libgfortran/ChangeLog

   1 2011-04-16  Janne Blomqvist  <jb@gcc.gnu.org>
   2
   3         * intrinsics/date_and_time.c (date_and_time): Remove sprintf CPP
   4         branch.
   5         * io/format.c (format_error): Use snprintf instead of sprintf.
   6         * io/list_read.c: Move snprintf fallback macro to libgfortran.h.
   7         (convert_integer): Use snprintf instead of sprintf.
   8         (parse_repeat): Likewise.
   9         (read_logical): Likewise.
  10         (read_integer): Likewise.
  11         (read_character): Likewise.
  12         (parse_real): Likewise.
  13         (read_complex): Likewise.
  14         (read_real): Likewise.
  15         (check_type): Likewise.
  16         (nml_parse_qualifier): Add string length argument, use snprintf
  17         instead of sprintf.
  18         (nml_get_obj_data): Use snprintf instead of sprintf.
  19         * io/open.c (new_unit): Remove sprintf CPP branch, use snprintf
  20         instead of sprintf.
  21         * io/transfer.c (require_type): Use snprintf instead of sprintf.
  22         * io/unix.c (tempfile): Likewise.
  23         * io/write.c (nml_write_obj): Likewise.
  24         * io/write_float.def (output_float): Remove sprintf CPP branch,
  25         use snprintf instead of sprintf.
  26         * libgfortran.h: Add fallback snprintf macro from io/list_read.c.
  27         * runtime/backtrace.c (show_backtrace): Remove sprintf CPP branch.
  28         * runtime/main.c (store_exe_path): Use snprintf instead of
  29         sprintf.
  30
  31 2011-04-15  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
  32
  33         PR libgfortran/48589
  34         * io/write_float.def (write_infnan): Set width properly for G0.
  35
  36 2011-04-15  Tobias Burnus  <burnus@net-b.de>
  37
  38         PR fortran/18918
  39         * Makefile.am: Build and install caf/single.c as
  40         libcaf_single.a.
  41         * Makefile.in: Regenerate.
  42
  43 2011-04-15  Janne Blomqvist  <jb@gcc.gnu.org>
  44
  45         PR libfortran/47571
  46         * configure: Regenerated.
  47         * config.h.in: Regenerated.
  48         * acinclude.m4: Add alpha*-dec-osf* to gthread blacklist.
  49         * configure.ac: Use separate symbol for clock_gettime in librt.
  50         * intrinsics/system_clock.c: Use weakrefs only when needed and
  51         supported.
  52
  53 2011-04-12  Janne Blomqvist  <jb@gcc.gnu.org>
  54
  55         * configure.ac: Use AC_TYPE_* to make sure we have (u)intptr_t,
  56         check for ptrdiff_t.
  57         * libgfortran.h: Remove (u)intptr_t definitions, use ptrdiff_t for
  58         index_type, change cshift0* prototypes.
  59         * config.h.in: Regenerated.
  60         * configure: Regenerated.
  61         * intrinsics/cshift0.c (cshift0): Use ptrdiff_t instead of ssize_t.
  62         * io/io.h (array_loop_spec): Use index_type instead of ssize_t.
  63         * io/list_read.c (nml_parse_qualifier): Likewise.
  64         * io/write.c (nml_write_obj): Likewise.
  65         * m4/cshift0.c (cshift0_'rtype_code`): Likewise.
  66         * generated/cshift0_*.c: Regenerated.
  67
  68 2011-04-12  Janne Blomqvist  <jb@gcc.gnu.org>
  69
  70         * Makefile.am: Remove _GNU_SOURCE from AM_CPPFLAGS.
  71         * Makefile.in: Regenerated.
  72
  73 2011-04-06  Tobias Burnus  <burnus@net-b.de>
  74
  75         PR fortran/18918
  76         * caf/mpi.c (_gfortran_caf_init, _gfortran_caf_finalize):
  77         Add global variable caf_mpi_initialized and use it for when
  78         finalizing.
  79
  80 2011-04-04  Tobias Burnus  <burnus@net-b.de>
  81
  82         * unix.c: Adapt stat DEFINEs since MinGW64 supports LFS.
  83         (fallback_access, open_internal4, compare_file_filename,
  84         find_file, file_size, inquire_sequential, inquire_direct,
  85         inquire_formatted): Use "struct stat" instead of gfstat_t.
