Add build metrics to rustbuild

Cargo.lock

@@ -223,6 +223,7 @@ dependencies = [
  "pretty_assertions",
  "serde",
  "serde_json",
+ "sysinfo",
  "tar",
  "toml",
  "winapi",
@@ -5057,6 +5058,21 @@ dependencies = [
  "unicode-xid",
 ]
 
+[[package]]
+name = "sysinfo"
+version = "0.24.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a6a8e71535da31837213ac114531d31def75d7aebd133264e420a3451fa7f703"
+dependencies = [
+ "cfg-if 1.0.0",
+ "core-foundation-sys",
+ "libc",
+ "ntapi",
+ "once_cell",
+ "rayon",
+ "winapi",
+]
+
 [[package]]
 name = "tar"
 version = "0.4.37"

config.toml.example

@@ -328,6 +328,12 @@ changelog-seen = 2
 # a Nix toolchain on non-NixOS distributions.
 #patch-binaries-for-nix = false
 
+# Collect information and statistics about the current build and writes it to
+# disk. Enabling this or not has no impact on the resulting build output. The
+# schema of the file generated by the build metrics feature is unstable, and
+# this is not intended to be used during local development.
+#metrics = false
+
 # =============================================================================
 # General install configuration options
 # =============================================================================

src/bootstrap/Cargo.toml

@@ -49,6 +49,9 @@ opener = "0.5"
 once_cell = "1.7.2"
 xz2 = "0.1"
 
+# Dependencies needed by the build-metrics feature
+sysinfo = { version = "0.24.1", optional = true }
+
 [target.'cfg(windows)'.dependencies.winapi]
 version = "0.3"
 features = [
@@ -64,3 +67,6 @@ features = [
 
 [dev-dependencies]
 pretty_assertions = "0.7"
+
+[features]
+build-metrics = ["sysinfo"]

src/bootstrap/bootstrap.py

@@ -896,6 +896,9 @@ def build_bootstrap(self):
             args.append("--locked")
         if self.use_vendored_sources:
             args.append("--frozen")
+        if self.get_toml("metrics", "build"):
+            args.append("--features")
+            args.append("build-metrics")
         run(args, env=env, verbose=self.verbose)
 
     def build_triple(self):

src/bootstrap/builder.rs

@@ -1757,6 +1757,9 @@ impl<'a> Builder<'a> {
             stack.push(Box::new(step.clone()));
         }
 
+        #[cfg(feature = "build-metrics")]
+        self.metrics.enter_step(&step);
+
         let (out, dur) = {
             let start = Instant::now();
             let zero = Duration::new(0, 0);
@@ -1780,6 +1783,9 @@ impl<'a> Builder<'a> {
             );
         }
 
+        #[cfg(feature = "build-metrics")]
+        self.metrics.exit_step();
+
         {
             let mut stack = self.stack.borrow_mut();
             let cur_step = stack.pop().expect("step stack empty");

src/bootstrap/config.rs

@@ -544,6 +544,7 @@ define_config! {
         dist_stage: Option<u32> = "dist-stage",
         bench_stage: Option<u32> = "bench-stage",
         patch_binaries_for_nix: Option<bool> = "patch-binaries-for-nix",
+        metrics: Option<bool> = "metrics",
     }
 }
 

src/bootstrap/lib.rs

@@ -149,6 +149,9 @@ mod tool;
 mod toolstate;
 pub mod util;
 
+#[cfg(feature = "build-metrics")]
+mod metrics;
+
 #[cfg(windows)]
 mod job;
 
@@ -311,6 +314,9 @@ pub struct Build {
     prerelease_version: Cell<Option<u32>>,
     tool_artifacts:
         RefCell<HashMap<TargetSelection, HashMap<String, (&'static str, PathBuf, Vec<String>)>>>,
+
+    #[cfg(feature = "build-metrics")]
+    metrics: metrics::BuildMetrics,
 }
 
 #[derive(Debug)]
@@ -500,6 +506,9 @@ impl Build {
             delayed_failures: RefCell::new(Vec::new()),
             prerelease_version: Cell::new(None),
             tool_artifacts: Default::default(),
+
+            #[cfg(feature = "build-metrics")]
+            metrics: metrics::BuildMetrics::init(),
         };
 
         build.verbose("finding compilers");
@@ -692,6 +701,9 @@ impl Build {
             }
             process::exit(1);
         }
+
+        #[cfg(feature = "build-metrics")]
+        self.metrics.persist(self);
     }
 
     /// Clear out `dir` if `input` is newer.

