[BACK]Return to tsqrt.c CVS log [TXT][DIR] Up to [local] / OpenXM_contrib / gmp / mpfr / tests

Annotation of OpenXM_contrib/gmp/mpfr/tests/tsqrt.c, Revision 1.1.1.1

1.1       ohara       1: /* Test file for mpfr_sqrt.
                      2:
                      3: Copyright 1999, 2001, 2002 Free Software Foundation, Inc.
                      4:
                      5: This file is part of the MPFR Library.
                      6:
                      7: The MPFR Library is free software; you can redistribute it and/or modify
                      8: it under the terms of the GNU Lesser General Public License as published by
                      9: the Free Software Foundation; either version 2.1 of the License, or (at your
                     10: option) any later version.
                     11:
                     12: The MPFR Library is distributed in the hope that it will be useful, but
                     13: WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
                     14: or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public
                     15: License for more details.
                     16:
                     17: You should have received a copy of the GNU Lesser General Public License
                     18: along with the MPFR Library; see the file COPYING.LIB.  If not, write to
                     19: the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston,
                     20: MA 02111-1307, USA. */
                     21:
                     22: #include <math.h>
                     23: #include <stdio.h>
                     24: #include <stdlib.h>
                     25: #include <time.h>
                     26: #include "gmp.h"
                     27: #include "gmp-impl.h"
                     28: #include "mpfr.h"
                     29: #include "mpfr-impl.h"
                     30: #include "mpfr-test.h"
                     31:
                     32: #define check(a,r) check3(a,r,-1.0)
                     33:
                     34: int maxulp=0;
                     35:
                     36: void check3 _PROTO((double, mp_rnd_t, double));
                     37: void check4 _PROTO((double, mp_rnd_t, char *));
                     38: void check24 _PROTO((float, mp_rnd_t, float));
                     39: void check_float _PROTO((void));
                     40: void special _PROTO((void));
                     41: void check_inexact _PROTO((mp_prec_t));
                     42: void check_nan _PROTO((void));
                     43:
                     44: void
                     45: check3 (double a, mp_rnd_t rnd_mode, double Q)
                     46: {
                     47:   mpfr_t q; double Q2; int ck,u;
                     48:
                     49:   ck = (Q!=-1.0); /* if ck=1, then Q is certified correct */
                     50:   mpfr_init2(q, 53);
                     51:   mpfr_set_d(q, a, rnd_mode);
                     52: #ifdef MPFR_HAVE_FESETROUND
                     53:   mpfr_set_machine_rnd_mode(rnd_mode);
                     54: #endif
                     55:   mpfr_sqrt(q, q, rnd_mode);
                     56:   if (ck==0) Q = sqrt(a);
                     57:   else {
                     58:     if (Q != sqrt(a) && (!isnan(Q) || !isnan(sqrt(a)))) {
                     59:       fprintf(stderr, "you've found a bug in your machine's sqrt for x=%1.20e\n", a);
                     60:       mpfr_clear(q);
                     61:       exit(1);
                     62:
                     63:     }
                     64:   }
                     65:   Q2 = mpfr_get_d1 (q);
                     66:   if (Q!=Q2 && (!isnan(Q) || !isnan(Q2))) {
                     67:     u = ulp(Q2,Q);
                     68:     if (ck) {
                     69:       printf("mpfr_sqrt failed for a=%1.20e, rnd_mode=%s\n",
                     70:             a, mpfr_print_rnd_mode(rnd_mode));
                     71:       printf("expected sqrt is %1.20e, got %1.20e (%d ulp)\n",Q,Q2,u);
                     72:       mpfr_clear(q);
                     73:       exit(1);
