[Danigamer05]

What

Add the ten missing stdc_leading_ones_{uc,us,ui,ul,ull} and stdc_trailing_ones_{uc,us,ui,ul,ull} libc exports. They count consecutive one bits from the most significant or least significant end using C++20 std::countl_one and std::countr_one. Zero returns 0; an all-ones input returns 8, 16, 32, 64 or 64. The guest ul argument is explicitly 64-bit on Windows. Each export has one public declaration and a separate regression case.

Contract evidence is PPSA03671 v01.001.005's bundled libc.prx (SHA256 483961e086b8a1fffa8138b3c0badf798987fb5f35ec908d828946408be2d8a6), read from disassembly and run directly on a Windows host. All ten functions are call-free leaves counting the corresponding zero run in the width-limited complement. The signatures and semantics also follow C23 draft N3220, sections 7.18.4 and 7.18.6 and the LLVM libc stdbit declarations.

Export	NID	Input bits	Original RVA
stdc_leading_ones_uc	Gb7WaJE6NVY	8	0x30440
stdc_leading_ones_us	xAkgpOJwa9k	16	0x30450
stdc_leading_ones_ui	n355wwngPkM	32	0x30460
stdc_leading_ones_ul	+e7Kan3oiag	64	0x30470
stdc_leading_ones_ull	kyMcGM6dDlk	64	0x30480
stdc_trailing_ones_uc	ai13B6znEco	8	0x304e0
stdc_trailing_ones_us	exfJv2yF4-Q	16	0x30500
stdc_trailing_ones_ui	L8ab0HIh9mY	32	0x30510
stdc_trailing_ones_ul	rn7utXuD-Yk	64	0x30520
stdc_trailing_ones_ull	RzigWMxfMMA	64	0x30530

No original binary, byte dump, third-party source, investigation notes or logs are committed. The remaining console-verification gap is recorded in libc's debt file.

Tested

Runtime-tested head: 1d64984245b43ccb6a13162fd748a44058c4d4e7, based on main 1829def9bf4a54a7df2449addfc2143016ed5b94. Published head: d7986b1fb6738b5971ac3c7b7e734207b81c414e, based on main 6db75895fbcb425edb3edddedc17308582e5ff1d (28 October 2026).

After runtime testing, main received AGC changes and other libraries' regression tests. All libc production sources, headers, CMake and the focused regression sources are byte-identical between the runtime-tested and published heads. The runtime and native-comparison results below belong to the stated tested head. Conventions, whitespace, strict main synchronization and the overlap comparison were checked on the published head.

