Skip to content

Commit a448f88

Browse files
Mark-SimulacrumMark-Simulacrum
committed
Utilize PGO for rustc linux dist builds
This implements support for applying PGO to the rustc compilation step (not standard library or any tooling, including rustdoc). Expanding PGO to more tools is not terribly difficult but will involve more work and greater CI time commitment. For the same reason of avoiding greater time commitment, this currently avoids implementing for platforms outside of x86_64-unknown-linux-gnu, though in practice it should be quite simple to extend over time to more platforms. The initial implementation is intentionally minimal here to avoid too much work investment before we start seeing wins for a subset of Rust users. The choice of workloads to profile here is somewhat arbitrary, but the general rationale was to aim for a small set that largely avoided time regressions on perf.rust-lang.org's full suite of crates. The set chosen is libcore, cargo (and its dependencies), and a few ad-hoc stress tests from perf.rlo. The stress tests are arguably the most controversial, but they benefit those cases (avoiding regressions) and do not really remove wins from other benchmarks. The primary next step after this PR lands is to implement support for PGO in LLVM. It is unclear whether we can afford a full LLVM rebuild in CI, though, so the approach taken there may need to be more staggered. rustc-only PGO seems well affordable on linux at least, giving us up to 20% wall time wins on some crates for 15 minutes of extra CI time (1 hour up from 45 minutes). The PGO data is uploaded to allow others to reuse it if attempting to reproduce the CI build or potentially, in the future, on other platforms where an off-by-one strategy is used for dist builds at minimal performance cost.
1 parent b32e6e6 commit a448f88

File tree

8 files changed

+176
-4
lines changed

8 files changed

+176
-4
lines changed

‎src/bootstrap/builder.rs

+1
Original file line numberDiff line numberDiff line change
@@ -471,6 +471,7 @@ impl<'a> Builder<'a> {
471471
dist::RustDev,
472472
dist::Extended,
473473
dist::BuildManifest,
474+
dist::ReproducibleArtifacts,
474475
),
475476
Kind::Install => describe!(
476477
install::Docs,

‎src/bootstrap/compile.rs

+36-1
Original file line numberDiff line numberDiff line change
@@ -501,6 +501,41 @@ impl Step for Rustc {
501501
let mut cargo = builder.cargo(compiler, Mode::Rustc, SourceType::InTree, target, "build");
502502
rustc_cargo(builder, &mut cargo, target);
503503

504+
if builder.config.rust_profile_use.is_some()
505+
&& builder.config.rust_profile_generate.is_some()
506+
{
507+
panic!("Cannot use and generate PGO profiles at the same time");
508+
}
509+
510+
let is_collecting = if let Some(path) = &builder.config.rust_profile_generate {
511+
if compiler.stage == 1 {
512+
cargo.rustflag(&format!("-Cprofile-generate={}", path));
513+
// Apparently necessary to avoid overflowing the counters during
514+
// a Cargo build profile
515+
cargo.rustflag("-Cllvm-args=-vp-counters-per-site=4");
516+
true
517+
} else {
518+
false
519+
}
520+
} else if let Some(path) = &builder.config.rust_profile_use {
521+
if compiler.stage == 1 {
522+
cargo.rustflag(&format!("-Cprofile-use={}", path));
523+
cargo.rustflag("-Cllvm-args=-pgo-warn-missing-function");
524+
true
525+
} else {
526+
false
527+
}
528+
} else {
529+
false
530+
};
531+
if is_collecting {
532+
// Ensure paths to Rust sources are relative, not absolute.
533+
cargo.rustflag(&format!(
534+
"-Cllvm-args=-static-func-strip-dirname-prefix={}",
535+
builder.config.src.components().count()
536+
));
537+
}
538+
504539
builder.info(&format!(
505540
"Building stage{} compiler artifacts ({} -> {})",
506541
compiler.stage, &compiler.host, target
@@ -752,7 +787,7 @@ fn copy_codegen_backends_to_sysroot(
752787
// Here we're looking for the output dylib of the `CodegenBackend` step and
753788
// we're copying that into the `codegen-backends` folder.
754789
let dst = builder.sysroot_codegen_backends(target_compiler);
755-
t!(fs::create_dir_all(&dst));
790+
t!(fs::create_dir_all(&dst), dst);
756791

757792
if builder.config.dry_run {
758793
return;

‎src/bootstrap/config.rs

+9
Original file line numberDiff line numberDiff line change
@@ -133,6 +133,8 @@ pub struct Config {
133133
pub rust_thin_lto_import_instr_limit: Option<u32>,
134134
pub rust_remap_debuginfo: bool,
135135
pub rust_new_symbol_mangling: bool,
136+
pub rust_profile_use: Option<String>,
137+
pub rust_profile_generate: Option<String>,
136138

137139
pub build: TargetSelection,
138140
pub hosts: Vec<TargetSelection>,
@@ -494,6 +496,8 @@ struct Rust {
494496
llvm_libunwind: Option<String>,
495497
control_flow_guard: Option<bool>,
496498
new_symbol_mangling: Option<bool>,
499+
profile_generate: Option<String>,
500+
profile_use: Option<String>,
497501
}
498502

