[ErenYeager1836]

What

Under IEEE 754 and FreeBSD's s_frexp.c and s_frexpf.c (libm), frexp and frexpf store 0 in *exp for zero, infinity, and NaN and return the argument unchanged without raising the FE_INVALID floating-point exception for a signaling NaN.

In core/libs/prx/libc/src/Math.cpp, frexp and frexpf forwarded directly to host std::frexp. On hosts where std::frexp evaluates floating-point arguments via FPU operations, passing a signaling NaN can raise FE_INVALID, and *exp is left unspecified by C17 7.12.6.4 for infinity and NaN.

This change:

• Uses bitwise checks (std::memcpy into integer words) to identify zero, infinity, and NaN without tripping host floating-point exception traps on signaling NaNs.
• Stores 0 in *exponent and returns the argument unchanged for these special values as FreeBSD libm does.
• Adds test coverage for special values, zeros, infinities, quiet NaNs, and signaling NaNs in GuestMath.cpp (CheckFrexpSpecialValues).
• Documents technical debt in docs/dev/debt/libc.md.

Tested

• Windows 11, GuestMath.cpp unit tests.
• Verified python tools/check_conventions.py --base upstream/main passes with no violations.

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
• Unblocks: none
• AI-assisted: yes

[xXPBSXx]

Windows test of this PR.

Environment: Windows 10 (19045), MinGW-w64 GCC 15.2.0 (WinLibs r7), Release, BUILD_TESTING=ON, -j2, libusb-cmake from a local checkout. Built from main ea2bbefb2 with this PR and six other open PRs (#4958, #4956, #4954, #4953, #4946, #4942, #4927) merged together; the merges had no conflict.

Result: guest_math (1 of 1; the test file is shared with #4942, both merged together) pass.

Not checked: whether the test fails without the change in the library code; the PR was not tested alone, only in the combined merge (the tests of the other PRs in the batch also pass); no title run; Linux not tested.

AI-assisted (Claude Code).

[nicolasfalesy]

Linux test (Arch, GCC 16.2.1, glibc, Release), this PR merged onto main 7e181fc together with #4963, #4953, #4946 and #4942 (no conflicts): guest_math passes. With this PR's Math.cpp change reverted it fails on Linux too, so the check guards the change on both systems.

[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
#5495	docs/dev/debt/libc.md L36-37

[oneandonlydean]

Closing: no title is known to reach this. With the queue at 700+ open PRs, please hold off on new PRs that don't name a title needing the change until it's down. If a title stops on this on current main, reopen with the title ID and the log line.