https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126824
Drea Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|WAITING |NEW
Keywords| |documentation
--- Comment #2 from Drea Pinski <pinskia at gcc dot gnu.org> ---
Oh wait this are for casting between float types.
https://gcc.gnu.org/onlinedocs/gccint/Soft-float-library-routines.html#index-_005f_005ftruncdfsf2
Most likely at the time the documentation was written different rounding modes
were not taken into account.