1.09
mainnet_e217614 - capella beforeProcessEpoch	83.413 ms/op	55.593 ms/op	1.50
mainnet_e217614 - capella processJustificationAndFinalization	6.9710 us/op	5.5390 us/op	1.26
mainnet_e217614 - capella processInactivityUpdates	17.146 ms/op	11.742 ms/op	1.46
mainnet_e217614 - capella processRewardsAndPenalties	97.378 ms/op	90.332 ms/op	1.08
mainnet_e217614 - capella processRegistryUpdates	4.5160 us/op	4.5830 us/op	0.99
mainnet_e217614 - capella processSlashings	138.00 ns/op	136.00 ns/op	1.01
mainnet_e217614 - capella processEth1DataReset	139.00 ns/op	132.00 ns/op	1.05
mainnet_e217614 - capella processEffectiveBalanceUpdates	20.822 ms/op	5.6406 ms/op	3.69
mainnet_e217614 - capella processSlashingsReset	704.00 ns/op	703.00 ns/op	1.00
mainnet_e217614 - capella processRandaoMixesReset	1.4200 us/op	1.1330 us/op	1.25
mainnet_e217614 - capella processHistoricalRootsUpdate	135.00 ns/op	138.00 ns/op	0.98
mainnet_e217614 - capella processParticipationFlagUpdates	456.00 ns/op	429.00 ns/op	1.06
mainnet_e217614 - capella afterProcessEpoch	109.72 ms/op	109.15 ms/op	1.01
phase0 processEpoch - mainnet_e58758	298.25 ms/op	274.72 ms/op	1.09
mainnet_e58758 - phase0 beforeProcessEpoch	69.002 ms/op	51.000 ms/op	1.35
mainnet_e58758 - phase0 processJustificationAndFinalization	6.2180 us/op	5.8590 us/op	1.06
mainnet_e58758 - phase0 processRewardsAndPenalties	16.399 ms/op	14.896 ms/op	1.10
mainnet_e58758 - phase0 processRegistryUpdates	2.2680 us/op	2.2970 us/op	0.99
mainnet_e58758 - phase0 processSlashings	138.00 ns/op	146.00 ns/op	0.95
mainnet_e58758 - phase0 processEth1DataReset	219.00 ns/op	146.00 ns/op	1.50
mainnet_e58758 - phase0 processEffectiveBalanceUpdates	849.26 us/op	1.1061 ms/op	0.77
mainnet_e58758 - phase0 processSlashingsReset	947.00 ns/op	844.00 ns/op	1.12
mainnet_e58758 - phase0 processRandaoMixesReset	1.3500 us/op	1.0860 us/op	1.24
mainnet_e58758 - phase0 processHistoricalRootsUpdate	137.00 ns/op	152.00 ns/op	0.90
mainnet_e58758 - phase0 processParticipationRecordUpdates	1.2700 us/op	1.0490 us/op	1.21
mainnet_e58758 - phase0 afterProcessEpoch	33.891 ms/op	33.054 ms/op	1.03
phase0 processEffectiveBalanceUpdates - 250000 normalcase	1.0250 ms/op	1.0143 ms/op	1.01
phase0 processEffectiveBalanceUpdates - 250000 worstcase 0.5	1.5923 ms/op	1.6602 ms/op	0.96
altair processInactivityUpdates - 250000 normalcase	10.835 ms/op	10.696 ms/op	1.01
altair processInactivityUpdates - 250000 worstcase	10.796 ms/op	10.736 ms/op	1.01
phase0 processRegistryUpdates - 250000 normalcase	2.1890 us/op	2.3190 us/op	0.94
phase0 processRegistryUpdates - 250000 badcase_full_deposits	145.69 us/op	150.19 us/op	0.97
phase0 processRegistryUpdates - 250000 worstcase 0.5	65.727 ms/op	63.709 ms/op	1.03
altair processRewardsAndPenalties - 250000 normalcase	16.215 ms/op	15.058 ms/op	1.08
altair processRewardsAndPenalties - 250000 worstcase	15.205 ms/op	14.581 ms/op	1.04
phase0 getAttestationDeltas - 250000 normalcase	5.3972 ms/op	5.4224 ms/op	1.00
phase0 getAttestationDeltas - 250000 worstcase	19.386 ms/op	5.4304 ms/op	3.57
phase0 processSlashings - 250000 worstcase	59.602 us/op	60.892 us/op	0.98
altair processSyncCommitteeUpdates - 250000	10.131 ms/op	10.318 ms/op	0.98
BeaconState.hashTreeRoot - No change	175.00 ns/op	185.00 ns/op	0.95
BeaconState.hashTreeRoot - 1 full validator	82.434 us/op	59.312 us/op	1.39
BeaconState.hashTreeRoot - 32 full validator	853.16 us/op	671.32 us/op	1.27
BeaconState.hashTreeRoot - 512 full validator	8.4053 ms/op	6.1997 ms/op	1.36
BeaconState.hashTreeRoot - 1 validator.effectiveBalance	92.533 us/op	73.403 us/op	1.26
BeaconState.hashTreeRoot - 32 validator.effectiveBalance	1.4520 ms/op	1.1053 ms/op	1.31
BeaconState.hashTreeRoot - 512 validator.effectiveBalance	18.236 ms/op	13.289 ms/op	1.37
