gcc-4_6-mobile/libgfortran/ChangeLog

   1 2011-07-13  Janne Blomqvist  <jb@gcc.gnu.org>
   2
   3         Backport from trunk:
   4         PR libfortran/49296
   5         * io/list_read.c (read_logical): Don't error out if a valid value
   6         is followed by EOF instead of a normal separator.
   7         (read_integer): Likewise.
   8
   9 2011-07-06  Thomas Koenig  <tkoenig@gcc.gnu.org>
  10
  11         Partial backport from trunk:
  12         PR fortran/49479
  13         * runtime/memory.c:  If size is zero, allocate a single byte.
  14         * m4/eoshift1.m4:  Remove double allocation.
  15         * m4/eoshift3.m4:  Likewise.
  16         * generated/eoshift1_4.c:  Regenerated.
  17         * generated/eoshift1_8.c:  Regenerated.
  18         * generated/eoshift1_16.c:  Regenerated.
  19         * generated/eoshift3_4.c:  Regenerated.
  20         * generated/eoshift3_8.c:  Regenerated.
  21         * generated/eoshift3_16.c:  Regenerated.
  22
  23 2011-07-03  Janne Blomqvist  <jb@gcc.gnu.org>
  24
  25         Backport from mainline:
  26         PR libfortran/49296
  27         * io/list_read.c (read_character): Accept EOF as a separator when
  28         reading string.
  29
  30 2011-06-27  Release Manager
  31
  32         * GCC 4.6.1 released.
  33
  34 2011-04-30  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
  35
  36         Backport from mainline:
  37         PR libgfortran/48030
  38         * io/read.c (read_x): Re-implement using fbuf_getc.
  39
  40 2011-04-18  Janne Blomqvist  <jb@gcc.gnu.org>
  41
  42         PR libfortran/47571
  43         * configure: Regenerated.
  44         * config.h.in: Regenerated.
  45         * acinclude.m4: Add alpha*-dec-osf* to gthread blacklist.
  46         * configure.ac: Use separate symbol for clock_gettime in librt.
  47         * intrinsics/system_clock.c: Use weakrefs only when needed and
  48         supported.
  49
  50 2011-03-28  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
  51
  52         Backport from mainline:
  53         2011-03-21  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
  54
  55         PR bootstrap/48135
  56         * configure.ac (gfortran_use_symver): Handle --disable-symvers.
  57         * configure: Regenerate.
  58
  59 2011-03-25  Release Manager
  60
  61         * GCC 4.6.0 released.
  62
  63 2011-03-17  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
  64
  65         PR fortran/47571
  66         * intrinsics/system_clock.c [__alpha__ && __osf__]
  67         (HAVE_CLOCK_GETTIME): Undef.
  68
  69 2011-03-12  Thomas Koenig  <tkoenig@gcc.gnu.org>
  70
  71         PR libfortran/48066
  72         * m4/ifunction.m4:  If return array is empty, return.
  73         * m4/ifunction_logical.m4:  Likewise.
  74         * generated/all_l16.c: Regenerated.
  75         * generated/all_l1.c: Regenerated.
  76         * generated/all_l2.c: Regenerated.
  77         * generated/all_l4.c: Regenerated.
  78         * generated/all_l8.c: Regenerated.
  79         * generated/any_l16.c: Regenerated.
  80         * generated/any_l1.c: Regenerated.
  81         * generated/any_l2.c: Regenerated.
  82         * generated/any_l4.c: Regenerated.
  83         * generated/any_l8.c: Regenerated.
  84         * generated/count_16_l.c: Regenerated.
  85         * generated/count_1_l.c: Regenerated.
  86         * generated/count_2_l.c: Regenerated.
  87         * generated/count_4_l.c: Regenerated.
  88         * generated/count_8_l.c: Regenerated.
  89         * generated/maxloc1_16_i16.c: Regenerated.
  90         * generated/maxloc1_16_i1.c: Regenerated.
  91         * generated/maxloc1_16_i2.c: Regenerated.
  92         * generated/maxloc1_16_i4.c: Regenerated.
  93         * generated/maxloc1_16_i8.c: Regenerated.
  94         * generated/maxloc1_16_r10.c: Regenerated.
  95         * generated/maxloc1_16_r16.c: Regenerated.
  96         * generated/maxloc1_16_r4.c: Regenerated.
  97         * generated/maxloc1_16_r8.c: Regenerated.
  98         * generated/maxloc1_4_i16.c: Regenerated.
  99         * generated/maxloc1_4_i1.c: Regenerated.
