d43cffdfe2
MPC is a C library for the arithmetic of complex numbers with arbitrarily high precision and correct rounding of the result. It is built upon and follows the same principles as MPFR. GCC >= 4.2 requires MPC.
191 lines
6.1 KiB
C
191 lines
6.1 KiB
C
/* tsqr -- test file for mpc_sqr.
|
|
|
|
Copyright (C) INRIA, 2002, 2005, 2008, 2010, 2011
|
|
|
|
This file is part of the MPC Library.
|
|
|
|
The MPC Library is free software; you can redistribute it and/or modify
|
|
it under the terms of the GNU Lesser General Public License as published by
|
|
the Free Software Foundation; either version 2.1 of the License, or (at your
|
|
option) any later version.
|
|
|
|
The MPC Library is distributed in the hope that it will be useful, but
|
|
WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
|
|
or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
|
|
License for more details.
|
|
|
|
You should have received a copy of the GNU Lesser General Public License
|
|
along with the MPC Library; see the file COPYING.LIB. If not, write to
|
|
the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston,
|
|
MA 02111-1307, USA. */
|
|
|
|
#include <stdio.h>
|
|
#include <stdlib.h>
|
|
#include "mpc-tests.h"
|
|
|
|
static void cmpsqr (mpc_srcptr x, mpc_rnd_t rnd)
|
|
/* computes the square of x with the specific function or by simple */
|
|
/* multiplication using the rounding mode rnd and compares the results */
|
|
/* and return values. */
|
|
/* In our current test suite, the real and imaginary parts of x have */
|
|
/* the same precision, and we use this precision also for the result. */
|
|
/* Furthermore, we check whether computing the square in the same */
|
|
/* place yields the same result. */
|
|
/* We also compute the result with four times the precision and check */
|
|
/* whether the rounding is correct. Error reports in this part of the */
|
|
/* algorithm might still be wrong, though, since there are two */
|
|
/* consecutive roundings. */
|
|
{
|
|
mpc_t z, t, u;
|
|
int inexact_z, inexact_t;
|
|
|
|
mpc_init2 (z, MPC_MAX_PREC (x));
|
|
mpc_init2 (t, MPC_MAX_PREC (x));
|
|
mpc_init2 (u, 4 * MPC_MAX_PREC (x));
|
|
|
|
inexact_z = mpc_sqr (z, x, rnd);
|
|
inexact_t = mpc_mul (t, x, x, rnd);
|
|
|
|
if (mpc_cmp (z, t))
|
|
{
|
|
fprintf (stderr, "sqr and mul differ for rnd=(%s,%s) \nx=",
|
|
mpfr_print_rnd_mode(MPC_RND_RE(rnd)),
|
|
mpfr_print_rnd_mode(MPC_RND_IM(rnd)));
|
|
mpc_out_str (stderr, 2, 0, x, MPC_RNDNN);
|
|
fprintf (stderr, "\nmpc_sqr gives ");
|
|
mpc_out_str (stderr, 2, 0, z, MPC_RNDNN);
|
|
fprintf (stderr, "\nmpc_mul gives ");
|
|
mpc_out_str (stderr, 2, 0, t, MPC_RNDNN);
|
|
fprintf (stderr, "\n");
|
|
exit (1);
|
|
}
|
|
if (inexact_z != inexact_t)
|
|
{
|
|
fprintf (stderr, "The return values of sqr and mul differ for rnd=(%s,%s) \nx= ",
|
|
mpfr_print_rnd_mode(MPC_RND_RE(rnd)),
|
|
mpfr_print_rnd_mode(MPC_RND_IM(rnd)));
|
|
mpc_out_str (stderr, 2, 0, x, MPC_RNDNN);
|
|
fprintf (stderr, "\nx^2=");
|
|
mpc_out_str (stderr, 2, 0, z, MPC_RNDNN);
|
|
fprintf (stderr, "\nmpc_sqr gives %i", inexact_z);
|
|
fprintf (stderr, "\nmpc_mul gives %i", inexact_t);
|
|
fprintf (stderr, "\n");
|
