synth_core/

backend.rs

//! Backend trait and registry for multi-backend compilation
//!
//! Every compiler backend (ARM, aWsm, wasker, w2c2) implements the `Backend`
//! trait, allowing the CLI and verification framework to treat them uniformly.

use crate::target::TargetSpec;
use crate::wasm_decoder::DecodedModule;
use crate::wasm_op::WasmOp;
use crate::wsc_facts::WscFact;
use std::collections::HashMap;
use thiserror::Error;

/// Errors from backend compilation
#[derive(Debug, Error)]
pub enum BackendError {
    #[error("compilation failed: {0}")]
    CompilationFailed(String),

    #[error("backend not available: {0}")]
    NotAvailable(String),

    #[error("unsupported configuration: {0}")]
    UnsupportedConfig(String),

    #[error("external tool error: {0}")]
    ExternalToolError(String),
}

/// Memory-bounds safety strategy. Phase 1 of `docs/binary-safety-design.md` §3.1.
///
/// - `Mpu`/PMP: rely on hardware (ARM MPU or RV32 PMP) — no inline check.
/// - `Software`: emit a `CMP/BHS Trap_Handler` (ARM) or `bgeu addr, mem_size, ebreak` (RV32)
///   before every load/store.
/// - `Mask`: emit `AND addr, addr, #(mem_size - 1)` — only valid when memory size
///   is a power of two. Wraps on OOB rather than trapping (fuzz-profile semantics).
/// - `None`: no bounds enforcement.
#[derive(Debug, Clone, Copy, PartialEq, Eq, Default)]
pub enum SafetyBounds {
    /// No bounds check (caller assumes the WASM module is trusted)
    #[default]
    None,
    /// ARM MPU / RV32 PMP — hardware enforcement, no inline guard
    Mpu,
    /// Software CMP/BHS (ARM) or BGEU+EBREAK (RV32) per access
    Software,
    /// AND-mask, requires power-of-two memory size
    Mask,
}

impl SafetyBounds {
    /// Parse the `--safety-bounds` argument value.
    pub fn parse(s: &str) -> std::result::Result<Self, String> {
        match s {
            "none" => Ok(SafetyBounds::None),
            "mpu" | "pmp" => Ok(SafetyBounds::Mpu),
            "software" | "soft" => Ok(SafetyBounds::Software),
            "mask" | "masking" => Ok(SafetyBounds::Mask),
            other => Err(format!(
                "unknown --safety-bounds value '{}'; expected one of: none, mpu, software, mask",
                other
            )),
        }
    }

    /// String form used in the safety manifest.
    pub fn as_str(self) -> &'static str {
        match self {
            SafetyBounds::None => "none",
            SafetyBounds::Mpu => "mpu",
            SafetyBounds::Software => "software",
            SafetyBounds::Mask => "mask",
        }
    }
}

/// The absolute SRAM address the OPTIMIZED (non-relocatable) ARM path
/// materializes as its linear-memory base (`MOVW/MOVT R12, #base` before each
/// const-address access, and the #468 base-CSE R11 hoist). Historical value:
/// 256 bytes above the SRAM start — the differential-harness contract for
/// optimized-path fixtures maps linmem here. `CompileConfig::linmem_base`
/// defaults to this; `--stack-layout=low` (#687) shifts it up by the reserved
/// stack size so the moved layout reaches user code, not just the startup.
pub const OPTIMIZED_LINMEM_BASE: u32 = 0x2000_0100;

/// Configuration for a compilation run
#[derive(Debug, Clone)]
pub struct CompileConfig {
    /// Optimization level (0 = none, 1 = fast, 2 = default, 3 = aggressive)
    pub opt_level: u8,
    /// Target specification
    pub target: TargetSpec,
    /// Legacy: enable software bounds checking for memory operations.
    /// Deprecated in favor of `safety_bounds`. When set, equivalent to
    /// `SafetyBounds::Software`. Kept for backwards compatibility with
    /// callers that haven't migrated yet.
    pub bounds_check: bool,
    /// Phase-1 unified safety-bounds knob. If `bounds_check` is `true` and
    /// this is `None`, the legacy field wins (back-compat). If both are set,
    /// `safety_bounds` wins.
    pub safety_bounds: SafetyBounds,
    /// Hardware profile name (e.g. "nrf52840", "stm32f407")
    pub hardware: String,
    /// Skip optimization passes (direct instruction selection)
    pub no_optimize: bool,
    /// Use Loom-compatible optimization preset
    pub loom_compat: bool,
    /// Number of imported functions (calls to indices below this use Meld dispatch)
    pub num_imports: u32,
    /// AAPCS integer-argument count per function, indexed by full WASM function
    /// index (imports first, then locals). Lets `Call` marshal the right number
    /// of operand-stack values into R0–R3 (issue #195). Empty = pass no args
    /// (pre-#195 behaviour).
    pub func_arg_counts: Vec<u32>,
    /// #851: result (return-value) count per function, indexed by full WASM
    /// function index (imports first). `0` = void, `1` = one value. The AArch64
    /// direct-`call` lowering needs the 0-vs-1 distinction to decide whether to
    /// push the `x0` result — `func_ret_i64/f32/f64` carry the result TYPE but
    /// conflate void and i32. Empty on backends/paths that do not lower calls
    /// this way (byte-invisible there).
    pub func_result_counts: Vec<u32>,
    /// AAPCS integer-argument count per function type, indexed by type index.
    /// Used by `call_indirect` (issue #195).
    pub type_arg_counts: Vec<u32>,
    /// Produce relocatable (ET_REL) host-link output. When set, the backend
    /// uses the direct instruction selector (`select_with_stack`) rather than
    /// the optimized path: the optimizer materializes an *absolute* linear-
    /// memory base (0x20000100) and does not preserve caller-saved registers
    /// across calls, both wrong for a host-linked object where the linmem base
    /// is supplied via `fp` at runtime and callees follow AAPCS. Imports are
    /// also emitted as direct `func_N` BLs (resolved to the wasm field name)
    /// instead of `__meld_dispatch_import`. (#197 — follow-up to #188/#171.)
    pub relocatable: bool,

