EIP-7990: RUNCODE Opcode – Execute arbitrary bytecode from memory within the same execution context

EIPs EIPs core
,
1

Summary

This EIP proposes a new RUNCODE opcode that enables execution of arbitrary bytecode from memory within the same execution context. This complements DELEGATECALL by providing an alternative for scenarios where dynamic code doesn’t need to be permanently stored on-chain, offering significant gas savings.

EIP-7990

Key Benefits

  • Alternative to DELEGATECALL: Provides option for temporary/generated code that doesn’t require permanent on-chain storage

  • Gas Optimization: Saves up to 2,600 gas per execution by eliminating cold address access costs

  • Reduced Blockchain Bloat: Avoids deploying temporary bytecode as contracts, reducing state growth

  • Deployment Savings: Eliminates need to deploy ephemeral code as contracts (~32,000+ gas savings per deployment)

  • Context Preservation: Maintains identical execution context as DELEGATECALL while operating from memory

Use Cases

  • Runtime code generation and execution without permanent deployment

  • One-time complex calculations that don’t justify permanent contract storage

  • Temporary computational libraries that change frequently

  • Meta-programming patterns where bytecode is generated dynamically

  • Mathematical computation libraries that generate optimized bytecode on-demand

Discussion Goals

Prior to opening a pull request to the EIPs repository, this proposal seeks community feedback on:

  • Real-world use cases where RUNCODE provides significant value
  • Any overlooked security implications compared to DELEGATECALL
  • Whether any opcode restrictions should be applied within RUNCODE
  • The appropriateness of the proposed 100 gas base cost

Comments, concerns, and alternative perspectives are highly encouraged.

— Alan Bojorquez (@camax)