Reapply "deps: Bump all sdk and spl crates to v3 compatibility (#7492…#7592
Reapply "deps: Bump all sdk and spl crates to v3 compatibility (#7492…#7592joncinque merged 2 commits intoanza-xyz:masterfrom
Conversation
|
If this PR represents a change to the public RPC API:
Thank you for keeping the RPC clients in sync with the server API @joncinque. |
|
The Firedancer team maintains a line-for-line reimplementation of the |
|
For your information, the |
|
I have this PR running on |
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## master #7592 +/- ##
========================================
Coverage 83.4% 83.4%
========================================
Files 813 813
Lines 365417 365537 +120
========================================
+ Hits 304871 304986 +115
- Misses 60546 60551 +5 🚀 New features to boost your workflow:
|
|
Thanks @willhickey ! |
joncinque
force-pushed
the
bump-sdk-the-reckoning
branch
from
e3bcb7e to
65ee12c
Compare
| feature = "frozen-abi", | ||
| derive(AbiEnumVisitor, AbiExample), | ||
| frozen_abi(digest = "9KN64WUT7XDYj9zZopS1hztGyAP9y4N4QznsyC4mqsGs") | ||
| frozen_abi(digest = "FGw38CCo7vg24qxe7TfGP11WdX2poe6T55BGN1r3XMFA") |
There was a problem hiding this comment.
why does the hash change here? Was the hash algo changed somehow?
There was a problem hiding this comment.
RepairRequestHeader contains a Pubkey, which is now just a type alias for Address. The name change is typically enough to change the frozen abi digest.
See some of the changes that were made in the PR for Address anza-xyz/solana-sdk#243
There was a problem hiding this comment.
wow I did not expect frozen_abi to care about filed names (as they are not part of the ABI). But ok I guess this is fine then=)
|
|
||
| #[test] | ||
| fn test_serialize_round_trip() { | ||
| // Unfortunately doing `Keypair::new_from_array(rng.gen())` gives |
There was a problem hiding this comment.
is it maybe preferable to let the code run and save the new hash values where they are checked? This lovely helper is not useful long-term.
There was a problem hiding this comment.
I'm happy to do that if you prefer! I wanted the code to produce exactly the same results, also as a check to make sure that the rand change was done correctly in the sdk.
Just let me know either way.
There was a problem hiding this comment.
I think now that we know the sdk change is ok (since test passes now) it is ok to use the new API directly and update the hashes. Or it could be a separate PR - up to you.
joncinque
force-pushed
the
bump-sdk-the-reckoning
branch
from
65ee12c to
ac427e5
Compare
brooksprumo
left a comment
brooksprumo
left a comment
There was a problem hiding this comment.
Approving for the changes to the serde/snapshot status cache.
| InstructionError::Immutable => Self::Immutable, | ||
| InstructionError::IncorrectAuthority => Self::IncorrectAuthority, | ||
| InstructionError::BorshIoError(_) => Self::BorshIoError(String::new()), | ||
| InstructionError::BorshIoError => Self::BorshIoError(String::new()), |
alexpyattaev
left a comment
alexpyattaev
left a comment
There was a problem hiding this comment.
LGTM on gossip, sigverify shreds and repair serialization.
|
Thanks for the quick reviews everyone! |
6 participants
…)" (#7556)
This reverts commit f149dec.
Problem
Agave should be using the SDK v3 crates before branching its own v3, for our own sanity. There's more information at #7492 and https://github.com/orgs/anza-xyz/projects/27
Summary of Changes
Pretty much reapplying the commit from #7492, and fixing a few merge conflicts:
solana-sbpfpulled in differently in Cargo.tomlTransactionContext