libgfortran/ChangeLog

   1 2011-03-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
   2
   3         PR libgfortran/48030
   4         * io/read.c (read_x): Re-implement using fbuf_getc.
   5
   6 2011-03-21  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
   7
   8         PR bootstrap/48135
   9         * configure.ac (gfortran_use_symver): Handle --disable-symvers.
  10         * configure: Regenerate.
  11
  12 2011-03-19  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
  13
  14         PR fortran/35667
  15         * io/io.h (convert_infnan): Declare.
  16         * io/read.c (convert_infnan): New.
  17         (read_f): Use convert_infnan to convert INFs and NANs.
  18         * list_read.c (parse_real, read_real): Likewise.
  19
  20 2011-03-19  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
  21
  22         PR libfortran/47439
  23         * io/unix.c (tempfile): Work around poor mktemp() implementations.
  24
  25 2011-03-16  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
  26
  27         PR libfortran/47883
  28         * acinclude.m4: Don't use AC_TRY_ macros.
  29         * configure.ac: Likewise.
  30         * configure: Regenerate.
  31
  32 2011-03-12  Thomas Koenig  <tkoenig@gcc.gnu.org>
  33
  34         PR libfortran/48066
  35         * m4/ifunction.m4:  If return array is empty, return.
  36         * m4/ifunction_logical.m4:  Likewise.
  37         * generated/all_l16.c: Regenerated.
  38         * generated/all_l1.c: Regenerated.
  39         * generated/all_l2.c: Regenerated.
  40         * generated/all_l4.c: Regenerated.
  41         * generated/all_l8.c: Regenerated.
  42         * generated/any_l16.c: Regenerated.
  43         * generated/any_l1.c: Regenerated.
  44         * generated/any_l2.c: Regenerated.
  45         * generated/any_l4.c: Regenerated.
  46         * generated/any_l8.c: Regenerated.
  47         * generated/count_16_l.c: Regenerated.
  48         * generated/count_1_l.c: Regenerated.
  49         * generated/count_2_l.c: Regenerated.
  50         * generated/count_4_l.c: Regenerated.
  51         * generated/count_8_l.c: Regenerated.
  52         * generated/maxloc1_16_i16.c: Regenerated.
  53         * generated/maxloc1_16_i1.c: Regenerated.
  54         * generated/maxloc1_16_i2.c: Regenerated.
  55         * generated/maxloc1_16_i4.c: Regenerated.
  56         * generated/maxloc1_16_i8.c: Regenerated.
  57         * generated/maxloc1_16_r10.c: Regenerated.
  58         * generated/maxloc1_16_r16.c: Regenerated.
  59         * generated/maxloc1_16_r4.c: Regenerated.
  60         * generated/maxloc1_16_r8.c: Regenerated.
  61         * generated/maxloc1_4_i16.c: Regenerated.
  62         * generated/maxloc1_4_i1.c: Regenerated.
  63         * generated/maxloc1_4_i2.c: Regenerated.
  64         * generated/maxloc1_4_i4.c: Regenerated.
  65         * generated/maxloc1_4_i8.c: Regenerated.
  66         * generated/maxloc1_4_r10.c: Regenerated.
  67         * generated/maxloc1_4_r16.c: Regenerated.
  68         * generated/maxloc1_4_r4.c: Regenerated.
  69         * generated/maxloc1_4_r8.c: Regenerated.
  70         * generated/maxloc1_8_i16.c: Regenerated.
  71         * generated/maxloc1_8_i1.c: Regenerated.
  72         * generated/maxloc1_8_i2.c: Regenerated.
  73         * generated/maxloc1_8_i4.c: Regenerated.
  74         * generated/maxloc1_8_i8.c: Regenerated.
  75         * generated/maxloc1_8_r10.c: Regenerated.
  76         * generated/maxloc1_8_r16.c: Regenerated.
  77         * generated/maxloc1_8_r4.c: Regenerated.
  78         * generated/maxloc1_8_r8.c: Regenerated.
  79         * generated/maxval_i16.c: Regenerated.
  80         * generated/maxval_i1.c: Regenerated.
  81         * generated/maxval_i2.c: Regenerated.
  82         * generated/maxval_i4.c: Regenerated.
