Commit Graph

284 Commits

Author SHA1 Message Date
ericsson 4431e399b1 Fix bugs:
- `range(len(col))`` instead of `range(col)`
- `beacon_parent_block.body.shard_transitions` instead of `beacon_parent_block.shard_transitions`
- `shard_states[len(shard_states)-1]` instead of `shard_states[-1]`
2020-05-13 16:13:33 +03:00
Hsiao-Wei Wang cdd0ed0f7b
Update to IETF BLS draft-irtf-cfrg-bls-signature-02 2020-05-09 11:48:48 +08:00
Danny Ryan 08652f543b
Merge branch 'v012x' into empty-bits-case 2020-05-07 10:17:26 -06:00
Danny Ryan 38f29ba0a8
remove PERSISTENT_COMMITTEE_PERIOD in favor of SHARD_COMMITTEE_PERIOD 2020-05-05 11:44:53 -06:00
Hsiao-Wei Wang 79b1b4bdbe
Add `(shard, shard_root)` to `LatestMessage` 2020-05-05 21:42:18 +08:00
Hsiao-Wei Wang fca1bbccb9
Remove `get_filtered_shard_block_tree` 2020-05-05 21:42:13 +08:00
Hsiao-Wei Wang 8fafb6a9e5
Make `ShardStore` an independent object 2020-05-05 21:42:06 +08:00
Hsiao-Wei Wang cddf9cf114
Refactor 2020-05-05 21:42:00 +08:00
Hsiao-Wei Wang dab5a936c4
wip shard fork choice rule 2020-05-05 21:41:46 +08:00
Hsiao-Wei Wang ee4c866575
Merge branch 'dev' into dankrad-custody-256bit-merging 2020-05-05 15:22:36 +08:00
Hsiao-Wei Wang 7a770186b5
Reorg beacon-chain spec a bit 2020-05-02 02:32:37 +08:00
Hsiao-Wei Wang 4558c7db4e
Reorg the file structure 2020-05-02 02:32:37 +08:00
Hsiao-Wei Wang b43e24acb6
specs/phase1/fraud-proofs.md -> specs/phase1/shard-transition.md 2020-05-02 02:32:37 +08:00
Hsiao-Wei Wang ff85025113
PR feedback from terence 2020-05-02 02:32:37 +08:00
Hsiao-Wei Wang e758fb76c2
Check `head_shard_root` of all `transition_attestations` 2020-05-02 02:32:36 +08:00
Hsiao-Wei Wang 524ba166d1
[squashed] shard chain updates wip
Fix wrong field names

Fix `build_attestation_data` and other PR feedback from Danny and
Terence

1. Rename `get_previous_slot` to `compute_previous_slot`
2. Break down `build_empty_block` into
`get_state_and_beacon_parent_root_at_slot`, use it in
`build_shard_block`
3. Set defult `slot` to `shard_state.slot + 1` in `build_shard_block`

Update `verify_shard_block_message`: check beacon_parent_root at fork
choice rule stage instead of state transition

Fix  `beacon-chain.md`

1. Fix typo `attestation.slot == state.slot` -> `attestation.data.slot == state.slot` in `is_winning_attestation`
2. Check `verify_shard_transition_false_positives` **after** `process_operations`
3. Fix `shard_attestations` filter in `process_crosslinks`: since attestations come from block, should use `attestation.data.slot + MIN_ATTESTATION_INCLUSION_DELAY == state.slot`
4. [TBD] Allow empty `light_client_signature` to make the tests pass
5. [TBD] Add `is_shard_attestation`, filter out empty `ShardTransition()`

Rework `test_process_crosslink`

Add basic phase 1 `test_blocks`

Add more test cases

Revert `is_shard_attestation` and fix test cases backward compatibility.

Remove `test_process_beacon_block_no_shard_transition` and consider it as invalid case.
2020-05-02 02:32:31 +08:00
Hsiao-Wei Wang c8a473ba24
Apply suggestions from code review
Co-Authored-By: Danny Ryan <dannyjryan@gmail.com>
2020-05-02 02:32:30 +08:00
Hsiao-Wei Wang 40483b587b
[squashed] shard chain updates wip
Use `ShardBlock` in `shard_state_transition`

PR feedback

1. Rename `ShardState.data` -> `ShardState.transition_digest`
2. Rename `compute_shard_transition_data` to `compute_shard_transition_digest`
3. Add `assert state.slot > PHASE_1_GENESIS_SLOT` just in case, may move it later

Add `get_post_shard_state` as a pure function wrapper
2020-05-02 02:32:24 +08:00
Hsiao-Wei Wang 85d5a9abaf
[squashed] shard chain updates wip
PR feedback from Danny and some refactor

1. Add stub `PHASE_1_GENESIS_SLOT`
2. Rename `get_updated_gasprice`  to `compute_updated_gasprice`
3. Rename `compute_shard_data_roots` to `compute_shard_body_roots`