src/bootstrap/metrics.rs

@@ -0,0 +1,208 @@
+//! This module is responsible for collecting metrics profiling information for the current build
+//! and dumping it to disk as JSON, to aid investigations on build and CI performance.
+//!
+//! As this module requires additional dependencies not present during local builds, it's cfg'd
+//! away whenever the `build.metrics` config option is not set to `true`.
+
+use crate::builder::Step;
+use crate::util::t;
+use crate::Build;
+use serde::{Deserialize, Serialize};
+use std::cell::RefCell;
+use std::fs::File;
+use std::io::BufWriter;
+use std::time::{Duration, Instant};
+use sysinfo::{CpuExt, System, SystemExt};
+
+pub(crate) struct BuildMetrics {
+    state: RefCell<MetricsState>,
+}
+
+impl BuildMetrics {
+    pub(crate) fn init() -> Self {
+        let state = RefCell::new(MetricsState {
+            finished_steps: Vec::new(),
+            running_steps: Vec::new(),
+
+            system_info: System::new(),
+            timer_start: None,
+            invocation_timer_start: Instant::now(),
+        });
+
+        BuildMetrics { state }
+    }
+
+    pub(crate) fn enter_step<S: Step>(&self, step: &S) {
+        let mut state = self.state.borrow_mut();
+
+        // Consider all the stats gathered so far as the parent's.
+        if !state.running_steps.is_empty() {
+            self.collect_stats(&mut *state);
+        }
+
+        state.system_info.refresh_cpu();
+        state.timer_start = Some(Instant::now());
+
+        state.running_steps.push(StepMetrics {
+            type_: std::any::type_name::<S>().into(),
+            debug_repr: format!("{step:?}"),
+
+            cpu_usage_time_sec: 0.0,
+            duration_excluding_children_sec: Duration::ZERO,
+
+            children: Vec::new(),
+        });
+    }
+
+    pub(crate) fn exit_step(&self) {
+        let mut state = self.state.borrow_mut();
+
+        self.collect_stats(&mut *state);
+
+        let step = state.running_steps.pop().unwrap();
+        if state.running_steps.is_empty() {
+            state.finished_steps.push(step);
+            state.timer_start = None;
+        } else {
+            state.running_steps.last_mut().unwrap().children.push(step);
+
+            // Start collecting again for the parent step.
+            state.system_info.refresh_cpu();
+            state.timer_start = Some(Instant::now());
+        }
+    }
+
+    fn collect_stats(&self, state: &mut MetricsState) {
+        let step = state.running_steps.last_mut().unwrap();
+
+        let elapsed = state.timer_start.unwrap().elapsed();
+        step.duration_excluding_children_sec += elapsed;
+
+        state.system_info.refresh_cpu();
+        let cpu = state.system_info.cpus().iter().map(|p| p.cpu_usage()).sum::<f32>();
+        step.cpu_usage_time_sec += cpu as f64 / 100.0 * elapsed.as_secs_f64();
+    }
+
+    pub(crate) fn persist(&self, build: &Build) {
+        let mut state = self.state.borrow_mut();
+        assert!(state.running_steps.is_empty(), "steps are still executing");
+
+        let dest = build.out.join("metrics.json");
+
+        let mut system = System::new();
+        system.refresh_cpu();
+        system.refresh_memory();
+
+        let system_stats = JsonInvocationSystemStats {
+            cpu_threads_count: system.cpus().len(),
+            cpu_model: system.cpus()[0].brand().into(),
+
+            memory_total_bytes: system.total_memory() * 1024,
+        };
+        let steps = std::mem::take(&mut state.finished_steps);
+
+        // Some of our CI builds consist of multiple independent CI invocations. Ensure all the
+        // previous invocations are still present in the resulting file.
+        let mut invocations = match std::fs::read(&dest) {
+            Ok(contents) => t!(serde_json::from_slice::<JsonRoot>(&contents)).invocations,
+            Err(err) => {
+                if err.kind() != std::io::ErrorKind::NotFound {
+                    panic!("failed to open existing metrics file at {}: {err}", dest.display());
+                }
+                Vec::new()
+            }
+        };
+        invocations.push(JsonInvocation {
+            duration_including_children_sec: state.invocation_timer_start.elapsed().as_secs_f64(),
+            children: steps.into_iter().map(|step| self.prepare_json_step(step)).collect(),
+        });
+
+        let json = JsonRoot { system_stats, invocations };
+
+        t!(std::fs::create_dir_all(dest.parent().unwrap()));
+        let mut file = BufWriter::new(t!(File::create(&dest)));
+        t!(serde_json::to_writer(&mut file, &json));
+    }
+
+    fn prepare_json_step(&self, step: StepMetrics) -> JsonNode {
+        JsonNode::RustbuildStep {
+            type_: step.type_,
+            debug_repr: step.debug_repr,
+
+            duration_excluding_children_sec: step.duration_excluding_children_sec.as_secs_f64(),
+            system_stats: JsonStepSystemStats {
+                cpu_utilization_percent: step.cpu_usage_time_sec * 100.0
+                    / step.duration_excluding_children_sec.as_secs_f64(),
+            },
+
+            children: step
+                .children
+                .into_iter()
+                .map(|child| self.prepare_json_step(child))
+                .collect(),
+        }
+    }
+}
+
+struct MetricsState {
+    finished_steps: Vec<StepMetrics>,
+    running_steps: Vec<StepMetrics>,
+
+    system_info: System,
+    timer_start: Option<Instant>,
+    invocation_timer_start: Instant,
+}
+
+struct StepMetrics {
+    type_: String,
+    debug_repr: String,
+
+    cpu_usage_time_sec: f64,
+    duration_excluding_children_sec: Duration,
+
+    children: Vec<StepMetrics>,
+}
+
+#[derive(Serialize, Deserialize)]
+#[serde(rename_all = "snake_case")]
+struct JsonRoot {
+    system_stats: JsonInvocationSystemStats,
+    invocations: Vec<JsonInvocation>,
+}
+
+#[derive(Serialize, Deserialize)]
+#[serde(rename_all = "snake_case")]
+struct JsonInvocation {
+    duration_including_children_sec: f64,
+    children: Vec<JsonNode>,
+}
+
+#[derive(Serialize, Deserialize)]
+#[serde(tag = "kind", rename_all = "snake_case")]
+enum JsonNode {
+    RustbuildStep {
+        #[serde(rename = "type")]
+        type_: String,
+        debug_repr: String,
+
+        duration_excluding_children_sec: f64,
+        system_stats: JsonStepSystemStats,
+
+        children: Vec<JsonNode>,
+    },
+}
+
+#[derive(Serialize, Deserialize)]
+#[serde(rename_all = "snake_case")]
+struct JsonInvocationSystemStats {
+    cpu_threads_count: usize,
+    cpu_model: String,
+
+    memory_total_bytes: u64,
+}
+
+#[derive(Serialize, Deserialize)]
+#[serde(rename_all = "snake_case")]
+struct JsonStepSystemStats {
+    cpu_utilization_percent: f64,
+}

