Always set the deployment target when building std #133092

madsmtm · 2024-11-16T04:30:04Z

cc has a bug/feature (I guess depending on how you look at it) where the default deployment target is taken from the SDK instead of from rustc. This causes compiler-builtins to build compiler-rt with the wrong deployment target on iOS.

I've been meaning to change how cc works in this regard, but that's a lengthy process, so let's fix it in bootstrap for now.

The behaviour can be seen locally with ./x build library --set build.optimized-compiler-builtins=true for various target triples, and then inspecting with otool -l build/host/stage1/lib/rustlib/*/lib/libcompiler_builtins-*.rlib | rg 'minos|version'. I have added a rmake test that ensures that we now have the same version everywhere.

Fixes #128419
Fixes rust-lang/compiler-builtins#650
Fixes #136523
See also rust-lang/cargo#13115, rust-lang/cc-rs#1171, #136113
See #133092 (comment) for a description of how the change works.

try-job: i686-gnu-1
try-job: i686-gnu-2
try-job: x86_64-apple-1
try-job: aarch64-apple
try-job: dist-apple-various
try-job: dist-aarch64-apple
try-job: dist-various-2
try-job: x86_64-fuchsia

rustbot · 2024-11-16T04:30:12Z

r? @Mark-Simulacrum

rustbot has assigned @Mark-Simulacrum.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

madsmtm · 2024-11-16T05:35:11Z

Okay... So this didn't actually work initially, since bootstrap is for some ~~godforsaken~~ reason calling cc to get default flags, and then later passing passing them to Cargo, which invokes the build script, which calls cc again to get default flags... So all the build flags end up getting duplicated (!), and the values in CFLAGS that were looked up previously end up being prioritized.

I've rebased and fixed that in the first and second commits, but it may break more than I think, so this should probably get a try build before merging.

rustbot · 2024-11-16T07:20:00Z

This PR changes how LLVM is built. Consider updating src/bootstrap/download-ci-llvm-stamp.

This PR modifies tests/run-make/. If this PR is trying to port a Makefile
run-make test to use rmake.rs, please update the
run-make port tracking issue
so we can track our progress. You can either modify the tracking issue
directly, or you can comment on the tracking issue and link this PR.

cc @jieyouxu

madsmtm · 2024-11-16T07:22:31Z

This PR changes how LLVM is built. Consider updating src/bootstrap/download-ci-llvm-stamp.

Not sure if I should? It changes in theory (we pass different flags), but in practice it shouldn't (because those flags were already set by cmake-rs, so they were just duplicated)?

src/bootstrap/src/core/build_steps/compile.rs

Mark-Simulacrum · 2024-11-24T18:50:02Z

r=me modulo comment

Mark-Simulacrum · 2024-11-28T22:09:53Z

@bors r+

bors · 2024-11-28T22:09:55Z

📌 Commit ff3dab4 has been approved by Mark-Simulacrum

It is now in the queue for this repository.

…t, r=Mark-Simulacrum Always set the deployment target when building std `cc` has [a bug/feature](rust-lang/cc-rs#1171) (I guess depending on how you look at it) where the default deployment target is taken from the SDK instead of from `rustc`. This causes `compiler-builtins` to build `compiler-rt` with the wrong deployment target on iOS. I've been meaning to change how `cc` works in this regard, but that's a lengthy process, so let's fix it in bootstrap for now. The behaviour be seen locally with `./x build library --set build.optimized-compiler-builtins=true` for various target triples, and then inspecting with `otool -l build/host/stage1/lib/rustlib/*/lib/libcompiler_builtins-*.rlib | rg 'minos|version'`. I have added a rmake test that ensures that we now have the same version everywhere. Fixes rust-lang#128419 Fixes rust-lang/compiler-builtins#650 See also rust-lang/cargo#13115 `@rustbot` label O-apple

Rollup of 7 pull requests Successful merges: - rust-lang#131323 (Support `clobber_abi` in AVR inline assembly) - rust-lang#133092 (Always set the deployment target when building std) - rust-lang#133134 (Don't use a SyntheticProvider for literally every type) - rust-lang#133538 (Better diagnostic for fn items in variadic functions) - rust-lang#133590 (Rename `-Zparse-only`) - rust-lang#133592 (Misc: better instructions for envrc, ignore `/build` instead of `build/`) - rust-lang#133608 (Revert rust-lang#133418 (Store coverage source regions as `Span`) due to regression rust-lang#133606) r? `@ghost` `@rustbot` modify labels: rollup