                     74:     }
                     75:     else if (u>maxulp || u<-maxulp) {
                     76:       maxulp = (u>maxulp) ? u : -u;
                     77:       printf("libm.a differs from mpfr_sqrt for a=%1.20e, rnd_mode=%s\n",
                     78:             a, mpfr_print_rnd_mode(rnd_mode));
                     79:       printf("libm.a gives %1.20e, mpfr_sqrt gives %1.20e (%d ulp)\n",Q,Q2,u);
                     80:     }
                     81:   }
                     82:   mpfr_clear(q);
                     83: }
                     84:
                     85: void
                     86: check4 (double a, mp_rnd_t rnd_mode, char *Q)
                     87: {
                     88:   mpfr_t q, res;
                     89:
                     90:   mpfr_init2(q, 53); mpfr_init2(res, 53);
                     91:   mpfr_set_d(q, a, rnd_mode);
                     92:   mpfr_sqrt(q, q, rnd_mode);
                     93:   mpfr_set_str(res, Q, 16, GMP_RNDN);
                     94:   if (mpfr_cmp(q, res)) {
                     95:       printf("mpfr_sqrt failed for a=%1.20e, rnd_mode=%s\n",
                     96:             a, mpfr_print_rnd_mode(rnd_mode));
                     97:       printf("expected "); mpfr_print_binary(res); putchar('\n');
                     98:       printf("got      "); mpfr_print_binary(q); putchar('\n');
                     99:       mpfr_clear(q); mpfr_clear(res);
                    100:       exit(1);
                    101:   }
                    102:   mpfr_clear(res);
                    103:   mpfr_clear(q);
                    104: }
                    105:
                    106: void
                    107: check24 (float a, mp_rnd_t rnd_mode, float Q)
                    108: {
                    109:   mpfr_t q; float Q2;
                    110:
                    111:   mpfr_init2(q, 24);
                    112:   mpfr_set_d(q, a, rnd_mode);
                    113:   mpfr_sqrt(q, q, rnd_mode);
                    114:   Q2 = mpfr_get_d1 (q);
                    115:   if (Q!=Q2) {
                    116:     printf("mpfr_sqrt failed for a=%1.10e, prec=24, rnd_mode=%s\n",
                    117:           a, mpfr_print_rnd_mode(rnd_mode));
                    118:     printf("expected sqrt is %1.10e, got %1.10e\n",Q,Q2);
                    119:     exit(1);
                    120:   }
                    121:   mpfr_clear(q);
                    122: }
                    123:
                    124: /* the following examples come from the paper "Number-theoretic Test
                    125:    Generation for Directed Rounding" from Michael Parks, Table 3 */
                    126: void
                    127: check_float (void)
                    128: {
                    129:   float b = 8388608.0; /* 2^23 */
                    130:
                    131:   check24(b*8388610.0, GMP_RNDN, 8.388609e6);
                    132:   check24(b*2.0*16777214.0, GMP_RNDN, 1.6777215e7);
                    133:   check24(b*8388612.0, GMP_RNDN, 8.388610e6);
                    134:   check24(b*2.0*16777212.0, GMP_RNDN, 1.6777214e7);
                    135:   check24(b*11946704.0, GMP_RNDN, 1.0010805e7);
                    136:   check24(b*14321479.0, GMP_RNDN, 1.0960715e7);
                    137:   check24(b*2.0*13689673.0, GMP_RNDN, 1.5155019e7);
                    138:   check24(b*8388614.0, GMP_RNDN, 8.388611e6);
                    139:   check24(b*2.0*16777210.0, GMP_RNDN, 1.6777213e7);
                    140:   check24(b*10873622.0, GMP_RNDN, 9.550631e6);
                    141:
                    142:   check24(b*8388610.0, GMP_RNDZ, 8.388608e6);
                    143:   check24(b*2.0*16777214.0, GMP_RNDZ, 1.6777214e7);
                    144:   check24(b*8388612.0, GMP_RNDZ, 8.388609e6);
                    145:   check24(b*2.0*16777212.0, GMP_RNDZ, 1.6777213e7);
                    146:   check24(b*11946704.0, GMP_RNDZ, 1.0010805e7);
                    147:   check24(b*14321479.0, GMP_RNDZ, 1.0960715e7);
                    148:   check24(b*2.0*13689673.0, GMP_RNDZ, 1.5155019e7);
