Files
rea/tests/conformance/ghidra
0baad58e60 feat(ghidra): admit RISC-V ELF analysis through a pinned generic lane (#2151)
* feat(ghidra): admit RISC-V ELF analysis through a pinned generic lane

Refs #2063.

Recognize `EM_RISCV` (243) as a provider-neutral family that carries its own ELF
facts, and bound the first Ghidra lane to little-endian `ET_EXEC` images that
declare the double-float ABI, pinned to `RISCV:LE:32:default` or
`RISCV:LE:64:default` with the `gcc` compiler specification.

Ghidra 12.1.4 is not reliable without the pin: its `riscv.opinion` pins
`variant="default"` for the 64-bit constraint and leaves the 32-bit constraint
with the comment that the loader "must choose between 'default' and 'AndeStar'
variants". Measured on the installed release tree, a source-owned generic RV32
ELF with no Andes declaration imports as `RISCV:LE:32:AndeStar_v5:gcc`, while
RV64 resolves to `RISCV:LE:64:default:gcc`. The lane therefore states the
language per width and refuses any caller-selected processor outside it instead
of inheriting the header guess.

Refusals name the declaration rather than falling back to a guessed profile:
big-endian images (Ghidra 12.1.4 ships no big-endian RISC-V language),
non-`ET_EXEC` images, `EF_RISCV_RVE`, `EF_RISCV_TSO`, and every float ABI other
than double, which the pinned generic compiler specifications model. The raw
e_flags bits stay in Evidence as declarations, not as instruction semantics.

Family recognition now has one owner for the ELF-metadata families:
`ELF_METADATA_ARCHITECTURES`, `isElfMetadataArchitecture`, and
`HeaderArchitecture` replace ten `Exclude<BinaryArchitecture, "mips">` sites that
would each have needed a second hand-edited member. `ghidraPinnedLanguage` is
that owner for the import commitment: the session readback previously re-derived
only the MIPS pin, so a profile committing any second pinned language failed with
a language-mismatch error before the bridge started.

Hopper and IDA refuse RISC-V by name, matching the MIPS wording. Caller-supplied
raw-binary entry keeps its existing four-family list; this lane is ELF-header
admission only.

Verification on macOS 15.7.7 arm64, Node 24.18.0, Ghidra 12.1.4 PUBLIC and JDK
21.0.12.1, with the host decompiler built from the release sources for
`os/mac_arm_64`: `npm run check` passes; `npm run test:fast` passes 396 test
files; `npm run verify:test-discovery` discovers all 917 test files exactly once.
Source-owned RV32 and RV64 double-float `ET_EXEC` fixtures complete
`rea analyze --provider ghidra` and `rea decompile rea_riscv_entry`, committing
`RISCV:LE:32:default` and `RISCV:LE:64:default` with `language_selection:
provider-pinned-v1`; the soft-float RV64 shape is refused with the named
float-ABI reason, and the full refusal matrix including the RV32 and
width-mismatch shapes is covered by unit tests. No `verify:ghidra:riscv` lane
script is included in this change; the runs above were performed manually against
the installed engine. Fork pull requests receive no CI check runs, so the
real-provider scope above is the evidence available for review.

* fix(ghidra): bound the RISC-V lane by declared flags and document it

Address review feedback on #2151.

`e_flags` bits outside `EF_RISCV_RVC | EF_RISCV_FLOAT_ABI` are now refused with
the offending value named, so a reserved or newer producer declaration cannot
enter the verified generic lane by falling through the RVE, TSO and float-ABI
checks.

Admission reads `e_type` only, so the refusal text no longer calls `ET_EXEC`
static: a fixed-address declaration does not establish static linkage, and
`PT_INTERP` and dynamic dependencies are not inspected. The session records that
boundary as a limitation rather than inferring a load model.

Add `docs/ghidra-riscv.md` in the shape of the MIPS lane guide (admission
boundary, refusal table, why the language is pinned per class, Evidence fields,
and what the fixtures do not establish), and correct
`docs/firmware-analysis.md`, which still told users that MIPS and RISC-V
decompilation had not been added.