Zalathar · 2024-11-29T06:19:47Z

Could this have caused the failure at #133609 (comment)?

jieyouxu · 2024-11-29T06:26:18Z

This looks most likely I agree.
@bors r-

workingjubilee · 2025-02-10T07:37:29Z

Yeaaah, I don't think that's spurious.

jieyouxu · 2025-02-10T07:44:02Z

Oh fuchsia... If you'd like help w/ that target, we can ping the fuchsia team.

madsmtm · 2025-02-10T07:51:38Z

I'll try to take a look myself first

madsmtm · 2025-02-10T09:22:27Z

Pretty sure I've fixed this now, the issue was that CI scripts were (correctly) setting CFLAGS_*, and bootstrap was then (incorrectly) overriding these preferences. It worked implicitly before by invoking cc-rs in bootstrap to get the default flags (which meant that cc-rs parsed the CFLAGS_*), now we do it manually.

@rustbot ready

jieyouxu · 2025-02-10T09:37:25Z

(Tidy moment)

jieyouxu

LGTM with tidy pleased

jieyouxu · 2025-02-10T10:07:40Z

The bootstrap diff LGTM.
@bors r=Mark-Simulacrum,jieyouxu

bors · 2025-02-10T10:07:42Z

📌 Commit ea68a47 has been approved by Mark-Simulacrum,jieyouxu

It is now in the queue for this repository.

bors · 2025-02-10T21:09:39Z

⌛ Testing commit ea68a47 with merge 6171d94...

bors · 2025-02-10T23:57:30Z

☀️ Test successful - checks-actions
Approved by: Mark-Simulacrum,jieyouxu
Pushing 6171d94 to master...

rust-timer · 2025-02-11T01:15:48Z

Finished benchmarking commit (6171d94): comparison URL.

Overall result: ❌ regressions - no action needed

@rustbot label: -perf-regression

Instruction count

This is the most reliable metric that we have; it was used to determine the overall result at the top of this comment. However, even this metric can sometimes exhibit noise.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.5%	[0.5%, 0.5%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary 3.4%, secondary 1.8%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	3.4%	[3.4%, 3.4%]	1
Regressions ❌ (secondary)	5.7%	[5.7%, 5.7%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.0%	[-2.0%, -2.0%]	1
All ❌✅ (primary)	3.4%	[3.4%, 3.4%]	1

Cycles

This benchmark run did not return any relevant results for this metric.

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 785.578s -> 786.626s (0.13%)
Artifact size: 348.28 MiB -> 348.30 MiB (0.00%)

…imulacrum,jieyouxu Always set the deployment target when building std `cc` has [a bug/feature](rust-lang/cc-rs#1171) (I guess depending on how you look at it) where the default deployment target is taken from the SDK instead of from `rustc`. This causes `compiler-builtins` to build `compiler-rt` with the wrong deployment target on iOS. I've been meaning to change how `cc` works in this regard, but that's a lengthy process, so let's fix it in bootstrap for now. The behaviour can be seen locally with `./x build library --set build.optimized-compiler-builtins=true` for various target triples, and then inspecting with `otool -l build/host/stage1/lib/rustlib/*/lib/libcompiler_builtins-*.rlib | rg 'minos|version'`. I have added a rmake test that ensures that we now have the same version everywhere. Fixes rust-lang/rust#128419 Fixes rust-lang/compiler-builtins#650 Fixes rust-lang/rust#136523 See also rust-lang/cargo#13115, rust-lang/cc-rs#1171, rust-lang/rust#136113 See rust-lang/rust#133092 (comment) for a description of how the change works. try-job: i686-gnu-1 try-job: i686-gnu-2 try-job: x86_64-apple-1 try-job: aarch64-apple try-job: dist-apple-various try-job: dist-aarch64-apple try-job: dist-various-2 try-job: x86_64-fuchsia

aeubanks · 2025-02-11T21:17:23Z

src/bootstrap/src/core/builder/cargo.rs

+                // Extend `CXXFLAGS_$TARGET` with our extra flags.
+                let env = format!("CXXFLAGS_{triple_underscored}");
+                let mut cxxflags =
+                    builder.cc_unhandled_cflags(target, GitRepo::Rustc, CLang::Cxx).join(" ");


@madsmtm this change here broke our rust build (http://crbug.com/395891130) and extending this with cc_handled_cflags() fixes the issue. was it intentional to leave out the cc handled flags here (and above)?