    /// #275: the SELF-CONTAINED Thumb-2 `--cortex-m` image path lowers
    /// `call_indirect` through a flash-resident funcref table addressed
    /// PC-RELATIVE (an `LdrSym` literal-pool pointer to
    /// [`FUNC_TABLE_SYMBOL`]) — NEVER through R11, which is the linear-memory
    /// base (the v0.42 #717 collision). Set by the CLI ONLY when the image
    /// builder that emits and patches that table
    /// (`build_multi_func_cortex_m_elf`) will run: Cortex-M family, not
    /// `--relocatable`, no imported functions. Every other self-contained
    /// configuration keeps the loud #275 decline. Default `false`.
    pub self_contained_funcref_table: bool,

    /// #687 (`--stack-layout=low`): the absolute linear-memory base the
    /// OPTIMIZED ARM path materializes into user code. Defaults to
    /// [`OPTIMIZED_LINMEM_BASE`] (`0x2000_0100`, byte-identical to every
    /// pre-#687 compile). Under the low stack layout the CLI shifts it up by
    /// the reserved stack size so const-address loads/stores land in the moved
    /// linear memory instead of the stack region. Only the optimized
    /// (non-relocatable) path consumes it — the direct selector is R11/fp
    /// - relative and follows the startup's R11 init instead.
    pub linmem_base: u32,

    /// #237: emit wasm function-static data as a base-independent `.data`
    /// section (`__synth_wasm_data`) addressed via MOVW/MOVT symbol relocations,
    /// so a host-pointer drop-in (linmem base = 0 for native `*ptr` derefs)
    /// doesn't mis-resolve the statics. Off by default — only the leaves'
    /// base-relative `[R11+const]` path is used unless explicitly requested.
    pub native_pointer_abi: bool,

    /// #237: wasm linear-memory minimum size in bytes — the full static-data
    /// extent (initialized `(data)` segments plus the zero-init/BSS region).
    /// Under `native_pointer_abi`, a const memory address below this is a wasm
    /// static → symbol-relative; any address beyond it is a runtime host pointer
    /// → `[R11=0 + addr]`.
    pub linear_memory_bytes: u32,

    /// VCR-MEM-002 phase 1 (#406): initial size in 64 KiB pages of EACH linear
    /// memory, indexed by memory index. Consulted only by the multi-memory
    /// lowering arms (loads/stores wrapped in `WasmOp::MultiMemory`,
    /// `memory.size`/`grow` with a non-zero index) — memory-0 lowering never
    /// reads it, so single-memory output is byte-identical whether it is set
    /// or empty. Empty (the default) means "no multi-memory context": any
    /// multi-memory op then declines loudly.
    pub memory_pages: Vec<u32>,

