libgfortran/ChangeLog

   1 2011-03-19  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
   2
   3         PR fortran/35667
   4         * io/io.h (convert_infnan): Declare.
   5         * io/read.c (convert_infnan): New.
   6         (read_f): Use convert_infnan to convert INFs and NANs.
   7         * list_read.c (parse_real, read_real): Likewise.
   8
   9 2011-03-19  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
  10
  11         PR libfortran/47439
  12         * io/unix.c (tempfile): Work around poor mktemp() implementations.
  13
  14 2011-03-16  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
  15
  16         PR libfortran/47883
  17         * acinclude.m4: Don't use AC_TRY_ macros.
  18         * configure.ac: Likewise.
  19         * configure: Regenerate.
  20
  21 2011-03-12  Thomas Koenig  <tkoenig@gcc.gnu.org>
  22
  23         PR libfortran/48066
  24         * m4/ifunction.m4:  If return array is empty, return.
  25         * m4/ifunction_logical.m4:  Likewise.
  26         * generated/all_l16.c: Regenerated.
  27         * generated/all_l1.c: Regenerated.
  28         * generated/all_l2.c: Regenerated.
  29         * generated/all_l4.c: Regenerated.
  30         * generated/all_l8.c: Regenerated.
  31         * generated/any_l16.c: Regenerated.
  32         * generated/any_l1.c: Regenerated.
  33         * generated/any_l2.c: Regenerated.
  34         * generated/any_l4.c: Regenerated.
  35         * generated/any_l8.c: Regenerated.
  36         * generated/count_16_l.c: Regenerated.
  37         * generated/count_1_l.c: Regenerated.
  38         * generated/count_2_l.c: Regenerated.
  39         * generated/count_4_l.c: Regenerated.
  40         * generated/count_8_l.c: Regenerated.
  41         * generated/maxloc1_16_i16.c: Regenerated.
  42         * generated/maxloc1_16_i1.c: Regenerated.
  43         * generated/maxloc1_16_i2.c: Regenerated.
  44         * generated/maxloc1_16_i4.c: Regenerated.
  45         * generated/maxloc1_16_i8.c: Regenerated.
  46         * generated/maxloc1_16_r10.c: Regenerated.
  47         * generated/maxloc1_16_r16.c: Regenerated.
  48         * generated/maxloc1_16_r4.c: Regenerated.
  49         * generated/maxloc1_16_r8.c: Regenerated.
  50         * generated/maxloc1_4_i16.c: Regenerated.
  51         * generated/maxloc1_4_i1.c: Regenerated.
  52         * generated/maxloc1_4_i2.c: Regenerated.
  53         * generated/maxloc1_4_i4.c: Regenerated.
  54         * generated/maxloc1_4_i8.c: Regenerated.
  55         * generated/maxloc1_4_r10.c: Regenerated.
  56         * generated/maxloc1_4_r16.c: Regenerated.
  57         * generated/maxloc1_4_r4.c: Regenerated.
  58         * generated/maxloc1_4_r8.c: Regenerated.
  59         * generated/maxloc1_8_i16.c: Regenerated.
  60         * generated/maxloc1_8_i1.c: Regenerated.
  61         * generated/maxloc1_8_i2.c: Regenerated.
  62         * generated/maxloc1_8_i4.c: Regenerated.
  63         * generated/maxloc1_8_i8.c: Regenerated.
  64         * generated/maxloc1_8_r10.c: Regenerated.
  65         * generated/maxloc1_8_r16.c: Regenerated.
  66         * generated/maxloc1_8_r4.c: Regenerated.
  67         * generated/maxloc1_8_r8.c: Regenerated.
  68         * generated/maxval_i16.c: Regenerated.
  69         * generated/maxval_i1.c: Regenerated.
  70         * generated/maxval_i2.c: Regenerated.
  71         * generated/maxval_i4.c: Regenerated.
  72         * generated/maxval_i8.c: Regenerated.
  73         * generated/maxval_r10.c: Regenerated.
  74         * generated/maxval_r16.c: Regenerated.
  75         * generated/maxval_r4.c: Regenerated.
  76         * generated/maxval_r8.c: Regenerated.
