libgfortran/ChangeLog

   1 2011-04-29  Janne Blomqvist  <jb@gcc.gnu.org>
   2
   3         PR libfortran/48488
   4         * io/write.c (write_real, write_real_g0): Update comments.
   5
   6 2011-04-29  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
   7             Janne Blomqvist  <jb@gcc.gnu.org>
   8
   9         PR libgfortran/48488
  10         PR libgfortran/48602
  11         PR libgfortran/48615
  12         PR libgfortran/48684
  13         PR libgfortran/48787
  14         * io/write.c (write_d, write_e, write_f, write_en,
  15         write_es): Add precision compemsation parameter to call.
  16         (set_fnode_default): Adjust default widths to assure
  17         round trip on write and read. (write_real): Adjust call to write_float.
  18         (write_real_g0): Calculate compensation for extra precision and adjust
  19         call to write_float.
  20         * io/write_float.def (output_float_FMT_G_): Use volatile rather than
  21         asm volatile to avoid optimization issue. Correctly calculate the
  22         number of blanks (nb) to be appended and simplify calculation logic.
  23         (write_float): Increase MIN_FIELD_WIDTH by one to accomodate the new
  24         default widths. Eliminate the code that attempted to reduce the
  25         the precision used in later sprintf functions.  Add call parameter to
  26         compensate for extra precision.
  27
  28 2011-04-20  Jim Meyering  <meyering@redhat.com>
  29
  30         * intrinsics/move_alloc.c (move_alloc): Remove useless
  31         if-before-free test.
  32         * io/fbuf.c (fbuf_destroy): Likewise.
  33         * io/format.c (save_parsed_format): Likewise.
  34         * io/open.c (already_open): Likewise.
  35         * io/unit.c (free_internal_unit, close_unit_1): Likewise.
  36         * io/unix.c (mem_close): Likewise.
  37
  38 2011-04-19  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
  39
  40         PR libgfortran/48602
  41         * io/write_float.def (output_float_FMT_G): Fix reversal in conditional.
  42         Use asm volatile to mark temp variable, avoiding optimization errors.
  43
  44 2011-04-17  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
  45
  46         PR libgfortran/48602
  47         * io/write_float.def (output_float_FMT_G): Use current rounding mode
  48         to set the rounding parameters. (output_float): Skip rounding
  49         if value is zero.
  50
  51 2011-04-16  Janne Blomqvist  <jb@gcc.gnu.org>
  52
  53         * intrinsics/date_and_time.c (date_and_time): Remove sprintf CPP
  54         branch.
  55         * io/format.c (format_error): Use snprintf instead of sprintf.
  56         * io/list_read.c: Move snprintf fallback macro to libgfortran.h.
  57         (convert_integer): Use snprintf instead of sprintf.
  58         (parse_repeat): Likewise.
  59         (read_logical): Likewise.
  60         (read_integer): Likewise.
  61         (read_character): Likewise.
  62         (parse_real): Likewise.
  63         (read_complex): Likewise.
  64         (read_real): Likewise.
  65         (check_type): Likewise.
  66         (nml_parse_qualifier): Add string length argument, use snprintf
  67         instead of sprintf.
  68         (nml_get_obj_data): Use snprintf instead of sprintf.
  69         * io/open.c (new_unit): Remove sprintf CPP branch, use snprintf
  70         instead of sprintf.
  71         * io/transfer.c (require_type): Use snprintf instead of sprintf.
  72         * io/unix.c (tempfile): Likewise.
  73         * io/write.c (nml_write_obj): Likewise.
  74         * io/write_float.def (output_float): Remove sprintf CPP branch,
  75         use snprintf instead of sprintf.
  76         * libgfortran.h: Add fallback snprintf macro from io/list_read.c.
  77         * runtime/backtrace.c (show_backtrace): Remove sprintf CPP branch.
  78         * runtime/main.c (store_exe_path): Use snprintf instead of
  79         sprintf.
  80
  81 2011-04-15  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
  82
  83         PR libgfortran/48589
  84         * io/write_float.def (write_infnan): Set width properly for G0.
  85
  86 2011-04-15  Tobias Burnus  <burnus@net-b.de>
  87
  88         PR fortran/18918
  89         * Makefile.am: Build and install caf/single.c as
  90         libcaf_single.a.
