cargo: Bump rustc to 1.89 by webbeef · Pull Request #36818 · servo/servo

webbeef · 2025-05-03T03:29:00Z

Update Rustc to 1.89.

Reviewable by commit.

Leftover work:

sagudev · 2025-05-03T11:54:40Z

macos unit test fails:

test fetch::test_fetch_redirect_updates_method has been running for over 60 seconds
  Warning: Attempt 1 failed. Reason: Timeout of 2400000ms hit

but there is no change to fetch(-related) code

atouchet · 2025-05-03T19:31:13Z

Should rust-version in Cargo.toml be updated as well?

webbeef · 2025-05-03T19:44:11Z

Should rust-version in Cargo.toml be updated as well?

Good catch! I changed it in the latest version.

jdm · 2025-05-04T20:55:35Z

Huh, I don't reproduce the hang with a debug/release build locally, but I do get this:

failures:

---- hsts::test_preload_hsts_domains_well_formed stdout ----

thread 'hsts::test_preload_hsts_domains_well_formed' panicked at components/net/tests/hsts.rs:395:5:
assertion failed: !hsts_list.entries_map.is_empty()


failures:
    hsts::test_preload_hsts_domains_well_formed

test result: FAILED. 432 passed; 1 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.51s

Filed #36837 since this also reproduces on main.

sagudev · 2025-05-05T14:20:56Z

my plan to fix this:

minimize command invocation for repro (fetch, httploader)
bisect rustc

sagudev · 2025-05-05T16:48:32Z

It works normally in my fork: https://github.com/sagudev/servo/actions/runs/14840532532/job/41662172500

I think the problem is in our self hosted runners. @delan we might need CLOBBER

jdm · 2025-05-06T03:32:38Z

This PR consistently breaks about:memory for me with:

null pointer dereference occurred (thread Script(2,3), at components/script_bindings/conversions.rs:248)
   0: backtrace::backtrace::libunwind::trace
             at /Users/jdm/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/backtrace-0.3.74/src/backtrace/libunwind.rs:116:5
      backtrace::backtrace::trace_unsynchronized
             at /Users/jdm/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/backtrace-0.3.74/src/backtrace/mod.rs:66:5
   1: <servoshell::backtrace::Print as core::fmt::Debug>::fmt
             at /Users/jdm/src/servo/ports/servoshell/backtrace.rs:65:13
   2: core::fmt::rt::Argument::fmt
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/core/src/fmt/rt.rs:177:76
      core::fmt::write
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/core/src/fmt/mod.rs:1449:21
   3: std::io::Write::write_fmt
             at /Users/jdm/.rustup/toolchains/1.86.0-aarch64-apple-darwin/lib/rustlib/src/rust/library/std/src/io/mod.rs:1890:15
   4: servoshell::backtrace::print
             at /Users/jdm/src/servo/ports/servoshell/backtrace.rs:18:5
   5: servoshell::panic_hook::panic_hook
             at /Users/jdm/src/servo/ports/servoshell/panic_hook.rs:40:17
   6: core::ops::function::Fn::call
             at /Users/jdm/.rustup/toolchains/1.86.0-aarch64-apple-darwin/lib/rustlib/src/rust/library/core/src/ops/function.rs:79:5
   7: <alloc::boxed::Box<F,A> as core::ops::function::Fn<Args>>::call
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/alloc/src/boxed.rs:1990:9
      std::panicking::rust_panic_with_hook
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/std/src/panicking.rs:839:13
   8: std::panicking::begin_panic_handler::{{closure}}
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/std/src/panicking.rs:697:13
   9: std::sys::backtrace::__rust_end_short_backtrace
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/std/src/sys/backtrace.rs:168:18
  10: rust_begin_unwind
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/std/src/panicking.rs:695:5
  11: core::panicking::panic_nounwind_fmt::runtime
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/core/src/panicking.rs:117:22
      core::panicking::panic_nounwind_fmt
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/core/src/intrinsics/mod.rs:3886:9
  12: core::panicking::panic_null_pointer_dereference
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/core/src/panicking.rs:304:5
  13: script_bindings::conversions::get_dom_class
             at /Users/jdm/src/servo/components/script_bindings/conversions.rs:248:19
  14: script::script_runtime::get_size
             at /Users/jdm/src/servo/components/script/script_runtime.rs:795:11
  15: _ZN15ServoDOMVisitor19sizeOfIncludingThisEP11nsISupports
             at /Users/runner/work/mozjs/mozjs/mozjs-sys/src/jsglue.cpp:472:16
  16: __ZL17StatsCellCallbackIL11Granularity1EEvP9JSRuntimePvN2JS9GCCellPtrEmRKNS4_15AutoRequireNoGCE
  17: __ZL35IterateRealmsArenasCellsUnbarrieredP9JSContextPN2JS4ZoneEPvPFvS0_S4_PNS1_5RealmERKNS1_15AutoRequireNoGCEEPFvP9JSRuntimeS4_PN2js2gc5ArenaENS1_9TraceKindEmS9_EPFvSD_S4_NS1_9GCCellPtrEmS9_ES9_
  18: __ZN2js22IterateHeapUnbarrieredEP9JSContextPvPFvP9JSRuntimeS2_PN2JS4ZoneERKNS5_15AutoRequireNoGCEEPFvS1_S2_PNS5_5RealmESA_EPFvS4_S2_PNS_2gc5ArenaENS5_9TraceKindEmSA_EPFvS4_S2_NS5_9GCCellPtrEmSA_E
  19: __ZL25CollectRuntimeStatsHelperP9JSContextPN2JS12RuntimeStatsEPNS1_20ObjectPrivateVisitorEbPFvP9JSRuntimePvNS1_9GCCellPtrEmRKNS1_15AutoRequireNoGCEE
  20: __ZN2JS14AddServoSizeOfEP9JSContextPFmPKvEPNS_20ObjectPrivateVisitorEPNS_10ServoSizesE
  21: CollectServoSizes
             at /Users/runner/work/mozjs/mozjs/mozjs-sys/src/jsglue.cpp:830:10
  22: <script_bindings::script_runtime::JSContext as script::script_runtime::JSContextHelper>::get_reports
             at /Users/jdm/src/servo/components/script/script_runtime.rs:917:17
  23: script::script_thread::ScriptThread::collect_reports::{{closure}}
             at /Users/jdm/src/servo/components/script/script_thread.rs:2445:28
  24: profile_traits::mem::perform_memory_report
             at /Users/jdm/src/servo/components/shared/profile/mem.rs:289:5
  25: script::script_thread::ScriptThread::collect_reports
             at /Users/jdm/src/servo/components/script/script_thread.rs:2436:9
  26: script::script_thread::ScriptThread::handle_msg_from_script
             at /Users/jdm/src/servo/components/script/script_thread.rs:1974:17
  27: script::script_thread::ScriptThread::handle_msgs::{{closure}}
             at /Users/jdm/src/servo/components/script/script_thread.rs:1502:60
  28: script::script_thread::ScriptThread::profile_event
             at /Users/jdm/src/servo/components/script/script_thread.rs:1714:13
  29: script::script_thread::ScriptThread::handle_msgs
             at /Users/jdm/src/servo/components/script/script_thread.rs:1493:27
  30: script::script_thread::ScriptThread::start
             at /Users/jdm/src/servo/components/script/script_thread.rs:988:15
  31: <script::script_thread::ScriptThread as script_layout_interface::ScriptThreadFactory>::create::{{closure}}::{{closure}}
             at /Users/jdm/src/servo/components/script/script_thread.rs:436:25
  32: profile_traits::mem::ProfilerChan::run_with_memory_reporting
             at /Users/jdm/src/servo/components/shared/profile/mem.rs:131:9
  33: <script::script_thread::ScriptThread as script_layout_interface::ScriptThreadFactory>::create::{{closure}}
             at /Users/jdm/src/servo/components/script/script_thread.rs:434:17
  34: std::sys::backtrace::__rust_begin_short_backtrace
             at /Users/jdm/.rustup/toolchains/1.86.0-aarch64-apple-darwin/lib/rustlib/src/rust/library/std/src/sys/backtrace.rs:152:18
  35: std::thread::Builder::spawn_unchecked_::{{closure}}::{{closure}}
             at /Users/jdm/.rustup/toolchains/1.86.0-aarch64-apple-darwin/lib/rustlib/src/rust/library/std/src/thread/mod.rs:559:17
  36: <core::panic::unwind_safe::AssertUnwindSafe<F> as core::ops::function::FnOnce<()>>::call_once
             at /Users/jdm/.rustup/toolchains/1.86.0-aarch64-apple-darwin/lib/rustlib/src/rust/library/core/src/panic/unwind_safe.rs:272:9
  37: std::panicking::try::do_call
             at /Users/jdm/.rustup/toolchains/1.86.0-aarch64-apple-darwin/lib/rustlib/src/rust/library/std/src/panicking.rs:587:40
  38: ___rust_try
  39: std::panicking::try
             at /Users/jdm/.rustup/toolchains/1.86.0-aarch64-apple-darwin/lib/rustlib/src/rust/library/std/src/panicking.rs:550:19
      std::panic::catch_unwind
             at /Users/jdm/.rustup/toolchains/1.86.0-aarch64-apple-darwin/lib/rustlib/src/rust/library/std/src/panic.rs:358:14
      std::thread::Builder::spawn_unchecked_::{{closure}}
             at /Users/jdm/.rustup/toolchains/1.86.0-aarch64-apple-darwin/lib/rustlib/src/rust/library/std/src/thread/mod.rs:557:30
  40: core::ops::function::FnOnce::call_once{{vtable.shim}}
             at /Users/jdm/.rustup/toolchains/1.86.0-aarch64-apple-darwin/lib/rustlib/src/rust/library/core/src/ops/function.rs:250:5
  41: <alloc::boxed::Box<F,A> as core::ops::function::FnOnce<Args>>::call_once
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/alloc/src/boxed.rs:1976:9
      <alloc::boxed::Box<F,A> as core::ops::function::FnOnce<Args>>::call_once
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/alloc/src/boxed.rs:1976:9
      std::sys::pal::unix::thread::Thread::new::thread_start
             at /rustc/05f9846f893b09a1be1fc8560e33fc3c815cfecb/library/std/src/sys/pal/unix/thread.rs:106:17
  42: __pthread_deallocate
[2025-05-06T03:31:50Z ERROR servoshell::panic_hook] null pointer dereference occurred
thread caused non-unwinding panic. aborting.
Redirecting call to abort() to mozalloc_abort

