[edubox]

What

Export std::setprecision(int) (_ZSt12setprecisioni, NID 1h8hFQghR7w). Dinkumware's setprecision returns the same 16-byte manipulator as setw: the setter goes in RAX and the 32-bit precision in RDX. The setter writes ios_base::precision at offset 0x1c, leaving flags and width alone.

Evidence: disassembled the libc.prx bundled in PPSA02530's sce_module. 1h8hFQghR7w is a 10-byte export that returns mov %esi,0x1c(%rdi); ret in RAX and the argument in EDX. _ZSt4setwi has the same shape and writes 0x20 (width); both offsets are now pinned with static_assert in GuestLocale.hpp.

Tested

• ctest -R '^guest_locale$' (Linux, CachyOS, GCC 16.2.1): passed. CheckStreamPrecision covers the manipulator pair, the precision write, flags/width untouched, zero and negative values, guard bytes and the null-stream throw.
• nm -D --defined-only build/core/libs/libs/libc.prx | grep 1h8hFQghR7w: exported.
• Real title: on the same implementation before the rebase onto current main (base 0a0ed7216), Minecraft Legends PPSA05510 used to stop at undefined symbol: 1h8hFQghR7w; it then stopped at undefined symbol: 8SDojuZyQaY (std::error_category::default_error_condition), so this import resolves and the title reaches the next known gap.
• python3 tools/check_conventions.py --base origin/main: pass.
• python3 tools/pr_overlap.py --branch feat/libc-setprecision --jobs 32: no overlap. Four open PRs share files but could not be compared because they conflict with the base (fix(libs): share Sony CRT handler and lconv state between libc and libSceAmpr #7081, feat(libc): add std::__sce_v2::_Random_device, _Fiopen and setw #9526, refactor(libSceLibcInternal): serve siginterrupt from libSceLibcInternal #10749, feat(libc): implement _Thrd_lt #10788); checked by hand, none implements or declares setprecision.

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 (removes one of the libc imports PPSA05510 needs; it still stops on the next one)
• GPUs tested: n/a (libc locale manipulation only)
• AI-assisted: yes

[github-actions]

Progress report (b90d009 → 93b0c59)

➖ System libraries: 50.09% (+0.0%, +2 implemented functions)

✅ 2 implemented

Library	Name
libc	SetStreamPrecision
libc	_ZSt12setprecisioni_nid_postfix

Changed libraries (implemented exports)

Library	Before	After	Change
libc	1348/1358 (99.26%)	1350/1360 (99.26%)	+0.00 pp

[MarcZX10]

Reviewed exact head 93b0c598706ac58645f7db2c7558859c70aaf19f. The new export follows the existing setw manipulator representation and updates only the measured IosBase::precision field; the focused test covers preservation of adjacent state, guard bytes, zero/negative inputs, and null-stream failure. git diff --check and current-main conventions pass; Linux register-context and relinker/ASan checks pass, with macOS and Windows jobs still pending. I could not run the full product target locally: the available build is relinker-only, and the repository rejected the manual full-build dispatch with HTTP 403. The author reports the focused guest_locale test passing on Linux; I did not independently rerun it.

[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 7081	core/libs/prx/libc/src/LocaleSupport.cpp
PR 7851	core/libs/prx/libc/include/Api.hpp
PR 9526	core/libs/prx/libc/src/LocaleSupport.cpp core/libs/tests/GuestLocale.cpp
PR 9772	core/libs/prx/libc/include/Api.hpp

[babyhuey]

Windows run of this head 93b0c598706, merged onto main 204c6747d0a (clean merge, local only), with the selection CI's skipped Windows product job uses: Windows 11, Ryzen 7 5800X, RTX 3080, WinLibs GCC 15.2, Release, BUILD_TESTING=ON. The tree was configured fresh after the merge, then select_tests.py select / ninja -f select.ninja selected / --target libs / select_tests.py ctest --parallel 4 --timeout 120, with check_skips.py --job windows (all skips expected).

• Selected 2,390 tests: 1,695 pass, 2 fail, pr_overlap_encoding and agc_host_import_cut_lifetime (agc_host_import_cut_lifetime fails on Windows since the merge of #9159 #10746), which also fail on 204c6747d0a itself on this machine, 693 skip. No failures from this PR.

AI-assisted: yes (Claude Code)