[Merged by Bors] - Update to spec v1.0.0-rc.0 and BLSv4 #1765

michaelsproul · 2020-10-13T06:02:11Z

Issue Addressed

Closes #1504
Closes #1505
Replaces #1703
Closes #1707

Proposed Changes

Update BLST and Milagro to versions compatible with BLSv4 spec
Update Lighthouse to spec v1.0.0-rc.0, and update EF test vectors
Use the v1.0.0 constants for MainnetEthSpec.
Rename InteropEthSpec -> V012LegacyEthSpec
- Change all constants to suit the mainnet v0.12.3 specification (i.e., Medalla).
Deprecate the --spec flag for the lighthouse binary
- This value is now obtained from the config_name field of the YamlConfig.
  - Built in testnet YAML files have been updated.
- Ignore the --spec value, if supplied, log a warning that it will be deprecated
- lcli still has the spec flag, that's fine because it's dev tooling.
Remove the E: EthSpec from YamlConfig
- This means we need to deser the genesis BeaconState on-demand, but this is fine.
Swap the old "minimal", "mainnet" strings over to the new EthSpecId enum.
Always require a CONFIG_NAME field in YamlConfig (it used to have a default).

Additional Info

Lots of breaking changes, do not merge! ~~We will likely need a Lighthouse v0.4.0 branch, and possibly a long-term v0.3.0 branch to keep Medalla alive~~.

michaelsproul · 2020-10-13T08:41:45Z

Likely need to delete the Altona, Medalla and Zinken configs from this branch, or else attempt some backwards-compatibility measures (that's why the tests are failing currently)

paulhauner · 2020-10-13T23:23:40Z

Likely need to delete the Altona, Medalla and Zinken configs from this branch, or else attempt some backwards-compatibility measures (that's why the tests are failing currently)

Would the incompatibilities cause a consensus split or are they just regarding our configs?

michaelsproul · 2020-10-13T23:44:35Z

Would the incompatibilities cause a consensus split or are they just regarding our configs?

I've thought about this some more and I think we would need to make an EthSpec instance for SpecV12, and a corresponding ChainSpec. That would get us most of the way there.

The only risk then would be that using BLSv4 on Medalla creates a chain split, because it is stricter about infinity pubkeys and signatures. As far as I understand it would be trivial for someone to create a pair of secret keys S and -S so that an aggregate signature of a message signed by S and a message signed by -S is infinite... although this is based on a very limited understanding and I'd like to summon @kirk-baird to confirm/deny.

If that is the case, I think the separate branch for Medalla compatibility makes sense, or maybe a feature flag...

kirk-baird · 2020-10-14T00:42:18Z

Yep it would be very trivial to cause a consensus split on Medalla if we update since we previously allowed secret key = 0 => public key = infinity and signing any message would have signature = infinity. Where we now reject any public key = infinity.

As a result it would be trivial to make a deposit for the public key = infinity and old version would accept the deposit and new ones reject it.

P.s.

As far as I understand it would be trivial for someone to create a pair of secret keys S and -S so that an aggregate signature of a message signed by S and a message signed by -S is infinite

This use case is still valid. We still need to accept AggregateSignature = infinity to handle this. We only reject PublicKey = infinity.

kirk-baird · 2020-10-20T07:05:14Z

This should also solve #1707

* Update blst and milagro_bls subgroup checking Signed-off-by: Kirk Baird <[email protected]> * cargo fmt Signed-off-by: Kirk Baird <[email protected]>

paulhauner · 2020-10-26T04:17:42Z

I'm comandeering this PR!

With @michaelsproul's permission, of course.

paulhauner · 2020-10-28T05:48:00Z

All ready! I can't select you as a reviewer though @michaelsproul 😕

michaelsproul

Looks good on the whole @paulhauner! Thanks for taking this over and getting it over the line.

I just have a few stylistic suggestions that should be addressed, then we can merge!

common/eth2_testnet_config/src/lib.rs

consensus/types/src/chain_spec.rs

consensus/types/src/eth_spec.rs

lighthouse/src/main.rs

testing/ef_tests/tests/tests.rs

paulhauner · 2020-10-28T06:57:26Z

Thanks! All comments addressed. I've been jigging around in my .vimrc, perhaps I caused those weird string wrappings..