src/ci/run.sh

@@ -45,6 +45,7 @@ fi
 
 if ! isCI || isCiBranch auto || isCiBranch beta || isCiBranch try; then
     RUST_CONFIGURE_ARGS="$RUST_CONFIGURE_ARGS --set build.print-step-timings --enable-verbose-tests"
+    RUST_CONFIGURE_ARGS="$RUST_CONFIGURE_ARGS --set build.metrics"
 fi
 
 RUST_CONFIGURE_ARGS="$RUST_CONFIGURE_ARGS --enable-sccache"

src/ci/scripts/upload-artifacts.sh

@@ -10,19 +10,24 @@ source "$(cd "$(dirname "$0")" && pwd)/../shared.sh"
 
 upload_dir="$(mktemp -d)"
 
+build_dir=build
+if isLinux; then
+    build_dir=obj/build
+fi
+
 # Release tarballs produced by a dist builder.
 if [[ "${DEPLOY-0}" -eq "1" ]] || [[ "${DEPLOY_ALT-0}" -eq "1" ]]; then
-    dist_dir=build/dist
-    if isLinux; then
-        dist_dir=obj/build/dist
-    fi
+    dist_dir="${build_dir}/dist"
     rm -rf "${dist_dir}/doc"
     cp -r "${dist_dir}"/* "${upload_dir}"
 fi
 
 # CPU usage statistics.
 cp cpu-usage.csv "${upload_dir}/cpu-${CI_JOB_NAME}.csv"
 
+# Build metrics generated by x.py.
+cp "${build_dir}/metrics.json" "${upload_dir}/metrics-${CI_JOB_NAME}.json"
+
 # Toolstate data.
 if [[ -n "${DEPLOY_TOOLSTATES_JSON+x}" ]]; then
     cp /tmp/toolstate/toolstates.json "${upload_dir}/${DEPLOY_TOOLSTATES_JSON}"