SIMD-0313: Drop unchained merkle shreds #313
Conversation
|
@bw-solana @alexpyattaev it won't let me add you as a reviewer so tagging here |
AshwinSekar
force-pushed
the
drop-unchained
branch
from
757b1ad to
0b322e2
Compare
| ## Security Considerations | ||
|
|
||
| None | ||
|
|
There was a problem hiding this comment.
Would it make sense to mention that this strictly improves security compared to the status quo?
There was a problem hiding this comment.
This SIMD by itself will not, but it leads to better security via alpenglow
| // the Merkle tree. The root of the Merkle tree is signed. | ||
|
|
||
| // Layout: {common, data} headers | data buffer | ||
| // | [Merkle root of the previous erasure batch if chained] |
There was a problem hiding this comment.
In context of this SIMD all shreds are chained, right?
There was a problem hiding this comment.
All shreds sent are now chained, but they used to not be chained. As specified below when the feature flag is active, any shred without the chained flag set will be dropped
|
I support it. We should be able to delete a lot of code |
|
@willhickey could we get this merged and resume activation on devnet? Thanks! |
Backfilling the SIMD for this feature https://github.com/anza-xyz/feature-gate-tracker/issues/80