man3/__setfpucw.3

   1 .\" Written Sat Mar  8 10:35:08 MEZ 1997 by
   2 .\" J. "MUFTI" Scheurich (mufti@csv.ica.uni-stuttgart.de)
   3 .\"
   4 .\" SPDX-License-Identifier: GPL-1.0-or-later
   5 .\"
   6 .TH __SETFPUCW 3 2021-03-22 "Linux man-pages (unreleased)"
   7 .SH NAME
   8 __setfpucw \- set FPU control word on i386 architecture (obsolete)
   9 .SH LIBRARY
  10 Standard C library
  11 .RI ( libc ", " \-lc )
  12 .SH SYNOPSIS
  13 .nf
  14 .B #include <i386/fpu_control.h>
  15 .PP
  16 .BI "void __setfpucw(unsigned short " control_word );
  17 .fi
  18 .SH DESCRIPTION
  19 .BR __setfpucw ()
  20 transfers
  21 .I control_word
  22 to the registers of the FPU (floating-point unit) on the i386 architecture.
  23 This was used to control floating-point precision,
  24 rounding and floating-point exceptions.
  25 .SH STANDARDS
  26 This function was a nonstandard GNU extension.
  27 .SH NOTES
  28 As of glibc 2.1 this function does not exist anymore.
  29 There are new functions from C99, with prototypes in
  30 .IR <fenv.h> ,
  31 to control FPU rounding modes, like
  32 .BR fegetround (3),
  33 .BR fesetround (3),
  34 and the floating-point environment, like
  35 .BR fegetenv (3),
  36 .BR feholdexcept (3),
  37 .BR fesetenv (3),
  38 .BR feupdateenv (3),
  39 and FPU exception handling, like
  40 .BR feclearexcept (3),
  41 .BR fegetexceptflag (3),
  42 .BR feraiseexcept (3),
  43 .BR fesetexceptflag (3),
  44 and
  45 .BR fetestexcept (3).
  46 .PP
  47 If direct access to the FPU control word is still needed, the
  48 .B _FPU_GETCW
  49 and
  50 .B _FPU_SETCW
  51 macros from
  52 .I <fpu_control.h>
  53 can be used.
  54 .SH EXAMPLES
  55 .B __setfpucw(0x1372)
  56 .PP
  57 Set FPU control word on the i386 architecture to
  58 .RS
  59 .PD 0
  60 .IP - 2
  61 extended precision
  62 .IP -
  63 rounding to nearest
  64 .IP -
  65 exceptions on overflow, zero divide and NaN
  66 .PD
  67 .RE
  68 .SH SEE ALSO
  69 .BR feclearexcept (3)
  70 .PP
  71 .I <fpu_control.h>