                    149:   check24(b*8388614.0, GMP_RNDZ, 8.38861e6);
                    150:   check24(b*2.0*16777210.0, GMP_RNDZ, 1.6777212e7);
                    151:   check24(b*10873622.0, GMP_RNDZ, 9.550631e6);
                    152:
                    153:   check24(b*8388610.0, GMP_RNDU, 8.388609e6);
                    154:   check24(b*2.0*16777214.0, GMP_RNDU, 1.6777215e7);
                    155:   check24(b*8388612.0, GMP_RNDU, 8.388610e6);
                    156:   check24(b*2.0*16777212.0, GMP_RNDU, 1.6777214e7);
                    157:   check24(b*11946704.0, GMP_RNDU, 1.0010806e7);
                    158:   check24(b*14321479.0, GMP_RNDU, 1.0960716e7);
                    159:   check24(b*2.0*13689673.0, GMP_RNDU, 1.515502e7);
                    160:   check24(b*8388614.0, GMP_RNDU, 8.388611e6);
                    161:   check24(b*2.0*16777210.0, GMP_RNDU, 1.6777213e7);
                    162:   check24(b*10873622.0, GMP_RNDU, 9.550632e6);
                    163:
                    164:   check24(b*8388610.0, GMP_RNDD, 8.388608e6);
                    165:   check24(b*2.0*16777214.0, GMP_RNDD, 1.6777214e7);
                    166:   check24(b*8388612.0, GMP_RNDD, 8.388609e6);
                    167:   check24(b*2.0*16777212.0, GMP_RNDD, 1.6777213e7);
                    168:   check24(b*11946704.0, GMP_RNDD, 1.0010805e7);
                    169:   check24(b*14321479.0, GMP_RNDD, 1.0960715e7);
                    170:   check24(b*2.0*13689673.0, GMP_RNDD, 1.5155019e7);
                    171:   check24(b*8388614.0, GMP_RNDD, 8.38861e6);
                    172:   check24(b*2.0*16777210.0, GMP_RNDD, 1.6777212e7);
                    173:   check24(b*10873622.0, GMP_RNDD, 9.550631e6);
                    174: }
                    175:
                    176: void
                    177: special (void)
                    178: {
                    179:   mpfr_t x, z;
                    180:   int inexact;
                    181:   mp_prec_t p;
                    182:
                    183:   mpfr_init (x);
                    184:   mpfr_init (z);
                    185:
                    186:   mpfr_set_prec (x, 27);
                    187:   mpfr_set_str_raw (x, "0.110100111010101000010001011");
                    188:   if ((inexact = mpfr_sqrt (x, x, GMP_RNDZ)) >= 0)
                    189:     {
                    190:       fprintf (stderr, "Wrong inexact flag: expected -1, got %d\n", inexact);
                    191:       exit (1);
                    192:     }
                    193:
                    194:   mpfr_set_prec (x, 2);
                    195:   for (p=2; p<1000; p++)
                    196:     {
                    197:       mpfr_set_prec (z, p);
                    198:       mpfr_set_ui (z, 1, GMP_RNDN);
                    199:       mpfr_add_one_ulp (z, GMP_RNDN);
                    200:       mpfr_sqrt (x, z, GMP_RNDU);
                    201:       if (mpfr_get_d1 (x) != 1.5)
                    202:        {
                    203:          fprintf (stderr, "Error: sqrt(1+ulp(1), up) should give 1.5 (prec=%u)\n", (unsigned) p);
                    204:          printf ("got "); mpfr_print_binary (x); putchar ('\n');
                    205:          exit (1);
                    206:        }
                    207:     }
                    208:
                    209:   /* check inexact flag */
                    210:   mpfr_set_prec (x, 5);
                    211:   mpfr_set_str_raw (x, "1.1001E-2");
                    212:   if ((inexact = mpfr_sqrt (x, x, GMP_RNDN)))
                    213:     {
                    214:       fprintf (stderr, "Wrong inexact flag: expected 0, got %d\n", inexact);
                    215:       exit (1);
                    216:     }
                    217:
                    218:   mpfr_set_prec (x, 2);
                    219:   mpfr_set_prec (z, 2);
                    220:
                    221:   /* checks the sign is correctly set */
                    222:   mpfr_set_d (x, 1.0, GMP_RNDN);
                    223:   mpfr_set_d (z, -1.0, GMP_RNDN);