BeaconState.hashTreeRoot - 1 balances	77.629 us/op	60.447 us/op	1.28
BeaconState.hashTreeRoot - 32 balances	757.00 us/op	568.15 us/op	1.33
BeaconState.hashTreeRoot - 512 balances	6.1304 ms/op	4.8226 ms/op	1.27
BeaconState.hashTreeRoot - 250000 balances	128.54 ms/op	116.87 ms/op	1.10
aggregationBits - 2048 els - zipIndexesInBitList	19.968 us/op	19.810 us/op	1.01
regular array get 100000 times	22.966 us/op	23.486 us/op	0.98
wrappedArray get 100000 times	22.985 us/op	23.329 us/op	0.99
arrayWithProxy get 100000 times	13.207 ms/op	10.487 ms/op	1.26
ssz.Root.equals	21.195 ns/op	21.961 ns/op	0.97
byteArrayEquals	21.017 ns/op	21.720 ns/op	0.97
Buffer.compare	9.2220 ns/op	9.0500 ns/op	1.02
processSlot - 1 slots	9.8560 us/op	8.5310 us/op	1.16
processSlot - 32 slots	2.0055 ms/op	1.5888 ms/op	1.26
getEffectiveBalanceIncrementsZeroInactive - 250000 vs - 7PWei	3.8443 ms/op	2.7737 ms/op	1.39
getCommitteeAssignments - req 1 vs - 250000 vc	1.6693 ms/op	1.7001 ms/op	0.98
getCommitteeAssignments - req 100 vs - 250000 vc	3.4065 ms/op	3.4569 ms/op	0.99
getCommitteeAssignments - req 1000 vs - 250000 vc	3.6674 ms/op	3.6988 ms/op	0.99
findModifiedValidators - 10000 modified validators	779.61 ms/op	697.06 ms/op	1.12
findModifiedValidators - 1000 modified validators	439.91 ms/op	425.50 ms/op	1.03
findModifiedValidators - 100 modified validators	285.12 ms/op	266.83 ms/op	1.07
findModifiedValidators - 10 modified validators	240.84 ms/op	151.29 ms/op	1.59
findModifiedValidators - 1 modified validators	184.90 ms/op	158.11 ms/op	1.17
findModifiedValidators - no difference	186.47 ms/op	155.62 ms/op	1.20
migrate state 1500000 validators, 3400 modified, 2000 new	3.1181 s/op	2.5643 s/op	1.22
RootCache.getBlockRootAtSlot - 250000 vs - 7PWei	3.7700 ns/op	3.7900 ns/op	0.99
state getBlockRootAtSlot - 250000 vs - 7PWei	406.18 ns/op	289.43 ns/op	1.40
computeProposerIndex 100000 validators	1.3852 ms/op	1.3077 ms/op	1.06
getNextSyncCommitteeIndices 1000 validators	2.8797 ms/op	2.8446 ms/op	1.01
getNextSyncCommitteeIndices 10000 validators	26.319 ms/op	25.072 ms/op	1.05
getNextSyncCommitteeIndices 100000 validators	91.300 ms/op	86.260 ms/op	1.06
computeProposers - vc 250000	575.00 us/op	551.63 us/op	1.04
computeEpochShuffling - vc 250000	39.710 ms/op	38.163 ms/op	1.04
getNextSyncCommittee - vc 250000	9.5754 ms/op	9.3464 ms/op	1.02
nodejs block root to RootHex using toHex	111.04 ns/op	117.78 ns/op	0.94
nodejs block root to RootHex using toRootHex	72.527 ns/op	76.711 ns/op	0.95
nodejs fromHex(blob)	873.80 us/op	738.65 us/op	1.18
nodejs fromHexInto(blob)	659.35 us/op	649.73 us/op	1.01
nodejs block root to RootHex using the deprecated toHexString	491.93 ns/op	486.00 ns/op	1.01
nodejs byteArrayEquals 32 bytes (block root)	26.230 ns/op	26.047 ns/op	1.01
nodejs byteArrayEquals 48 bytes (pubkey)	37.939 ns/op	37.581 ns/op	1.01
nodejs byteArrayEquals 96 bytes (signature)	36.563 ns/op	38.402 ns/op	0.95
nodejs byteArrayEquals 1024 bytes	42.789 ns/op	42.093 ns/op	1.02
nodejs byteArrayEquals 131072 bytes (blob)	1.7807 us/op	1.7619 us/op	1.01
browser block root to RootHex using toHex	146.86 ns/op	146.56 ns/op	1.00
browser block root to RootHex using toRootHex	130.52 ns/op	130.38 ns/op	1.00
browser fromHex(blob)	1.6634 ms/op	1.5155 ms/op	1.10
browser fromHexInto(blob)	654.81 us/op	654.63 us/op	1.00
browser block root to RootHex using the deprecated toHexString	337.67 ns/op	342.92 ns/op	0.98
browser byteArrayEquals 32 bytes (block root)	30.459 ns/op	28.069 ns/op	1.09
browser byteArrayEquals 48 bytes (pubkey)	43.844 ns/op	39.599 ns/op	1.11
browser byteArrayEquals 96 bytes (signature)	83.357 ns/op	74.179 ns/op	1.12
browser byteArrayEquals 1024 bytes	816.25 ns/op	753.96 ns/op	1.08
browser byteArrayEquals 131072 bytes (blob)	95.958 us/op	95.144 us/op	1.01