    /// #237: the wasm stack-pointer global as `(index, init_value)`, if the
    /// module has one. Under `native_pointer_abi` the backend register-promotes
    /// it: `global.get` materializes `__synth_wasm_data + init` (the real stack
    /// top) and the init value doubles as the static-data base that separates
    /// pointer consts (`>= init`) from frame-size scalars (`< init`).
    pub stack_pointer_global: Option<(u32, i32)>,
    /// #311: per-function (full index) / per-type "returns i64" — the call
    /// lowering must tag i64 results as a register pair or the hi half is
    /// invisible to liveness.
    pub func_ret_i64: Vec<bool>,
    pub type_ret_i64: Vec<bool>,
    /// #643: byte width of each defined global's storage slot, indexed by
    /// global index — 4 for i32/f32, 8 for i64/f64, 16 for v128 (from the
    /// module's global section). The globals table is laid out by SUMMING
    /// these widths: an i64 global needs a register-PAIR store/load at
    /// `[R9, off]`/`[R9, off+4]`, and every later global's offset shifts.
    /// Empty ⇒ every global assumed 4 bytes (the legacy `idx * 4` layout;
    /// hand-built op streams and i32-only modules are byte-identical).
    pub global_widths: Vec<u32>,
    /// #359: declared parameter widths per *function* (full index, imports
    /// first): `func_params_i64[f][k]` is true when param `k` of function `f` is
    /// i64/f64. The AAPCS stack-argument path needs the *declared* widths
    /// (op-stream inference can't see an unused i64 param that still shifts the
    /// incoming-stack layout). The source of truth — a per-function driver loop
    /// (`compile_module` / the CLI loop) indexes it by `func.index` and copies
    /// the slice into [`current_func_params_i64`] before each `compile_function`.
    /// Empty → every param assumed i32 (the legacy path; keeps every function
    /// with <=4 params, or all-i32 params, byte-identical).
    pub func_params_i64: Vec<Vec<bool>>,
    /// #359: declared parameter widths of the function CURRENTLY being compiled
    /// — `current_func_params_i64[k]` is true when param `k` is i64/f64. Set per
    /// function (a cheap clone of the config) from [`func_params_i64`] by the
    /// driver loop, because `compile_function` is shared across backends and
    /// carries no function index. Empty → assume i32.
    pub current_func_params_i64: Vec<bool>,
    /// GI-FPU-002 (#619/#369): per-function declared f32-param mask (full index,
    /// imports first). The driver copies `func_params_f32[f]` into
    /// [`current_func_params_f32`] before each `compile_function`. Empty ⇒
    /// all-non-f32 (byte-identical to before).
    pub func_params_f32: Vec<Vec<bool>>,
    /// GI-FPU-002: declared f32-param mask of the function CURRENTLY being
    /// compiled — `current_func_params_f32[k]` is true when param `k` is f32.
    /// Set per function from [`func_params_f32`], mirroring
    /// [`current_func_params_i64`]. Empty ⇒ no f32 params.
    pub current_func_params_f32: Vec<bool>,
    /// GI-FPU-002 phase 2 (#369): per-function declared f64-param mask (full
    /// index, imports first) and the CURRENT function's slice. Hard-float
    /// targets decline f64-param functions loudly — the legacy width
    /// inference treats an f64 param as an i64 CORE-register pair, which
    /// reads the wrong registers under AAPCS-VFP (the caller put it in a
    /// D-register). Empty ⇒ no f64 params (byte-identical legacy path).
    pub func_params_f64: Vec<Vec<bool>>,
    /// See [`func_params_f64`](Self::func_params_f64).
    pub current_func_params_f64: Vec<bool>,
    /// GI-FPU-002 phase 2 (#719/#369): whether the function CURRENTLY being
    /// compiled returns f32. Set per function from the decoder's `func_ret_f32`.
    /// The direct selector's epilogue uses it to loudly decline a result that
    /// reaches the return in a core register instead of an S-register (a call
    /// that returned f32 as integer-tagged R0 would otherwise be a silent
    /// miscompile — the AAPCS-VFP caller reads S0). `false` for hand-built op
    /// streams / non-f32 returns (byte-identical to before).
    pub current_func_ret_f32: bool,
    /// GI-FPU-002 phase 2 (#719/#369): whether the function CURRENTLY being
    /// compiled returns f64 (D0 under AAPCS-VFP). Same epilogue-soundness role.
    pub current_func_ret_f64: bool,
    /// GI-FPU-002 phase 2 (#719/#369): per-function (full index, imports first)
    /// "returns f32/f64" tables. The direct selector declines a `call` to an
    /// f32/f64-returning callee LOUDLY at the call site — the result arrives in
    /// S0/D0 (AAPCS-VFP), which this increment does not marshal into the operand
    /// stack; tagging it as an integer R0 would be a silent miscompile. Also the
    /// source for [`current_func_ret_f32`]/[`current_func_ret_f64`] in the
    /// per-function driver loops. Empty ⇒ callees assumed non-float-returning
    /// (hand-built op streams; byte-identical legacy behaviour).
    pub func_ret_f32: Vec<bool>,
    /// See [`func_ret_f32`](Self::func_ret_f32).
    pub func_ret_f64: Vec<bool>,
    /// GI-FPU-002 phase 2 (#719/#369): per-type "returns f32/f64" — the
    /// `call_indirect` analogue of [`func_ret_f32`](Self::func_ret_f32).
    pub type_ret_f32: Vec<bool>,
    /// See [`type_ret_f32`](Self::type_ret_f32).
    pub type_ret_f64: Vec<bool>,
    /// #457: DECLARED parameter count of the function CURRENTLY being compiled,
    /// from the module's type section (`func_arg_counts[func.index]`). Set per
    /// function by the driver loops like [`current_func_params_i64`].
    ///
    /// The backends otherwise INFER the param count from local-access patterns
    /// (`count_params`: a local whose first access is a read is assumed to be a
    /// param) — which cannot distinguish a param from a read-before-write
    /// non-param local. WASM zero-initializes non-param locals, so such a local
    /// must read 0; the inference instead homed it in a parameter register and
    /// read caller garbage (#457). The backends cap the inferred count at this
    /// declared count when it is present, which reclassifies exactly the
    /// read-before-write locals (an inferred count can only exceed the declared
    /// one via a read-first index >= the declared count) and leaves every other
    /// function's codegen byte-identical.
    ///
    /// `None` → declared signature unknown (hand-built op streams, direct
    /// `compile_function` callers) → pure inference, the legacy behaviour.
    pub current_func_param_count: Option<u32>,
    /// (#778 phase 4 / #49) The WASM index of the function CURRENTLY being compiled,
    /// so the WCET pass can identify this function's OWN `func_<idx>` self-call label
    /// (a self-recursive `BL func_N` where N == this index) and prove/decline the
    /// self-recursion depth. Set per function by the driver loop (like
    /// [`current_func_params_i64`]). `None` → unknown (hand-built op streams, direct
    /// `compile_function` callers) → no self-recursion certificate is attempted.
    pub current_func_index: Option<u32>,
    /// #509: blocktype-arity side-table of the function CURRENTLY being compiled
    /// — `(param_count, result_count)` of the k-th `Block`/`Loop`/`If` in its op
    /// stream (ordinal-keyed; see [`FunctionOps::block_arity`]). Set per function
    /// by the driver loop (like [`current_func_params_i64`]). The direct selector
    /// uses it to land a value carried by `br`/`br_if`/`br_table` in the target
    /// block's designated result register instead of dropping it. Empty → every
    /// block treated as void (the legacy lowering; hand-built op streams).
    ///
    /// [`FunctionOps::block_arity`]: crate::wasm_decoder::FunctionOps::block_arity
    pub current_func_block_arity: Vec<(u8, u8)>,

