fix(libc): find a relinked module by stem when the exact name misses (carries #6226) by xXPBSXx · Pull Request #8626 · boykopovar/AnyPS5

xXPBSXx · 2026-10-20T18:53:18Z

What

Carries #6226 (@lendinglibrary) onto current main, keeping the original commit and its author (a3c4cdcdbb), and adds one commit that addresses the review points on #6226 (the author has not pushed since 15 October). The problem and the Unity IL2CPP evidence are those described in #6226: a title asks for a module by its console name, the relinker names the converted module after the ELF it was built from, so RelinkedModulePath misses and sceKernelLoadStartModule fails with 0x80020002; after the exact <name>.guest.prx and the case-insensitive lookup both miss, the lookup takes the sibling with the same stem that ends with .guest.prx.

Changes on top of the original commit:

The stem is the name without its last extension only (Engine.Render.prx no longer matches Engine.Core.prx.guest.prx); the review on fix(libc): find a relinked module by stem when the exact name misses #6226 asked for this (cut at the first dot before).
The variant-names test requests a dotted name that must not load the sibling fixture (guest_module_fixture.other.prx).
Lookup order as discussed for fix(libc): prefer nested prx/*.guest.prx for RelinkedModulePath #6544: exact, same directory ignoring case, nested, stem last.
Rebased onto main b85d972f42; the only conflict was the shared line in docs/dev/debt/libc.md (the original addition is kept).

Fixes: none. #6226 stays open so its author can answer; if the author prefers to push the review fixes there, this PR can be closed.

Tested

Windows 10 (19045), MinGW-w64 GCC 15.2.0, Release, on main + this branch:

guest_dynamic_loader_variant_names: fails (failed at GuestDynamicLoaderVariants.cpp:22, exit 0xc0000409) with the cut at the first dot (new case kept) and passes with the last extension; guest_dynamic_loader and guest_filesystem pass, 3 of 3 runs.
python tools/check_conventions.py --base upstream/main: 0 errors, 1 warning (debt-length: the libc.md entry over 1000 characters comes from the original commit).
Overlap: git merge-tree of this branch with fix(libc): find a relinked module by stem when the exact name misses #6226 has three conflicts (the same code, expected); with upstream/main it is clean. pr_overlap was not run.
Not tested: Linux/macOS, a title run (the Unity run is the evidence in fix(libc): find a relinked module by stem when the exact name misses #6226).

Checklist

AI-assisted (Claude Code).

🤖 Generated with Claude Code

A title asks for a module by its console name and the relinker names the converted module after the ELF it was built from: Unity IL2CPP titles load Media/Modules/<name>.prx and Media/Plugins/<name>.prx while the dump holds <name>.elf.guest.prx, so sceKernelLoadStartModule failed with 0x80020002 and the title stopped 3 s after start. After the exact and the case-insensitive <name>.guest.prx lookups miss, take the sibling whose stem matches the request and ends with .guest.prx, shortest name first.

…n only The stem of a request and of a relinked module name was cut at the first dot, so a request for Engine.Render.prx that missed could load Engine.Core.prx.guest.prx. Only the last extension is removed now, as the review of the stem lookup asked, and the variant-names test requests a dotted name that must not load the sibling fixture module. Co-Authored-By: lendinglibrary <340867775+lendinglibrary@users.noreply.github.com> Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

github-actions · 2026-10-20T19:07:59Z

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 6226			`core/libs/tests/GuestDynamicLoaderVariants.cpp`
PR 6544	`core/libs/prx/libc/src/General.cpp` L677-680

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	`core/libs/prx/libc/src/General.cpp` `core/libs/prx/libkernel/CMakeLists.txt` `docs/dev/debt/libc.md`
PR 6044	`core/libs/prx/libkernel/CMakeLists.txt`
PR 6226	`core/libs/prx/libc/src/General.cpp` `core/libs/prx/libkernel/CMakeLists.txt` `core/libs/tests/GuestDynamicLoaderVariants.cpp` `docs/dev/debt/libc.md`
PR 6469	`core/libs/prx/libkernel/CMakeLists.txt` `docs/dev/debt/libc.md`
PR 6819	`docs/dev/debt/libc.md`
PR 6936	`core/libs/prx/libkernel/CMakeLists.txt`
PR 6977	`core/libs/prx/libkernel/CMakeLists.txt`
PR 7081	`docs/dev/debt/libc.md`
PR 7192	`core/libs/prx/libkernel/CMakeLists.txt`
PR 7201	`core/libs/prx/libc/src/General.cpp` `core/libs/prx/libkernel/CMakeLists.txt`
PR 7532	`core/libs/prx/libkernel/CMakeLists.txt`
PR 7648	`docs/dev/debt/libc.md`
PR 7715	`docs/dev/debt/libc.md`
PR 8150	`docs/dev/debt/libc.md`
PR 8520	`core/libs/prx/libkernel/CMakeLists.txt` `docs/dev/debt/libc.md`
PR 8711	`docs/dev/debt/libc.md`

MarcZX10 · 2026-10-20T19:30:53Z

            WORKING_DIRECTORY "$<TARGET_FILE_DIR:guest_module_fixture>" RESOURCE_LOCK guest_dynamic_loader_files)
    endif()
+    file(MAKE_DIRECTORY "${CMAKE_CURRENT_BINARY_DIR}/variant_names")
+    file(WRITE "${CMAKE_CURRENT_BINARY_DIR}/variant_names/guest_module_fixture.sprx.elf.GUEST.PRX" "not a module\n")


This fixture no longer exercises the shortest-match or dangling-symlink-before-valid-candidate paths after the last-extension change. For the request guest_module_fixture.prx, SameRelinkedStem strips .GUEST.PRX and then only the last extension: guest_module_fixture.sprx.elf.GUEST.PRX yields stem guest_module_fixture.sprx, and the symlink name ending in .zprx.elf.GUEST.PRX yields guest_module_fixture.zprx. Neither matches guest_module_fixture; only the copied .elf.GUEST.PRX file matches. Could the decoy and symlink instead be named e.g. guest_module_fixture.sprx.GUEST.PRX and guest_module_fixture.zprx.GUEST.PRX? Those share the requested stem under this rule and would make the selection/error-path coverage real.