Caught signal 11 in thread "Script(2,3)"

jdm · 2025-05-06T04:44:51Z

Ah, we're hitting https://blog.rust-lang.org/2025/04/03/Rust-1.86.0/#debug-assertions-that-pointers-are-non-null-when-required-for-soundness rather than a miscompilation. I've opened #36869 that avoids this.

…ndler custom data (#36869) While the vast majority of DOM proxy objects created have a non-null pointer in the handler's extra data field, there is one place we create a proxy object that has a null pointer: https://github.com/servo/servo/blob/8b05b7449d3ceaeb6ae834246012ac04a4f048c6/components/script/window_named_properties.rs#L76 . Before #36818, dereferencing this null pointer was undefined behaviour that was silently being ignored; now that Rust 1.86 adds debug pointer validity checks, we get a panic when trying to dereference it. Testing: Tested about:memory with rustc 1.86. --------- Signed-off-by: Josh Matthews <[email protected]>

…ndler custom data (servo#36869) While the vast majority of DOM proxy objects created have a non-null pointer in the handler's extra data field, there is one place we create a proxy object that has a null pointer: https://github.com/servo/servo/blob/8b05b7449d3ceaeb6ae834246012ac04a4f048c6/components/script/window_named_properties.rs#L76 . Before servo#36818, dereferencing this null pointer was undefined behaviour that was silently being ignored; now that Rust 1.86 adds debug pointer validity checks, we get a panic when trying to dereference it. Testing: Tested about:memory with rustc 1.86. --------- Signed-off-by: Josh Matthews <[email protected]>

webbeef · 2025-05-16T16:48:40Z

And now rustc 1.87 is available :)

github-actions · 2025-08-19T10:09:17Z

🔨 Triggering try run (#17066407600) for MacOS

github-actions · 2025-08-19T10:12:17Z

🔨 Triggering try run (#17066477502) for Linux (Build libservo)

github-actions · 2025-08-19T10:20:26Z

⚠️ Try run (#17066477502) failed.