                    224:   mpfr_sqrt (z, x, GMP_RNDN);
                    225:   if (mpfr_cmp_ui (z, 0) < 0) {
                    226:     fprintf (stderr, "Error: square root of %e gives %e\n",
                    227:             mpfr_get_d1 (x), mpfr_get_d1 (z));
                    228:     exit (1);
                    229:   }
                    230:
                    231:   mpfr_set_prec (x, 192);
                    232:   mpfr_set_prec (z, 160);
                    233:   mpfr_set_str_raw (z, "0.1011010100000100100100100110011001011100100100000011000111011001011101101101110000110100001000100001100001011000E1");
                    234:   mpfr_set_prec (x, 160);
                    235:   mpfr_sqrt(x, z, GMP_RNDN);
                    236:   mpfr_sqrt(z, x, GMP_RNDN);
                    237:
                    238:   mpfr_clear (x);
                    239:   mpfr_clear (z);
                    240: }
                    241:
                    242: void
                    243: check_inexact (mp_prec_t p)
                    244: {
                    245:   mpfr_t x, y, z;
                    246:   mp_rnd_t rnd;
                    247:   int inexact, sign;
                    248:
                    249:   mpfr_init2 (x, p);
                    250:   mpfr_init2 (y, p);
                    251:   mpfr_init2 (z, 2*p);
                    252:   mpfr_random (x);
                    253:   rnd = LONG_RAND() % 4;
                    254:   inexact = mpfr_sqrt (y, x, rnd);
                    255:   if (mpfr_mul (z, y, y, rnd)) /* exact since prec(z) = 2*prec(y) */
                    256:     {
                    257:       fprintf (stderr, "Error: multiplication should be exact\n");
                    258:       exit (1);
                    259:     }
                    260:   mpfr_sub (z, z, x, rnd); /* exact also */
                    261:   sign = mpfr_cmp_ui (z, 0);
                    262:   if (((inexact == 0) && (sign)) ||
                    263:       ((inexact > 0) && (sign <= 0)) ||
                    264:       ((inexact < 0) && (sign >= 0)))
                    265:     {
                    266:       fprintf (stderr, "Error: wrong inexact flag, expected %d, got %d\n",
                    267:               sign, inexact);
                    268:       printf ("x=");
                    269:       mpfr_print_binary (x);
                    270:       printf (" rnd=%s\n", mpfr_print_rnd_mode (rnd));
                    271:       printf ("y="); mpfr_print_binary (y); putchar ('\n');
                    272:       exit (1);
                    273:     }
                    274:   mpfr_clear (x);
                    275:   mpfr_clear (y);
                    276:   mpfr_clear (z);
                    277: }
                    278:
                    279: void
                    280: check_nan (void)
                    281: {
                    282:   mpfr_t  x, got;
                    283:
                    284:   mpfr_init2 (x, 100L);
                    285:   mpfr_init2 (got, 100L);
                    286:
                    287:   /* sqrt(NaN) == NaN */
                    288:   MPFR_CLEAR_FLAGS (x);
                    289:   MPFR_SET_NAN (x);
                    290:   ASSERT_ALWAYS (mpfr_sqrt (got, x, GMP_RNDZ) == 0); /* exact */
                    291:   ASSERT_ALWAYS (mpfr_nan_p (got));
                    292:
                    293:   /* sqrt(-1) == NaN */
                    294:   mpfr_set_si (x, -1L, GMP_RNDZ);
                    295:   ASSERT_ALWAYS (mpfr_sqrt (got, x, GMP_RNDZ) == 0); /* exact */
                    296:   ASSERT_ALWAYS (mpfr_nan_p (got));
                    297:
                    298:   /* sqrt(+inf) == +inf */
                    299:   MPFR_CLEAR_FLAGS (x);
                    300:   MPFR_SET_INF (x);
                    301:   MPFR_SET_POS (x);
                    302:   ASSERT_ALWAYS (mpfr_sqrt (got, x, GMP_RNDZ) == 0); /* exact */
                    303:   ASSERT_ALWAYS (mpfr_inf_p (got));
                    304:
                    305:   /* sqrt(-inf) == NaN */
                    306:   MPFR_CLEAR_FLAGS (x);