  77         * generated/minloc1_16_i16.c: Regenerated.
  78         * generated/minloc1_16_i1.c: Regenerated.
  79         * generated/minloc1_16_i2.c: Regenerated.
  80         * generated/minloc1_16_i4.c: Regenerated.
  81         * generated/minloc1_16_i8.c: Regenerated.
  82         * generated/minloc1_16_r10.c: Regenerated.
  83         * generated/minloc1_16_r16.c: Regenerated.
  84         * generated/minloc1_16_r4.c: Regenerated.
  85         * generated/minloc1_16_r8.c: Regenerated.
  86         * generated/minloc1_4_i16.c: Regenerated.
  87         * generated/minloc1_4_i1.c: Regenerated.
  88         * generated/minloc1_4_i2.c: Regenerated.
  89         * generated/minloc1_4_i4.c: Regenerated.
  90         * generated/minloc1_4_i8.c: Regenerated.
  91         * generated/minloc1_4_r10.c: Regenerated.
  92         * generated/minloc1_4_r16.c: Regenerated.
  93         * generated/minloc1_4_r4.c: Regenerated.
  94         * generated/minloc1_4_r8.c: Regenerated.
  95         * generated/minloc1_8_i16.c: Regenerated.
  96         * generated/minloc1_8_i1.c: Regenerated.
  97         * generated/minloc1_8_i2.c: Regenerated.
  98         * generated/minloc1_8_i4.c: Regenerated.
  99         * generated/minloc1_8_i8.c: Regenerated.
 100         * generated/minloc1_8_r10.c: Regenerated.
 101         * generated/minloc1_8_r16.c: Regenerated.
 102         * generated/minloc1_8_r4.c: Regenerated.
 103         * generated/minloc1_8_r8.c: Regenerated.
 104         * generated/minval_i16.c: Regenerated.
 105         * generated/minval_i1.c: Regenerated.
 106         * generated/minval_i2.c: Regenerated.
 107         * generated/minval_i4.c: Regenerated.
 108         * generated/minval_i8.c: Regenerated.
 109         * generated/minval_r10.c: Regenerated.
 110         * generated/minval_r16.c: Regenerated.
 111         * generated/minval_r4.c: Regenerated.
 112         * generated/minval_r8.c: Regenerated.
 113         * generated/product_c10.c: Regenerated.
 114         * generated/product_c16.c: Regenerated.
 115         * generated/product_c4.c: Regenerated.
 116         * generated/product_c8.c: Regenerated.
 117         * generated/product_i16.c: Regenerated.
 118         * generated/product_i1.c: Regenerated.
 119         * generated/product_i2.c: Regenerated.
 120         * generated/product_i4.c: Regenerated.
 121         * generated/product_i8.c: Regenerated.
 122         * generated/product_r10.c: Regenerated.
 123         * generated/product_r16.c: Regenerated.
 124         * generated/product_r4.c: Regenerated.
 125         * generated/product_r8.c: Regenerated.
 126         * generated/sum_c10.c: Regenerated.
 127         * generated/sum_c16.c: Regenerated.
 128         * generated/sum_c4.c: Regenerated.
 129         * generated/sum_c8.c: Regenerated.
 130         * generated/sum_i16.c: Regenerated.
 131         * generated/sum_i1.c: Regenerated.
 132         * generated/sum_i2.c: Regenerated.
 133         * generated/sum_i4.c: Regenerated.
 134         * generated/sum_i8.c: Regenerated.
 135         * generated/sum_r10.c: Regenerated.
 136         * generated/sum_r16.c: Regenerated.
 137         * generated/sum_r4.c: Regenerated.
 138         * generated/sum_r8.c: Regenerated.
 139
 140 2011-03-11  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 141
 142         PR libgfortran/48047
 143         * io/write_float.def (write_float): Change MIN_FIELD_WIDTH to 48.
 144
 145 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 146
 147         PR libfortran/47802
 148         * intrinsics/ctime.c (strctime): Use builtins to check localtime_r
 149         return type.
 150
 151 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 152
 153         PR libfortran/47802
 154         * intrinsics/ctime.c (strctime): Don't use return value of
 155         localtime_r.
