[win] Use a dash instead of slash for linker to avoid breaking lld by dpaoliello · Pull Request #140977 · rust-lang/rust

dpaoliello · 2025-05-13T21:06:39Z

#140758 added the /arm64hazardfree linker arg for aarch64-pc-windows-msvc, unfortunately this breaks linking with lld as it 1) doesn't understand the arguments and 2) assumes that unknown args that start with / are input files.

Fix is to use a dash for the arg: -arm64hazardfree

r? @wesleywiser

rustbot · 2025-05-13T21:06:44Z

These commits modify compiler targets.
(See the Target Tier Policy.)

dpaoliello · 2025-05-13T21:08:12Z

r? @jieyouxu

(Sorry, referenced the wrong PR as causing the regression)

jieyouxu

Thanks

jieyouxu · 2025-05-14T00:18:07Z

r? jieyouxu @bors r+ rollup

bors · 2025-05-14T00:18:10Z

📌 Commit a1ea0ce has been approved by jieyouxu

It is now in the queue for this repository.

bors · 2025-05-14T05:39:10Z

⌛ Testing commit a1ea0ce with merge 55429e2...

[win] Use a dash instead of slash for linker to avoid breaking lld rust-lang#140758 added the `/arm64hazardfree` linker arg for aarch64-pc-windows-msvc, unfortunately this breaks linking with `lld` as it 1) doesn't understand the arguments and 2) assumes that unknown args that start with `/` are input files. Fix is to use a dash for the arg: `-arm64hazardfree` r? `@wesleywiser`

rust-log-analyzer · 2025-05-14T07:38:45Z

The job x86_64-msvc-1 failed! Check out the build log: (web) (plain)

Click to see the possible cause of the failure (guessed by this bot)

test [rustdoc] tests\rustdoc\private\files-creation-private.rs ... FAILED
test [rustdoc] tests\rustdoc\private\hidden-private.rs ... FAILED
test [rustdoc] tests\rustdoc\private\inline-private-with-intermediate-doc-hidden.rs ... FAILED
test [rustdoc] tests\rustdoc\private\inner-private-110422.rs ... FAILED
test [rustdoc] tests\rustdoc\private\macro-document-private-duplicate.rs ... ignored, ignored always ((fa2025-05-14T07:38:33.7790497Z ##[error]There is not enough space on the disk. : 'C:\actions-runner\cached\_diag\blocks\5e069e9a-a3b8-445c-8b4c-268021635ec2_27eb41a7-2ed4-56f7-ee04-8c5e17bd5c5c.2'
##[error]There is not enough space on the disk. : 'C:\actions-runner\cached\_diag\pages\5e069e9a-a3b8-445c-8b4c-268021635ec2_ccc4059e-4ea2-4996-9631-737428769f20_1.log'

bors · 2025-05-14T07:39:23Z

💔 Test failed - checks-actions

jieyouxu · 2025-05-14T08:17:20Z

@bors retry (not enough space)

[win] Use a dash instead of slash for linker to avoid breaking lld rust-lang#140758 added the `/arm64hazardfree` linker arg for aarch64-pc-windows-msvc, unfortunately this breaks linking with `lld` as it 1) doesn't understand the arguments and 2) assumes that unknown args that start with `/` are input files. Fix is to use a dash for the arg: `-arm64hazardfree` r? `@wesleywiser`

Rollup of 4 pull requests Successful merges: - rust-lang#140953 (Fix a compiletest blessing message) - rust-lang#140973 (Update rustix to 1.0.7 for bootstrap) - rust-lang#140976 (Add `Ipv4Addr` and `Ipv6Addr` diagnostic items) - rust-lang#140977 ([win] Use a dash instead of slash for linker to avoid breaking lld) r? `@ghost` `@rustbot` modify labels: rollup

glandium · 2025-05-15T00:20:48Z

Fix is to use a dash for the arg: -arm64hazardfree

For a value of "fix" that means lld emitting a warning about an unknown argument.