 100         * generated/maxloc1_4_i2.c: Regenerated.
 101         * generated/maxloc1_4_i4.c: Regenerated.
 102         * generated/maxloc1_4_i8.c: Regenerated.
 103         * generated/maxloc1_4_r10.c: Regenerated.
 104         * generated/maxloc1_4_r16.c: Regenerated.
 105         * generated/maxloc1_4_r4.c: Regenerated.
 106         * generated/maxloc1_4_r8.c: Regenerated.
 107         * generated/maxloc1_8_i16.c: Regenerated.
 108         * generated/maxloc1_8_i1.c: Regenerated.
 109         * generated/maxloc1_8_i2.c: Regenerated.
 110         * generated/maxloc1_8_i4.c: Regenerated.
 111         * generated/maxloc1_8_i8.c: Regenerated.
 112         * generated/maxloc1_8_r10.c: Regenerated.
 113         * generated/maxloc1_8_r16.c: Regenerated.
 114         * generated/maxloc1_8_r4.c: Regenerated.
 115         * generated/maxloc1_8_r8.c: Regenerated.
 116         * generated/maxval_i16.c: Regenerated.
 117         * generated/maxval_i1.c: Regenerated.
 118         * generated/maxval_i2.c: Regenerated.
 119         * generated/maxval_i4.c: Regenerated.
 120         * generated/maxval_i8.c: Regenerated.
 121         * generated/maxval_r10.c: Regenerated.
 122         * generated/maxval_r16.c: Regenerated.
 123         * generated/maxval_r4.c: Regenerated.
 124         * generated/maxval_r8.c: Regenerated.
 125         * generated/minloc1_16_i16.c: Regenerated.
 126         * generated/minloc1_16_i1.c: Regenerated.
 127         * generated/minloc1_16_i2.c: Regenerated.
 128         * generated/minloc1_16_i4.c: Regenerated.
 129         * generated/minloc1_16_i8.c: Regenerated.
 130         * generated/minloc1_16_r10.c: Regenerated.
 131         * generated/minloc1_16_r16.c: Regenerated.
 132         * generated/minloc1_16_r4.c: Regenerated.
 133         * generated/minloc1_16_r8.c: Regenerated.
 134         * generated/minloc1_4_i16.c: Regenerated.
 135         * generated/minloc1_4_i1.c: Regenerated.
 136         * generated/minloc1_4_i2.c: Regenerated.
 137         * generated/minloc1_4_i4.c: Regenerated.
 138         * generated/minloc1_4_i8.c: Regenerated.
 139         * generated/minloc1_4_r10.c: Regenerated.
 140         * generated/minloc1_4_r16.c: Regenerated.
 141         * generated/minloc1_4_r4.c: Regenerated.
 142         * generated/minloc1_4_r8.c: Regenerated.
 143         * generated/minloc1_8_i16.c: Regenerated.
 144         * generated/minloc1_8_i1.c: Regenerated.
 145         * generated/minloc1_8_i2.c: Regenerated.
 146         * generated/minloc1_8_i4.c: Regenerated.
 147         * generated/minloc1_8_i8.c: Regenerated.
 148         * generated/minloc1_8_r10.c: Regenerated.
 149         * generated/minloc1_8_r16.c: Regenerated.
 150         * generated/minloc1_8_r4.c: Regenerated.
 151         * generated/minloc1_8_r8.c: Regenerated.
 152         * generated/minval_i16.c: Regenerated.
 153         * generated/minval_i1.c: Regenerated.
 154         * generated/minval_i2.c: Regenerated.
 155         * generated/minval_i4.c: Regenerated.
 156         * generated/minval_i8.c: Regenerated.
 157         * generated/minval_r10.c: Regenerated.
 158         * generated/minval_r16.c: Regenerated.
 159         * generated/minval_r4.c: Regenerated.
 160         * generated/minval_r8.c: Regenerated.
 161         * generated/product_c10.c: Regenerated.
 162         * generated/product_c16.c: Regenerated.
 163         * generated/product_c4.c: Regenerated.
 164         * generated/product_c8.c: Regenerated.
 165         * generated/product_i16.c: Regenerated.
 166         * generated/product_i1.c: Regenerated.
 167         * generated/product_i2.c: Regenerated.
 168         * generated/product_i4.c: Regenerated.
 169         * generated/product_i8.c: Regenerated.
 170         * generated/product_r10.c: Regenerated.
 171         * generated/product_r16.c: Regenerated.
 172         * generated/product_r4.c: Regenerated.
 173         * generated/product_r8.c: Regenerated.