 156
 157 2011-02-28  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 158
 159         PR libgfortran/47567
 160         * io/write_float.def (output_float): Move handling of w = 0 to after
 161         output rounding. Check for zero and set zero_flag accordingly. Set
 162         width according to zero_flag. Add better comments.
 163
 164 2011-02-27  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 165
 166         PR libgfortran/47778
 167         * io/list_read.c (namelist_read): Intialize the error string buffere.
 168         If pprev_nl was used during the previous namelist read and the rank
 169         was zero, reset the pointer to NULL for the next namelist read.
 170
 171 2011-02-26  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 172
 173         PR libfortran/45165
 174         * unix.c (fallback_access): Fix file descriptor leaks.
 175
 176 2011-02-25  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 177
 178         * acinclude.m4 (LIBGFOR_CHECK_FPSETMASK): Set shell variable
 179         tested in configure.host.
 180         * configure: Regenerate.
 181
 182 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 183
 184         PR libfortran/47802
 185         * config.h.in: Regenerated.
 186         * configure: Regenerated.
 187         * configure.ac: Remove checks for ctime and ctime_r, add check for
 188         strftime.
 189         * intrinsics/date_and_time.c (localtime_r): Move fallback
 190         implementation to time_1.h.
 191         * intrinsics/time_1.h (localtime_r): Fallback implementation.
 192         * intrinsics/ctime.c: Include time_1.h.
 193         (ctime_r): Remove fallback implementation.
 194         (strctime): New function.
 195         (fdate): Use strctime instead of ctime_r.
 196         (fdate_sub): Likewise.
 197         (ctime): Likewise.
 198         (ctime_sub): Likewise.
 199
 200 2011-02-24  Jakub Jelinek  <jakub@redhat.com>
 201
 202         PR fortran/47878
 203         * io/transfer.c (read_sf): Call fbuf_getptr only at the end,
 204         and subtract n, dtp->u.p.sf_seen_eor and seen_comma from it.
 205
 206 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 207
 208         PR libfortran/47802
 209         * configure.ac: Add test for POSIX getpwuid_r.
 210         * intrinsics/getlog.c (getlog): CPP test for
 211         HAVE_POSIX_GETPWUID_R.
 212         * config.h.in: Regenerated.
 213         * configure: Regenerated.
 214
 215 2011-02-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 216
 217         PR libgfortran/47567
 218         * io/write_float.def (output_float): Remove special case handling of
 219         zero with width 1.
 220
 221 2011-02-23  Janne Blomqvist  <jb@gcc.gnu.org>
 222             Jerry DeLisle    <jvdelisle@gcc.gnu.org>
 223
 224         PR libfortran/47694
 225         * io/fbuf.h (fbuf_getptr): New inline function.
 226         * io/transfer.c (read_sf): Use fbuf_getptr and fbuf_getc to scan
 227         through the string instead of fbuf_read.
 228
 229 2011-02-22  Tobias Burnus  <burnus@net-b.de>
 230             Kai-Uwe Eckhardt  <kuehro@gmx.de>
 231
 232         PR libfortran/47830
 233         * intrinsics/c99_functions.c (roundl): Make C valid for
 234         HAVE_NEXTAFTERL.
 235
 236 2011-02-19  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 237
 238         PR libgfortran/47567
 239         * io/write_float.def (output_float): Adjust width for F0.d to
 240         allow space for negative signs on zero.
 241
 242 2011-02-16  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 243
 244         PR libgfortran/47667
 245         * io/list_read.c (read_logical): Check for end of line before calling
 246         eat_line. (read_integer): Likewise. (parse_real): Don't unget the
 247         separator. Check for end of line before calling eat_line.
 248         (read_complex): Allow line-end before and after parenthesis and comma.
 249         Check for end of line before calling eat_line. (read_real): Check for
 250         end of line before calling eat_line.
 251
 252 2011-02-16  Jakub Jelinek  <jakub@redhat.com>
 253
 254         PR libfortran/47757
 255         * gfortran.map (GFORTRAN_1.4): Export
 256         _gfortran_{m,s}i{all,any,parity}_i{1,2,4,8,16} and
 257         _gfortran_{cshift0,eoshift{0,2}}_16_char4.
