[mugurc]

What

guest_path_name_length (#8164) fails on macOS at its check that a guest path of exactly 1023 bytes is not too long. The guest check passes it; the host does not. macOS fails a host path of 1024 bytes or more with ENAMETOOLONG, and the host form of a guest path is the root directory followed by the guest path, so a 1023-byte guest path is always longer than that there. In the test the root is the 73-byte build directory, so the host path has 1097 bytes. Part of the macOS series announced on #1 (#1 (comment)).

• The test skips that one check on macOS. Every other check still runs there, including the 1024-byte path that must fail.
• A new debt entry gives the macOS limit. The entry for guest path names already says that the host's own limit applies below FreeBSD's and gives the Windows one; it is at the 1000-character limit, so the macOS one has its own entry.

Tested

macOS 26.5.1, Apple M5 Max, Apple clang 21, x86-64 build under Rosetta 2. The libraries do not link on macOS without the rest of the macOS series, so the run is on main 3fd1cc3 with the series (ci/macos-series-2 on my fork), #8279, the open macOS fixes #7524 and #7937, and three changes from #7332.

• guest_path_name_length aborts without this PR with "a path of exactly 1023 bytes must not be too long (errno 63)" and passes with it.
• A probe with stat on macOS: an absolute path of 1023 bytes gives ENOENT, one of 1024 bytes ENAMETOOLONG.
• guest_working_directory, guest_path_case and guest_directory pass. guest_filesystem fails on macOS with and without this PR; it is among the macOS failures feat(libkernel): direct memory mappings on macOS hosts #7631 lists, from before fix(libc): fail an oversized guest path name with ENAMETOOLONG #8164.
• Linux and Windows: not run. Outside macOS the test is unchanged.

Checklist

• Based on current main; no other open PR implements the same functions
• One topic per PR; follow-ups go in a new PR
• No comments in code except technical debt
• Unimplemented paths throw (NotImplemented_nid_no_patch); silent stubs are listed under Silent stubs in their library's technical debt file
• No notes, investigation .md files or images added to the repository (attach them to this PR)
• New third-party code is a submodule built from source (none added)
• Generic behaviour, not specific to one title
• Depends on: none
• Fixes: none
• Unblocks: none
• GPUs tested: n/a
• AI-assisted: yes

[github-actions]

Other open pull requests touch the same code. Merging one will break the other, or both implement the same thing. If this one builds on them, list them in Depends on.

Pull request	Conflicting files (lines on main)	Same exports	Same new files
PR 9873	docs/dev/debt/libc.md L77

Some comparisons could not be completed because a pull request conflicts with the base branch. Review these shared files manually before concluding that there is no overlap.

Pull request	Shared files not compared
PR 3965	docs/dev/debt/libc.md
PR 5021	docs/dev/debt/libc.md
PR 5303	docs/dev/debt/libc.md
PR 7081	docs/dev/debt/libc.md
PR 7648	docs/dev/debt/libc.md
PR 7715	docs/dev/debt/libc.md
PR 8150	docs/dev/debt/libc.md
PR 8306	docs/dev/debt/libc.md
PR 8965	docs/dev/debt/libc.md
PR 9133	docs/dev/debt/libc.md
PR 9408	docs/dev/debt/libc.md
PR 9447	docs/dev/debt/libc.md
PR 9526	docs/dev/debt/libc.md
PR 9527	docs/dev/debt/libc.md
PR 9552	docs/dev/debt/libc.md
PR 9669	docs/dev/debt/libc.md
PR 9929	docs/dev/debt/libc.md

[ycbuzi]

Windows run of 381c3b9 merged onto main ec7eacc (f24e2dec3a, clean), one of the hosts the description lists as not run. Windows 11 Home 10.0.26200, MinGW-w64 GCC 15.2.0 (ucrt-posix-seh r7), Release, BUILD_TESTING=ON.

Built guest_path_name_length_tests with no errors or warnings; ctest --repeat until-fail:2 on guest_path_name_length: passes on both repeats. Not run: Linux.