 174         * generated/sum_c10.c: Regenerated.
 175         * generated/sum_c16.c: Regenerated.
 176         * generated/sum_c4.c: Regenerated.
 177         * generated/sum_c8.c: Regenerated.
 178         * generated/sum_i16.c: Regenerated.
 179         * generated/sum_i1.c: Regenerated.
 180         * generated/sum_i2.c: Regenerated.
 181         * generated/sum_i4.c: Regenerated.
 182         * generated/sum_i8.c: Regenerated.
 183         * generated/sum_r10.c: Regenerated.
 184         * generated/sum_r16.c: Regenerated.
 185         * generated/sum_r4.c: Regenerated.
 186         * generated/sum_r8.c: Regenerated.
 187
 188 2011-03-11  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 189
 190         PR libgfortran/48047
 191         * io/write_float.def (write_float): Change MIN_FIELD_WIDTH to 48.
 192
 193 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 194
 195         PR libfortran/47802
 196         * intrinsics/ctime.c (strctime): Use builtins to check localtime_r
 197         return type.
 198
 199 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 200
 201         PR libfortran/47802
 202         * intrinsics/ctime.c (strctime): Don't use return value of
 203         localtime_r.
 204
 205 2011-02-28  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 206
 207         PR libgfortran/47567
 208         * io/write_float.def (output_float): Move handling of w = 0 to after
 209         output rounding. Check for zero and set zero_flag accordingly. Set
 210         width according to zero_flag. Add better comments.
 211
 212 2011-02-27  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 213
 214         PR libgfortran/47778
 215         * io/list_read.c (namelist_read): Intialize the error string buffere.
 216         If pprev_nl was used during the previous namelist read and the rank
 217         was zero, reset the pointer to NULL for the next namelist read.
 218
 219 2011-02-26  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 220
 221         PR libfortran/45165
 222         * unix.c (fallback_access): Fix file descriptor leaks.
 223
 224 2011-02-25  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 225
 226         * acinclude.m4 (LIBGFOR_CHECK_FPSETMASK): Set shell variable
 227         tested in configure.host.
 228         * configure: Regenerate.
 229
 230 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 231
 232         PR libfortran/47802
 233         * config.h.in: Regenerated.
 234         * configure: Regenerated.
 235         * configure.ac: Remove checks for ctime and ctime_r, add check for
 236         strftime.
 237         * intrinsics/date_and_time.c (localtime_r): Move fallback
 238         implementation to time_1.h.
 239         * intrinsics/time_1.h (localtime_r): Fallback implementation.
 240         * intrinsics/ctime.c: Include time_1.h.
 241         (ctime_r): Remove fallback implementation.
 242         (strctime): New function.
 243         (fdate): Use strctime instead of ctime_r.
 244         (fdate_sub): Likewise.
 245         (ctime): Likewise.
 246         (ctime_sub): Likewise.
 247
 248 2011-02-24  Jakub Jelinek  <jakub@redhat.com>
 249
 250         PR fortran/47878
 251         * io/transfer.c (read_sf): Call fbuf_getptr only at the end,
 252         and subtract n, dtp->u.p.sf_seen_eor and seen_comma from it.
 253
 254 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 255
 256         PR libfortran/47802
 257         * configure.ac: Add test for POSIX getpwuid_r.
 258         * intrinsics/getlog.c (getlog): CPP test for
 259         HAVE_POSIX_GETPWUID_R.
 260         * config.h.in: Regenerated.
 261         * configure: Regenerated.
 262
 263 2011-02-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 264
 265         PR libgfortran/47567
 266         * io/write_float.def (output_float): Remove special case handling of
 267         zero with width 1.
 268
 269 2011-02-23  Janne Blomqvist  <jb@gcc.gnu.org>
 270             Jerry DeLisle    <jvdelisle@gcc.gnu.org>
 271
 272         PR libfortran/47694
 273         * io/fbuf.h (fbuf_getptr): New inline function.
 274         * io/transfer.c (read_sf): Use fbuf_getptr and fbuf_getc to scan
 275         through the string instead of fbuf_read.
 276
 277 2011-02-22  Tobias Burnus  <burnus@net-b.de>
 278             Kai-Uwe Eckhardt  <kuehro@gmx.de>
 279
 280         PR libfortran/47830
 281         * intrinsics/c99_functions.c (roundl): Make C valid for
 282         HAVE_NEXTAFTERL.
 283
 284 2011-02-19  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 285
 286         PR libgfortran/47567
 287         * io/write_float.def (output_float): Adjust width for F0.d to
 288         allow space for negative signs on zero.