  86
  87 2011-03-27  Tobias Burnus  <burnus@net-b.de>
  88
  89         PR fortran/18918
  90         * caf/libcaf.h: New - coarray communication library.
  91         * caf/mpi.c: New.
  92         * caf/single.c: New.
  93
  94 2011-03-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
  95
  96         PR libgfortran/48030
  97         * io/read.c (read_x): Re-implement using fbuf_getc.
  98
  99 2011-03-21  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
 100
 101         PR bootstrap/48135
 102         * configure.ac (gfortran_use_symver): Handle --disable-symvers.
 103         * configure: Regenerate.
 104
 105 2011-03-19  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
 106
 107         PR fortran/35667
 108         * io/io.h (convert_infnan): Declare.
 109         * io/read.c (convert_infnan): New.
 110         (read_f): Use convert_infnan to convert INFs and NANs.
 111         * list_read.c (parse_real, read_real): Likewise.
 112
 113 2011-03-19  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 114
 115         PR libfortran/47439
 116         * io/unix.c (tempfile): Work around poor mktemp() implementations.
 117
 118 2011-03-16  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 119
 120         PR libfortran/47883
 121         * acinclude.m4: Don't use AC_TRY_ macros.
 122         * configure.ac: Likewise.
 123         * configure: Regenerate.
 124
 125 2011-03-12  Thomas Koenig  <tkoenig@gcc.gnu.org>
 126
 127         PR libfortran/48066
 128         * m4/ifunction.m4:  If return array is empty, return.
 129         * m4/ifunction_logical.m4:  Likewise.
 130         * generated/all_l16.c: Regenerated.
 131         * generated/all_l1.c: Regenerated.
 132         * generated/all_l2.c: Regenerated.
 133         * generated/all_l4.c: Regenerated.
 134         * generated/all_l8.c: Regenerated.
 135         * generated/any_l16.c: Regenerated.
 136         * generated/any_l1.c: Regenerated.
 137         * generated/any_l2.c: Regenerated.
 138         * generated/any_l4.c: Regenerated.
 139         * generated/any_l8.c: Regenerated.
 140         * generated/count_16_l.c: Regenerated.
 141         * generated/count_1_l.c: Regenerated.
 142         * generated/count_2_l.c: Regenerated.
 143         * generated/count_4_l.c: Regenerated.
 144         * generated/count_8_l.c: Regenerated.
 145         * generated/maxloc1_16_i16.c: Regenerated.
 146         * generated/maxloc1_16_i1.c: Regenerated.
 147         * generated/maxloc1_16_i2.c: Regenerated.
 148         * generated/maxloc1_16_i4.c: Regenerated.
 149         * generated/maxloc1_16_i8.c: Regenerated.
 150         * generated/maxloc1_16_r10.c: Regenerated.
 151         * generated/maxloc1_16_r16.c: Regenerated.
 152         * generated/maxloc1_16_r4.c: Regenerated.
 153         * generated/maxloc1_16_r8.c: Regenerated.
 154         * generated/maxloc1_4_i16.c: Regenerated.
 155         * generated/maxloc1_4_i1.c: Regenerated.
 156         * generated/maxloc1_4_i2.c: Regenerated.
 157         * generated/maxloc1_4_i4.c: Regenerated.
 158         * generated/maxloc1_4_i8.c: Regenerated.
 159         * generated/maxloc1_4_r10.c: Regenerated.
 160         * generated/maxloc1_4_r16.c: Regenerated.
 161         * generated/maxloc1_4_r4.c: Regenerated.
 162         * generated/maxloc1_4_r8.c: Regenerated.
 163         * generated/maxloc1_8_i16.c: Regenerated.
 164         * generated/maxloc1_8_i1.c: Regenerated.
 165         * generated/maxloc1_8_i2.c: Regenerated.
 166         * generated/maxloc1_8_i4.c: Regenerated.
 167         * generated/maxloc1_8_i8.c: Regenerated.
 168         * generated/maxloc1_8_r10.c: Regenerated.
 169         * generated/maxloc1_8_r16.c: Regenerated.
 170         * generated/maxloc1_8_r4.c: Regenerated.
 171         * generated/maxloc1_8_r8.c: Regenerated.