|
exit (1);
|
|
}
|
|
|
|
mpc_set (t, x, MPC_RNDNN);
|
|
inexact_t = mpc_sqr (t, t, rnd);
|
|
if (mpc_cmp (z, t))
|
|
{
|
|
fprintf (stderr, "sqr and sqr in place differ for rnd=(%s,%s) \nx=",
|
|
mpfr_print_rnd_mode(MPC_RND_RE(rnd)),
|
|
mpfr_print_rnd_mode(MPC_RND_IM(rnd)));
|
|
mpc_out_str (stderr, 2, 0, x, MPC_RNDNN);
|
|
fprintf (stderr, "\nmpc_sqr gives ");
|
|
mpc_out_str (stderr, 2, 0, z, MPC_RNDNN);
|
|
fprintf (stderr, "\nmpc_sqr in place gives ");
|
|
mpc_out_str (stderr, 2, 0, t, MPC_RNDNN);
|
|
fprintf (stderr, "\n");
|
|
exit (1);
|
|
}
|
|
if (inexact_z != inexact_t)
|
|
{
|
|
fprintf (stderr, "The return values of sqr and sqr in place differ for rnd=(%s,%s) \nx= ",
|
|
mpfr_print_rnd_mode(MPC_RND_RE(rnd)),
|
|
mpfr_print_rnd_mode(MPC_RND_IM(rnd)));
|
|
mpc_out_str (stderr, 2, 0, x, MPC_RNDNN);
|
|
fprintf (stderr, "\nx^2=");
|
|
mpc_out_str (stderr, 2, 0, z, MPC_RNDNN);
|
|
fprintf (stderr, "\nmpc_sqr gives %i", inexact_z);
|
|
fprintf (stderr, "\nmpc_sqr in place gives %i", inexact_t);
|
|
fprintf (stderr, "\n");
|
|
exit (1);
|
|
}
|
|
|
|
mpc_sqr (u, x, rnd);
|
|
mpc_set (t, u, rnd);
|
|
if (mpc_cmp (z, t))
|
|
{
|
|
fprintf (stderr, "rounding in sqr might be incorrect for rnd=(%s,%s) \nx=",
|
|
mpfr_print_rnd_mode(MPC_RND_RE(rnd)),
|
|
mpfr_print_rnd_mode(MPC_RND_IM(rnd)));
|
|
mpc_out_str (stderr, 2, 0, x, MPC_RNDNN);
|
|
fprintf (stderr, "\nmpc_sqr gives ");
|
|
mpc_out_str (stderr, 2, 0, z, MPC_RNDNN);
|
|
fprintf (stderr, "\nmpc_sqr quadruple precision gives ");
|
|
mpc_out_str (stderr, 2, 0, u, MPC_RNDNN);
|
|
fprintf (stderr, "\nand is rounded to ");
|
|
mpc_out_str (stderr, 2, 0, t, MPC_RNDNN);
|
|
fprintf (stderr, "\n");
|
|
exit (1);
|
|
}
|
|
|
|
mpc_clear (z);
|
|
mpc_clear (t);
|
|
mpc_clear (u);
|
|
}
|
|
|
|
|
|
static void
|
|
testsqr (long a, long b, mpfr_prec_t prec, mpc_rnd_t rnd)
|
|
{
|
|
mpc_t x;
|
|
|
|
mpc_init2 (x, prec);
|
|
|
|
mpc_set_si_si (x, a, b, rnd);
|
|
|
|
cmpsqr (x, rnd);
|
|
|
|
mpc_clear (x);
|
|
}
|
|
|
|
|
|
static void
|
|
reuse_bug (void)
|
|
{
|
|
mpc_t z1;
|
|
|
|
/* reuse bug found by Paul Zimmermann 20081021 */
|
|
mpc_init2 (z1, 2);
|
|
/* RE (z1^2) overflows, IM(z^2) = -0 */
|
|
mpfr_set_str (MPC_RE (z1), "0.11", 2, GMP_RNDN);
|
|
mpfr_mul_2si (MPC_RE (z1), MPC_RE (z1), mpfr_get_emax (), GMP_RNDN);
|
|
mpfr_set_ui (MPC_IM (z1), 0, GMP_RNDN);
|
|
mpc_conj (z1, z1, MPC_RNDNN);
|
|
mpc_sqr (z1, z1, MPC_RNDNN);
|
|
if (!mpfr_inf_p (MPC_RE (z1)) || mpfr_signbit (MPC_RE (z1))
|
|
||!mpfr_zero_p (MPC_IM (z1)) || !mpfr_signbit (MPC_IM (z1)))
|
|
{
|
|
printf ("Error: Regression, bug 20081021 reproduced\n");
|
|
MPC_OUT (z1);
|
|
exit (1);
|
|
}
|
|
|
|
mpc_clear (z1);
|
|
}
|
|
|
|
int
|
|
main (void)
|
|
{
|
|
DECL_FUNC (CC, f, mpc_sqr);
|
|
test_start ();
|
|
|
|
testsqr (247, -65, 8, 24);
|
|
testsqr (5, -896, 3, 2);
|
|
testsqr (-3, -512, 2, 16);
|
|
testsqr (266013312, 121990769, 27, 0);
|
|
testsqr (170, 9, 8, 0);
|
|
testsqr (768, 85, 8, 16);
|
|
testsqr (145, 1816, 8, 24);
|
|
testsqr (0, 1816, 8, 24);
|
|
testsqr (145, 0, 8, 24);
|
|
|
|
data_check (f, "sqr.dat");
|
|
tgeneric (f, 2, 1024, 1, 0);
|
|
|
|
reuse_bug ();
|
|
|
|
return 0;
|
|
}
|