Document ForkDigest-context for EIP-4844 BlobSidecar in Electra
#3850
Conversation
Currently, documentation does not state what determines the `ForkDigest` for `BlobSidecar`. It makes most sense to use the corresponding slot as that one determines the `ForkDigest` of the corresponding beacon block. Furthermore, add "and later" remarks so that `ELECTRA_FORK_VERSION` is also associated with the `deneb.BlobSidecar` type. This is in line with how `light-client/p2p-interface.md` is done.
specs/deneb/p2p-interface.md
Outdated
| The gossip `ForkDigest`-context is determined based on `compute_fork_version(compute_epoch_at_slot(blob_sidecar.signed_block_header.message.slot))`. | ||
|
|
||
| Per `context = compute_fork_digest(fork_version, genesis_validators_root)`: | ||
|
|
||
| [0]: # (eth2spec: skip) | ||
|
|
||
| | `fork_version` | Chunk SSZ type | | ||
| |--------------------------------|-------------------------------| | ||
| | `DENEB_FORK_VERSION` and later | `deneb.BlobSidecar` | | ||
|
|
There was a problem hiding this comment.
The ForkDigest-context thing is only for Req/Resp, not GossipSub topics. GossipSub has nothing to do with it.
There was a problem hiding this comment.
ForkDigest is also part of Gossipsub (as part of the topic name): https://github.com/ethereum/consensus-specs/blob/a42d6706d89c414764eda7e2d0103e19f1e23761/specs/phase0/p2p-interface.md#topics-and-messages
There was a problem hiding this comment.
Let's call it ForkDigestValue then.
This is new to the spec (except light-client), but, given that |
| @@ -241,7 +251,7 @@ No more than `MAX_REQUEST_BLOCKS_DENEB` may be requested at a time. | |||
|
|
|||
| Per `context = compute_fork_digest(fork_version, genesis_validators_root)`: | |||
|
|
|||
| [1]: # (eth2spec: skip) | |||
| [0]: # (eth2spec: skip) | |||
There was a problem hiding this comment.
I'm a bit curious when to use [0] and when to use [1] here. cc: @hwwhww
There was a problem hiding this comment.
Likewise, it's inconsistent inside the same document. Overall, [0] seems to be more.. "dominant" throughout the codebase, for these table annotations.
There was a problem hiding this comment.
Agreed. It shouldn't really matter either way, it's only used to tell pyspec not to parse the table.
Currently, documentation does not state what determines the
ForkDigestforBlobSidecar. It makes most sense to use the corresponding slot as that one determines theForkDigestof the corresponding beacon block.Furthermore, add "and later" remarks so that
ELECTRA_FORK_VERSIONis also associated with thedeneb.BlobSidecartype.This is in line with how
light-client/p2p-interface.mdis done.