 172         * generated/maxval_i16.c: Regenerated.
 173         * generated/maxval_i1.c: Regenerated.
 174         * generated/maxval_i2.c: Regenerated.
 175         * generated/maxval_i4.c: Regenerated.
 176         * generated/maxval_i8.c: Regenerated.
 177         * generated/maxval_r10.c: Regenerated.
 178         * generated/maxval_r16.c: Regenerated.
 179         * generated/maxval_r4.c: Regenerated.
 180         * generated/maxval_r8.c: Regenerated.
 181         * generated/minloc1_16_i16.c: Regenerated.
 182         * generated/minloc1_16_i1.c: Regenerated.
 183         * generated/minloc1_16_i2.c: Regenerated.
 184         * generated/minloc1_16_i4.c: Regenerated.
 185         * generated/minloc1_16_i8.c: Regenerated.
 186         * generated/minloc1_16_r10.c: Regenerated.
 187         * generated/minloc1_16_r16.c: Regenerated.
 188         * generated/minloc1_16_r4.c: Regenerated.
 189         * generated/minloc1_16_r8.c: Regenerated.
 190         * generated/minloc1_4_i16.c: Regenerated.
 191         * generated/minloc1_4_i1.c: Regenerated.
 192         * generated/minloc1_4_i2.c: Regenerated.
 193         * generated/minloc1_4_i4.c: Regenerated.
 194         * generated/minloc1_4_i8.c: Regenerated.
 195         * generated/minloc1_4_r10.c: Regenerated.
 196         * generated/minloc1_4_r16.c: Regenerated.
 197         * generated/minloc1_4_r4.c: Regenerated.
 198         * generated/minloc1_4_r8.c: Regenerated.
 199         * generated/minloc1_8_i16.c: Regenerated.
 200         * generated/minloc1_8_i1.c: Regenerated.
 201         * generated/minloc1_8_i2.c: Regenerated.
 202         * generated/minloc1_8_i4.c: Regenerated.
 203         * generated/minloc1_8_i8.c: Regenerated.
 204         * generated/minloc1_8_r10.c: Regenerated.
 205         * generated/minloc1_8_r16.c: Regenerated.
 206         * generated/minloc1_8_r4.c: Regenerated.
 207         * generated/minloc1_8_r8.c: Regenerated.
 208         * generated/minval_i16.c: Regenerated.
 209         * generated/minval_i1.c: Regenerated.
 210         * generated/minval_i2.c: Regenerated.
 211         * generated/minval_i4.c: Regenerated.
 212         * generated/minval_i8.c: Regenerated.
 213         * generated/minval_r10.c: Regenerated.
 214         * generated/minval_r16.c: Regenerated.
 215         * generated/minval_r4.c: Regenerated.
 216         * generated/minval_r8.c: Regenerated.
 217         * generated/product_c10.c: Regenerated.
 218         * generated/product_c16.c: Regenerated.
 219         * generated/product_c4.c: Regenerated.
 220         * generated/product_c8.c: Regenerated.
 221         * generated/product_i16.c: Regenerated.
 222         * generated/product_i1.c: Regenerated.
 223         * generated/product_i2.c: Regenerated.
 224         * generated/product_i4.c: Regenerated.
 225         * generated/product_i8.c: Regenerated.
 226         * generated/product_r10.c: Regenerated.
 227         * generated/product_r16.c: Regenerated.
 228         * generated/product_r4.c: Regenerated.
 229         * generated/product_r8.c: Regenerated.
 230         * generated/sum_c10.c: Regenerated.
 231         * generated/sum_c16.c: Regenerated.
 232         * generated/sum_c4.c: Regenerated.
 233         * generated/sum_c8.c: Regenerated.
 234         * generated/sum_i16.c: Regenerated.
 235         * generated/sum_i1.c: Regenerated.
 236         * generated/sum_i2.c: Regenerated.
 237         * generated/sum_i4.c: Regenerated.
 238         * generated/sum_i8.c: Regenerated.
 239         * generated/sum_r10.c: Regenerated.
 240         * generated/sum_r16.c: Regenerated.
 241         * generated/sum_r4.c: Regenerated.
 242         * generated/sum_r8.c: Regenerated.
 243
 244 2011-03-11  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 245
 246         PR libgfortran/48047
 247         * io/write_float.def (write_float): Change MIN_FIELD_WIDTH to 48.