• Windows 11 / WinLibs MinGW-w64 GCC 15.2.0 and Ubuntu 24.04 WSL / GCC 13.3.0, Release, existing full-project configurations: built production libc, guest_one_runs_tests, guest_count_ones_tests and guest_integer_builtins_tests. ctest --test-dir build -R "^(guest_one_runs_(leading|trailing)_(uc|us|ui|ul|ull)|guest_count_ones_(uc|us|ui|ul|ull)|guest_integer_builtins|public_headers)$" -V --output-on-failure passes 17/17 on both platforms on the runtime-tested head. Windows also rebuilt the NID-patched production libc used below.
• The ten new cases check zero/all-ones, every single zero-bit position and pair of zero-bit positions, complements and guest errno preservation. The reference examines bit positions without the production counting primitives. Every 8- and 16-bit input is covered in both directions; each wider export also gets 100,000 deterministic random inputs and their complements. Each case has a separate 20-second timeout.
• Negative control on Windows: all ten new cases fail with their production bodies replaced by NotImplemented_nid_no_patch(__func__). Restoring production makes them pass with the same test executable; the negative, positive and final runtime-tested executable are byte-identical (SHA256 805138cc256c58bf2d92d58250f44644692e876f3f10978bbfef297317d0f65c).
• Direct original-versus-production comparison on Windows: 269,763 input vectors, 2,697,630 original calls and 2,697,630 production calls, zero mismatches, also checked against an independent bit-position reference. Production is reached through all ten patched NIDs; their names were independently hashed and resolved in the original dynamic export table. Only the ten original call-free leaves were executed from mapped load segments; no original initialization, relocation, other export or title execution was performed. This is host evidence, not a console measurement.
• Conventions, PR whitespace and strict main synchronization pass on the published head: one commit ahead, zero behind the stated main. No new dependency or comments in code.
• The required official full overlap scan was followed by an incremental update covering all 1367 open PR heads on the published main base. Classifications of unchanged heads with no shared file/export were reused; 26 new/updated heads were probed and 163 shared branches were normalized again using the official comparison functions. No confirmed line, export or new-file overlap and no duplicate exports on main. 40 already-conflicting branches could not be fully compared: feat(aot): consolidate native ARM64 compiler, runtime and consumers #5021, feat(libc): export _FDsign/_Dsign like MSVC CRT (PPSA34015) #6085, fix(libc): use the default heap for an executable without process parameters #6713, fix(libs): share Sony CRT handler and lconv state between libc and libSceAmpr #7081, feat(libc): scan 16-bit wide conversions on Linux #7386, feat(libc): support archive-backed stdio reads #7447, feat(libc): implement std::_Throw_C_error #7648, fix(libc): provide ASOBI_ROOT environment variable for title path resolution #7715, fix(libc): support character reads from random device streams #8150, feat(libc): implement fesetenv #8937, fix(libc): follow a Windows symbolic link before .. in a guest path #9371, feat(libc): add the std::__sce_v2 thread start, timed wait, clock and processor count calls #9447, feat(libc): add std::__sce_v2::_Random_device, _Fiopen and setw #9526, fix(libc): count clock() as the process time, like the console #9677, test(libc): cover the unwind context accessors #9775, fix(libc): compute tanhf with FreeBSD msun's algorithm, as titles' libc.prx does #9867, fix(libc): restore syslog program-name fallback #9929, fix(libc): compute expf with FreeBSD msun's algorithm, as titles' libc.prx does #10113, fix(libc): compute expm1f with FreeBSD msun's algorithm, as titles' libc.prx does #10222, fix(libc): compute erff and erfcf with fdlibm's algorithm, as titles' libc.prx does #10226, fix(libc): compute atan and atan2 with FreeBSD msun's algorithm, as titles' libc.prx does #10229, fix(libc): compute log with FreeBSD msun's algorithm, as titles' libc.prx does #10245, fix(libc): compute log10 with fdlibm's algorithm, as titles' libc.prx does #10247, fix(libc): compute sin, cos and tan with FreeBSD msun's algorithm, as titles' libc.prx does #10250, fix(libc): compute asin and acos with FreeBSD msun's algorithm, as titles' libc.prx does #10253, fix(libc): compute pow with fdlibm's algorithm, as titles' libc.prx does #10265, feat(libc): track stream orientation and implement fwide #10324, fix(libc): compute remainder and remquo with fprem1, as titles' libc.prx does #10349, fix(libc): open a directory with freopen in a read mode on Windows #10433, fix(libc): reopen a stream onto /dev/random or /dev/urandom with freopen #10620, fix(libc): bound narrow formatting fields #11332, feat(libc): implement strmode #11664, feat(libc): implement fcloseall with tracked stream lifetimes #11831, feat(libc): implement freezero #11848, fix(libc): return an aligned empty block from reallocalign to 0 bytes #11850, fix(libc): reject a malloc_stats_fast block that is not size 0x28 version 1 #11870, feat(libkernel): cancel blocked POSIX reads safely #11872, feat(libkernel): cancel blocked POSIX writes safely #11969, fix(libc): compute _FSinh as titles do #12053, fix(libc): compute cosh with fdlibm's algorithm, as titles' libc.prx does #12122. Their shared-file changes were inspected from their own merge bases; none adds these ten names/NIDs or their test. Those stale branches are not claimed merge-clean.
• This contribution, count-zeros feat(libc): add C23 count-zeros exports #12187 and zero-runs feat(libc): add C23 leading and trailing zero counts #12193 normalize onto the same main and merge cleanly in all six orders, producing one content tree with thirty unique bit-function declarations and definitions, including count-ones already on main. They are independent. This is a content check; the combined tree was not runtime-tested.

The full CTest suite, all standalone-header compilation batches, macOS, console, title and GPU execution were not tested. No performance or title-unblock claim.

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
• Generic behaviour, not specific to one title
• Depends on: none
• Fixes: none
• Unblocks: none
• GPUs tested: n/a
• AI-assisted: yes

[github-actions]

Progress report (061cdb6 → d7986b1)

📈 System libraries: 46.62% (+0.03%, +10 implemented functions)

✅ 10 implemented

Library	Name
libc	stdc_leading_ones_uc_nid_postfix
libc	stdc_leading_ones_ui_nid_postfix
libc	stdc_leading_ones_ul_nid_postfix
libc	stdc_leading_ones_ull_nid_postfix
libc	stdc_leading_ones_us_nid_postfix
libc	stdc_trailing_ones_uc_nid_postfix
libc	stdc_trailing_ones_ui_nid_postfix
libc	stdc_trailing_ones_ul_nid_postfix
libc	stdc_trailing_ones_ull_nid_postfix
libc	stdc_trailing_ones_us_nid_postfix