  83         * generated/maxval_i8.c: Regenerated.
  84         * generated/maxval_r10.c: Regenerated.
  85         * generated/maxval_r16.c: Regenerated.
  86         * generated/maxval_r4.c: Regenerated.
  87         * generated/maxval_r8.c: Regenerated.
  88         * generated/minloc1_16_i16.c: Regenerated.
  89         * generated/minloc1_16_i1.c: Regenerated.
  90         * generated/minloc1_16_i2.c: Regenerated.
  91         * generated/minloc1_16_i4.c: Regenerated.
  92         * generated/minloc1_16_i8.c: Regenerated.
  93         * generated/minloc1_16_r10.c: Regenerated.
  94         * generated/minloc1_16_r16.c: Regenerated.
  95         * generated/minloc1_16_r4.c: Regenerated.
  96         * generated/minloc1_16_r8.c: Regenerated.
  97         * generated/minloc1_4_i16.c: Regenerated.
  98         * generated/minloc1_4_i1.c: Regenerated.
  99         * generated/minloc1_4_i2.c: Regenerated.
 100         * generated/minloc1_4_i4.c: Regenerated.
 101         * generated/minloc1_4_i8.c: Regenerated.
 102         * generated/minloc1_4_r10.c: Regenerated.
 103         * generated/minloc1_4_r16.c: Regenerated.
 104         * generated/minloc1_4_r4.c: Regenerated.
 105         * generated/minloc1_4_r8.c: Regenerated.
 106         * generated/minloc1_8_i16.c: Regenerated.
 107         * generated/minloc1_8_i1.c: Regenerated.
 108         * generated/minloc1_8_i2.c: Regenerated.
 109         * generated/minloc1_8_i4.c: Regenerated.
 110         * generated/minloc1_8_i8.c: Regenerated.
 111         * generated/minloc1_8_r10.c: Regenerated.
 112         * generated/minloc1_8_r16.c: Regenerated.
 113         * generated/minloc1_8_r4.c: Regenerated.
 114         * generated/minloc1_8_r8.c: Regenerated.
 115         * generated/minval_i16.c: Regenerated.
 116         * generated/minval_i1.c: Regenerated.
 117         * generated/minval_i2.c: Regenerated.
 118         * generated/minval_i4.c: Regenerated.
 119         * generated/minval_i8.c: Regenerated.
 120         * generated/minval_r10.c: Regenerated.
 121         * generated/minval_r16.c: Regenerated.
 122         * generated/minval_r4.c: Regenerated.
 123         * generated/minval_r8.c: Regenerated.
 124         * generated/product_c10.c: Regenerated.
 125         * generated/product_c16.c: Regenerated.
 126         * generated/product_c4.c: Regenerated.
 127         * generated/product_c8.c: Regenerated.
 128         * generated/product_i16.c: Regenerated.
 129         * generated/product_i1.c: Regenerated.
 130         * generated/product_i2.c: Regenerated.
 131         * generated/product_i4.c: Regenerated.
 132         * generated/product_i8.c: Regenerated.
 133         * generated/product_r10.c: Regenerated.
 134         * generated/product_r16.c: Regenerated.
 135         * generated/product_r4.c: Regenerated.
 136         * generated/product_r8.c: Regenerated.
 137         * generated/sum_c10.c: Regenerated.
 138         * generated/sum_c16.c: Regenerated.
 139         * generated/sum_c4.c: Regenerated.
 140         * generated/sum_c8.c: Regenerated.
 141         * generated/sum_i16.c: Regenerated.
 142         * generated/sum_i1.c: Regenerated.
 143         * generated/sum_i2.c: Regenerated.
 144         * generated/sum_i4.c: Regenerated.
 145         * generated/sum_i8.c: Regenerated.
 146         * generated/sum_r10.c: Regenerated.
 147         * generated/sum_r16.c: Regenerated.
 148         * generated/sum_r4.c: Regenerated.
 149         * generated/sum_r8.c: Regenerated.
 150
 151 2011-03-11  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 152
 153         PR libgfortran/48047
 154         * io/write_float.def (write_float): Change MIN_FIELD_WIDTH to 48.
 155
 156 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 157
 158         PR libfortran/47802
 159         * intrinsics/ctime.c (strctime): Use builtins to check localtime_r
 160         return type.