Could you please double check I set the proportional_slashing_multiplier correctly in the YAML configs, please? :)

michaelsproul · 2020-10-28T07:10:03Z

Yep! All good! And my Medalla node is happy with them too ☺️

michaelsproul · 2020-10-28T07:10:12Z

Merge at will!

paulhauner · 2020-10-28T22:15:51Z

bors r+

## Issue Addressed Closes #1504 Closes #1505 Replaces #1703 Closes #1707 ## Proposed Changes * Update BLST and Milagro to versions compatible with BLSv4 spec * Update Lighthouse to spec v1.0.0-rc.0, and update EF test vectors * Use the v1.0.0 constants for `MainnetEthSpec`. * Rename `InteropEthSpec` -> `V012LegacyEthSpec` * Change all constants to suit the mainnet `v0.12.3` specification (i.e., Medalla). * Deprecate the `--spec` flag for the `lighthouse` binary * This value is now obtained from the `config_name` field of the `YamlConfig`. * Built in testnet YAML files have been updated. * Ignore the `--spec` value, if supplied, log a warning that it will be deprecated * `lcli` still has the spec flag, that's fine because it's dev tooling. * Remove the `E: EthSpec` from `YamlConfig` * This means we need to deser the genesis `BeaconState` on-demand, but this is fine. * Swap the old "minimal", "mainnet" strings over to the new `EthSpecId` enum. * Always require a `CONFIG_NAME` field in `YamlConfig` (it used to have a default). ## Additional Info Lots of breaking changes, do not merge! ~~We will likely need a Lighthouse v0.4.0 branch, and possibly a long-term v0.3.0 branch to keep Medalla alive~~. Co-authored-by: Kirk Baird <[email protected]> Co-authored-by: Paul Hauner <[email protected]>

paulhauner · 2020-10-28T22:16:56Z

bors r-

bors · 2020-10-28T22:16:57Z

Canceled.

paulhauner · 2020-10-28T22:18:33Z

I had to merge #1836 into this to fix a merge conflict.

paulhauner · 2020-10-28T22:19:08Z

bors r+

## Issue Addressed Closes #1504 Closes #1505 Replaces #1703 Closes #1707 ## Proposed Changes * Update BLST and Milagro to versions compatible with BLSv4 spec * Update Lighthouse to spec v1.0.0-rc.0, and update EF test vectors * Use the v1.0.0 constants for `MainnetEthSpec`. * Rename `InteropEthSpec` -> `V012LegacyEthSpec` * Change all constants to suit the mainnet `v0.12.3` specification (i.e., Medalla). * Deprecate the `--spec` flag for the `lighthouse` binary * This value is now obtained from the `config_name` field of the `YamlConfig`. * Built in testnet YAML files have been updated. * Ignore the `--spec` value, if supplied, log a warning that it will be deprecated * `lcli` still has the spec flag, that's fine because it's dev tooling. * Remove the `E: EthSpec` from `YamlConfig` * This means we need to deser the genesis `BeaconState` on-demand, but this is fine. * Swap the old "minimal", "mainnet" strings over to the new `EthSpecId` enum. * Always require a `CONFIG_NAME` field in `YamlConfig` (it used to have a default). ## Additional Info Lots of breaking changes, do not merge! ~~We will likely need a Lighthouse v0.4.0 branch, and possibly a long-term v0.3.0 branch to keep Medalla alive~~. Co-authored-by: Kirk Baird <[email protected]> Co-authored-by: Paul Hauner <[email protected]>

bors · 2020-10-28T23:33:03Z

Pull request successfully merged into master.

Build succeeded:

michaelsproul added spec_change A change related to the Eth2 spec work-in-progress PR is a work-in-progress do-not-merge consensus An issue/PR that touches consensus code, such as state_processing or block verification. crypto An issue/PR that touches cryptography code. A1 labels Oct 13, 2020

michaelsproul mentioned this pull request Oct 13, 2020

Update BLST, add force-adx support (take II) #1703

Closed

michaelsproul force-pushed the spec-v1.0.0-rc branch from dd34bec to 083db9c Compare October 22, 2020 06:33

michaelsproul and others added 2 commits October 26, 2020 15:16

Update to spec v1.0.0-rc.0 and BLSv4

3625001

Update blst and milagro_bls subgroup checking (#1793)