… for the stem With the stem cut at the last extension, the decoy guest_module_fixture.sprx.elf.GUEST.PRX and the dangling link guest_module_fixture.zprx.elf.GUEST.PRX had the stems guest_module_fixture.sprx and guest_module_fixture.zprx, so they no longer matched the request and the shortest-name rule was not exercised. They are now guest_module_fixture.sprx.GUEST.PRX and guest_module_fixture.zprx.GUEST.PRX, which have the request's stem and are longer than the real module copy. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>

xXPBSXx · 2026-10-20T20:18:02Z

Good catch, thank you. After the cut at the last extension the decoy and the dangling link no longer had the request's stem, so the shortest-name rule was not exercised any more. I renamed them as you suggested (guest_module_fixture.sprx.GUEST.PRX, guest_module_fixture.zprx.GUEST.PRX; commit c983777). To check that they compete again I flipped the length comparison in RelinkedModulePath's stem lookup: guest_dynamic_loader_variant_names then fails at the first dlopen, and it passes with the real comparison (Windows 10, MinGW-w64 GCC 15.2, Release). The dangling-link variant only exists on non-Windows hosts and I did not run it.
AI-assisted (Claude Code).

oneandonlydean · 2026-10-21T01:22:52Z

@xXPBSXx I merged current main into this branch (2de9386). Besides a keep-both in General.cpp, the merge commit adds one guard in MatchNames.

#7467 added an early return to MatchNamesWindows for case-insensitive directories (the NTFS default). That path looks up each name exactly and never reaches SameRelinkedStem, so on Windows this PR's stem fallback would never run. The merge now takes the Windows fast path only when no requested name has stem set:

bool stem = false;
for (const auto& name : names) stem = stem || name.stem;
if (!stem && CaseInsensitiveDirectoryWindows(directory)) return MatchNamesWindows(directory, names);

ResolveHostPath never sets stem, so #7467's fast path is unchanged for ordinary lookups. Only RelinkedModulePath goes back to the iterator. guest_dynamic_loader_variant_names passing on Windows CI will confirm it.

Co-authored-by: xXPBSXx <20614131+xXPBSXx@users.noreply.github.com>

oneandonlydean · 2026-10-21T06:29:26Z

Note: I force-pushed the merge commit as 8cb7251b2e to correct its committer identity. The tree is identical to the previous head 2de93860ab, so nothing in the code changed. CI on the new head is next.

lendinglibrary and others added 2 commits October 20, 2026 20:42

MarcZX10 reviewed Oct 20, 2026

View reviewed changes

github-actions Bot added libc libkernel build docs labels Oct 20, 2026

xXPBSXx mentioned this pull request Oct 20, 2026

fix(libkernel): fail unlink, remove, rename and open of a file named with a trailing slash with ENOTDIR on Windows #8680

Closed

10 tasks done

ArjenS-LA mentioned this pull request Oct 20, 2026

fix(libkernel): classify Windows directory links by app0 scope #8726

Merged

12 tasks done

github-actions Bot added the needs rebase Conflicts with main, rebase needed label Oct 21, 2026

github-actions Bot removed the needs rebase Conflicts with main, rebase needed label Oct 21, 2026

Merge origin/main into PR 8626

8cb7251

Co-authored-by: xXPBSXx <20614131+xXPBSXx@users.noreply.github.com>

oneandonlydean force-pushed the carry/6226-relinker-stem branch from 2de9386 to 8cb7251 Compare October 21, 2026 06:28

oneandonlydean merged commit 5b5d2a8 into boykopovar:main Oct 21, 2026
17 checks passed

This was referenced Oct 21, 2026

fix(libc): prefer nested prx/*.guest.prx for RelinkedModulePath #6544

Closed

fix(libc): find a relinked module by stem when the exact name misses #6226

Closed

xXPBSXx mentioned this pull request Oct 24, 2026

fix(libc): look in prx, sce_module and sce_modules for a relinked module after its own directory #10494

Merged

12 tasks done

Navigation Menu

Repository navigation

fix(libc): find a relinked module by stem when the exact name misses (carries #6226) - #8626

fix(libc): find a relinked module by stem when the exact name misses (carries #6226)#8626
oneandonlydean merged 4 commits into
boykopovar:mainfrom
xXPBSXx:carry/6226-relinker-stem

xXPBSXx commented Oct 20, 2026

Uh oh!

github-actions Bot commented Oct 20, 2026 •

edited

Loading

Uh oh!

MarcZX10 Oct 20, 2026

Uh oh!

xXPBSXx commented Oct 20, 2026

Uh oh!

oneandonlydean commented Oct 21, 2026

Uh oh!

oneandonlydean commented Oct 21, 2026

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Conversation

xXPBSXx commented Oct 20, 2026

What

Tested

Checklist

Uh oh!

github-actions Bot commented Oct 20, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

MarcZX10 Oct 20, 2026

Choose a reason for hiding this comment

Uh oh!

xXPBSXx commented Oct 20, 2026

Uh oh!

oneandonlydean commented Oct 21, 2026

Uh oh!

oneandonlydean commented Oct 21, 2026

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

github-actions Bot commented Oct 20, 2026 •

edited

Loading