Fee Market Updates #17
Conversation
Inphi
force-pushed
the
inphi/fee-hdr
branch
2 times, most recently
from
e14ad39
to
8e7bdb1
Compare
- and update the blob fee rules to the full version
Inphi
force-pushed
the
inphi/fee-hdr
branch
from
8e7bdb1
to
9a9fe09
Compare
|
cc: @dgcoffman |
| @@ -139,8 +139,8 @@ func Transaction(ctx *cli.Context) error { | |||
| r.Address = sender | |||
| } | |||
| // Check intrinsic gas | |||
| if gas, err := core.IntrinsicGas(tx.Data(), tx.AccessList(), len(tx.BlobVersionedHashes()), tx.To() == nil, | |||
| chainConfig.IsHomestead(new(big.Int)), chainConfig.IsIstanbul(new(big.Int))); err != nil { | |||
| if gas, err := core.IntrinsicGas(tx.Data(), tx.AccessList(), len(tx.BlobVersionedHashes()), 0, tx.To() == nil, | |||
There was a problem hiding this comment.
I don't know a lot about this utility (looks like it's for testing?) but thought I'd ask if we might get into trouble by always using 0 for excess blobs. Should there maybe at least be a comment saying we always set it to 0?
There was a problem hiding this comment.
The utility is used to validate transactions. Up until EIP-4844, the intrinsic gas did not rely on the prestate of the execution. I imagine the utility may be updated in the future to take an excessBlobs parameter, but for now I think using zero is a nice default.
I agree that there should be a comment explaining this. Will add.
| return adjusted - params.TargetBlobsPerBlock | ||
| } | ||
|
|
||
| // FakeExponential approximates 2 ** (num / denom) |
There was a problem hiding this comment.
Should we adopt George's proposed taylor expansion version already or will you update if/when it gets merged into the spec?
There was a problem hiding this comment.
Is there a PR for the spec changes? I'll update this once there's one.
There was a problem hiding this comment.
don't see one yet... this was where I saw it, but makes sense to wait if/when. adietrichs/EIPs@5f91002
Implements the full fee market spec.