499503
/// TOML representation of how each build target is configured.
@@ -871,6 +875,11 @@ impl Config {
871875

872876
config.rust_codegen_units = rust.codegen_units.map(threads_from_config);
873877
config.rust_codegen_units_std = rust.codegen_units_std.map(threads_from_config);
878+
config.rust_profile_use = flags.rust_profile_use.or(rust.profile_use);
879+
config.rust_profile_generate = flags.rust_profile_generate.or(rust.profile_generate);
880+
} else {
881+
config.rust_profile_use = flags.rust_profile_use;
882+
config.rust_profile_generate = flags.rust_profile_generate;
874883
}
875884

876885
if let Some(t) = toml.target {

‎src/bootstrap/dist.rs

+69
Original file line numberDiff line numberDiff line change
@@ -2664,3 +2664,72 @@ impl Step for BuildManifest {
26642664
distdir(builder).join(format!("{}-{}.tar.gz", name, self.target.triple))
26652665
}
26662666
}
2667+
2668+
/// Tarball containing artifacts necessary to reproduce the build of rustc.
2669+
///
2670+
/// Currently this is the PGO profile data.
2671+
///
2672+
/// Should not be considered stable by end users.
2673+
#[derive(Clone, Debug, Eq, Hash, PartialEq)]
2674+
pub struct ReproducibleArtifacts {
2675+
pub target: TargetSelection,
2676+
}
2677+
2678+
impl Step for ReproducibleArtifacts {
2679+
type Output = Option<PathBuf>;
2680+
const DEFAULT: bool = true;
2681+
const ONLY_HOSTS: bool = true;
2682+
2683+
fn should_run(run: ShouldRun<'_>) -> ShouldRun<'_> {
2684+
run.path("reproducible")
2685+
}
2686+
2687+
fn make_run(run: RunConfig<'_>) {
2688+
run.builder.ensure(ReproducibleArtifacts { target: run.target });
2689+
}
2690+
2691+
fn run(self, builder: &Builder<'_>) -> Self::Output {
2692+
let name = pkgname(builder, "reproducible-artifacts");
2693+
let tmp = tmpdir(builder);
2694+
2695+
// Prepare the image.
2696+
let image = tmp.join("reproducible-artifacts-image");
2697+
let _ = fs::remove_dir_all(&image);
2698+
2699+
if let Some(path) = &builder.config.rust_profile_use {
2700+
builder.install(std::path::Path::new(path), &image, 0o644);
2701+
} else {
2702+
return None;
2703+
}
2704+
2705+
// Prepare the overlay.
2706+
let overlay = tmp.join("reproducible-artifacts-overlay");
2707+
let _ = fs::remove_dir_all(&overlay);
2708+
builder.create_dir(&overlay);
2709+
builder.create(&overlay.join("version"), &builder.rust_version());
2710+
for file in &["COPYRIGHT", "LICENSE-APACHE", "LICENSE-MIT", "README.md"] {
2711+
builder.install(&builder.src.join(file), &overlay, 0o644);
2712+
}
2713+
2714+
// Create the final tarball.
2715+
let mut cmd = rust_installer(builder);
2716+
cmd.arg("generate")
2717+
.arg("--product-name=Rust")
2718+
.arg("--rel-manifest-dir=rustlib")
2719+
.arg("--success-message=reproducible-artifacts installed.")
2720+
.arg("--image-dir")
2721+
.arg(&image)
2722+
.arg("--work-dir")
2723+
.arg(&tmpdir(builder))
2724+
.arg("--output-dir")
2725+
.arg(&distdir(builder))
2726+
.arg("--non-installed-overlay")
2727+
.arg(&overlay)
2728+
.arg(format!("--package-name={}-{}", name, self.target.triple))
2729+
.arg("--legacy-manifest-dirs=rustlib,cargo")
2730+
.arg("--component-name=reproducible-artifacts");
2731+
2732+
builder.run(&mut cmd);
2733+
Some(distdir(builder).join(format!("{}-{}.tar.gz", name, self.target.triple)))
2734+
}
2735+
}

‎src/bootstrap/flags.rs

+7
Original file line numberDiff line numberDiff line change
@@ -68,6 +68,9 @@ pub struct Flags {
6868
pub deny_warnings: Option<bool>,
6969

7070
pub llvm_skip_rebuild: Option<bool>,
71+
72+
pub rust_profile_use: Option<String>,
73+
pub rust_profile_generate: Option<String>,
7174
}
7275

7376
pub enum Subcommand {
@@ -219,6 +222,8 @@ To learn more about a subcommand, run `./x.py <subcommand> -h`",
219222
VALUE overrides the skip-rebuild option in config.toml.",
220223
"VALUE",
221224
);
225+
opts.optopt("", "rust-profile-generate", "rustc error format", "FORMAT");
226+
opts.optopt("", "rust-profile-use", "rustc error format", "FORMAT");
222227