                    307:   MPFR_SET_INF (x);
                    308:   MPFR_SET_NEG (x);
                    309:   ASSERT_ALWAYS (mpfr_sqrt (got, x, GMP_RNDZ) == 0); /* exact */
                    310:   ASSERT_ALWAYS (mpfr_nan_p (got));
                    311:
                    312:   /* sqrt(-0) == 0 */
                    313:   mpfr_set_si (x, 0L, GMP_RNDZ);
                    314:   MPFR_SET_NEG (x);
                    315:   ASSERT_ALWAYS (mpfr_sqrt (got, x, GMP_RNDZ) == 0); /* exact */
                    316:   ASSERT_ALWAYS (mpfr_number_p (got));
                    317:   ASSERT_ALWAYS (mpfr_cmp_ui (got, 0L) == 0);
                    318:
                    319:   mpfr_clear (x);
                    320:   mpfr_clear (got);
                    321: }
                    322:
                    323: double five = 5.0;
                    324:
                    325: int
                    326: main (void)
                    327: {
                    328:   double a;
                    329:   mp_prec_t p;
                    330:   int k;
                    331: #ifdef MPFR_HAVE_FESETROUND
                    332:   int i;
                    333:
                    334:   mpfr_test_init ();
                    335:
                    336:   /* On Debian potato glibc 2.1.3-18, sqrt() doesn't seem to respect
                    337:      fesetround. */
                    338:   {
                    339:     double  a, b;
                    340:     mpfr_set_machine_rnd_mode (GMP_RNDU);
                    341:     a = sqrt (five);
                    342:     mpfr_set_machine_rnd_mode (GMP_RNDD);
                    343:     b = sqrt (five);
                    344:     if (a == b)
                    345:       {
                    346:         printf ("Tests suppressed, mpfr_set_machine_rnd_mode doesn't affect sqrt()\n");
                    347:         goto nogood;
                    348:       }
                    349:   }
                    350:
                    351:   SEED_RAND (time(NULL));
                    352:   for (i=0;i<100000;i++)
                    353:     {
                    354:       a = drand();
                    355:       if (a < 0.0) a = -a; /* ensures a is positive */
                    356:       check (a, LONG_RAND() % 4);
                    357:     }
                    358:  nogood:
                    359: #endif
                    360:
                    361:   check_nan ();
                    362:
                    363:   for (p=2; p<200; p++)
                    364:     for (k=0; k<200; k++)
                    365:       check_inexact (p);
                    366:   special ();
                    367:   check_float();
                    368: #ifdef HAVE_INFS
                    369:   check3 (DBL_NAN, GMP_RNDN, DBL_NAN);
                    370:   check3 (-1.0, GMP_RNDN, DBL_NAN);
                    371:   check3 (DBL_POS_INF, GMP_RNDN, DBL_POS_INF);
                    372:   check3 (DBL_NEG_INF, GMP_RNDN, DBL_NAN);
                    373: #endif
                    374:   check3(-0.0, GMP_RNDN, 0.0);
                    375:   check4(6.37983013646045901440e+32, GMP_RNDN, "5.9bc5036d09e0c@13");
                    376:   check4(1.0, GMP_RNDN, "1");
                    377:   check4(1.0, GMP_RNDZ, "1");
                    378:   check4(3.725290298461914062500000e-9, GMP_RNDN, "4@-4");
                    379:   check4(3.725290298461914062500000e-9, GMP_RNDZ, "4@-4");
                    380:   a=1190456976439861.0;
                    381:   check4(a, GMP_RNDZ, "2.0e7957873529a@6");
                    382:   check4(1024.0*a, GMP_RNDZ, "4.1cf2af0e6a534@7");
                    383:   /* the following examples are bugs in Cygnus compiler/system, found by
                    384:      Fabrice Rouillier while porting mpfr to Windows */
                    385:   check4(9.89438396044940256501e-134, GMP_RNDU, "8.7af7bf0ebbee@-56");
                    386:   check4(7.86528588050363751914e+31, GMP_RNDZ, "1.f81fc40f32062@13");
                    387:   check4(0.99999999999999988897, GMP_RNDN, "f.ffffffffffff8@-1");