e3096c2

* Update blst and milagro_bls subgroup checking Signed-off-by: Kirk Baird <[email protected]> * cargo fmt Signed-off-by: Kirk Baird <[email protected]>

paulhauner force-pushed the spec-v1.0.0-rc branch from 083db9c to e3096c2 Compare October 26, 2020 04:18

paulhauner added 13 commits October 26, 2020 15:40

Update Cargo.lock

0c2d879

Update InteropEthSpec to new values

c5cda05

Rename interop in CLI

a45eda4

Rename InteropEthSpec

50d3c80

Tidy uses of interop spec

998f7ca

Fix lcli, boot_node

10c456c

Merge branch 'master' into spec-v1.0.0-rc

b9eb978

Add features to task_executor

79e2dd7

Fix failing test

cc28ce5

Remove EthSpec from Eth2TestnetConfig

778c245

Remove --spec flag from lighthouse

96feb86

Add warning for --spec flag

9234141

Use testnet names in YamlConfig

2494546

paulhauner added 4 commits October 28, 2020 15:22

Introduce EthSpecId enum

b43d42e

Remove text spec constants

7a5b43c

Tidy

33588f4

Rename legacy spec id

2eced35

paulhauner added ready-for-review The code is ready for review and removed work-in-progress PR is a work-in-progress labels Oct 28, 2020

michaelsproul commented Oct 28, 2020

View reviewed changes

paulhauner mentioned this pull request Oct 28, 2020

Fix disabled Eth1Block test #1835

Closed

Address review comments

f996e1b

paulhauner added 2 commits October 28, 2020 18:19

Bump version tags

f29f3ec

Run cargo update

1495642

paulhauner mentioned this pull request Oct 28, 2020

Bump version to v0.3.2 #1836

Closed

Merge branch 'cut-v0.3.2' into spec-v1.0.0-rc

489b9c9

bors bot changed the title ~~Update to spec v1.0.0-rc.0 and BLSv4~~ [Merged by Bors] - Update to spec v1.0.0-rc.0 and BLSv4 Oct 28, 2020

bors bot closed this Oct 28, 2020

michaelsproul deleted the spec-v1.0.0-rc branch December 1, 2020 01:21

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Merged by Bors] - Update to spec v1.0.0-rc.0 and BLSv4 #1765

[Merged by Bors] - Update to spec v1.0.0-rc.0 and BLSv4 #1765

michaelsproul commented Oct 13, 2020 •

edited by paulhauner

Loading

michaelsproul commented Oct 13, 2020

paulhauner commented Oct 13, 2020

michaelsproul commented Oct 13, 2020

kirk-baird commented Oct 14, 2020

kirk-baird commented Oct 20, 2020

paulhauner commented Oct 26, 2020

paulhauner commented Oct 28, 2020

michaelsproul left a comment

paulhauner commented Oct 28, 2020

michaelsproul commented Oct 28, 2020

michaelsproul commented Oct 28, 2020

paulhauner commented Oct 28, 2020

paulhauner commented Oct 28, 2020

bors bot commented Oct 28, 2020

paulhauner commented Oct 28, 2020

paulhauner commented Oct 28, 2020

bors bot commented Oct 28, 2020

[Merged by Bors] - Update to spec v1.0.0-rc.0 and BLSv4 #1765

[Merged by Bors] - Update to spec v1.0.0-rc.0 and BLSv4 #1765

Conversation

michaelsproul commented Oct 13, 2020 • edited by paulhauner Loading

Issue Addressed

Proposed Changes

Additional Info

michaelsproul commented Oct 13, 2020

paulhauner commented Oct 13, 2020

michaelsproul commented Oct 13, 2020

kirk-baird commented Oct 14, 2020

kirk-baird commented Oct 20, 2020

paulhauner commented Oct 26, 2020

paulhauner commented Oct 28, 2020

michaelsproul left a comment

Choose a reason for hiding this comment

paulhauner commented Oct 28, 2020

michaelsproul commented Oct 28, 2020

michaelsproul commented Oct 28, 2020

paulhauner commented Oct 28, 2020

paulhauner commented Oct 28, 2020

bors bot commented Oct 28, 2020

paulhauner commented Oct 28, 2020

paulhauner commented Oct 28, 2020

bors bot commented Oct 28, 2020

michaelsproul commented Oct 13, 2020 •

edited by paulhauner

Loading