Skip to main content
Technical proposal SIMD-0459 Review

Syscall Parameter Address Restrictions

  1. Idea
  2. Draft
  3. Review
  4. Accept
  5. Implement
  6. Active

Decision brief

Why this proposal matters

Syscall parameter related changes split off from SIMD-0219: Removing pitfalls and foot-guns from the ABI (including syscalls) and runtime.

Source-backed reading aid · source language: EN

Proposal at a glance

What changes

  • The following pointers must be on the stack or heap, meaning their virtual address is inside 0x200000000..0x400000000, otherwise SyscallError::InvalidPointer must be thrown:
  • The destination address of all sysvar related syscalls
  • The pointer in the array of &[AccountInfo] / SolAccountInfo

Stakeholder map

Who is affected

Builders & client teams Medium impact

Splitting SIMD-0219 should have no impact on dApp developers or validators. The additional changes are continuously tested (over a month) and monitored on mainnet-beta to not cause any existing dApps to behave differently.

Action may be required
Validators & operators Medium impact

Splitting SIMD-0219 should have no impact on dApp developers or validators. The additional changes are continuously tested (over a month) and monitored on mainnet-beta to not cause any existing dApps to behave differently.

Action may be required
Users & stakers Impact unknown

No proposal-specific evidence was found for this group.

Action requirement unknown
Governance & ecosystem Impact unknown

No proposal-specific evidence was found for this group.

Action requirement unknown

Exact source revision

Full proposal document

Source-language document. Technical identifiers and evidence remain unchanged. 0616093b2952

Summary

Syscall parameter related changes split off from SIMD-0219: Removing pitfalls and foot-guns from the ABI (including syscalls) and runtime.

Motivation

In a recent meeting between the Agave and Firedancer core developers it was decided that SIMD-0219 should be split into two feature gates. This SIMD covers the first half which is focused on reducing the attack surface and fix long standing issues with CPI in general.

The reasoning for splitting SIMD-0219 is to de-risk its deployment and to be able to make progress on shipping earlier parts even if later parts have to be delayed further. The reason for cutting SIMD-0219 at this specific boundary is that this first half is not expected to have any impact on performance, while the second half on the other hand will.

Additionally, some consensus relevant code reordering is added to reduce the attack surface of around CU metering and make the CPI logic more comprehensible after this SIMDs feature gates are cleaned up.

New Terminology

None.

Detailed Design

Changes inherited from SIMD-0219

  • The following pointers must be on the stack or heap, meaning their virtual address is inside 0x200000000..0x400000000, otherwise SyscallError::InvalidPointer must be thrown:
    • The destination address of all sysvar related syscalls
    • The pointer in the array of &[AccountInfo] / SolAccountInfo*
    • The AccountInfo::data field, which is a RefCell<&[u8]> in sol_invoke_signed_rust
    • The AccountInfo::lamports field, which is a RefCell<&u64> in sol_invoke_signed_rust
  • The following pointers must point to what was originally serialized in the input regions by the program runtime, otherwise SyscallError::InvalidPointer must be thrown:
    • AccountInfo::key / SolAccountInfo::key
    • AccountInfo::owner / SolAccountInfo::owner
    • AccountInfo::lamports / SolAccountInfo::lamports
    • AccountInfo::data::ptr / SolAccountInfo::data

Additional changes

InstructionError::InvalidRealloc must be thrown on the CPI call edge if the caller requested an account length which does not fit the payload address space of the account in the caller. This check must occur after the address of the account length is translated but before the address of the account payload is translated.

InstructionError::InvalidRealloc must be thrown on the CPI return edge if the callee requested an account length which does not fit the payload address space of the account in the caller. This check currently allows for +10 KiB even in ABIv0 which is incorrect and must be reduced to 0 growth allowed in ABIv0.

All sites in the CPI code which perform guest to host address translation first and then perform pointer arithmetic on the host must be swapped such that they perform pointer arithmetic in the virtual address space first followed by the address translation second. This specifically affects the account length field.

As a consequence of the changes to the address translation of the account length field, the CU charging for the account length in CPI must be moved to occur after the address translation of the account length field.

Modifications of the callee accounts on the CPI call edge must all happen together after every account info and account meta is translated, instead of being interleaved with these.

Alternatives Considered

Leaving SIMD-0219 as is.

Impact

Splitting SIMD-0219 should have no impact on dApp developers or validators. The additional changes are continuously tested (over a month) and monitored on mainnet-beta to not cause any existing dApps to behave differently.

Security Considerations

None.

Evidence graph

Related proposals and rollout

One or more sources are unavailable.

Upstream review record

Upstream discussion & review

GitHub review is editorial context, not evidence of on-chain support, voting, or outcome.

PR #459 · merged SIMD-0459: Syscall Parameter Address Restrictions 19 comments and reviews · May 4, 2026
@Lichtso

The changes added in this SIMD over SIMD-0219 are covered in https://github.com/anza-xyz/agave/pull/9995 and have been tested since then (a month ago).

GitHub ↗
@Lichtso

I'm also not entirely clear which charge is being moved around: the charge for address translation? Or the entire translation+access operation? We never charged for the address translation, we charge for the length of the accounts as their data needs to be copied and compared.

GitHub ↗
@simd-botbot

Thanks, ripatel-fd(https://github.com/ripatel-fd)! ⚠️ Status: Cannot merge yet - Missing approval from: Anza (t-nelson(https://github.com/t-nelson), sakridge(https://github.com/sakridge), or bw-solana(https://github.com/bw-solana))

GitHub ↗
@simd-botbot

Thanks, topointon-jump(https://github.com/topointon-jump)! ⚠️ Status: Cannot merge yet - Missing approval from: Anza (t-nelson(https://github.com/t-nelson), sakridge(https://github.com/sakridge), or bw-solana(https://github.com/bw-solana))

GitHub ↗
@simd-botbot

✅ All approvals received! @Lichtso, you can now merge this by commenting /merge. ✅ Status: Ready to merge

GitHub ↗
@simd-botbot

✅ Merge successful! Lichtso(https://github.com/Lichtso)'s PR has been merged.

GitHub ↗
@t-nelson

would have been ideal had this been written as a standalone document with only ref to 0219 being the supersedes field

GitHub ↗

simd.watch community discussion · SIMD-0459

Powered by Giscus · Sign in with GitHub to comment

Community comments load when this section approaches the viewport.

Provenance

Evidence & technical details

Rollout or chain data, source revisions, freshness and integrity.
1

SIMD

Deployment Status

Document lifecycleReview
CategoryCore protocol

Feature Gate not yet created

Exact source revision

Sources & integrity

  • Proposal document pinned_commit_blob
    0616093b2952ed6de52c4a27d66aadd11d48d4f9
  • simd-document document · current · Sep 11, 2026
    0616093b2952ed6de52c4a27d66aadd11d48d4f9

One or more sources are unavailable.