 248
 249 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 250
 251         PR libfortran/47802
 252         * intrinsics/ctime.c (strctime): Use builtins to check localtime_r
 253         return type.
 254
 255 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 256
 257         PR libfortran/47802
 258         * intrinsics/ctime.c (strctime): Don't use return value of
 259         localtime_r.
 260
 261 2011-02-28  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 262
 263         PR libgfortran/47567
 264         * io/write_float.def (output_float): Move handling of w = 0 to after
 265         output rounding. Check for zero and set zero_flag accordingly. Set
 266         width according to zero_flag. Add better comments.
 267
 268 2011-02-27  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 269
 270         PR libgfortran/47778
 271         * io/list_read.c (namelist_read): Intialize the error string buffere.
 272         If pprev_nl was used during the previous namelist read and the rank
 273         was zero, reset the pointer to NULL for the next namelist read.
 274
 275 2011-02-26  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 276
 277         PR libfortran/45165
 278         * unix.c (fallback_access): Fix file descriptor leaks.
 279
 280 2011-02-25  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 281
 282         * acinclude.m4 (LIBGFOR_CHECK_FPSETMASK): Set shell variable
 283         tested in configure.host.
 284         * configure: Regenerate.
 285
 286 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 287
 288         PR libfortran/47802
 289         * config.h.in: Regenerated.
 290         * configure: Regenerated.
 291         * configure.ac: Remove checks for ctime and ctime_r, add check for
 292         strftime.
 293         * intrinsics/date_and_time.c (localtime_r): Move fallback
 294         implementation to time_1.h.
 295         * intrinsics/time_1.h (localtime_r): Fallback implementation.
 296         * intrinsics/ctime.c: Include time_1.h.
 297         (ctime_r): Remove fallback implementation.
 298         (strctime): New function.
 299         (fdate): Use strctime instead of ctime_r.
 300         (fdate_sub): Likewise.
 301         (ctime): Likewise.
 302         (ctime_sub): Likewise.
 303
 304 2011-02-24  Jakub Jelinek  <jakub@redhat.com>
 305
 306         PR fortran/47878
 307         * io/transfer.c (read_sf): Call fbuf_getptr only at the end,
 308         and subtract n, dtp->u.p.sf_seen_eor and seen_comma from it.
 309
 310 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 311
 312         PR libfortran/47802
 313         * configure.ac: Add test for POSIX getpwuid_r.
 314         * intrinsics/getlog.c (getlog): CPP test for
 315         HAVE_POSIX_GETPWUID_R.
 316         * config.h.in: Regenerated.
 317         * configure: Regenerated.
 318
 319 2011-02-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 320
 321         PR libgfortran/47567
 322         * io/write_float.def (output_float): Remove special case handling of
 323         zero with width 1.
 324
 325 2011-02-23  Janne Blomqvist  <jb@gcc.gnu.org>
 326             Jerry DeLisle    <jvdelisle@gcc.gnu.org>
 327
 328         PR libfortran/47694
 329         * io/fbuf.h (fbuf_getptr): New inline function.
 330         * io/transfer.c (read_sf): Use fbuf_getptr and fbuf_getc to scan
 331         through the string instead of fbuf_read.
 332
 333 2011-02-22  Tobias Burnus  <burnus@net-b.de>
 334             Kai-Uwe Eckhardt  <kuehro@gmx.de>
 335
 336         PR libfortran/47830
 337         * intrinsics/c99_functions.c (roundl): Make C valid for
 338         HAVE_NEXTAFTERL.
 339
 340 2011-02-19  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 341
 342         PR libgfortran/47567
 343         * io/write_float.def (output_float): Adjust width for F0.d to
 344         allow space for negative signs on zero.
 345
 346 2011-02-16  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 347
 348         PR libgfortran/47667
 349         * io/list_read.c (read_logical): Check for end of line before calling
 350         eat_line. (read_integer): Likewise. (parse_real): Don't unget the
 351         separator. Check for end of line before calling eat_line.
 352         (read_complex): Allow line-end before and after parenthesis and comma.
 353         Check for end of line before calling eat_line. (read_real): Check for
 354         end of line before calling eat_line.