Signed-off-by: sagudev <[email protected]>

github-actions · 2025-08-19T10:35:14Z

🔨 Triggering try run (#17067002229) for Linux (Build libservo)

github-actions · 2025-08-19T10:44:57Z

✨ Try run (#17067002229) succeeded.

jschwe · 2025-08-19T11:40:56Z

Thanks @webbeef and @sagudev! Great to have this landed :)

…38780) This is a follow up to #36818 (comment). Since openxr is a feature only available on Windows, it was omitted. Testing: Fixes ``` warning: hiding a lifetime that's elided elsewhere is confusing --> components\webxr\openxr\input.rs:366:9 | 366 | &self, | ^^^^^ the lifetime is elided here ... 371 | ) -> Vec<Binding> { | ------- the same lifetime is hidden here | = help: the same lifetime is referred to in inconsistent ways, making the signature confusing = note: `#[warn(mismatched_lifetime_syntaxes)]` on by default help: use `'_` for type paths | 371 | ) -> Vec<Binding<'_>> { ``` Signed-off-by: Euclid Ye <[email protected]>

Update Rustc to 1.89. Reviewable by commit. Leftover work: - servo#37330 - servo#38777 --------- Signed-off-by: sagudev <[email protected]> Co-authored-by: sagudev <[email protected]>

…ervo#38780) This is a follow up to servo#36818 (comment). Since openxr is a feature only available on Windows, it was omitted. Testing: Fixes ``` warning: hiding a lifetime that's elided elsewhere is confusing --> components\webxr\openxr\input.rs:366:9 | 366 | &self, | ^^^^^ the lifetime is elided here ... 371 | ) -> Vec<Binding> { | ------- the same lifetime is hidden here | = help: the same lifetime is referred to in inconsistent ways, making the signature confusing = note: `#[warn(mismatched_lifetime_syntaxes)]` on by default help: use `'_` for type paths | 371 | ) -> Vec<Binding<'_>> { ``` Signed-off-by: Euclid Ye <[email protected]>

Update Rustc to 1.89. Reviewable by commit. Leftover work: - servo#37330 - servo#38777 --------- Signed-off-by: sagudev <[email protected]> Co-authored-by: sagudev <[email protected]>

…ervo#38780) This is a follow up to servo#36818 (comment). Since openxr is a feature only available on Windows, it was omitted. Testing: Fixes ``` warning: hiding a lifetime that's elided elsewhere is confusing --> components\webxr\openxr\input.rs:366:9 | 366 | &self, | ^^^^^ the lifetime is elided here ... 371 | ) -> Vec<Binding> { | ------- the same lifetime is hidden here | = help: the same lifetime is referred to in inconsistent ways, making the signature confusing = note: `#[warn(mismatched_lifetime_syntaxes)]` on by default help: use `'_` for type paths | 371 | ) -> Vec<Binding<'_>> { ``` Signed-off-by: Euclid Ye <[email protected]>

Update Rustc to 1.89. Reviewable by commit. Leftover work: - servo#37330 - servo#38777 --------- Signed-off-by: sagudev <[email protected]> Co-authored-by: sagudev <[email protected]>

…ervo#38780) This is a follow up to servo#36818 (comment). Since openxr is a feature only available on Windows, it was omitted. Testing: Fixes ``` warning: hiding a lifetime that's elided elsewhere is confusing --> components\webxr\openxr\input.rs:366:9 | 366 | &self, | ^^^^^ the lifetime is elided here ... 371 | ) -> Vec<Binding> { | ------- the same lifetime is hidden here | = help: the same lifetime is referred to in inconsistent ways, making the signature confusing = note: `#[warn(mismatched_lifetime_syntaxes)]` on by default help: use `'_` for type paths | 371 | ) -> Vec<Binding<'_>> { ``` Signed-off-by: Euclid Ye <[email protected]>

Loirooriol · 2025-09-04T18:57:12Z

The mismatched_lifetime_syntaxes lint is complaining a lot when I compile Servo patched with a local Stylo. Was there a way to automatically address the warnings, or does it need to be manual?

webbeef · 2025-09-05T02:23:58Z

The mismatched_lifetime_syntaxes lint is complaining a lot when I compile Servo patched with a local Stylo. Was there a way to automatically address the warnings, or does it need to be manual?

