Workspace and crate topology

Maps the four-package Cargo workspace, source ownership, dependency direction, release profile, and responsibilities of each executable or library.

Experimental
Last verified
2026-06-25 00:00 UTC
Updated
Reading time
2 minutes

Maps the four-package Cargo workspace, source ownership, dependency direction, release profile, and responsibilities of each executable or library.

Implementation evidence: this topic is grounded in the reviewed GGUF.MiRust.com source snapshot. It documents observed code and artifacts without claiming broad deployment, model quality, or production readiness.

Workspace members

  • tinyrustlm/runtime: cdylib and rlib runtime.
  • tinyrustlm/tools/slm_pack: model writer, validator, converter, manifest and quality-gate CLI.
  • tinyrustlm/tools/local_server: loopback static server.
  • tinyrustlm/tools/browser_harness: static and HTTP contract audit.

Shared package policy

All four packages inherit version 0.1.0, Rust edition 2021, a Proprietary license declaration, repository metadata, missing_docs = deny, and unsafe_code = warn.

Release profile

The workspace uses one codegen unit, link-time optimization, size-oriented opt-level = "z", and panic = "abort". This favors compact WASM and deterministic panic termination over unwind recovery.

Dependency direction

The lockfile lists only local workspace packages. Shared behavior is copied or kept inside each crate rather than factored through external dependencies. Any future shared crate should preserve clear ownership and avoid creating a circular tool/runtime dependency.

Scope

This starter page defines the questions, boundaries, evidence, and failure modes that should be recorded before a capability is presented as supported.

Engineering considerations

  • Identify the source, version, target environment, and owner.
  • Separate observed values from estimates and externally reported values.
  • Record trade-offs, unsupported cases, and fallback behavior.
  • Link performance statements to a compatible benchmark methodology.

Verification questions

  • What exact artifact, revision, backend, and environment were reviewed?
  • Which assumptions could change the result?
  • Which data should be retained so another engineer can reproduce the conclusion?