Apply shard transition for the skipped slots

Refactor `shard_state_transition`

Get `beacon_parent_root` from offset slot

Add more test

Add `verify_shard_block_message`

Add `> 0`

Keep `beacon_parent_block` unchanged in `is_valid_fraud_proof`

Remove some lines

Fix type

Refactor + simplify skipped slot processing
2020-05-02 02:31:54 +08:00
Hsiao-Wei Wang 9724cb832d
Apply suggestions from code review from @djrtwo
Co-Authored-By: Danny Ryan <dannyjryan@gmail.com>
2020-05-02 02:31:53 +08:00
Hsiao-Wei Wang afa12caf1f
Refactor `get_shard_state_transition_result` 2020-05-02 02:31:53 +08:00
Hsiao-Wei Wang 4e8a7ff115
[squashed] shard transition wip
Fix the wrong `get_shard_proposer_index` parameters order

Phase 1 WIP

Add shard transition basic test

Fix lint error

Fix
2020-05-02 02:31:10 +08:00
Hsiao-Wei Wang 849d3f83bf
Apply @terencechain 's review feedback
Co-Authored-By: terence tsao <terence@prysmaticlabs.com>
2020-05-02 02:31:09 +08:00
Hsiao-Wei Wang 247a6c8fca
Add `verify_fraud_proof` function 2020-05-02 02:31:09 +08:00
Hsiao-Wei Wang be50020bf8
Refactor `get_light_client_committee` to similar to
`get_shard_committee`
2020-05-02 02:31:08 +08:00
Hsiao-Wei Wang 5f69afea38
Make `shard_state_transition` more like beacon state_transition function 2020-05-02 02:31:08 +08:00
Hsiao-Wei Wang e9f1e4186d
Add `proposer_index` to shard block 2020-05-02 02:31:08 +08:00
Hsiao-Wei Wang feb27a14be
beacon-chain.md: Replace block wrapper with signable pattern 2020-05-02 02:31:08 +08:00
protolambda 09cae4b3cc
Handle empty-aggregation-bits case, and add tests. See #1713 2020-05-01 15:17:41 +02:00
Dankrad Feist 964bf42335
Fix type 2020-05-01 00:32:02 +01:00
Dankrad Feist d30f11a781
Fix lint 2020-05-01 00:16:00 +01:00
Dankrad Feist d58d7627b7
Fix toc 2020-04-30 19:25:18 +01:00
Dankrad Feist 0e2931b9b3
All tests passed 2020-04-28 01:09:20 +01:00
Dankrad Feist 2449db1bb6
Phase 1 block tests are working 2020-04-27 16:08:49 +01:00
Dankrad Feist ab2ee0e2c2
Restoring chunk challenges and testing 2020-04-24 17:06:27 +01:00
Dankrad Feist 907c56dabd
Fix ToC 2020-04-05 15:47:59 +01:00
Dankrad Feist c3c24b4fc4
Fix lint 2020-04-05 15:35:11 +01:00
Dankrad Feist bf34fdf023
Fix ToC 2020-04-05 15:10:09 +01:00
Dankrad Feist ca6af0c2e9
256-bit custody atoms for better alignment with rest of the spec and greater efficiency 2020-04-05 14:39:00 +01:00
Danny Ryan f135eff021
add lookahed for shard subnets for beacon committee in validator guide 2020-04-03 13:23:38 -06:00
Danny Ryan e86c5ef41d
final PR nitpicks 2020-04-03 10:29:35 -06:00
Danny Ryan 246b46771e
address @hwwhww feedback 2020-04-03 09:46:57 -06:00
Danny Ryan f2c2da95ed
add compute_offset_slots 2020-04-03 09:19:56 -06:00
Danny Ryan d61b2991a0
fix lint 2020-04-02 16:58:39 -06:00
Danny Ryan 6067c511c5
add light client to phase 1 validator 2020-04-02 16:48:02 -06:00
Danny Ryan d789f3d32d
getting phase 1 val guide in place 2020-04-02 15:09:45 -06:00
Danny Ryan 613f368c00
fix call to get_beacon_committee in process_crosslink_for_shard 2020-04-01 12:20:32 -06:00
Danny Ryan 16208790a5
Merge pull request #1699 from ethereum/hwwhww/unpack_compact_validator
Add `unpack_compact_validator` back
2020-03-31 10:40:45 -06:00
Hsiao-Wei Wang f82cdb7e68
beacon-chain.md: add `unpack_compact_validator` for `light-client-sync.md` 2020-03-31 22:13:48 +08:00
Hsiao-Wei Wang 071f6b5126
Gasprice rework: use `MIN_GASPRICE` as the initial gasprice and change `MIN_GASPRICE` to 8 gwei 2020-03-31 22:09:20 +08:00
Danny Ryan 073f78efa1
Merge branch 'dev' into phase1-tests 2020-03-29 17:04:25 -06:00
Danny Ryan d299b06a1c
fix custody bit calculation format 2020-03-16 09:52:27 -06:00
Dankrad Feist 9b7e0ab2be
Fix error in custody bit computation 2020-03-13 17:15:25 +00:00
Danny Ryan 1293320675
Merge branch 'dev' into phase1-tests 2020-03-12 07:12:41 -06:00
Danny Ryan e2918c6364
Merge pull request #1626 from ethereum/proposer-index
add proposer index to BeaconBlock
2020-03-10 15:10:07 -06:00
Danny Ryan 3b7704a78f
Merge pull request #1649 from ethereum/eth1-voting-period-in-epochs
Eth1 voting period in epochs [updated for configs/phase1/tests compat.]
2020-03-10 13:24:03 -06:00
protolambda 2d7a292d36
eth1 vote period constant in epochs: update configs, phase1, tests 2020-03-10 18:36:53 +01:00
Danny Ryan 92eef0e00b
fix light client sig verification in phase 1 2020-03-09 14:52:30 -06:00
Danny Ryan 7e04989e29
add genesis_validators_root to beaconstate and utilize in sig domain separation as well as fork separation 2020-03-05 09:21:32 -07:00
Hsiao-Wei Wang b4c7481b35
Fix the misc table 2020-03-03 01:28:58 +01:00
Danny Ryan 186d4258b6
fix shard offsets 2020-02-28 13:20:37 -06:00
Danny Ryan 9718d206a7
fix attester slahsing test 2020-02-26 11:20:19 -06:00
Danny Ryan 721f605a91
Merge branch 'dev' into phase1-tests 2020-02-22 12:10:35 -06:00
Danny Ryan 4c1fc9bffa
work through phase 1 tests 2020-02-22 12:06:31 -06:00
Danny Ryan 97fa3741af
working through test issues 2020-02-22 09:30:33 -06:00
Danny Ryan ceb6633eb9
working through phase 1 attestation testing 2020-02-22 09:24:14 -06:00
Danny Ryan d414aac933
rework process_attestation and work through tests 2020-02-22 09:22:49 -06:00
Danny Ryan 757f5a31dd
add proposer index and add/modify tests 2020-02-18 11:38:17 -06:00
Danny Ryan 129aa02cb3
support tests with SLOTS_PER_EPOCH * 256 vals 2020-02-10 17:56:05 -07:00
terence tsao fe58c78da8
Fix table 2020-01-29 14:08:48 -08:00
vbuterin 52fb929978
Update specs/core/1_beacon-chain.md 2020-01-28 17:32:57 -07:00
vbuterin 2a91b43eaf
Remove shard block chunking
Only store a 32 byte root for every shard block