by benchmarkbot/action

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 6ff9833288

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

[chatgpt-codex-connector]

Avoid caching future-slot head during prepareNextSlot

Passing prepareSlot into recomputeForkChoiceHead makes fork choice recompute and cache this.head using next-slot semantics while the node is still in the current slot. After this runs (~67% into slot N), any code that reads the cached head via forkChoice.getHead() before the next slot tick (for example late validator/API flows) can observe a head selected with slot N+1 FULL-vs-EMPTY tie-breaker instead of slot N rules, which can produce premature head-dependent outputs.

Useful? React with 👍 / 👎.

[twoeths]

after 67% of the slot n I don't see any flows want to use head
if there is in the future, it should compute head again and cache it, especially when there are more ptc messages received
so caching here is not an issue

we can make an ephemeral head in this case, but it will make the current design more complex
the above mechanism should work

[nflaig]

doesn't this cause the head to be recomputed so any caller to forkChoice.getHead() in slot N would get the head of the next slot N+1?

[twoeths]

doesn't this cause the head to be recomputed so any caller to forkChoice.getHead() in slot N would get the head of the next slot N+1?

yes. From getting both EMPTY + FULL till the next slot, we need to fairly choose one of them, recompute head and cache. My interpretation would be: when preparing for the next slot, if we have both EMPTY and FULL variant, we need to get through tie-breaker logic to get the winner

[nflaig]

More general, we run prepareNextSlot at 8 seconds into the slot but with gloas, the payload timeliness deadline and ptc vote deadline is at 9 seconds (PAYLOAD_ATTESTATION_DUE_BPS = 7500), usually if the payload is timely it will arrive before and PTC should vote early, so in the happy case we likley get payload + all ptc votes by 8 seconds into the slot. But if the builder does not reveal and it's EMPTY, then PTC will only cast their votes at 9 seconds into the slot, and we will very likely only receive them at ~10 seconds, or later.

We need to carefully review how if the timings here make sense to call recomputeForkChoiceHead and also if we are the next proposer and issue an fcu to the EL this timing is relevant so we prepare the payload on the correct block hash

[twoeths]

good call, that was already a TODO GLOAS
given this epoch transition metric on mainnet, I think it's safe to run this 10s into the slot, we should recheck when we're close to mainnet because gloas could introduce some epoch transition time

[wemeetagain]

Discussed this offline with @nflaig

would prefer an interface like so:

recomputeForkChoiceHead(
  caller: ForkchoiceCaller,
  opts?: {prepareNextSlot?: boolean}
): ProtoBlock;

// Call site
const headBlock = this.chain.recomputeForkChoiceHead(
  ForkchoiceCaller.prepareNextSlot, {prepareNextSlot: true}
);

Pass optional opts to forkChoice.updateHead and do the const currentSlot = opts?.prepareNextSlot ? store.currentSlot + 1 : store.curentSlot; there

[nflaig]

just realized we already have ForkchoiceCaller.prepareNextSlot, so the boolean here seems redundant

[nflaig]

after my spec changes we can keep previous PREPARE_NEXT_SLOT_BPS but the fcu() call needs to be delayed or we need to reach ptc threshold

something like this could work

await Promise.race([
      sleep(SOME_BPS_VALUE - this.clock.msFromSlot(slot), signal),
      this.emitter.waitForPtcThreshold(slot),
    ]);