(also nit: typo in cc_handled_clags)

ah I see it was intentional based on the comments, but it's dropping flags we're passing via CFLAGS/CXXFLAGS/LDFLAGS/RUSTFLAGS_BOOTSTRAP/RUSTFLAGS_NOT_BOOTSTRAP (here) like --sysroot= in the stage1 build. am I holding it wrong? should I be using some other method to pass flags?

= note: ld.lld: error: undefined reference: __libc_single_threaded >>> referenced by /usr/home/aeubanks/repos/chromium/src/third_party/rust-src/build/x86_64-unknown-linux-gnu/stage0-rustc/x86_64-unknown-linux-gnu/release/deps/librustc_driver-d74657ee7dc4cb63.so (disallowed by --no-allow-shlib-undefined) ld.lld: error: undefined reference: std::__throw_bad_array_new_length() >>> referenced by /usr/home/aeubanks/repos/chromium/src/third_party/rust-src/build/x86_64-unknown-linux-gnu/stage0-rustc/x86_64-unknown-linux-gnu/release/deps/librustc_driver-d74657ee7dc4cb63.so (disallowed by --no-allow-shlib-undefined) ld.lld: error: undefined reference: std::ios_base_library_init() >>> referenced by /usr/home/aeubanks/repos/chromium/src/third_party/rust-src/build/x86_64-unknown-linux-gnu/stage0-rustc/x86_64-unknown-linux-gnu/release/deps/librustc_driver-d74657ee7dc4cb63.so (disallowed by --no-allow-shlib-undefined) clang: error: linker command failed with exit code 1 (use -v to see invocation) error: could not compile `rustc-main` (bin "rustc-main") due to 1 previous error

@madsmtm having stared at this change a bit more, it seems like there's no longer a way to pass flags to stage1 links, which seems like a big regression?

…imulacrum,jieyouxu Always set the deployment target when building std `cc` has [a bug/feature](rust-lang/cc-rs#1171) (I guess depending on how you look at it) where the default deployment target is taken from the SDK instead of from `rustc`. This causes `compiler-builtins` to build `compiler-rt` with the wrong deployment target on iOS. I've been meaning to change how `cc` works in this regard, but that's a lengthy process, so let's fix it in bootstrap for now. The behaviour can be seen locally with `./x build library --set build.optimized-compiler-builtins=true` for various target triples, and then inspecting with `otool -l build/host/stage1/lib/rustlib/*/lib/libcompiler_builtins-*.rlib | rg 'minos|version'`. I have added a rmake test that ensures that we now have the same version everywhere. Fixes rust-lang/rust#128419 Fixes rust-lang/compiler-builtins#650 Fixes rust-lang/rust#136523 See also rust-lang/cargo#13115, rust-lang/cc-rs#1171, rust-lang/rust#136113 See rust-lang/rust#133092 (comment) for a description of how the change works. try-job: i686-gnu-1 try-job: i686-gnu-2 try-job: x86_64-apple-1 try-job: aarch64-apple try-job: dist-apple-various try-job: dist-aarch64-apple try-job: dist-various-2 try-job: x86_64-fuchsia