    /// #543 Phase 1 — integrator-marked volatile linear-memory segments (the DMA
    /// transfer window). Each range `[base, base+len)` names a region of the fused
    /// linear memory that an EXTERNAL agent (the DMA engine, modelled by gale as a
    /// Component-Model `own<buffer>` handoff — gale decision `DD-DMA-REGION-001`,
    /// gale#124) rewrites out-of-band. Loads and stores whose address falls inside
    /// a marked range must eventually be treated as VOLATILE: not cached, hoisted,
    /// or reordered across the transfer boundary.
    ///
    /// PHASE-2 CONTRACT (implemented — issue #543): the optimizer's
    /// address-caching passes HONOR these ranges. Consumption points:
    ///  - the #468 base-CSE / const-address-fold
    ///    (`optimizer_bridge::plan_base_cse`, DEFAULT-ON, opt-out
    ///    `SYNTH_BASE_CSE=0`): a const-address access whose 4-byte window
    ///    intersects a marked range is EXCLUDED from the fold set — it keeps
    ///    its verbatim per-access materialize-and-access codegen, while
    ///    accesses outside the range still fold;
    ///  - const-CSE (`liveness::apply_const_cse` wired in `arm_backend.rs`,
    ///    DEFAULT-ON, opt-out `SYNTH_CONST_CSE=0`; the former bridge-level
    ///    inline cache is retired, #242): declines WHOLESALE while any range is
    ///    marked — a cached constant cannot be classified address-vs-data at
    ///    that level, so the conservative stance for statically-unknown
    ///    addressing is to re-materialize every constant at each occurrence.
    ///
    /// Passes that only touch SP-relative frame slots (stack-reload forwarding,
    /// frame-slot DCE, spill re-choice) are unaffected by design: these ranges
    /// are LINEAR-MEMORY addresses, and frame slots are never linmem. Nothing on
    /// the pipeline deletes, forwards, or reorders a linear-memory access (IR CSE
    /// deliberately never CSEs `MemLoad`s; DCE removes only unreachable blocks),
    /// so every marked access is issued verbatim, in program order.
    ///
    /// Empty (the default): zero behavior change by construction — every gate
    /// reduces to the pre-#543 path, so the emitted `.text` is byte-identical
    /// with or without this code (the frozen-codegen gate holds). See rivet
    /// `VCR-DMA-001`.
    pub volatile_segments: Vec<VolatileRange>,

    /// #778 phase 2 — the parsed `--wcet-hints` file (UNTRUSTED per-function
    /// loop-bound hints, the scry seam). Consulted ONLY by the WCET sidecar
    /// computation over the final instruction stream; NEVER by codegen — the
    /// emitted bytes are byte-identical with or without hints. Every hint is
    /// soundly verified before use and rejected with a machine reason
    /// otherwise.
    pub wcet_hints: Option<crate::wcet::WcetHints>,