 161
 162 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 163
 164         PR libfortran/47802
 165         * intrinsics/ctime.c (strctime): Don't use return value of
 166         localtime_r.
 167
 168 2011-02-28  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 169
 170         PR libgfortran/47567
 171         * io/write_float.def (output_float): Move handling of w = 0 to after
 172         output rounding. Check for zero and set zero_flag accordingly. Set
 173         width according to zero_flag. Add better comments.
 174
 175 2011-02-27  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 176
 177         PR libgfortran/47778
 178         * io/list_read.c (namelist_read): Intialize the error string buffere.
 179         If pprev_nl was used during the previous namelist read and the rank
 180         was zero, reset the pointer to NULL for the next namelist read.
 181
 182 2011-02-26  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 183
 184         PR libfortran/45165
 185         * unix.c (fallback_access): Fix file descriptor leaks.
 186
 187 2011-02-25  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 188
 189         * acinclude.m4 (LIBGFOR_CHECK_FPSETMASK): Set shell variable
 190         tested in configure.host.
 191         * configure: Regenerate.
 192
 193 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 194
 195         PR libfortran/47802
 196         * config.h.in: Regenerated.
 197         * configure: Regenerated.
 198         * configure.ac: Remove checks for ctime and ctime_r, add check for
 199         strftime.
 200         * intrinsics/date_and_time.c (localtime_r): Move fallback
 201         implementation to time_1.h.
 202         * intrinsics/time_1.h (localtime_r): Fallback implementation.
 203         * intrinsics/ctime.c: Include time_1.h.
 204         (ctime_r): Remove fallback implementation.
 205         (strctime): New function.
 206         (fdate): Use strctime instead of ctime_r.
 207         (fdate_sub): Likewise.
 208         (ctime): Likewise.
 209         (ctime_sub): Likewise.
 210
 211 2011-02-24  Jakub Jelinek  <jakub@redhat.com>
 212
 213         PR fortran/47878
 214         * io/transfer.c (read_sf): Call fbuf_getptr only at the end,
 215         and subtract n, dtp->u.p.sf_seen_eor and seen_comma from it.
 216
 217 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 218
 219         PR libfortran/47802
 220         * configure.ac: Add test for POSIX getpwuid_r.
 221         * intrinsics/getlog.c (getlog): CPP test for
 222         HAVE_POSIX_GETPWUID_R.
 223         * config.h.in: Regenerated.
 224         * configure: Regenerated.
 225
 226 2011-02-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 227
 228         PR libgfortran/47567
 229         * io/write_float.def (output_float): Remove special case handling of
 230         zero with width 1.
 231
 232 2011-02-23  Janne Blomqvist  <jb@gcc.gnu.org>
 233             Jerry DeLisle    <jvdelisle@gcc.gnu.org>
 234
 235         PR libfortran/47694
 236         * io/fbuf.h (fbuf_getptr): New inline function.
 237         * io/transfer.c (read_sf): Use fbuf_getptr and fbuf_getc to scan
 238         through the string instead of fbuf_read.
 239
 240 2011-02-22  Tobias Burnus  <burnus@net-b.de>
 241             Kai-Uwe Eckhardt  <kuehro@gmx.de>
 242
 243         PR libfortran/47830
 244         * intrinsics/c99_functions.c (roundl): Make C valid for
 245         HAVE_NEXTAFTERL.
 246
 247 2011-02-19  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 248
 249         PR libgfortran/47567
 250         * io/write_float.def (output_float): Adjust width for F0.d to
 251         allow space for negative signs on zero.
 252
 253 2011-02-16  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 254
 255         PR libgfortran/47667
 256         * io/list_read.c (read_logical): Check for end of line before calling
 257         eat_line. (read_integer): Likewise. (parse_real): Don't unget the
 258         separator. Check for end of line before calling eat_line.
 259         (read_complex): Allow line-end before and after parenthesis and comma.
 260         Check for end of line before calling eat_line. (read_real): Check for
 261         end of line before calling eat_line.