…5-g7b473dfe-1 / d4bdd1ed551fed0c951eb47b4be2c79d7a02d181-1 : 9fcc9cf4a202aadfe1f44722b39c83536eba3dba-2 https://chromium.googlesource.com/external/github.com/llvm/llvm-project/+log/ba476d0b..7b473dfe https://chromium.googlesource.com/external/github.com/rust-lang/rust/+log/d4bdd1ed551f..9fcc9cf4a202 Ran: ./tools/clang/scripts/upload_revision.py 7b473dfe84c17319930d4019ab3f6ca0cfc03416 Revert rust-lang/rust#133092 since it breaks our Linux Rust build. Bug: 395520092, 395891130 Change-Id: I91789dec738622e88746cbdc1ec7f5ffb401cc93 Disable-Rts: True Cq-Include-Trybots: chromium/try:chromeos-amd64-generic-cfi-thin-lto-rel Cq-Include-Trybots: chromium/try:dawn-win10-x86-deps-rel Cq-Include-Trybots: chromium/try:linux-chromeos-dbg Cq-Include-Trybots: chromium/try:linux_chromium_cfi_rel_ng Cq-Include-Trybots: chromium/try:linux_chromium_chromeos_msan_rel_ng Cq-Include-Trybots: chromium/try:linux_chromium_msan_rel_ng Cq-Include-Trybots: chromium/try:mac11-arm64-rel,mac_chromium_asan_rel_ng Cq-Include-Trybots: chromium/try:ios-catalyst,win-asan,android-official Cq-Include-Trybots: chromium/try:fuchsia-arm64-cast-receiver-rel Cq-Include-Trybots: chromium/try:mac-official,linux-official Cq-Include-Trybots: chromium/try:win-official,win32-official Cq-Include-Trybots: chromium/try:win-arm64-rel Cq-Include-Trybots: chromium/try:linux-swangle-try-x64,win-swangle-try-x86 Cq-Include-Trybots: chromium/try:android-cronet-mainline-clang-arm64-dbg Cq-Include-Trybots: chromium/try:android-cronet-mainline-clang-arm64-rel Cq-Include-Trybots: chromium/try:android-cronet-mainline-clang-riscv64-dbg Cq-Include-Trybots: chromium/try:android-cronet-mainline-clang-riscv64-rel Cq-Include-Trybots: chromium/try:android-cronet-mainline-clang-x86-dbg Cq-Include-Trybots: chromium/try:android-cronet-mainline-clang-x86-rel Cq-Include-Trybots: chromium/try:android-cronet-riscv64-dbg Cq-Include-Trybots: chromium/try:android-cronet-riscv64-rel Cq-Include-Trybots: chrome/try:iphone-device,ipad-device Cq-Include-Trybots: chrome/try:linux-chromeos-chrome Cq-Include-Trybots: chrome/try:win-chrome,win64-chrome,linux-chrome,mac-chrome Cq-Include-Trybots: chrome/try:linux-pgo,mac-pgo,win32-pgo,win64-pgo Cq-Include-Trybots: luci.chromium.try:linux-cast-x64-rel Cq-Include-Trybots: chromium/try:android-rust-arm32-rel Cq-Include-Trybots: chromium/try:android-rust-arm64-dbg Cq-Include-Trybots: chromium/try:android-rust-arm64-rel Cq-Include-Trybots: chromium/try:linux-rust-x64-dbg Cq-Include-Trybots: chromium/try:linux-rust-x64-rel Cq-Include-Trybots: chromium/try:mac-rust-x64-dbg Cq-Include-Trybots: chromium/try:win-rust-x64-dbg Cq-Include-Trybots: chromium/try:win-rust-x64-rel Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6261055 Reviewed-by: Nico Weber <thakis@chromium.org> Cr-Commit-Position: refs/heads/main@{#1420430}

rustbot assigned Mark-Simulacrum Nov 16, 2024

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-bootstrap Relevant to the bootstrap subteam: Rust's build system (x.py and src/bootstrap) O-apple Operating system: Apple (macOS, iOS, tvOS, visionOS, watchOS) labels Nov 16, 2024

This was referenced Nov 16, 2024

Print name of env var in --print=deployment-target #133041

Merged

"was built for iOS 16.4" warning after updating from 1.79 to 1.80 #128419

Closed

madsmtm force-pushed the bootstrap-deployment-target branch from fd382a0 to dd8ca01 Compare November 16, 2024 05:34

madsmtm force-pushed the bootstrap-deployment-target branch 3 times, most recently from 50812ff to 2294b05 Compare November 16, 2024 06:55

rustbot added the A-run-make Area: port run-make Makefiles to rmake.rs label Nov 16, 2024

madsmtm force-pushed the bootstrap-deployment-target branch from 2294b05 to f816d33 Compare November 16, 2024 07:05

madsmtm marked this pull request as ready for review November 16, 2024 07:19

Mark-Simulacrum reviewed Nov 24, 2024

View reviewed changes

src/bootstrap/src/core/build_steps/compile.rs Outdated Show resolved Hide resolved

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Nov 28, 2024

Zalathar mentioned this pull request Nov 29, 2024

Rollup of 7 pull requests #133609

Closed

bors added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Nov 29, 2024

madsmtm mentioned this pull request Feb 10, 2025

Simplify the error output on failed Command invocation rust-lang/cc-rs#1397

Merged

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Feb 10, 2025

This comment has been minimized.