[win] Use a dash instead of slash for linker to avoid breaking lld rust-lang#140758 added the `/arm64hazardfree` linker arg for aarch64-pc-windows-msvc, unfortunately this breaks linking with `lld` as it 1) doesn't understand the arguments and 2) assumes that unknown args that start with `/` are input files. Fix is to use a dash for the arg: `-arm64hazardfree` r? `@wesleywiser`

dpaoliello · 2025-05-15T17:55:41Z

Fix is to use a dash for the arg: -arm64hazardfree

For a value of "fix" that means lld emitting a warning about an unknown argument.

Unfortunately, #133409 forced the lld and non-lld args to be the same, and there doesn't seem to be a standard way to indicate that an argument is intentionally different between the two.

@RalfJung @petrochenkov what's the correct way to add an MSVC linker ONLY arg that SHOULDN'T be passed to lld?

dpaoliello · 2025-05-15T18:07:00Z

Nevermind, the more that I think about this, the more that I believe that using the undocumented flag is the incorrect solution.

I'm going to disable the test instead, and I've filed a bug against the MSVC team to remove this check from the linker: https://developercommunity.microsoft.com/t/Remove-checking-for-and-fixing-Cortex-A/10905134

dpaoliello · 2025-05-15T18:20:59Z

Abandoning in favor of #141045

…mpiler-errors Rollup of 14 pull requests Successful merges: - rust-lang#139749 (docs(library/core/src/pin): fix typo "necessarily" -> "necessary") - rust-lang#140130 (Add LLDB providers for BTreeMap and BTreeSet) - rust-lang#140685 (Simplify `Vec::as_non_null` implementation and make it `const`) - rust-lang#140712 (normalization: avoid incompletely constraining GAT args) - rust-lang#140768 (Improve `dangerous_implicit_aurorefs` diagnostic output) - rust-lang#140834 (move (or remove) some impl Trait tests) - rust-lang#140910 (Remove `stable` attribute from wasi fs (read_exact|write_all)_at) - rust-lang#140947 (Flush errors before deep normalize in `dropck_outlives`) - rust-lang#140966 (Remove #![feature(let_chains)] from library and src/librustdoc) - rust-lang#140977 ([win] Use a dash instead of slash for linker to avoid breaking lld) - rust-lang#140990 (VxWorks: updates from recent libc versions) - rust-lang#141003 (Improve ternary operator recovery) - rust-lang#141013 (Implement methods to set STARTUPINFO flags for Command API on Windows) - rust-lang#141026 (rustc-dev-guide subtree update) r? `@ghost` `@rustbot` modify labels: rollup

[win][arm64] Remove 'Arm64 Hazard' undocumented MSVC option and instead disable problematic test PR rust-lang#140758 added the undocumented `/arm64hazardfree` MSVC linker flag to work around a test failure where LLVM generated code that would trip a hazard in an outdated ARM processor. Adding this flag caused issues with LLD, as it doesn't recognize it. Rethinking the issue, using the undocumented flag seems like the incorrect solution: there's no guarantee that the flag won't be removed in the future, or change its meaning. Instead, I've disabled the problematic test for Arm64 Windows and have filed a bug with the MSVC team to have the check removed: <https://developercommunity.microsoft.com/t/Remove-checking-for-and-fixing-Cortex-A/10905134> This PR supersedes rust-lang#140977 r? `@jieyouxu`

[win][arm64] Remove 'Arm64 Hazard' undocumented MSVC option and instead disable problematic test PR rust-lang#140758 added the undocumented `/arm64hazardfree` MSVC linker flag to work around a test failure where LLVM generated code that would trip a hazard in an outdated ARM processor. Adding this flag caused issues with LLD, as it doesn't recognize it. Rethinking the issue, using the undocumented flag seems like the incorrect solution: there's no guarantee that the flag won't be removed in the future, or change its meaning. Instead, I've disabled the problematic test for Arm64 Windows and have filed a bug with the MSVC team to have the check removed: <https://developercommunity.microsoft.com/t/Remove-checking-for-and-fixing-Cortex-A/10905134> This PR supersedes rust-lang#140977 r? ``@jieyouxu``

