Use `symlink_dir` to create junctions on Windows instead of trying to use symbolic links in `copy_link_internal` by Zoxc · Pull Request #152433 · rust-lang/rust

Zoxc · 2026-02-10T10:33:59Z

Symbolic links require extra privileges on Windows, so this falls back to creating junctions in symlink_file when operating on a directory.

rustbot · 2026-02-10T10:34:04Z

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

Why was this reviewer chosen?

The reviewer was selected based on:

Owners of files modified in this PR: bootstrap
bootstrap expanded to 6 candidates
Random selection from Mark-Simulacrum, Zalathar, clubby789, jieyouxu

ChrisDenton · 2026-02-10T10:42:06Z

Junction points only work on directories, not files. A hard link or a file copy would be necessary if symlinks aren't available.

Zoxc · 2026-02-11T09:40:42Z

Well this seems to only be called to create junctions to directories (on Windows), otherwise I'd be seeing more privilege errors.

ChrisDenton · 2026-02-11T09:43:15Z

So it sounds like symlink_dir should be being called and it's a bug that symlink_file is.

Mark-Simulacrum · 2026-02-14T20:57:29Z

@rustbot author

I don't have much context on the details here (especially Windows) but generally speaking it seems plausible that this is a bug in the callsite?

rustbot · 2026-02-14T20:57:33Z

Reminder, once the PR becomes ready for a review, use @rustbot ready.

rustbot · 2026-08-21T14:29:48Z

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

Zoxc · 2026-08-21T14:30:55Z

@rustbot ready

I have moved the logic to the callsite.

ChrisDenton · 2026-08-21T15:10:23Z

I should be able to look at this later today

r? ChrisDenton

ChrisDenton · 2026-08-21T19:38:14Z

lgtm

@bors r+ rollup

rust-bors · 2026-08-21T19:38:16Z

📌 Commit bbd34d2 has been approved by ChrisDenton

It is now in the queue for this repository.

Use `symlink_dir` to create junctions on Windows instead of trying to use symbolic links in `copy_link_internal` Symbolic links require extra privileges on Windows, so this falls back to creating junctions in `symlink_file` when operating on a directory.

JonathanBrouwer · 2026-08-22T08:46:14Z

@bors try jobs=x86_64-gnu-distcheck

Use `symlink_dir` to create junctions on Windows instead of trying to use symbolic links in `copy_link_internal` try-job: x86_64-gnu-distcheck

JonathanBrouwer · 2026-08-22T08:47:13Z

💔 I suspect this PR failed tests as part of a rollup
@bors r-

After fixing the problem, consider running a try job for the failed job before re-approving.

Link to failure: #161494 (comment)

rust-bors · 2026-08-22T08:47:19Z

This pull request was unapproved.

This PR was contained in a rollup (#161501), which was unapproved.

View changes since this unapproval

rust-bors · 2026-08-22T08:50:29Z

💔 Test for 35a958f failed: CI. Failed job:

try - x86_64-gnu-distcheck (web logs, enhanced plaintext logs)

Zoxc · 2026-08-23T17:19:48Z

I changed this to use is_symlink_dir on Windows, as that doesn't require us to look up the target. CI failed as the previous code tried to follow link not using the file as the absolute base.

… use symbolic links in `copy_link_internal`

JonathanBrouwer · 2026-08-23T19:39:06Z

@bors try jobs=x86_64-gnu-distcheck

Use `symlink_dir` to create junctions on Windows instead of trying to use symbolic links in `copy_link_internal` try-job: x86_64-gnu-distcheck

rust-bors · 2026-08-23T21:56:57Z

☀️ Try build successful (CI)
Build commit: 4c92e91 (4c92e91dc6006e06b90a88520c88bf9d9bfef997)
Base parent: 9a4ad59 (9a4ad59ae3073b013cd62f53f8349ddc61a012e8)

ChrisDenton · 2026-08-24T00:12:39Z

@bors r+

rust-bors · 2026-08-24T00:12:42Z

📌 Commit 9c78500 has been approved by ChrisDenton

It is now in the queue for this repository.