 262
 263 2011-02-16  Jakub Jelinek  <jakub@redhat.com>
 264
 265         PR libfortran/47757
 266         * gfortran.map (GFORTRAN_1.4): Export
 267         _gfortran_{m,s}i{all,any,parity}_i{1,2,4,8,16} and
 268         _gfortran_{cshift0,eoshift{0,2}}_16_char4.
 269
 270 2011-02-15  Tobias Burnus  <burnus@net-b.de>
 271
 272         PR fortran/47716
 273         PR fortran/47648
 274         * acinclude.m4 (LIBGFOR_CHECK_FLOAT128): Use check from
 275         libquadmath, which uses more features.
 276         * configure: Regenerate.
 277
 278 2011-02-14  Jakub Jelinek  <jakub@redhat.com>
 279
 280         PR fortran/47642
 281         * io/write_float.def (DTOAQ): Use quadmath_snprintf instead of
 282         quadmath_flt128tostr.
 283         * io/transfer128.c (tmp2): Initialize to quadmath_snprintf instead
 284         of quadmath_flt128tostr.
 285
 286 2011-02-13  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 287
 288         * Makefile.in: Regenerate.
 289         * aclocal.m4: Likewise.
 290         * configure: Likewise.
 291
 292 2011-02-05  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 293
 294         PR libgfortran/47567
 295         * io/write_float.def (output_float): Eliminate some redundant code.
 296         Adjust width for case of F0.X for values of zero and all other values.
 297         Expand cases where '*' is set to give cleaner results.
 298
 299 2011-02-05  Janne Blomqvist  <jb@gcc.gnu.org>
 300
 301         PR libfortran/47571
 302         * intrinsics/time_1.h (GF_CLOCK_MONOTONIC): Move to system_clock.c.
 303         (weak_gettime): Likewise.
 304         (gf_gettime): Change API, move weak_gettime() usage to
 305         system_clock.c
 306         * intrinsics/system_clock.c (GTHREAD_USE_WEAK): Define.
 307         (gf_gettime_mono): New function.
 308         (system_clock_4): Use gf_gettime_mono().
 309         (system_clock_8): Likewise.
 310         * intrinsics/date_and_time.c (date_and_time): Update gf_gettime()
 311         usage.
 312
 313 2011-02-02  Janne Blomqvist  <jb@gcc.gnu.org>
 314
 315         PR libfortran/47571
 316         * configure: Regenerated.
 317         * configure.ac: Don't add librt to LIBS.
 318         * intrinsics/time_1.h (weak_gettime): Weakref trickery for
 319         clock_gettime().
 320         (gf_gettime): Use weak_gettime() instead of clock_gettime().
 321
 322 2011-02-01  Janne Blomqvist  <jb@gcc.gnu.org>
 323
 324         * intrinsics/time_1.h: Include errno.h needed by fallbacks.
 325
 326 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 327
 328         * intrinsics/time_1.h: Fix definition of GF_CLOCK_MONOTONIC macro.
 329
 330 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 331
 332         * configure.ac: Check for clock_gettime().
 333         * configure: Regenerated.
 334         * config.h.in: Regenerated.
 335         * intrinsics/time_1.h (__time_1): Rename to gf_cputime, add
 336         times() fallback.
 337         (gf_gettime): New function.
 338         * intrinsics/cpu_time.c (__cpu_time_1): Update to call gf_cputime.
 339         * intrinsics/date_and_time.c (date_and_time): Use gf_gettime.
 340         * intrinsics/dtime.c (dtime_sub): Use gf_cputime.
 341         * intrinsics/etime.c (etime_sub): Use gf_cputime.
 342         * intrinsics/system_clock.c (system_clock_4): Use gf_gettime.
 343         (system_clock_8): Use gf_gettime, increase count rate to allow
 344         nanosecond precision, remove overflow prone branch.
 345
 346 2011-01-29  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 347
 348         PR libgfortran/47434
 349         * io/write_float.def (write_infnan): Use calculate_sign to determine
 350         if the sign should be given and check field widths accordingly.
 351
 352 2011-01-29  Kai Tietz  <kai.tietz@onevision.com>
 353
 354         * intrinsics/ctime.c (ctime_r): Improve implementation.
 355
 356 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 357
 358         PR libfortran/47431
 359         * config.h.in: Regenerated.
 360         * configure: Regenerated.
 361         * configure.ac: Add check for ctime_r().
