Ecmul Precompile #1829
Ecmul Precompile #1829
Conversation
EIPS/eip-ecmul-precompile.md
Outdated
| discussions-to: https://ethereum-magicians.org/t/ewasm-precompile-for-general-elliptic-curve-math/2581 | ||
| status: Draft | ||
| type: Core | ||
| replaces: EIP-196 (in some cases), EIP 665, EIP 1108 (partially), https://github.com/ethereum/EIPs/issues/603 |
There was a problem hiding this comment.
Unfortunately this can only use an EIP number and that must be merged (cannot be an unmerged draft). For #603 I'd suggest you refer to it in a "References" or "Prior art" section.
There was a problem hiding this comment.
I actually don't think the replaces field is appropriate here. Perhaps it is as field not applicable to the EIP process we're following.
EIPS/eip-ecmul-precompile.md
Outdated
| @@ -0,0 +1,151 @@ | |||
| --- | |||
| eip: To be assigned | |||
There was a problem hiding this comment.
You can change this to 1829 and also rename the file.
EIPS/eip-ecmul-precompile.md
Outdated
| status: Draft | ||
| type: Core | ||
| replaces: EIP-196 (in some cases), EIP 665, EIP 1108 (partially), https://github.com/ethereum/EIPs/issues/603 | ||
| author: Remco Bloemen <[email protected]> |
There was a problem hiding this comment.
Could order these fields the same way the template has it?
EIPS/eip-ecmul-precompile.md
Outdated
| MUL_GAS = ... | ||
| ``` | ||
|
|
||
| The total gas cost is `BASE_GAS` plus `ADD_GAS` for each $s_i$ that is $1$ and `MUL_GAS` for each $s_i > 1$ ($s_0 = 0$ is free). |
There was a problem hiding this comment.
Not sure markdown renders $..$ but you could use backticks instead?
|
@axic Thank you! I believe I have addressed all the comments. |
|
The Travis error does not seem related to the PR. |
|
The thing that has always stopped me from proposing a similar thing is, what happens if someone submits a non-prime modulus? There are different ways to implement elliptic curve addition and multiplication, and in general I don't expect that the different libraries have been tested to ensure that they fail in the same cases when the modulus ends up being composite, and this seems like a large and deep potential source of consensus failures. Unfortunately primality tests that take into account weird edge cases like Carmichael numbers but are also clean and efficient don't really exist. |
|
EDIT: Response moved to the discussion thread to keep the conversation in one place: https://ethereum-magicians.org/t/precompile-for-general-elliptic-curve-linear-combinations/2581/14?u=recmo |
No description provided.