[win][arm64] Remove 'Arm64 Hazard' undocumented MSVC option and instead disable problematic test PR rust-lang#140758 added the undocumented `/arm64hazardfree` MSVC linker flag to work around a test failure where LLVM generated code that would trip a hazard in an outdated ARM processor. Adding this flag caused issues with LLD, as it doesn't recognize it. Rethinking the issue, using the undocumented flag seems like the incorrect solution: there's no guarantee that the flag won't be removed in the future, or change its meaning. Instead, I've disabled the problematic test for Arm64 Windows and have filed a bug with the MSVC team to have the check removed: <https://developercommunity.microsoft.com/t/Remove-checking-for-and-fixing-Cortex-A/10905134> This PR supersedes rust-lang#140977 r? ```@jieyouxu```

Rollup merge of rust-lang#141045 - dpaoliello:noarmhazard, r=jieyouxu [win][arm64] Remove 'Arm64 Hazard' undocumented MSVC option and instead disable problematic test PR rust-lang#140758 added the undocumented `/arm64hazardfree` MSVC linker flag to work around a test failure where LLVM generated code that would trip a hazard in an outdated ARM processor. Adding this flag caused issues with LLD, as it doesn't recognize it. Rethinking the issue, using the undocumented flag seems like the incorrect solution: there's no guarantee that the flag won't be removed in the future, or change its meaning. Instead, I've disabled the problematic test for Arm64 Windows and have filed a bug with the MSVC team to have the check removed: <https://developercommunity.microsoft.com/t/Remove-checking-for-and-fixing-Cortex-A/10905134> This PR supersedes rust-lang#140977 r? ```@jieyouxu```

[win] Use a dash instead of slash for linker to avoid breaking lld

a1ea0ce

rustbot assigned wesleywiser May 13, 2025

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels May 13, 2025

rustbot assigned jieyouxu and unassigned wesleywiser May 13, 2025

dpaoliello mentioned this pull request May 13, 2025

[win][arm64] Disable MSVC Linker 'Arm Hazard' warning #140758

Merged

jieyouxu approved these changes May 14, 2025

View reviewed changes

This comment has been minimized.

Sign in to view

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 May 14, 2025

bors added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels May 14, 2025

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 May 14, 2025

jieyouxu mentioned this pull request May 14, 2025

Rollup of 4 pull requests #140993

Closed

compiler-errors mentioned this pull request May 15, 2025

Rollup of 14 pull requests #141041

Closed

dpaoliello mentioned this pull request May 15, 2025

[win][arm64] Remove 'Arm64 Hazard' undocumented MSVC option and instead disable problematic test #141045

Merged

dpaoliello closed this May 15, 2025

dpaoliello deleted the dash branch August 5, 2025 00:09

Uh oh!

Conversation

dpaoliello commented May 13, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented May 13, 2025

Uh oh!

dpaoliello commented May 13, 2025

Uh oh!

jieyouxu left a comment

Choose a reason for hiding this comment

Uh oh!

jieyouxu commented May 14, 2025

Uh oh!

This comment has been minimized.

bors commented May 14, 2025

Uh oh!

bors commented May 14, 2025

Uh oh!

rust-log-analyzer commented May 14, 2025

Uh oh!

bors commented May 14, 2025

Uh oh!

jieyouxu commented May 14, 2025

Uh oh!

glandium commented May 15, 2025

Uh oh!

dpaoliello commented May 15, 2025

Uh oh!

dpaoliello commented May 15, 2025

Uh oh!

dpaoliello commented May 15, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

dpaoliello commented May 13, 2025 •

edited

Loading