 258
 259 2011-02-15  Tobias Burnus  <burnus@net-b.de>
 260
 261         PR fortran/47716
 262         PR fortran/47648
 263         * acinclude.m4 (LIBGFOR_CHECK_FLOAT128): Use check from
 264         libquadmath, which uses more features.
 265         * configure: Regenerate.
 266
 267 2011-02-14  Jakub Jelinek  <jakub@redhat.com>
 268
 269         PR fortran/47642
 270         * io/write_float.def (DTOAQ): Use quadmath_snprintf instead of
 271         quadmath_flt128tostr.
 272         * io/transfer128.c (tmp2): Initialize to quadmath_snprintf instead
 273         of quadmath_flt128tostr.
 274
 275 2011-02-13  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 276
 277         * Makefile.in: Regenerate.
 278         * aclocal.m4: Likewise.
 279         * configure: Likewise.
 280
 281 2011-02-05  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 282
 283         PR libgfortran/47567
 284         * io/write_float.def (output_float): Eliminate some redundant code.
 285         Adjust width for case of F0.X for values of zero and all other values.
 286         Expand cases where '*' is set to give cleaner results.
 287
 288 2011-02-05  Janne Blomqvist  <jb@gcc.gnu.org>
 289
 290         PR libfortran/47571
 291         * intrinsics/time_1.h (GF_CLOCK_MONOTONIC): Move to system_clock.c.
 292         (weak_gettime): Likewise.
 293         (gf_gettime): Change API, move weak_gettime() usage to
 294         system_clock.c
 295         * intrinsics/system_clock.c (GTHREAD_USE_WEAK): Define.
 296         (gf_gettime_mono): New function.
 297         (system_clock_4): Use gf_gettime_mono().
 298         (system_clock_8): Likewise.
 299         * intrinsics/date_and_time.c (date_and_time): Update gf_gettime()
 300         usage.
 301
 302 2011-02-02  Janne Blomqvist  <jb@gcc.gnu.org>
 303
 304         PR libfortran/47571
 305         * configure: Regenerated.
 306         * configure.ac: Don't add librt to LIBS.
 307         * intrinsics/time_1.h (weak_gettime): Weakref trickery for
 308         clock_gettime().
 309         (gf_gettime): Use weak_gettime() instead of clock_gettime().
 310
 311 2011-02-01  Janne Blomqvist  <jb@gcc.gnu.org>
 312
 313         * intrinsics/time_1.h: Include errno.h needed by fallbacks.
 314
 315 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 316
 317         * intrinsics/time_1.h: Fix definition of GF_CLOCK_MONOTONIC macro.
 318
 319 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 320
 321         * configure.ac: Check for clock_gettime().
 322         * configure: Regenerated.
 323         * config.h.in: Regenerated.
 324         * intrinsics/time_1.h (__time_1): Rename to gf_cputime, add
 325         times() fallback.
 326         (gf_gettime): New function.
 327         * intrinsics/cpu_time.c (__cpu_time_1): Update to call gf_cputime.
 328         * intrinsics/date_and_time.c (date_and_time): Use gf_gettime.
 329         * intrinsics/dtime.c (dtime_sub): Use gf_cputime.
 330         * intrinsics/etime.c (etime_sub): Use gf_cputime.
 331         * intrinsics/system_clock.c (system_clock_4): Use gf_gettime.
 332         (system_clock_8): Use gf_gettime, increase count rate to allow
 333         nanosecond precision, remove overflow prone branch.
 334
 335 2011-01-29  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 336
 337         PR libgfortran/47434
 338         * io/write_float.def (write_infnan): Use calculate_sign to determine
 339         if the sign should be given and check field widths accordingly.
 340
 341 2011-01-29  Kai Tietz  <kai.tietz@onevision.com>
 342
 343         * intrinsics/ctime.c (ctime_r): Improve implementation.
 344
 345 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 346
 347         PR libfortran/47431
 348         * config.h.in: Regenerated.
 349         * configure: Regenerated.
 350         * configure.ac: Add check for ctime_r().
 351         * intrinsics/ctime.c (ctime_r): Fallback implementation.