 289
 290 2011-02-16  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 291
 292         PR libgfortran/47667
 293         * io/list_read.c (read_logical): Check for end of line before calling
 294         eat_line. (read_integer): Likewise. (parse_real): Don't unget the
 295         separator. Check for end of line before calling eat_line.
 296         (read_complex): Allow line-end before and after parenthesis and comma.
 297         Check for end of line before calling eat_line. (read_real): Check for
 298         end of line before calling eat_line.
 299
 300 2011-02-16  Jakub Jelinek  <jakub@redhat.com>
 301
 302         PR libfortran/47757
 303         * gfortran.map (GFORTRAN_1.4): Export
 304         _gfortran_{m,s}i{all,any,parity}_i{1,2,4,8,16} and
 305         _gfortran_{cshift0,eoshift{0,2}}_16_char4.
 306
 307 2011-02-15  Tobias Burnus  <burnus@net-b.de>
 308
 309         PR fortran/47716
 310         PR fortran/47648
 311         * acinclude.m4 (LIBGFOR_CHECK_FLOAT128): Use check from
 312         libquadmath, which uses more features.
 313         * configure: Regenerate.
 314
 315 2011-02-14  Jakub Jelinek  <jakub@redhat.com>
 316
 317         PR fortran/47642
 318         * io/write_float.def (DTOAQ): Use quadmath_snprintf instead of
 319         quadmath_flt128tostr.
 320         * io/transfer128.c (tmp2): Initialize to quadmath_snprintf instead
 321         of quadmath_flt128tostr.
 322
 323 2011-02-13  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 324
 325         * Makefile.in: Regenerate.
 326         * aclocal.m4: Likewise.
 327         * configure: Likewise.
 328
 329 2011-02-05  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 330
 331         PR libgfortran/47567
 332         * io/write_float.def (output_float): Eliminate some redundant code.
 333         Adjust width for case of F0.X for values of zero and all other values.
 334         Expand cases where '*' is set to give cleaner results.
 335
 336 2011-02-05  Janne Blomqvist  <jb@gcc.gnu.org>
 337
 338         PR libfortran/47571
 339         * intrinsics/time_1.h (GF_CLOCK_MONOTONIC): Move to system_clock.c.
 340         (weak_gettime): Likewise.
 341         (gf_gettime): Change API, move weak_gettime() usage to
 342         system_clock.c
 343         * intrinsics/system_clock.c (GTHREAD_USE_WEAK): Define.
 344         (gf_gettime_mono): New function.
 345         (system_clock_4): Use gf_gettime_mono().
 346         (system_clock_8): Likewise.
 347         * intrinsics/date_and_time.c (date_and_time): Update gf_gettime()
 348         usage.
 349
 350 2011-02-02  Janne Blomqvist  <jb@gcc.gnu.org>
 351
 352         PR libfortran/47571
 353         * configure: Regenerated.
 354         * configure.ac: Don't add librt to LIBS.
 355         * intrinsics/time_1.h (weak_gettime): Weakref trickery for
 356         clock_gettime().
 357         (gf_gettime): Use weak_gettime() instead of clock_gettime().
 358
 359 2011-02-01  Janne Blomqvist  <jb@gcc.gnu.org>
 360
 361         * intrinsics/time_1.h: Include errno.h needed by fallbacks.
 362
 363 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 364
 365         * intrinsics/time_1.h: Fix definition of GF_CLOCK_MONOTONIC macro.
 366
 367 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 368
 369         * configure.ac: Check for clock_gettime().
 370         * configure: Regenerated.
 371         * config.h.in: Regenerated.
 372         * intrinsics/time_1.h (__time_1): Rename to gf_cputime, add
 373         times() fallback.
 374         (gf_gettime): New function.
 375         * intrinsics/cpu_time.c (__cpu_time_1): Update to call gf_cputime.
 376         * intrinsics/date_and_time.c (date_and_time): Use gf_gettime.
 377         * intrinsics/dtime.c (dtime_sub): Use gf_cputime.
 378         * intrinsics/etime.c (etime_sub): Use gf_cputime.
 379         * intrinsics/system_clock.c (system_clock_4): Use gf_gettime.
 380         (system_clock_8): Use gf_gettime, increase count rate to allow
 381         nanosecond precision, remove overflow prone branch.
 382
 383 2011-01-29  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 384
 385         PR libgfortran/47434
 386         * io/write_float.def (write_infnan): Use calculate_sign to determine
 387         if the sign should be given and check field widths accordingly.