    /// VCR-PERF-002 Phase 1 (#494) — proven invariants forwarded by loom in
    /// the `wsc.facts` custom section (encoding:
    /// `docs/design/wsc-facts-encoding.md`; program:
    /// `docs/design/proof-carrying-specialization.md`), whole-module table
    /// keyed by `(func_index, value_id)`. The compile driver copies the
    /// current function's slice into [`current_func_facts`] (the
    /// `func_params_i64` → `current_func_params_i64` pattern), because
    /// `compile_function` carries no function index.
    ///
    /// PHASE-1 CONTRACT: threaded but NOT consumed — no codegen path reads
    /// facts, so emitted bytes are unchanged whether or not the module
    /// carries the section (locked by `wsc_facts_ingestion_494.rs`). Phase 2
    /// turns each fact into a premise for a flag-gated (`SYNTH_FACT_SPEC`),
    /// per-elision ordeal-validated specialization; the facts-absent compile
    /// stays byte-identical by construction (empty ⇒ every gate vacuous).
    ///
    /// [`current_func_facts`]: CompileConfig::current_func_facts
    pub wsc_facts: Vec<WscFact>,
    /// VCR-PERF-002 Phase 1 (#494): the `wsc.facts` invariants of the function
    /// CURRENTLY being compiled (`fact.func_index == func.index`), set per
    /// function by the driver loops like [`current_func_params_i64`]. This is
    /// the field a Phase-2 selector pass will read its premises from. Empty →
    /// no facts → no specialization may ever fire (the fail-safe default).
    ///
    /// [`current_func_params_i64`]: CompileConfig::current_func_params_i64
    pub current_func_facts: Vec<WscFact>,
    /// VCR-PERF-002 Phase 2b (#494, divisor-nonzero): op indices (into the op
    /// stream passed to `compile_function`) of `div`/`rem` ops whose
    /// DIVIDE-BY-ZERO trap guard is proven dead — the fact-spec pass
    /// discharged `UNSAT(P ∧ divisor == 0)` per site through the
    /// certificate-checked ordeal solver BEFORE the driver set this field.
    /// Consumed by the ARM direct selector (`select_with_stack`); every other
    /// path ignores it (guards stay — sound). Empty (the default) ⇒ every
    /// guard is emitted, byte-identical to today.
    pub fact_div_zero_elide: Vec<usize>,
    /// VCR-PERF-002 Phase 2b (#494): op indices of `div_s` ops whose
    /// `INT_MIN / -1` OVERFLOW trap guard is proven dead — a SEPARATE
    /// obligation (`UNSAT(P ∧ dividend == INT_MIN ∧ divisor == -1)`). A
    /// divisor-nonzero fact alone NEVER lands here: divisor ≠ 0 does not
    /// exclude -1 (#633/#634 two-guard distinction). Empty ⇒ guard emitted.
    pub fact_div_ovf_elide: Vec<usize>,
    /// #494 bounds-elision (#390 `guard_bool`): op indices of i32 memory
    /// accesses whose `--safety-bounds software` inline guard is proven dead
    /// — the fact-spec pass discharged
    /// `UNSAT(P ∧ trap_mem_oob(zext64(index) + offset, size,
    /// min_memory_bytes))` per site through the certificate-checked ordeal
    /// solver BEFORE the driver set this field (ordeal 0.9.1 `trap_mem_oob`
    /// shape, wraparound-safe 64-bit extension). Consumed by the ARM direct
    /// selector (`select_with_stack`); every other path ignores it (guards
    /// stay — sound). Empty (the default) ⇒ every guard is emitted,
    /// byte-identical to today.
    pub fact_mem_bounds_elide: Vec<usize>,
    /// #642: `call_indirect` guard inputs — the compile-time table size for
    /// the runtime bounds check and the per-expected-type closed-world type
    /// verdicts — computed from the decoded module by
    /// [`crate::wasm_decoder::DecodedModule::call_indirect_guards`] and set by
    /// the driver loops. The default (`table_size: None`, empty verdicts)
    /// DECLINES every `call_indirect` lowering: an unchecked indirect branch
    /// is never emitted (WASM Core §4.4.8 requires OOB/type-mismatch traps).
    pub call_indirect_guards: crate::wasm_decoder::CallIndirectGuards,
    /// #851 lane L3: result count per FUNCTION TYPE (see
    /// [`crate::wasm_decoder::DecodedModule::type_result_counts`]). The aarch64
    /// `call_indirect` lowering needs the 0-vs-1 result distinction for a callee
    /// it knows only by its static type.
    pub type_result_counts: Vec<u32>,
    /// #851 lane L3: the STRUCTURAL signature class id per function type (see
    /// [`crate::wasm_decoder::DecodedModule::structural_type_class_ids`]). The
    /// aarch64 `call_indirect` type check compares this, not the raw type index
    /// — WASM type equality is structural. Distinct from
    /// `call_indirect_guards.type_class_ids`, which the ARM path populates only
    /// when its heterogeneous-table sidecar exists.
    pub type_class_ids: Vec<u32>,
    /// #851 lane L3, aarch64 only — the driver has EMITTED the module-level
    /// substrate the globals and `call_indirect` lowerings address: the `.data`
    /// globals image (`__synth_globals`) and the `.text` funcref table
    /// (`__synth_func_table`), both produced by
    /// `synth_backend_aarch64::substrate::plan`.
    ///
    /// FAIL-SAFE BY DEFAULT (`false`): the aarch64 selector LOUD-DECLINES
    /// `global.get`/`global.set`/`call_indirect` unless this is set, so a driver
    /// that compiles function bodies but never emits the regions cannot ship
    /// code addressing a symbol that does not exist. Set only on the two paths
    /// that call `plan()` and place its output in the object.
    pub a64_substrate_emitted: bool,
}

/// #543 — an integrator-marked volatile linear-memory segment (the DMA transfer
/// window): the half-open byte range `[base, base + len)` of the fused linear
/// memory that an external agent rewrites out-of-band. Parsed from the CLI
/// `--volatile-segment <base>:<len>` flag. See [`CompileConfig::volatile_segments`]
/// for the Phase-1/Phase-2 split.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub struct VolatileRange {
    /// Start address of the volatile region, in linear-memory bytes.
    pub base: u32,
    /// Length of the volatile region, in bytes. The region is `[base, base+len)`.
    pub len: u32,
}

impl CompileConfig {
    /// Resolve the effective safety-bounds setting, honouring the legacy
    /// `bounds_check` field as a fallback. Used by backends to pick the
    /// inline-check shape.
    pub fn effective_safety_bounds(&self) -> SafetyBounds {
        match (self.safety_bounds, self.bounds_check) {
            (SafetyBounds::None, true) => SafetyBounds::Software,
            (s, _) => s,
        }
    }
}