Setting RUSTFLAGS="-A mismatched_lifetime_syntaxes" in your environment should do the trick.

Loirooriol · 2025-09-08T09:52:38Z

@webbeef Yes, that's what I'm doing for now, but would be better to fix the code in Stylo. Doing it manually will be quite tedious, though.

Was fb17b63 done in some automated way?

sagudev · 2025-09-08T09:54:20Z

Not really, I did do find and replace on most common types tho.

webbeef requested review from Loirooriol, gterzian, mrobinson and nicoburns as code owners May 3, 2025 03:29

webbeef marked this pull request as draft May 3, 2025 04:00

webbeef force-pushed the rustc-1.86 branch from ee8c516 to 1c1f24b Compare May 3, 2025 04:14

webbeef marked this pull request as ready for review May 3, 2025 04:50

sagudev previously approved these changes May 3, 2025

View reviewed changes

jdm added this pull request to the merge queue May 3, 2025

github-merge-queue bot removed this pull request from the merge queue due to failed status checks May 3, 2025

sagudev added this pull request to the merge queue May 3, 2025

github-merge-queue bot removed this pull request from the merge queue due to failed status checks May 3, 2025

webbeef force-pushed the rustc-1.86 branch from 1c1f24b to 287456c Compare May 3, 2025 19:43

This comment was marked as resolved.

Sign in to view

jdm mentioned this pull request May 6, 2025

script_bindings: Check for null pointer before dereferencing proxy handler custom data #36869

Merged

mrobinson previously approved these changes May 16, 2025

View reviewed changes

mrobinson added this pull request to the merge queue May 16, 2025

github-actions bot removed the T-macos Do a try run on macOS label Aug 19, 2025

sagudev added the T-linux-build-libservo label Aug 19, 2025

github-actions bot removed the T-linux-build-libservo label Aug 19, 2025

another fixup

470b3ed

Signed-off-by: sagudev <[email protected]>

sagudev added the T-linux-build-libservo label Aug 19, 2025

github-actions bot removed the T-linux-build-libservo label Aug 19, 2025

sagudev enabled auto-merge August 19, 2025 10:48

sagudev added this pull request to the merge queue Aug 19, 2025

Merged via the queue into servo:main with commit 3225d19 Aug 19, 2025
35 checks passed

yezhizhen mentioned this pull request Aug 19, 2025

webxr: Fix compile warning for Windows after upgrade to rustc 1.89 #38780

Merged

jschwe mentioned this pull request Aug 27, 2025

Tracking issue: Binary size regressions #38966

Open

Loirooriol mentioned this pull request Oct 13, 2025

Warning: New version for cargo-deny available: 0.18.5 #39836

Closed

Uh oh!

Conversation

webbeef commented May 3, 2025 • edited by sagudev Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

sagudev commented May 3, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

atouchet commented May 3, 2025

Uh oh!

webbeef commented May 3, 2025

Uh oh!

jdm commented May 4, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

This comment was marked as resolved.

This comment was marked as resolved.

This comment was marked as resolved.

sagudev commented May 5, 2025

Uh oh!

sagudev commented May 5, 2025

Uh oh!

jdm commented May 6, 2025

Uh oh!

jdm commented May 6, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

webbeef commented May 16, 2025

Uh oh!

github-actions bot commented Aug 19, 2025

Uh oh!

github-actions bot commented Aug 19, 2025

Uh oh!

github-actions bot commented Aug 19, 2025

Uh oh!

github-actions bot commented Aug 19, 2025

Uh oh!

github-actions bot commented Aug 19, 2025

Uh oh!

Uh oh!

jschwe commented Aug 19, 2025

Uh oh!

Loirooriol commented Sep 4, 2025

Uh oh!

webbeef commented Sep 5, 2025

Uh oh!

Loirooriol commented Sep 8, 2025

Uh oh!

sagudev commented Sep 8, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

9 participants

webbeef commented May 3, 2025 •

edited by sagudev

Loading

sagudev commented May 3, 2025 •

edited

Loading

jdm commented May 4, 2025 •

edited

Loading

jdm commented May 6, 2025 •

edited

Loading