 355
 356 2011-02-16  Jakub Jelinek  <jakub@redhat.com>
 357
 358         PR libfortran/47757
 359         * gfortran.map (GFORTRAN_1.4): Export
 360         _gfortran_{m,s}i{all,any,parity}_i{1,2,4,8,16} and
 361         _gfortran_{cshift0,eoshift{0,2}}_16_char4.
 362
 363 2011-02-15  Tobias Burnus  <burnus@net-b.de>
 364
 365         PR fortran/47716
 366         PR fortran/47648
 367         * acinclude.m4 (LIBGFOR_CHECK_FLOAT128): Use check from
 368         libquadmath, which uses more features.
 369         * configure: Regenerate.
 370
 371 2011-02-14  Jakub Jelinek  <jakub@redhat.com>
 372
 373         PR fortran/47642
 374         * io/write_float.def (DTOAQ): Use quadmath_snprintf instead of
 375         quadmath_flt128tostr.
 376         * io/transfer128.c (tmp2): Initialize to quadmath_snprintf instead
 377         of quadmath_flt128tostr.
 378
 379 2011-02-13  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 380
 381         * Makefile.in: Regenerate.
 382         * aclocal.m4: Likewise.
 383         * configure: Likewise.
 384
 385 2011-02-05  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 386
 387         PR libgfortran/47567
 388         * io/write_float.def (output_float): Eliminate some redundant code.
 389         Adjust width for case of F0.X for values of zero and all other values.
 390         Expand cases where '*' is set to give cleaner results.
 391
 392 2011-02-05  Janne Blomqvist  <jb@gcc.gnu.org>
 393
 394         PR libfortran/47571
 395         * intrinsics/time_1.h (GF_CLOCK_MONOTONIC): Move to system_clock.c.
 396         (weak_gettime): Likewise.
 397         (gf_gettime): Change API, move weak_gettime() usage to
 398         system_clock.c
 399         * intrinsics/system_clock.c (GTHREAD_USE_WEAK): Define.
 400         (gf_gettime_mono): New function.
 401         (system_clock_4): Use gf_gettime_mono().
 402         (system_clock_8): Likewise.
 403         * intrinsics/date_and_time.c (date_and_time): Update gf_gettime()
 404         usage.
 405
 406 2011-02-02  Janne Blomqvist  <jb@gcc.gnu.org>
 407
 408         PR libfortran/47571
 409         * configure: Regenerated.
 410         * configure.ac: Don't add librt to LIBS.
 411         * intrinsics/time_1.h (weak_gettime): Weakref trickery for
 412         clock_gettime().
 413         (gf_gettime): Use weak_gettime() instead of clock_gettime().
 414
 415 2011-02-01  Janne Blomqvist  <jb@gcc.gnu.org>
 416
 417         * intrinsics/time_1.h: Include errno.h needed by fallbacks.
 418
 419 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 420
 421         * intrinsics/time_1.h: Fix definition of GF_CLOCK_MONOTONIC macro.
 422
 423 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 424
 425         * configure.ac: Check for clock_gettime().
 426         * configure: Regenerated.
 427         * config.h.in: Regenerated.
 428         * intrinsics/time_1.h (__time_1): Rename to gf_cputime, add
 429         times() fallback.
 430         (gf_gettime): New function.
 431         * intrinsics/cpu_time.c (__cpu_time_1): Update to call gf_cputime.
 432         * intrinsics/date_and_time.c (date_and_time): Use gf_gettime.
 433         * intrinsics/dtime.c (dtime_sub): Use gf_cputime.
 434         * intrinsics/etime.c (etime_sub): Use gf_cputime.
 435         * intrinsics/system_clock.c (system_clock_4): Use gf_gettime.
 436         (system_clock_8): Use gf_gettime, increase count rate to allow
 437         nanosecond precision, remove overflow prone branch.
 438
 439 2011-01-29  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 440
 441         PR libgfortran/47434
 442         * io/write_float.def (write_infnan): Use calculate_sign to determine
 443         if the sign should be given and check field widths accordingly.
 444
 445 2011-01-29  Kai Tietz  <kai.tietz@onevision.com>
 446
 447         * intrinsics/ctime.c (ctime_r): Improve implementation.
 448
 449 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 450
 451         PR libfortran/47431
 452         * config.h.in: Regenerated.
 453         * configure: Regenerated.
