fix: prevent oracle failure on tag computation for invalid recipient#22163
Conversation
benesjan
left a comment
benesjan
left a comment
There was a problem hiding this comment.
Why not just tackle F-512 straight away instead of needing to clutter the code with the branching logic?
Just checking that upon registration would probably have even smaller diff (and we would not waste time wit this)
This fixes two problems, only one of which is addressed by F-512 - this still needs to account for invalid recipients, so the PR is required regardless (and this is the more severe issue). I didn't want to mix those two concerns here. |
Co-authored-by: Jan Beneš <janbenes1234@gmail.com>
benesjan
left a comment
benesjan
left a comment
There was a problem hiding this comment.
NIce.
The suggestions I made are uncontroversial so I will apply them and merge this.
Co-authored-by: Jan Beneš <janbenes1234@gmail.com>
|
✅ Successfully backported to backport-to-v4-next-staging #22205. |
BEGIN_COMMIT_OVERRIDE feat!: remove ALL_SCOPES (#22136) chore: syncing oracle versions (#22131) fix(aztec-nr): use registered accounts as capsule test scopes (#22171) chore: reverting accidentally merged AI slop (#22175) feat: move event size check from declaration to private emission (#22168) fix: prevent oracle failure on tag computation for invalid recipient (#22163) feat: disable noir test cache on merge-train/fairies (#22206) END_COMMIT_OVERRIDE
BEGIN_COMMIT_OVERRIDE cherry-pick: feat: move event size check from declaration to private emission (#22168) fix: prevent oracle failure on tag computation for invalid recipient (#22163) feat: move event size check from declaration to private emission (#22168) [v4-next backport] (#22182) fix(cli-wallet): peek claim stack instead of popping for estimate-gas-only (#22196) fix: use Fr.fromString for CLI wallet claim params to handle decimal values (#22197) fix: indefinite retry for prover node and agent broker communication (#22202) END_COMMIT_OVERRIDE
3 participants
From https://github.com/AztecProtocol/barretenberg-claude/issues/1903.
This fixes a king-of-the-hill vector in which an attempt to deliver to an invalid address resulted in a simulation crash. The contract correctly handles this already during encryption, but the PXE tagging code did not. Future versions of PXE will not be vulnerable to this.