Use `symlink_dir` to create junctions on Windows instead of trying to use symbolic links in `copy_link_internal` Symbolic links require extra privileges on Windows, so this falls back to creating junctions in `symlink_file` when operating on a directory.

Rollup of 24 pull requests Successful merges: - #161586 (miri subtree update) - #161187 (add `Complex<T>` layout tests for straightforward targets) - #161524 (Put back `tests/rustdoc-gui/search-result-display.goml`) - #161592 (core: refactor tests/pattern.rs tests) - #161602 (Fix flakyness issue for `tests/rustdoc-gui/headers-color.goml`) - #152433 (Use `symlink_dir` to create junctions on Windows instead of trying to use symbolic links in `copy_link_internal`) - #155254 (Recover on attribute in use tree) - #158695 (Replace `CrateDump` with the Debug impl from `CStore`) - #159456 (Fix the reporting of layout normalization failures in transmute checks) - #159940 (Eliminate some buggy `unreachable!()`s in `expand_[option_]env()`) - #160452 (-Ctarget-feature is not unsafe (any more)) - #161007 (tests/ui/union/union-nodrop.rs: fix typo "expressios") - #161166 (add crashtests [5/N], remove unused aux files ) - #161294 (add crashtests [6/N]) - #161552 (also trigger overflow FCW when going from overflow -> error) - #161569 (Bump cfg_aliases to 0.2.2) - #161573 (re-bless `pretty-std` on windows) - #161580 (Add codegen test for redundant bit mask after a range check) - #161588 (update eyre) - #161589 (std: reduce visibility of some internal OsStr related types) - #161598 ([Bootstrap] Pass exact CI `llvm-config` executable path) - #161601 (Path: use optimized is_empty() method) - #161604 (coretests: Add a few tests for backward multibyte predicate) - #161615 (interpret: make validate_c_variadic_compatible_ty public)

Use `symlink_dir` to create junctions on Windows instead of trying to use symbolic links in `copy_link_internal` Symbolic links require extra privileges on Windows, so this falls back to creating junctions in `symlink_file` when operating on a directory.

Rollup of 22 pull requests Successful merges: - #161586 (miri subtree update) - #161187 (add `Complex<T>` layout tests for straightforward targets) - #161524 (Put back `tests/rustdoc-gui/search-result-display.goml`) - #161592 (core: refactor tests/pattern.rs tests) - #161602 (Fix flakyness issue for `tests/rustdoc-gui/headers-color.goml`) - #152433 (Use `symlink_dir` to create junctions on Windows instead of trying to use symbolic links in `copy_link_internal`) - #155254 (Recover on attribute in use tree) - #158695 (Replace `CrateDump` with the Debug impl from `CStore`) - #159456 (Fix the reporting of layout normalization failures in transmute checks) - #159940 (Eliminate some buggy `unreachable!()`s in `expand_[option_]env()`) - #160452 (-Ctarget-feature is not unsafe (any more)) - #161007 (tests/ui/union/union-nodrop.rs: fix typo "expressios") - #161552 (also trigger overflow FCW when going from overflow -> error) - #161569 (Bump cfg_aliases to 0.2.2) - #161573 (re-bless `pretty-std` on windows) - #161580 (Add codegen test for redundant bit mask after a range check) - #161588 (update eyre) - #161589 (std: reduce visibility of some internal OsStr related types) - #161598 ([Bootstrap] Pass exact CI `llvm-config` executable path) - #161601 (Path: use optimized is_empty() method) - #161604 (coretests: Add a few tests for backward multibyte predicate) - #161615 (interpret: make validate_c_variadic_compatible_ty public)

rustbot assigned Mark-Simulacrum Feb 10, 2026

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) labels Feb 10, 2026