 454         * configure.ac: Add check for ctime_r().
 455         * intrinsics/ctime.c (ctime_r): Fallback implementation.
 456         (fdate): Use ctime_r() instead of ctime().
 457         (fdate_sub): Likewise.
 458         (ctime): Likewise.
 459         (ctime_sub): Likewise.
 460
 461 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 462
 463         PR libfortran/47432
 464         * config.h.in: Regenerated.
 465         * configure: Regenerated.
 466         * configure.ac: Add check for ttyname_r.
 467         * io/unix.h: Add TTY_NAME_MAX, change stream_ttyname prototype.
 468         * io/unix.c (stream_ttyname): Use ttyname_r if available, conform
 469         to new prototype.
 470         * io/inquire.c (inquire_via_unit): Use changed stream_ttyname.
 471         * io/intrinsics.c (ttynam_sub): Likewise.
 472         (ttynam): Likewise.
 473
 474 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 475
 476         PR libfortran/47491
 477         * configure.ac: Call AC_USE_SYSTEM_EXTENSIONS to enable common
 478         extensions.
 479         * config.h.in: Regenerate.
 480         * configure: Regenerate.
 481
 482 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 483
 484         PR libgfortran/47285
 485         * io/write_float.def (write_infnan): Adjust processor selected width
 486         to 3 if NaN.
 487
 488 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 489
 490         PR libgfortran/47285
 491         * io/write_float.def (output_float): Return SUCCESS or FAILURE and use
 492         the result to set the padding.
 493
 494 2011-01-26  Kai Tietz  <kai.tietz@onevision.com>
 495
 496         * intrinsics/getlog.c (getlog): Fix label/statement issue.
 497
 498 2011-01-25  Janne Blomqvist  <jb@gcc.gnu.org>
 499
 500         PR libfortran/47375
 501         * config.h.in: Regenerated.
 502         * configure: Regenerated.
 503         * configure.ac: Add check for getpwuid_r.
 504         * intrinsics/getlog.c (getlog): Use getpwuid_r() if available.
 505
 506 2011-01-22  Janne Blomqvist  <jb@gcc.gnu.org>
 507
 508         PR libfortran/46267
 509         * config.h.in: Regenerated.
 510         * configure: Regenerated.
 511         * configure.ac: Check presence of strerror_r.
 512         * intrinsics/gerror.c (gerror): Use gf_strerror, modify logic.
 513         * io/unix.c (get_oserror): Remove.
 514         * libgfortran.h (gf_strerror): Add prototype.
 515         (get_oserror): Remove prototype.
 516         * runtime/error.c (gf_strerror): New function.
 517         (os_error): Use gf_strerror instead of get_oserror.
 518         (generate_errror): Likewise.
 519
 520 2011-01-17  Janne Blomqvist  <jb@gcc.gnu.org>
 521
 522         PR libfortran/47296
 523         * io/unix.c (tempfile): Set opp->file and opp->file_len also if an
 524         error occurs.
 525
 526 2011-01-16  Jakub Jelinek  <jakub@redhat.com>
 527
 528         PR fortran/46625
 529         * io/write_float.def (DTOAQ): Use quadmath_flt128tostr
 530         instead of quadmath_dtoa.
 531         * io/transfer128.c (tmp1, tmp2): New variables, bring in
 532         strtoflt128 and quadmath_flt128tostr.
 533         (transfer_real128, transfer_real128_write, transfer_complex128,
 534         transfer_complex128_write): Remove tmp1/tmp2 variables.
 535         * io/read.c (convert_real): Use strtoflt128 instead of
 536         quadmath_strtopQ, adjust for the changed arguments and return
 537         value.
 538
 539 2011-01-14  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 540
 541         PR libgfortran/47296
 542         * io/unix.c (unpack_filename): Return non-zero if the filename passed
 543         in is NULL.
 544
 545 2011-01-04  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 546
 547         PR libgfortran/47154
 548         * io/list_read.c (namelist_read): Remove calls to hit_eof to avoid the
 549         duplicate calls via next_record.
 550
 551 \f
 552 Copyright (C) 2011 Free Software Foundation, Inc.
 553
 554 Copying and distribution of this file, with or without modification,
 555 are permitted in any medium without royalty provided the copyright
 556 notice and this notice are preserved.