impl Default for CompileConfig {
    fn default() -> Self {
        Self {
            opt_level: 2,
            target: TargetSpec::cortex_m4(),
            bounds_check: false,
            safety_bounds: SafetyBounds::None,
            hardware: String::new(),
            no_optimize: false,
            loom_compat: false,
            num_imports: 0,
            func_arg_counts: Vec::new(),
            func_result_counts: Vec::new(),
            type_arg_counts: Vec::new(),
            relocatable: false,
            // #275: self-contained funcref-table dispatch is opt-in by the
            // CLI's cortex-m image path; everything else keeps the decline.
            self_contained_funcref_table: false,
            // #687: the historical optimized-path absolute base — every
            // default compile stays byte-identical.
            linmem_base: OPTIMIZED_LINMEM_BASE,
            native_pointer_abi: false,
            linear_memory_bytes: 0,
            // #406: empty ⇒ no multi-memory context ⇒ multi-memory ops decline
            // loudly; memory-0 lowering never reads it.
            memory_pages: Vec::new(),
            stack_pointer_global: None,
            func_ret_i64: Vec::new(),
            type_ret_i64: Vec::new(),
            // #643: empty ⇒ legacy all-4-byte global slots (i32-only modules).
            global_widths: Vec::new(),
            func_params_i64: Vec::new(),
            current_func_params_i64: Vec::new(),
            func_params_f32: Vec::new(),
            current_func_params_f32: Vec::new(),
            // GI-FPU-002 phase 2 (#719/#369): false ⇒ non-float return (or a
            // hand-built op stream); driver loops set it per function.
            current_func_ret_f32: false,
            current_func_ret_f64: false,
            // GI-FPU-002 phase 2 (#719/#369): empty ⇒ callees assumed
            // non-float-returning (hand-built op streams).
            func_params_f64: Vec::new(),
            current_func_params_f64: Vec::new(),
            func_ret_f32: Vec::new(),
            func_ret_f64: Vec::new(),
            type_ret_f32: Vec::new(),
            type_ret_f64: Vec::new(),
            // #457: None ⇒ declared signature unknown ⇒ param-count inference
            // only (unit tests / hand-built op streams); driver loops fill it.
            current_func_param_count: None,
            current_func_index: None,
            // #509: empty ⇒ legacy void-block lowering (unit tests / hand-built
            // op streams); the driver loops fill it per function.
            current_func_block_arity: Vec::new(),
            // #543 Phase 1: no volatile segments unless the CLI flag names them.
            // Empty ⇒ inert ⇒ emitted bytes unchanged.
            volatile_segments: Vec::new(),
            // VCR-PERF-002 Phase 1 (#494): no facts unless the module carries
            // a parseable `wsc.facts` section. Empty ⇒ inert (and Phase 1 has
            // no consumer anyway) ⇒ emitted bytes unchanged.
            wsc_facts: Vec::new(),
            current_func_facts: Vec::new(),
            // VCR-PERF-002 Phase 2b (#494): no guard-elision marks unless the
            // fact-spec pass discharged the per-site obligations. Empty ⇒
            // every div/rem trap guard is emitted, byte-identical.
            fact_div_zero_elide: Vec::new(),
            fact_div_ovf_elide: Vec::new(),
            fact_mem_bounds_elide: Vec::new(),
            // #642: no guard inputs ⇒ every call_indirect lowering declines
            // loudly (never an unchecked indirect branch). Driver loops fill
            // this from the decoded module.
            call_indirect_guards: crate::wasm_decoder::CallIndirectGuards::default(),
            type_result_counts: Vec::new(),
            type_class_ids: Vec::new(),
            a64_substrate_emitted: false,
            // #778 phase 2: no --wcet-hints file ⇒ no hints. Consulted ONLY by
            // the WCET sidecar computation — never by codegen (the emitted
            // bytes are byte-identical with or without hints).
            wcet_hints: None,
        }
    }
}

/// #275: the base symbol of the SELF-CONTAINED funcref table — the
/// flash-resident region `build_multi_func_cortex_m_elf` appends after the
/// function code: one 4-byte code pointer per table slot across ALL tables in
/// declaration order (the same contiguous layout the `--relocatable` R11
/// contract uses — `TableGuards::base_byte_offset` stays valid verbatim),
/// null slots as ZERO words (#664), followed by the #676 type-id sidecar at
/// `type_ids_byte_offset` when a heterogeneous table needs it. The dispatch
/// reaches it through an `LdrSym` literal-pool word (an `Abs32` reloc against
/// this symbol) that the image builder patches post-layout — never through
/// R11, which is the linear-memory base (the #717 collision).
pub const FUNC_TABLE_SYMBOL: &str = "__synth_func_table";

/// A relocation entry produced during compilation
///
/// Records that a BL instruction at `offset` bytes into the function's code
/// targets an external symbol (e.g., `__meld_dispatch_import`). The linker
/// resolves these when combining the Synth object with the Kiln bridge.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub enum RelocKind {
    /// R_ARM_THM_CALL — a Thumb BL call site (the default; #167).
    ThmCall,
    /// R_ARM_MOVW_ABS_NC — the MOVW half of a symbol-relative address (#237).
    MovwAbs,
    /// R_ARM_MOVT_ABS — the MOVT half of a symbol-relative address (#237).
    MovtAbs,
    /// R_ARM_ABS32 — a 32-bit absolute address held in a `.text` literal-pool
    /// word, loaded via `LDR rX, [pc, #off]` (#345). The link-survivable
    /// replacement for the inline-immediate MOVW/MOVT-ABS pair: `ld`/bfd patches
    /// the data word at link time (`S + A`, the addend living in the word, REL
    /// semantics), which survives placement into a large multi-object image —
    /// whereas an inline-instruction MOVW_ABS immediate can be mangled.
    Abs32,
    /// R_AARCH64_CALL26 (ELF type 283) — an AArch64 `BL` call site (#851). The
    /// AArch64 analogue of [`RelocKind::ThmCall`]: the linker patches the 26-bit
    /// word-offset immediate of the `bl` at `offset` to reach the target symbol.
    /// Emitted only by the `EM_AARCH64` backend's `.rela.text`.
    AArch64Call26,
    /// R_AARCH64_JUMP26 (ELF type 282) — an AArch64 `B` (tail-branch) site
    /// (#851 lane L3). Same 26-bit word-offset immediate as
    /// [`RelocKind::AArch64Call26`], but for a branch that does NOT set `x30`:
    /// the aarch64 `call_indirect` funcref table is a `.text`-resident array of
    /// `b func_N` trampolines, so the dispatch's `blr` sets the return address
    /// and the trampoline tail-branches into the callee (which returns straight
    /// to the dispatcher).
    AArch64Jump26,
    /// R_AARCH64_ADR_PREL_PG_HI21 (ELF type 275) — the `adrp` half of a
    /// PC-relative symbol address (#851 lane L3). Patches the 21-bit page delta
    /// (`immlo`[30:29] + `immhi`[23:5]) so `adrp xd, sym` reaches the 4 KiB page
    /// containing `sym`. Always paired with an
    /// [`RelocKind::AArch64AddAbsLo12Nc`] on the next instruction. This pair is
    /// how aarch64 reaches a synth-EMITTED region (the globals `.data` image,
    /// the funcref table) with NO dedicated base register — so neither feature
    /// adds an embedder precondition alongside `x28`.
    AArch64AdrPrelPgHi21,
    /// R_AARCH64_ADD_ABS_LO12_NC (ELF type 277) — the `add xd, xd, :lo12:sym`
    /// half of a PC-relative symbol address (#851 lane L3). Patches the 12-bit
    /// immediate field [21:10] with `(S + A) & 0xFFF`.
    AArch64AddAbsLo12Nc,
    /// R_RISCV_CALL_PLT (ELF type 19) — a RISC-V `auipc`+`jalr` call pair
    /// (#871). The RV32 analogue of [`RelocKind::ThmCall`]: `offset` points at
    /// the `auipc` of an 8-byte `auipc ra, 0 ; jalr ra, 0(ra)` placeholder and
    /// the linker patches BOTH instructions' immediates to reach the target
    /// symbol (the modern form; `R_RISCV_CALL` is deprecated). Emitted only by
    /// the `EM_RISCV` backend's `.rela.text`.
    RiscvCallPlt,
}