Changed libraries (implemented exports)

Library	Before	After	Change
libc	1410/1423 (99.09%)	1420/1433 (99.09%)	+0.01 pp

[MarcZX10]

I reviewed the implementation and regression coverage. The ten functions use width-correct unsigned arguments and C++20 countl_one/countr_one; the test includes exhaustive 8/16-bit inputs, boundary and complement cases, deterministic wider-input coverage, and errno preservation. The description also distinguishes the host comparison from console verification and records that remaining gap. On the exact head, conventions, whitespace, the Linux/Debug and Linux/Release register-context jobs, Ubuntu relinker and relinker ASan passed; Windows, Windows relinker, and macOS jobs remain in progress, while full Linux is skipped. The author reports focused runtime/reference checks on Windows and Ubuntu WSL.

I’m requesting an update to current main before approval as required by CONTRIBUTING. The branch is 19 commits behind 7fe69e276fc4e000ccb2a360ea8faeb3d7fcc137; those changes do not overlap this PR and the merge preview is clean. Please rebase/update and rerun the affected checks on the resulting head. I found no functional issue in the implementation as reviewed.

[github-actions]

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 5021	core/libs/prx/libc/CMakeLists.txt docs/dev/debt/libc.md
PR 6085	core/libs/prx/libc/CMakeLists.txt
PR 6713	core/libs/prx/libc/CMakeLists.txt docs/dev/debt/libc.md
PR 7081	core/libs/prx/libc/CMakeLists.txt docs/dev/debt/libc.md
PR 7386	docs/dev/debt/libc.md
PR 7447	core/libs/prx/libc/CMakeLists.txt docs/dev/debt/libc.md
PR 7715	core/libs/prx/libc/CMakeLists.txt docs/dev/debt/libc.md
PR 8150	core/libs/prx/libc/CMakeLists.txt docs/dev/debt/libc.md
PR 8937	core/libs/prx/libc/CMakeLists.txt core/libs/prx/libc/include/Api.hpp docs/dev/debt/libc.md
PR 9371	docs/dev/debt/libc.md
PR 9447	docs/dev/debt/libc.md
PR 9526	docs/dev/debt/libc.md
PR 9688	docs/dev/debt/libc.md
PR 9775	core/libs/prx/libc/CMakeLists.txt
PR 9929	core/libs/prx/libc/CMakeLists.txt docs/dev/debt/libc.md
PR 10253	docs/dev/debt/libc.md
PR 10324	core/libs/prx/libc/CMakeLists.txt core/libs/prx/libc/include/Api.hpp docs/dev/debt/libc.md
PR 10433	docs/dev/debt/libc.md
PR 10620	docs/dev/debt/libc.md
PR 11270	docs/dev/debt/libc.md
PR 11332	docs/dev/debt/libc.md
PR 11625	core/libs/prx/libc/CMakeLists.txt docs/dev/debt/libc.md
PR 11849	docs/dev/debt/libc.md
PR 11872	core/libs/prx/libc/CMakeLists.txt core/libs/prx/libc/include/Api.hpp docs/dev/debt/libc.md
PR 11892	docs/dev/debt/libc.md
PR 11969	core/libs/prx/libc/CMakeLists.txt core/libs/prx/libc/include/Api.hpp docs/dev/debt/libc.md
PR 12397	docs/dev/debt/libc.md
PR 12598	docs/dev/debt/libc.md
PR 12605	docs/dev/debt/libc.md

[MarcZX10]

The ten definitions use the correct width-specific unsigned types, and countl_one/countr_one match the described semantics. The focused host tests, original-versus-production comparisons, overlap analysis, and hosted checks are extensive; I found no functional defect in the reviewed diff. The published branch is based on 6db75895, while current main is 0c5401dc (43 commits ahead). CONTRIBUTING requires branches to be based on current main; please update the branch and rerun the focused libc cases and conventions check on that head before approval.

[MarcZX10]

The new exports use width-specific unsigned arguments and C++20 countl_one/countr_one, matching the described C23 operations; the Windows unsigned long ABI is explicitly fixed to 64 bits. Tests include zero, all ones, boundaries and reference-count comparisons, and the ABI evidence is identified as title libc.prx disassembly/runtime results. Current registration, Windows, and cross-platform relinker checks pass. No blocking issue found.