223228
// We can't use getopt to parse the options until we have completed specifying which
224229
// options are valid, but under the current implementation, some options are conditional on
@@ -674,6 +679,8 @@ Arguments:
674679
color: matches
675680
.opt_get_default("color", Color::Auto)
676681
.expect("`color` should be `always`, `never`, or `auto`"),
682+
rust_profile_use: matches.opt_str("rust-profile-use"),
683+
rust_profile_generate: matches.opt_str("rust-profile-generate"),
677684
}
678685
}
679686
}

‎src/ci/docker/host-x86_64/dist-x86_64-linux/Dockerfile

+6-3
Original file line numberDiff line numberDiff line change
@@ -85,6 +85,8 @@ ENV CC=clang CXX=clang++
8585
COPY scripts/sccache.sh /scripts/
8686
RUN sh /scripts/sccache.sh
8787

88+
ENV PGO_HOST=x86_64-unknown-linux-gnu
89+
8890
ENV HOSTS=x86_64-unknown-linux-gnu
8991

9092
ENV RUST_CONFIGURE_ARGS \
@@ -98,9 +100,10 @@ ENV RUST_CONFIGURE_ARGS \
98100
--set llvm.thin-lto=true \
99101
--set llvm.ninja=false \
100102
--set rust.jemalloc
101-
ENV SCRIPT python2.7 ../x.py dist --host $HOSTS --target $HOSTS \
102-
--include-default-paths \
103-
src/tools/build-manifest
103+
ENV SCRIPT ../src/ci/pgo.sh python2.7 ../x.py dist \
104+
--host $HOSTS --target $HOSTS \
105+
--include-default-paths \
106+
src/tools/build-manifest
104107
ENV CARGO_TARGET_X86_64_UNKNOWN_LINUX_GNU_LINKER=clang
105108

106109
# This is the only builder which will create source tarballs

‎src/ci/pgo.sh

+47
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,47 @@
1+
#!/bin/bash
2+
3+
set -euxo pipefail
4+
5+
rm -rf /tmp/rustc-pgo
6+
7+
python2.7 ../x.py build --target=$PGO_HOST --host=$PGO_HOST \
8+
--stage 2 library/std --rust-profile-generate=/tmp/rustc-pgo
9+
10+
./build/$PGO_HOST/stage2/bin/rustc --edition=2018 \
11+
--crate-type=lib ../library/core/src/lib.rs
12+
13+
# Download and build a single-file stress test benchmark on perf.rust-lang.org.
14+
function pgo_perf_benchmark {
15+
local PERF=e095f5021bf01cf3800f50b3a9f14a9683eb3e4e
16+
local github_prefix=https://raw.githubusercontent.com/rust-lang/rustc-perf/$PERF
17+
local name=$1
18+
curl -o /tmp/$name.rs $github_prefix/collector/benchmarks/$name/src/lib.rs
19+
./build/$PGO_HOST/stage2/bin/rustc --edition=2018 --crate-type=lib /tmp/$name.rs
20+
}
21+
22+
pgo_perf_benchmark externs
23+
pgo_perf_benchmark ctfe-stress-4
24+
25+
cp -pri ../src/tools/cargo /tmp/cargo
26+
27+
# Build cargo (with some flags)
28+
function pgo_cargo {
29+
RUSTC=./build/$PGO_HOST/stage2/bin/rustc \
30+
./build/$PGO_HOST/stage0/bin/cargo $@ \
31+
--manifest-path /tmp/cargo/Cargo.toml
32+
}
33+
34+
# Build a couple different variants of Cargo
35+
CARGO_INCREMENTAL=1 pgo_cargo check
36+
echo 'pub fn barbarbar() {}' >> /tmp/cargo/src/cargo/lib.rs
37+
CARGO_INCREMENTAL=1 pgo_cargo check
38+
touch /tmp/cargo/src/cargo/lib.rs
39+
CARGO_INCREMENTAL=1 pgo_cargo check
40+
pgo_cargo build --release
41+
42+
# Merge the profile data we gathered
43+
./build/$PGO_HOST/llvm/bin/llvm-profdata \
44+
merge -o /tmp/rustc-pgo.profdata /tmp/rustc-pgo
45+
46+
# This produces the actual final set of artifacts.
47+
$@ --rust-profile-use=/tmp/rustc-pgo.profdata

‎src/tools/build-manifest/src/main.rs

+1
Original file line numberDiff line numberDiff line change
@@ -299,6 +299,7 @@ impl Builder {
299299
let mut package = |name, targets| self.package(name, &mut manifest.pkg, targets);
300300
package("rustc", HOSTS);
301301
package("rustc-dev", HOSTS);
302+
package("reproducible-artifacts", HOSTS);
302303
package("rustc-docs", HOSTS);
303304
package("cargo", HOSTS);
304305
package("rust-mingw", MINGW);

0 commit comments

Comments
 (0)