#[derive(Debug, Clone, PartialEq, Eq)]
pub struct CodeRelocation {
    /// Byte offset within the function's machine code where the reloc applies
    pub offset: u32,
    /// Target symbol name (e.g., "__meld_dispatch_import", "__synth_wasm_data")
    pub symbol: String,
    /// Which ARM relocation type to emit for this site.
    pub kind: RelocKind,
}

/// VCR-DBG-001: a per-instruction source map — `(machine_offset_within_code,
/// wasm_op_index)` pairs, one per emitted machine instruction. A `None` op-index
/// marks an instruction with no originating wasm op (prologue/epilogue, literal
/// pool). Consumed by the DWARF `.debug_line` emitter; empty when no source map
/// was produced.
pub type LineMap = Vec<(u32, Option<usize>)>;

/// VCR-DEC-003 (#396, witness#130): the object-level control-flow class of one
/// emitted machine instruction, captured at encode time alongside [`LineMap`].
/// It is the piece post-hoc CLI derivation cannot recover — `line_map` records
/// which wasm op an instruction came from, but not whether that instruction IS a
/// conditional branch, an unconditional branch, or a predicated (IT-block) move.
/// The `synth-provenance-v1` emitter needs it to enumerate the ACTUAL object
/// conditional branches (so it can prove "every object branch resolves to a
/// source condition", not just "every source branch has an object PC").
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub enum BranchClass {
    /// A conditional branch (`Bcc`/`Blo`/`Bhs`/`BCondOffset`) — an object-level
    /// decision point MC/DC must account for.
    CondBranch,
    /// An unconditional branch (`B`/`BOffset`) — control flow, not a decision.
    UncondBranch,
    /// A predicated conditional move (`SelectMove`, the IT-block form the
    /// cmp→select fuse produces) — a folded decision with no branch.
    Predicated,
    /// Anything else (data-processing, load/store, call, prologue/epilogue).
    Other,
}

/// VCR-DEC-003: per-instruction object-branch class, parallel to [`LineMap`]
/// (same length, same order — one entry per emitted machine instruction).
/// `(machine_offset_within_code, class)`. Empty when provenance is not being
/// produced (never serialized into `.text`; frozen-safe additive metadata).
pub type BranchMap = Vec<(u32, BranchClass)>;

/// A single compiled function
#[derive(Debug, Clone)]
pub struct CompiledFunction {
    /// Function name (from WASM export or generated)
    pub name: String,
    /// Raw machine code bytes
    pub code: Vec<u8>,
    /// Original WASM ops (retained for verification)
    pub wasm_ops: Vec<WasmOp>,
    /// Relocations for external symbol references (BL to bridge functions)
    pub relocations: Vec<CodeRelocation>,
    /// VCR-DBG-001: per-instruction source map for DWARF `.debug_line` emission —
    /// `(machine_offset_within_code, wasm_op_index)` captured at encode time, one
    /// entry per emitted machine instruction. A `None` op-index marks an
    /// instruction with no originating wasm op (prologue/epilogue, literal-pool
    /// word). This is purely additive metadata: it is never serialized unless
    /// `.debug_line` emission is requested, so the emitted `.text` is
    /// byte-identical with or without it. Empty for backends/paths that do not
    /// yet produce a source map (RISC-V, the optimized ARM path).
    pub line_map: LineMap,
    /// VCR-DEC-003 (#396): per-instruction object-branch class, parallel to
    /// `line_map`. Lets the `synth-provenance-v1` emitter enumerate the real
    /// object conditional branches (not just re-walk the wasm branch ops).
    /// Purely additive metadata: never serialized into `.text`, so emitted bytes
    /// are byte-identical with or without it. Empty for backends/paths that do
    /// not produce it (RISC-V, the optimized ARM path).
    pub branch_map: BranchMap,
    /// #778 (v0.46): the SOUND static worst-case-cycle bound for this function,
    /// or a loud decline, computed over the final Thumb-2 instruction stream (see
    /// [`crate::wcet`]). `Some` only when the ARM backend produced it (the RISC-V
    /// and AArch64 backends carry no cycle model yet → `None`). Purely additive
    /// metadata: derived from the already-decided instruction list, never
    /// serialized into `.text`, so emitted bytes are byte-identical with or
    /// without it (frozen-safe). Emitted as the `<output>.wcet.json` sidecar only
    /// under `--emit-wcet`.
    pub wcet: Option<crate::wcet::WcetFunction>,
    /// #778 phase 3: the per-function WCET INTERMEDIATE (own-body cycles + direct
    /// call sites, or a composition-independent decline) BEFORE inter-procedural
    /// composition. The module driver composes these across the direct call graph
    /// into the final per-function bounds (a caller's bound = its own body + each
    /// direct callee's bound × the call site's proven execution count). `Some` only
    /// on the Thumb-2 path that produced `wcet`. Purely additive, `.text`-invisible
    /// (frozen-safe) — derived from the already-decided instruction list.
    pub wcet_intermediate: Option<crate::wcet::WcetIntermediate>,
}