Rationale: originally, I added shard block chunking (store 4 chunks for every shard block instead of one root) to facilitate construction of data availability roots. However, it turns out that there is an easier technique. Set the width of the data availability rectangle's rows to be 1/4 the max size of a shard block, so each block would fill multiple rows. Then, non-full blocks will generally create lots of zero rows. For example if the block bodies are `31415926535` and `897932` with a max size of 24 bytes, the rows might look like this:

```
31415926
53500000
00000000
89793200
00000000
00000000
```
Zero rows would extend rightward to complete zero rows, and when extending downward we can count the number of zero rows, and reduce the number of extra rows that we make, so we only make a new row for every nonzero row in the original data. This way we get only a close-to-optimal ~4-5x blowup in the data even if the data has zero rows in the middle.
2020-01-28 17:31:51 -07:00
protolambda d9f62f9303
Remerkleable - merkle tree based ssz for better and faster spec 2020-01-25 00:43:43 +01:00
Danny Ryan 7a412534d9
remove test_shard_blocks (outdated) and reduce PERSISTENT_COMMITTEE_PERIOD in minimal config 2020-01-15 18:17:07 -07:00
Danny Ryan c0b69e531f
cycle through committee indexes instead of through active shards when forming crosslinks 2020-01-15 17:43:11 -07:00
Danny Ryan 3c07b2c954
Make phase 0 fork choice more modular to more easily adopt for slight modifications in phase 1 2020-01-15 16:40:50 -07:00
protolambda 5785b4fc5b
custody bits temporary solution 2020-01-14 01:59:01 +01:00
protolambda f04a686db7
doctoc 2020-01-14 01:42:19 +01:00
protolambda f6f8bd5350
no custody bits fallback 2020-01-14 01:36:16 +01:00
protolambda 702b253361
update configs for phase1 2020-01-13 19:50:36 +01:00
protolambda 419b6a3250
config change, need more space for worst-case reveals 2020-01-13 19:00:24 +01:00
protolambda 507a9afbfb
apply custody bit fix suggestion from Dankrad 2020-01-13 18:57:56 +01:00
protolambda 4732c7beb1
merge in dev (v0.10) and fix reorg/lint issues 2020-01-13 18:55:21 +01:00
Danny Ryan 676e216beb
reorg specs by fork and move ssz out to own folder. make all of the build and link changes to support move 2020-01-10 11:55:13 -07:00