[GH-ISSUE #2353] [Feature] Decode recorded Linux AArch64 ELF core evidence #778

Closed
opened 2026-10-15 18:28:16 +00:00 by bsakaguchi · 0 comments
Owner

Originally created by @matbrgz on GitHub (Oct 12, 2026).
Original GitHub issue: https://github.com/morluto/rea/issues/2353

Analyst question and current scope

docs/recorded-crashes.md and src/domain/native/recordedCrash.ts declare an initial Linux ELF64 x86-64 little-endian core profile. Legitimate recorded Linux AArch64 crashes need their own ABI-bound register/note interpretation rather than reuse of the amd64 layout.

Acceptance criteria

  • Specify supported Linux/AArch64 ELF/ABI variants and register-note layouts, keeping unfamiliar note owners/kinds opaque.
  • Bind every decoded thread/register/signal fact to validated source byte ranges and retain the complete recorded ABI register set.
  • Keep historical PIDs and mapping names as recorded metadata, never as selection/attach authority for a live process.
  • Verify original fixtures, high register values, malformed notes, unsupported ABIs, cancellation, resource bounds and both CLI/MCP adapters.

Extend the established recorded-crash contract from #969 with a separately verified architecture profile. No live-debugging behavior is requested; #1267 owns that broader work.

Priority: P2. The current surface was checked at upstream 1c3791430, and open/closed issues were searched before filing. Proposed file-analysis features need their own spec/plan/tasks under docs/specs/. If implementation requires a new provider or expands execution authority, obtain the maintainer scope decision required by CONTRIBUTING.md before integration; this proposal itself does not authorize that expansion.

Originally created by @matbrgz on GitHub (Oct 12, 2026). Original GitHub issue: https://github.com/morluto/rea/issues/2353 ## Analyst question and current scope docs/recorded-crashes.md and src/domain/native/recordedCrash.ts declare an initial Linux ELF64 x86-64 little-endian core profile. Legitimate recorded Linux AArch64 crashes need their own ABI-bound register/note interpretation rather than reuse of the amd64 layout. ## Acceptance criteria - [ ] Specify supported Linux/AArch64 ELF/ABI variants and register-note layouts, keeping unfamiliar note owners/kinds opaque. - [ ] Bind every decoded thread/register/signal fact to validated source byte ranges and retain the complete recorded ABI register set. - [ ] Keep historical PIDs and mapping names as recorded metadata, never as selection/attach authority for a live process. - [ ] Verify original fixtures, high register values, malformed notes, unsupported ABIs, cancellation, resource bounds and both CLI/MCP adapters. ## Source and related work - https://github.com/ARM-software/abi-aa Extend the established recorded-crash contract from #969 with a separately verified architecture profile. No live-debugging behavior is requested; #1267 owns that broader work. Priority: P2. The current surface was checked at upstream 1c3791430, and open/closed issues were searched before filing. Proposed file-analysis features need their own spec/plan/tasks under docs/specs/. If implementation requires a new provider or expands execution authority, obtain the maintainer scope decision required by CONTRIBUTING.md before integration; this proposal itself does not authorize that expansion.
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
morluto/rea#778
No description provided.