Sign in to view

jieyouxu approved these changes Feb 10, 2025

View reviewed changes

bootstrap: Don't overwrite CFLAGS_* set in the environment

ea68a47

madsmtm force-pushed the bootstrap-deployment-target branch from 148ebac to ea68a47 Compare February 10, 2025 10:03

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Feb 10, 2025

bors added the merged-by-bors This PR was explicitly merged by bors. label Feb 10, 2025

bors merged commit 6171d94 into rust-lang:master Feb 10, 2025
7 checks passed

rustbot added this to the 1.86.0 milestone Feb 10, 2025

madsmtm deleted the bootstrap-deployment-target branch February 11, 2025 07:00

madsmtm mentioned this pull request Feb 11, 2025

"was built for iOS 16.4" warning rust-lang/compiler-builtins#650

Closed

aeubanks reviewed Feb 11, 2025

View reviewed changes

aeubanks mentioned this pull request Feb 13, 2025

Cannot pass flags to bootstrap links after #133092 #136984

Open

madsmtm mentioned this pull request Feb 14, 2025

Always read from all CFLAGS-style flags rust-lang/cc-rs#1401

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Always set the deployment target when building std #133092

Always set the deployment target when building std #133092

madsmtm commented Nov 16, 2024 •

edited by workingjubilee

Loading

rustbot commented Nov 16, 2024

madsmtm commented Nov 16, 2024 •

edited

Loading

rustbot commented Nov 16, 2024

madsmtm commented Nov 16, 2024

Mark-Simulacrum commented Nov 24, 2024

Mark-Simulacrum commented Nov 28, 2024

bors commented Nov 28, 2024

Zalathar commented Nov 29, 2024

jieyouxu commented Nov 29, 2024

workingjubilee commented Feb 10, 2025

jieyouxu commented Feb 10, 2025 •

edited

Loading

madsmtm commented Feb 10, 2025

madsmtm commented Feb 10, 2025 •

edited

Loading

This comment has been minimized.

jieyouxu commented Feb 10, 2025

jieyouxu left a comment •

edited

Loading

jieyouxu commented Feb 10, 2025

bors commented Feb 10, 2025

bors commented Feb 10, 2025

bors commented Feb 10, 2025

rust-timer commented Feb 11, 2025

aeubanks Feb 11, 2025

aeubanks Feb 11, 2025 •

edited

Loading

aeubanks Feb 12, 2025

Always set the deployment target when building std #133092

Always set the deployment target when building std #133092

Conversation

madsmtm commented Nov 16, 2024 • edited by workingjubilee Loading

rustbot commented Nov 16, 2024

madsmtm commented Nov 16, 2024 • edited Loading

rustbot commented Nov 16, 2024

madsmtm commented Nov 16, 2024

Mark-Simulacrum commented Nov 24, 2024

Mark-Simulacrum commented Nov 28, 2024

bors commented Nov 28, 2024

Zalathar commented Nov 29, 2024

jieyouxu commented Nov 29, 2024

workingjubilee commented Feb 10, 2025

jieyouxu commented Feb 10, 2025 • edited Loading

madsmtm commented Feb 10, 2025

madsmtm commented Feb 10, 2025 • edited Loading

This comment has been minimized.

jieyouxu commented Feb 10, 2025

jieyouxu left a comment • edited Loading

Choose a reason for hiding this comment

jieyouxu commented Feb 10, 2025

bors commented Feb 10, 2025

bors commented Feb 10, 2025

bors commented Feb 10, 2025

rust-timer commented Feb 11, 2025

Overall result: ❌ regressions - no action needed

Instruction count

Max RSS (memory usage)

Cycles

Binary size

aeubanks Feb 11, 2025

Choose a reason for hiding this comment

aeubanks Feb 11, 2025 • edited Loading

Choose a reason for hiding this comment

aeubanks Feb 12, 2025

Choose a reason for hiding this comment

madsmtm commented Nov 16, 2024 •

edited by workingjubilee

Loading

madsmtm commented Nov 16, 2024 •

edited

Loading

jieyouxu commented Feb 10, 2025 •

edited

Loading

madsmtm commented Feb 10, 2025 •

edited

Loading

jieyouxu left a comment •

edited

Loading

aeubanks Feb 11, 2025 •

edited

Loading