[Bug 294719] lib/msun: Added fmaximum_mag_num

[email protected] Fri, 01 May 2026 17:59:04 +0000
Newsgroups gmane.os.freebsd.devel.standards
Message-ID <[email protected]/bugzilla/>
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D294719

--- Comment #12 from Steve Kargl <[email protected]> ---
(In reply to Robert Clausecker from comment #11)
Sure, go ahead.  Commit the changes.

I'll note that I'm not a big fan of the proliferation of
USE_BUILTIN_* in the code.  libm should contain implementations
that do the job without *explicitly* using __builtin_*.  If
a compiler supports a particular builtin, it will use that
builtin when a user compiles code that includes math.h (unless
the user specifically disables that feature with a compiler
option).

Are we really going to have CFLAGS look like

CFLAGS +=3D -DUSE_BUILTIN_FMINIMUM_NUM -DUSE_BUILTIN_FMINIMUM_NUMF \
          -DUSE_BUILTIN_FMINIMUM_MAG_NUM -DUSE_BUILTIN_FMINIMUM_MAG_NUMF \
          -DUSE_BUILTIN_FMINIMUM_MAG -DUSE_BUILTIN_FMINIMUM_MAGF \
          -DUSE_BUILTIN_FMAXIMUM_NUM -DUSE_BUILTIN_FMAXIMUM_NUMF \
          -DUSE_BUILTIN_FMAXIMUM_MAG_NUM -DUSE_BUILTIN_FMAXIMUM_MAG_NUMF \
          -DUSE_BUILTIN_FMAXIMUM_MAG -DUSE_BUILTIN_FMAXIMUM_MAGF


If you're going to support the above, then why are the
long double builtins missing?=20
          -DUSE_BUILTIN_FMAXIMUM_MAGL \
          -DUSE_BUILTIN_FMINIMUM_NUML \
          -DUSE_BUILTIN_FMINIMUM_MAG_NUML \
          -DUSE_BUILTIN_FMINIMUM_MAGL \
          -DUSE_BUILTIN_FMAXIMUM_NUML \
          -DUSE_BUILTIN_FMAXIMUM_MAG_NUML

gcc.info does not list fmaximum* and fminimum* as=20
having builtins.  I cannot locate a list of math
builtins for clang.  It seems pointless to clutter
the code.  If you're going to insist on the use
of builtins, then make it an all or nothing macro
-DUSE_BUILTINS.

--=20
You are receiving this mail because:
You are the assignee for the bug.=