                    388:   check4(1.00000000000000022204, GMP_RNDN, "1");
                    389:   /* the following examples come from the paper "Number-theoretic Test
                    390:    Generation for Directed Rounding" from Michael Parks, Table 4 */
                    391:   a = 4503599627370496.0; /* 2^52 */
                    392:
                    393:   check4(a*2.0*8732221479794286.0, GMP_RNDN, "1.f81fc40f32063@13");
                    394:   check4(a*8550954388695124.0, GMP_RNDN, "1.60c012a92fc65@13");
                    395:   check4(a*7842344481681754.0, GMP_RNDN, "1.51d17526c7161@13");
                    396:   check4(a*5935035262218600.0, GMP_RNDN, "1.25e19302f7e51@13");
                    397:   check4(a*5039650445085418.0, GMP_RNDN, "1.0ecea7dd2ec3d@13");
                    398:   check4(a*5039721545366078.0, GMP_RNDN, "1.0ecf250e8e921@13");
                    399:   check4(a*8005963117781324.0, GMP_RNDN, "1.5552f3eedcf33@13");
                    400:   check4(a*6703494707970582.0, GMP_RNDN, "1.3853ee10c9c99@13");
                    401:   check4(a*8010323124937260.0, GMP_RNDN, "1.556abe212b56f@13");
                    402:   check4(a*2.0*8010776873384260.0, GMP_RNDN, "1.e2d9a51977e6e@13");
                    403:
                    404:   check4(a*2.0*8732221479794286.0, GMP_RNDZ, "1.f81fc40f32062@13");
                    405:   check4(a*8550954388695124.0, GMP_RNDZ, "1.60c012a92fc64@13");
                    406:   check4(a*7842344481681754.0, GMP_RNDZ, "1.51d17526c716@13");
                    407:   check4(a*5935035262218600.0, GMP_RNDZ, "1.25e19302f7e5@13");
                    408:   check4(a*5039650445085418.0, GMP_RNDZ, "1.0ecea7dd2ec3c@13");
                    409:   check4(a*5039721545366078.0, GMP_RNDZ, "1.0ecf250e8e92@13");
                    410:   check4(a*8005963117781324.0, GMP_RNDZ, "1.5552f3eedcf32@13");
                    411:   check4(a*6703494707970582.0, GMP_RNDZ, "1.3853ee10c9c98@13");
                    412:   check4(a*8010323124937260.0, GMP_RNDZ, "1.556abe212b56e@13");
                    413:   check4(a*2.0*8010776873384260.0, GMP_RNDZ, "1.e2d9a51977e6d@13");
                    414:
                    415:   check4(a*2.0*8732221479794286.0, GMP_RNDU, "1.f81fc40f32063@13");
                    416:   check4(a*8550954388695124.0, GMP_RNDU, "1.60c012a92fc65@13");
                    417:   check4(a*7842344481681754.0, GMP_RNDU, "1.51d17526c7161@13");
                    418:   check4(a*5935035262218600.0, GMP_RNDU, "1.25e19302f7e51@13");
                    419:   check4(a*5039650445085418.0, GMP_RNDU, "1.0ecea7dd2ec3d@13");
                    420:   check4(a*5039721545366078.0, GMP_RNDU, "1.0ecf250e8e921@13");
                    421:   check4(a*8005963117781324.0, GMP_RNDU, "1.5552f3eedcf33@13");
                    422:   check4(a*6703494707970582.0, GMP_RNDU, "1.3853ee10c9c99@13");
                    423:   check4(a*8010323124937260.0, GMP_RNDU, "1.556abe212b56f@13");
                    424:   check4(a*2.0*8010776873384260.0, GMP_RNDU, "1.e2d9a51977e6e@13");
                    425:
                    426:   check4(a*2.0*8732221479794286.0, GMP_RNDD, "1.f81fc40f32062@13");
                    427:   check4(a*8550954388695124.0, GMP_RNDD, "1.60c012a92fc64@13");
                    428:   check4(a*7842344481681754.0, GMP_RNDD, "1.51d17526c716@13");
                    429:   check4(a*5935035262218600.0, GMP_RNDD, "1.25e19302f7e5@13");
                    430:   check4(a*5039650445085418.0, GMP_RNDD, "1.0ecea7dd2ec3c@13");
                    431:   check4(a*5039721545366078.0, GMP_RNDD, "1.0ecf250e8e92@13");
                    432:   check4(a*8005963117781324.0, GMP_RNDD, "1.5552f3eedcf32@13");
                    433:   check4(a*6703494707970582.0, GMP_RNDD, "1.3853ee10c9c98@13");
                    434:   check4(a*8010323124937260.0, GMP_RNDD, "1.556abe212b56e@13");
                    435:   check4(a*2.0*8010776873384260.0, GMP_RNDD, "1.e2d9a51977e6d@13");
                    436:
                    437:   return 0;
                    438: }

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>