Struct BenchReport 

pub struct BenchReport {
    pub scenario: ScenarioMetadata,
    pub backend: BackendInfo,
    pub host: HostInfo,
    pub iterations: IterationStats,
    pub response_bytes: Option<usize>,
    pub expected_match: Option<bool>,
    pub passes_applied: Vec<String>,
    pub compiler_visible_allocs: Option<usize>,
}

Fields

scenario: ScenarioMetadata

backend: BackendInfo

Identifies the build that ran the bench: aver version, build profile, target backend, plus optional version strings for per-target runtimes (e.g. wasmtime for wasm-local).

host: HostInfo

OS / architecture / process identity. Same JSON shape across targets; downstream tools join on host.os + host.arch + backend.name to compare like-for-like across runs.

iterations: IterationStats

response_bytes: Option<usize>

UTF-8 byte count of the last iteration’s “result”. Semantics vary by target:

• vm: byte length of main’s return value rendered through aver_display (same path Console.print uses). None when main returns Unit — those scenarios print for side effect, and bench mode silences the console.
• wasm-local: total bytes the guest tried to write through fd_write (sum of iovec lengths) during the last iteration. 0 when the guest never called fd_write (most scenarios that don’t print).
• rust: actual stdout byte count from the spawned binary’s subprocess output. 0 when the binary printed nothing.

VM and wasm-local/rust use different shapes (“rendered return value” vs “actual stdout”). aver bench --compare only ever matches same-target baselines, so the divergence doesn’t break gating — the field is exact-match within a target.

expected_match: Option<bool>

true when the run satisfied every [expected] constraint in the manifest. null when the manifest has no expectations.

passes_applied: Vec<String>

Pipeline stages that actually fired. Sourced from the pipeline’s on_after_pass hook so it reflects what ran, not what was requested.

compiler_visible_allocs: Option<usize>

IR-level allocation counter. null in 0.15.1 — pending the aver compile --explain-allocations work in 0.15.2.

Trait Implementations

impl Clone for BenchReport

fn clone(&self) -> BenchReport

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for BenchReport

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for BenchReport

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for BenchReport

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.