 362         * intrinsics/ctime.c (ctime_r): Fallback implementation.
 363         (fdate): Use ctime_r() instead of ctime().
 364         (fdate_sub): Likewise.
 365         (ctime): Likewise.
 366         (ctime_sub): Likewise.
 367
 368 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 369
 370         PR libfortran/47432
 371         * config.h.in: Regenerated.
 372         * configure: Regenerated.
 373         * configure.ac: Add check for ttyname_r.
 374         * io/unix.h: Add TTY_NAME_MAX, change stream_ttyname prototype.
 375         * io/unix.c (stream_ttyname): Use ttyname_r if available, conform
 376         to new prototype.
 377         * io/inquire.c (inquire_via_unit): Use changed stream_ttyname.
 378         * io/intrinsics.c (ttynam_sub): Likewise.
 379         (ttynam): Likewise.
 380
 381 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 382
 383         PR libfortran/47491
 384         * configure.ac: Call AC_USE_SYSTEM_EXTENSIONS to enable common
 385         extensions.
 386         * config.h.in: Regenerate.
 387         * configure: Regenerate.
 388
 389 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 390
 391         PR libgfortran/47285
 392         * io/write_float.def (write_infnan): Adjust processor selected width
 393         to 3 if NaN.
 394
 395 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 396
 397         PR libgfortran/47285
 398         * io/write_float.def (output_float): Return SUCCESS or FAILURE and use
 399         the result to set the padding.
 400
 401 2011-01-26  Kai Tietz  <kai.tietz@onevision.com>
 402
 403         * intrinsics/getlog.c (getlog): Fix label/statement issue.
 404
 405 2011-01-25  Janne Blomqvist  <jb@gcc.gnu.org>
 406
 407         PR libfortran/47375
 408         * config.h.in: Regenerated.
 409         * configure: Regenerated.
 410         * configure.ac: Add check for getpwuid_r.
 411         * intrinsics/getlog.c (getlog): Use getpwuid_r() if available.
 412
 413 2011-01-22  Janne Blomqvist  <jb@gcc.gnu.org>
 414
 415         PR libfortran/46267
 416         * config.h.in: Regenerated.
 417         * configure: Regenerated.
 418         * configure.ac: Check presence of strerror_r.
 419         * intrinsics/gerror.c (gerror): Use gf_strerror, modify logic.
 420         * io/unix.c (get_oserror): Remove.
 421         * libgfortran.h (gf_strerror): Add prototype.
 422         (get_oserror): Remove prototype.
 423         * runtime/error.c (gf_strerror): New function.
 424         (os_error): Use gf_strerror instead of get_oserror.
 425         (generate_errror): Likewise.
 426
 427 2011-01-17  Janne Blomqvist  <jb@gcc.gnu.org>
 428
 429         PR libfortran/47296
 430         * io/unix.c (tempfile): Set opp->file and opp->file_len also if an
 431         error occurs.
 432
 433 2011-01-16  Jakub Jelinek  <jakub@redhat.com>
 434
 435         PR fortran/46625
 436         * io/write_float.def (DTOAQ): Use quadmath_flt128tostr
 437         instead of quadmath_dtoa.
 438         * io/transfer128.c (tmp1, tmp2): New variables, bring in
 439         strtoflt128 and quadmath_flt128tostr.
 440         (transfer_real128, transfer_real128_write, transfer_complex128,
 441         transfer_complex128_write): Remove tmp1/tmp2 variables.
 442         * io/read.c (convert_real): Use strtoflt128 instead of
 443         quadmath_strtopQ, adjust for the changed arguments and return
 444         value.
 445
 446 2011-01-14  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 447
 448         PR libgfortran/47296
 449         * io/unix.c (unpack_filename): Return non-zero if the filename passed
 450         in is NULL.
 451
 452 2011-01-04  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 453
 454         PR libgfortran/47154
 455         * io/list_read.c (namelist_read): Remove calls to hit_eof to avoid the
 456         duplicate calls via next_record.
 457
 458 \f
 459 Copyright (C) 2011 Free Software Foundation, Inc.
 460
 461 Copying and distribution of this file, with or without modification,
 462 are permitted in any medium without royalty provided the copyright
 463 notice and this notice are preserved.