 388
 389 2011-01-29  Kai Tietz  <kai.tietz@onevision.com>
 390
 391         * intrinsics/ctime.c (ctime_r): Improve implementation.
 392
 393 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 394
 395         PR libfortran/47431
 396         * config.h.in: Regenerated.
 397         * configure: Regenerated.
 398         * configure.ac: Add check for ctime_r().
 399         * intrinsics/ctime.c (ctime_r): Fallback implementation.
 400         (fdate): Use ctime_r() instead of ctime().
 401         (fdate_sub): Likewise.
 402         (ctime): Likewise.
 403         (ctime_sub): Likewise.
 404
 405 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 406
 407         PR libfortran/47432
 408         * config.h.in: Regenerated.
 409         * configure: Regenerated.
 410         * configure.ac: Add check for ttyname_r.
 411         * io/unix.h: Add TTY_NAME_MAX, change stream_ttyname prototype.
 412         * io/unix.c (stream_ttyname): Use ttyname_r if available, conform
 413         to new prototype.
 414         * io/inquire.c (inquire_via_unit): Use changed stream_ttyname.
 415         * io/intrinsics.c (ttynam_sub): Likewise.
 416         (ttynam): Likewise.
 417
 418 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 419
 420         PR libfortran/47491
 421         * configure.ac: Call AC_USE_SYSTEM_EXTENSIONS to enable common
 422         extensions.
 423         * config.h.in: Regenerate.
 424         * configure: Regenerate.
 425
 426 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 427
 428         PR libgfortran/47285
 429         * io/write_float.def (write_infnan): Adjust processor selected width
 430         to 3 if NaN.
 431
 432 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 433
 434         PR libgfortran/47285
 435         * io/write_float.def (output_float): Return SUCCESS or FAILURE and use
 436         the result to set the padding.
 437
 438 2011-01-26  Kai Tietz  <kai.tietz@onevision.com>
 439
 440         * intrinsics/getlog.c (getlog): Fix label/statement issue.
 441
 442 2011-01-25  Janne Blomqvist  <jb@gcc.gnu.org>
 443
 444         PR libfortran/47375
 445         * config.h.in: Regenerated.
 446         * configure: Regenerated.
 447         * configure.ac: Add check for getpwuid_r.
 448         * intrinsics/getlog.c (getlog): Use getpwuid_r() if available.
 449
 450 2011-01-22  Janne Blomqvist  <jb@gcc.gnu.org>
 451
 452         PR libfortran/46267
 453         * config.h.in: Regenerated.
 454         * configure: Regenerated.
 455         * configure.ac: Check presence of strerror_r.
 456         * intrinsics/gerror.c (gerror): Use gf_strerror, modify logic.
 457         * io/unix.c (get_oserror): Remove.
 458         * libgfortran.h (gf_strerror): Add prototype.
 459         (get_oserror): Remove prototype.
 460         * runtime/error.c (gf_strerror): New function.
 461         (os_error): Use gf_strerror instead of get_oserror.
 462         (generate_errror): Likewise.
 463
 464 2011-01-17  Janne Blomqvist  <jb@gcc.gnu.org>
 465
 466         PR libfortran/47296
 467         * io/unix.c (tempfile): Set opp->file and opp->file_len also if an
 468         error occurs.
 469
 470 2011-01-16  Jakub Jelinek  <jakub@redhat.com>
 471
 472         PR fortran/46625
 473         * io/write_float.def (DTOAQ): Use quadmath_flt128tostr
 474         instead of quadmath_dtoa.
 475         * io/transfer128.c (tmp1, tmp2): New variables, bring in
 476         strtoflt128 and quadmath_flt128tostr.
 477         (transfer_real128, transfer_real128_write, transfer_complex128,
 478         transfer_complex128_write): Remove tmp1/tmp2 variables.
 479         * io/read.c (convert_real): Use strtoflt128 instead of
 480         quadmath_strtopQ, adjust for the changed arguments and return
 481         value.
 482
 483 2011-01-14  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 484
 485         PR libgfortran/47296
 486         * io/unix.c (unpack_filename): Return non-zero if the filename passed
 487         in is NULL.
 488
 489 2011-01-04  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 490
 491         PR libgfortran/47154
 492         * io/list_read.c (namelist_read): Remove calls to hit_eof to avoid the
 493         duplicate calls via next_record.
 494
 495 \f
 496 Copyright (C) 2011 Free Software Foundation, Inc.
 497
 498 Copying and distribution of this file, with or without modification,
 499 are permitted in any medium without royalty provided the copyright
 500 notice and this notice are preserved.