/// Result of compiling a full module
#[derive(Debug)]
pub struct CompilationResult {
    /// Compiled functions
    pub functions: Vec<CompiledFunction>,
    /// Complete ELF binary (if backend produces one directly)
    pub elf: Option<Vec<u8>>,
    /// Name of the backend that produced this result
    pub backend_name: String,
}

/// What a backend can and cannot do
#[derive(Debug, Clone)]
pub struct BackendCapabilities {
    /// Backend produces complete ELF files (external backends like aWsm)
    pub produces_elf: bool,
    /// Backend supports per-rule verification (only our custom ARM backend)
    pub supports_rule_verification: bool,
    /// Backend supports binary-level verification (all backends via disassembly)
    pub supports_binary_verification: bool,
    /// Backend is an external tool (not a library)
    pub is_external: bool,
}

/// Trait that every compilation backend implements
pub trait Backend: Send + Sync {
    /// Human-readable backend name
    fn name(&self) -> &str;

    /// What this backend can do
    fn capabilities(&self) -> BackendCapabilities;

    /// Which targets this backend supports
    fn supported_targets(&self) -> Vec<TargetSpec>;

    /// Compile an entire decoded WASM module
    fn compile_module(
        &self,
        module: &DecodedModule,
        config: &CompileConfig,
    ) -> std::result::Result<CompilationResult, BackendError>;

    /// Compile a single function from WASM ops to machine code
    fn compile_function(
        &self,
        name: &str,
        ops: &[WasmOp],
        config: &CompileConfig,
    ) -> std::result::Result<CompiledFunction, BackendError>;

    /// Check if this backend is available (external tools installed, etc.)
    fn is_available(&self) -> bool;
}

/// Registry of available backends
pub struct BackendRegistry {
    backends: HashMap<String, Box<dyn Backend>>,
}

impl BackendRegistry {
    pub fn new() -> Self {
        Self {
            backends: HashMap::new(),
        }
    }

    /// Register a backend under its name
    pub fn register(&mut self, backend: Box<dyn Backend>) {
        let name = backend.name().to_string();
        self.backends.insert(name, backend);
    }

    /// Get a backend by name
    pub fn get(&self, name: &str) -> Option<&dyn Backend> {
        self.backends.get(name).map(|b| b.as_ref())
    }

    /// List all registered backends
    pub fn list(&self) -> Vec<&dyn Backend> {
        self.backends.values().map(|b| b.as_ref()).collect()
    }

    /// List backends that are actually available (installed and working)
    pub fn available(&self) -> Vec<&dyn Backend> {
        self.backends
            .values()
            .filter(|b| b.is_available())
            .map(|b| b.as_ref())
            .collect()
    }
}

impl Default for BackendRegistry {
    fn default() -> Self {
        Self::new()
    }
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_registry_empty() {
        let reg = BackendRegistry::new();
        assert!(reg.list().is_empty());
        assert!(reg.available().is_empty());
        assert!(reg.get("arm").is_none());
    }

    #[test]
    fn test_compile_config_default() {
        let config = CompileConfig::default();
        assert_eq!(config.opt_level, 2);
        assert!(!config.bounds_check);
        assert_eq!(config.safety_bounds, SafetyBounds::None);
        assert!(!config.no_optimize);
    }

    #[test]
    fn safety_bounds_parse_round_trip() {
        for s in ["none", "mpu", "software", "mask"] {
            let sb = SafetyBounds::parse(s).unwrap();
            assert_eq!(sb.as_str(), s);
        }
        assert_eq!(SafetyBounds::parse("pmp").unwrap(), SafetyBounds::Mpu);
        assert_eq!(SafetyBounds::parse("soft").unwrap(), SafetyBounds::Software);
        assert!(SafetyBounds::parse("nonsense").is_err());
    }

    #[test]
    fn effective_safety_bounds_legacy_promotes_to_software() {
        let cfg = CompileConfig {
            bounds_check: true,
            ..Default::default()
        };
        assert_eq!(cfg.effective_safety_bounds(), SafetyBounds::Software);
    }

    #[test]
    fn effective_safety_bounds_new_field_wins() {
        let cfg = CompileConfig {
            bounds_check: true,
            safety_bounds: SafetyBounds::Mpu,
            ..Default::default()
        };
        assert_eq!(cfg.effective_safety_bounds(), SafetyBounds::Mpu);
    }
}