inefficient hashing

[thedavidmeister]

note[asm-keccak256]: use of inefficient hashing mechanism; consider using inline assembly
--> src/lib/state/LibInterpreterState.sol:28:16
|
28 | return keccak256(abi.encode(state));
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
= help: https://book.getfoundry.sh/reference/forge/forge-lint#asm-keccak256

[thedavidmeister]

⏺ Bash(cp /Users/thedavidmeister/Code/rain.interpreter/tmp/KeccakTest.sol /Users/thedavidmeister/Code/rain.interpreter/src/tmp_KeccakTest.sol && nix develop -c forge i…)
⎿  Warning: This is a nightly build of Foundry. It is recommended to use the latest stable version. To mute this warning set FOUNDRY_DISABLE_NIGHTLY_WARNING in your environment.

 /// @use-src 0:"src/tmp_KeccakTest.sol"                                                                                                                                                         
 … +42 lines (ctrl+o to expand)                                                                                                                                                                  

⎿  (timeout 1m)

⏺ Interesting — the optimizer produces identical code for both functions:

keccak256(add(_1, 0x20), mload(_1))

The Solidity compiler is smart enough to optimize keccak256(data) into the same keccak256(ptr+0x20, length) assembly pattern. No extra allocation from keccak256 itself.

So for the simple keccak256(bytes memory) case, there's zero overhead. The lint is a false positive here — or at least, it's flagging a pattern the optimizer already handles. The real cost in the
fingerprint() function is entirely from abi.encode(state), not from keccak256.