 352         (fdate): Use ctime_r() instead of ctime().
 353         (fdate_sub): Likewise.
 354         (ctime): Likewise.
 355         (ctime_sub): Likewise.
 356
 357 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 358
 359         PR libfortran/47432
 360         * config.h.in: Regenerated.
 361         * configure: Regenerated.
 362         * configure.ac: Add check for ttyname_r.
 363         * io/unix.h: Add TTY_NAME_MAX, change stream_ttyname prototype.
 364         * io/unix.c (stream_ttyname): Use ttyname_r if available, conform
 365         to new prototype.
 366         * io/inquire.c (inquire_via_unit): Use changed stream_ttyname.
 367         * io/intrinsics.c (ttynam_sub): Likewise.
 368         (ttynam): Likewise.
 369
 370 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 371
 372         PR libfortran/47491
 373         * configure.ac: Call AC_USE_SYSTEM_EXTENSIONS to enable common
 374         extensions.
 375         * config.h.in: Regenerate.
 376         * configure: Regenerate.
 377
 378 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 379
 380         PR libgfortran/47285
 381         * io/write_float.def (write_infnan): Adjust processor selected width
 382         to 3 if NaN.
 383
 384 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 385
 386         PR libgfortran/47285
 387         * io/write_float.def (output_float): Return SUCCESS or FAILURE and use
 388         the result to set the padding.
 389
 390 2011-01-26  Kai Tietz  <kai.tietz@onevision.com>
 391
 392         * intrinsics/getlog.c (getlog): Fix label/statement issue.
 393
 394 2011-01-25  Janne Blomqvist  <jb@gcc.gnu.org>
 395
 396         PR libfortran/47375
 397         * config.h.in: Regenerated.
 398         * configure: Regenerated.
 399         * configure.ac: Add check for getpwuid_r.
 400         * intrinsics/getlog.c (getlog): Use getpwuid_r() if available.
 401
 402 2011-01-22  Janne Blomqvist  <jb@gcc.gnu.org>
 403
 404         PR libfortran/46267
 405         * config.h.in: Regenerated.
 406         * configure: Regenerated.
 407         * configure.ac: Check presence of strerror_r.
 408         * intrinsics/gerror.c (gerror): Use gf_strerror, modify logic.
 409         * io/unix.c (get_oserror): Remove.
 410         * libgfortran.h (gf_strerror): Add prototype.
 411         (get_oserror): Remove prototype.
 412         * runtime/error.c (gf_strerror): New function.
 413         (os_error): Use gf_strerror instead of get_oserror.
 414         (generate_errror): Likewise.
 415
 416 2011-01-17  Janne Blomqvist  <jb@gcc.gnu.org>
 417
 418         PR libfortran/47296
 419         * io/unix.c (tempfile): Set opp->file and opp->file_len also if an
 420         error occurs.
 421
 422 2011-01-16  Jakub Jelinek  <jakub@redhat.com>
 423
 424         PR fortran/46625
 425         * io/write_float.def (DTOAQ): Use quadmath_flt128tostr
 426         instead of quadmath_dtoa.
 427         * io/transfer128.c (tmp1, tmp2): New variables, bring in
 428         strtoflt128 and quadmath_flt128tostr.
 429         (transfer_real128, transfer_real128_write, transfer_complex128,
 430         transfer_complex128_write): Remove tmp1/tmp2 variables.
 431         * io/read.c (convert_real): Use strtoflt128 instead of
 432         quadmath_strtopQ, adjust for the changed arguments and return
 433         value.
 434
 435 2011-01-14  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 436
 437         PR libgfortran/47296
 438         * io/unix.c (unpack_filename): Return non-zero if the filename passed
 439         in is NULL.
 440
 441 2011-01-04  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 442
 443         PR libgfortran/47154
 444         * io/list_read.c (namelist_read): Remove calls to hit_eof to avoid the
 445         duplicate calls via next_record.
 446
 447 \f
 448 Copyright (C) 2011 Free Software Foundation, Inc.
 449
 450 Copying and distribution of this file, with or without modification,
 451 are permitted in any medium without royalty provided the copyright
 452 notice and this notice are preserved.