Verification after the change: `npm run check` and `npm run docs:check` pass; the
RISC-V module tests pass (17 lane cases plus 9 parser cases); on Ghidra 12.1.4
with JDK 21.0.12.1 the RV32 and RV64 double-float fixtures still analyze and
decompile with `RISCV:LE:32:default` / `RISCV:LE:64:default` and
`language_selection: provider-pinned-v1`, the soft-float fixture is still refused
by name, and the new linkage limitation appears in the emitted Evidence.

* fix(ghidra): align RISC-V pinning with admission

Reuse the complete lane admission boundary when selecting a language, retain unsigned unsupported flag bits in diagnostics, and derive float ABI metadata from the source declaration.

* test(ghidra): verify RISC-V CLI and MCP workflows

Build source-owned RV32/RV64 fixtures with Clang and LLD, establish independent ELF and symbol facts with LLVM readelf, and verify real provider evidence, instruction bytes, calls, decompilation, schema projections, snapshot replay, refusal diagnostics, source immutability and cleanup.

Verified on macOS ARM64 with Ghidra 12.1.4, JDK 21 and LLVM/LLD 23.1.2. Dynamic linkage, big-endian, Andes and RVE/TSO workflows remain unverified; ISA attributes are not inspected.

---------

Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
2026-10-13 00:11:35 +08:00
..

Ghidra conformance corpus

These source-owned fixtures are compiled at verification time; generated binaries and Ghidra projects are never committed.

  • ../c/riscv.c and riscv-probe.S produce source-owned RV32/RV64 ELF images with double-float ABI declarations and an independently known instruction. npm run verify:ghidra:riscv exercises real CLI/MCP discovery, decompilation, output schemas, profile and snapshot identity, refusal diagnostics, immutable source bytes, and owned cleanup. It requires BYO Ghidra 12.1.4/JDK 21+ and RISC-V Clang/LLD/LLVM readelf; REA_RISCV_CLANG, REA_RISCV_LLD and REA_RISCV_READELF select existing tool commands.

  • inventory.c produces separate x86-64 debug and stripped ELF executables. It covers imports, an external puts function, linker thunks, source symbols, direct calls, a targetless callback call, two referenced strings, and a multi-block branch.

  • entry-aliases.c preserves secondary imported entry labels, a bare-hex alias, and an interior label. The native CLI/MCP lane verifies entry selection, annotation, interior-label rejection, and unchanged source bytes.

  • cross-format.c is freestanding so the verifier can produce AArch64 ELF, x86-64 PE, and x86-64 Mach-O targets from the same semantics. It preserves an exported entry, direct and indirect calls, a volatile string reference, and a multi-block branch across loaders.

  • no-return.c and ReaNoReturnProbe.java inject incorrect import flags into a real ELF or Mach-O analysis database, verify caller recovery, propagated flag repair and signature preservation, and retain genuine/unknown no-return calls and independently identified entries. Run npm run build:cached, then node scripts/verify-real-ghidra-noreturn.mjs with the Ghidra/JDK environment below on Linux or macOS x64/arm64. On Linux, use a procps-compatible process table that exposes REA's current process for owned cleanup. It also checks named limitations through the real read-only TCP bridge and provider Evidence projection; default Unix-socket transport, Windows, and the reported Amethyst artifact need separate verification. Fixture target code is never executed.

scripts/verify-real-ghidra.mjs uses cc, clang, and lld-link by default; REA_CC, REA_CLANG, and REA_LLD_LINK can select alternate commands. The verifier requires the verified bring-your-own Ghidra 12.1.4 installation through GHIDRA_INSTALL_DIR and validates header classification before starting the provider. Provider admission accepts the rest of the 12.1.x line; this lane does not.

Conformance compares semantic facts: provider/profile identity, function classification, resolved call edges, typed references, strings/xrefs, and CFG topology. It only requires provider pseudocode and assembly to be non-empty, bounded, and address-bearing; it never compares their text with Hopper. The callback fixture also proves that an unresolved targetless flow is not silently promoted to a direct callee. Every target runs in a separate owned project and must leave no process, socket, project, or runtime root after close.

relative-switch-signed.S covers signed halfword offsets, the CBZ zero-entry route, and selector biases of 0, 2 and 10 in the AArch64 jump-table lane. The nonzero biases exercise overlapping and disjoint typed label/index ranges. REA retains Ghidra's typed labels when the compared selector has an earlier register definition; additional labels remain unknown until their relationship to the typed switch selector can be proved. The oracle rejects extra numeric labels and verifies each retained target against the source instruction immediate.