  91         * Makefile.in: Regenerate.
  92
  93 2011-04-15  Janne Blomqvist  <jb@gcc.gnu.org>
  94
  95         PR libfortran/47571
  96         * configure: Regenerated.
  97         * config.h.in: Regenerated.
  98         * acinclude.m4: Add alpha*-dec-osf* to gthread blacklist.
  99         * configure.ac: Use separate symbol for clock_gettime in librt.
 100         * intrinsics/system_clock.c: Use weakrefs only when needed and
 101         supported.
 102
 103 2011-04-12  Janne Blomqvist  <jb@gcc.gnu.org>
 104
 105         * configure.ac: Use AC_TYPE_* to make sure we have (u)intptr_t,
 106         check for ptrdiff_t.
 107         * libgfortran.h: Remove (u)intptr_t definitions, use ptrdiff_t for
 108         index_type, change cshift0* prototypes.
 109         * config.h.in: Regenerated.
 110         * configure: Regenerated.
 111         * intrinsics/cshift0.c (cshift0): Use ptrdiff_t instead of ssize_t.
 112         * io/io.h (array_loop_spec): Use index_type instead of ssize_t.
 113         * io/list_read.c (nml_parse_qualifier): Likewise.
 114         * io/write.c (nml_write_obj): Likewise.
 115         * m4/cshift0.c (cshift0_'rtype_code`): Likewise.
 116         * generated/cshift0_*.c: Regenerated.
 117
 118 2011-04-12  Janne Blomqvist  <jb@gcc.gnu.org>
 119
 120         * Makefile.am: Remove _GNU_SOURCE from AM_CPPFLAGS.
 121         * Makefile.in: Regenerated.
 122
 123 2011-04-06  Tobias Burnus  <burnus@net-b.de>
 124
 125         PR fortran/18918
 126         * caf/mpi.c (_gfortran_caf_init, _gfortran_caf_finalize):
 127         Add global variable caf_mpi_initialized and use it for when
 128         finalizing.
 129
 130 2011-04-04  Tobias Burnus  <burnus@net-b.de>
 131
 132         * unix.c: Adapt stat DEFINEs since MinGW64 supports LFS.
 133         (fallback_access, open_internal4, compare_file_filename,
 134         find_file, file_size, inquire_sequential, inquire_direct,
 135         inquire_formatted): Use "struct stat" instead of gfstat_t.
 136
 137 2011-03-27  Tobias Burnus  <burnus@net-b.de>
 138
 139         PR fortran/18918
 140         * caf/libcaf.h: New - coarray communication library.
 141         * caf/mpi.c: New.
 142         * caf/single.c: New.
 143
 144 2011-03-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 145
 146         PR libgfortran/48030
 147         * io/read.c (read_x): Re-implement using fbuf_getc.
 148
 149 2011-03-21  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
 150
 151         PR bootstrap/48135
 152         * configure.ac (gfortran_use_symver): Handle --disable-symvers.
 153         * configure: Regenerate.
 154
 155 2011-03-19  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
 156
 157         PR fortran/35667
 158         * io/io.h (convert_infnan): Declare.
 159         * io/read.c (convert_infnan): New.
 160         (read_f): Use convert_infnan to convert INFs and NANs.
 161         * list_read.c (parse_real, read_real): Likewise.
 162
 163 2011-03-19  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 164
 165         PR libfortran/47439
 166         * io/unix.c (tempfile): Work around poor mktemp() implementations.
 167
 168 2011-03-16  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 169
 170         PR libfortran/47883
 171         * acinclude.m4: Don't use AC_TRY_ macros.
 172         * configure.ac: Likewise.
 173         * configure: Regenerate.
 174
 175 2011-03-12  Thomas Koenig  <tkoenig@gcc.gnu.org>
 176
 177         PR libfortran/48066
 178         * m4/ifunction.m4:  If return array is empty, return.
 179         * m4/ifunction_logical.m4:  Likewise.
 180         * generated/all_l16.c: Regenerated.
 181         * generated/all_l1.c: Regenerated.
 182         * generated/all_l2.c: Regenerated.
 183         * generated/all_l4.c: Regenerated.
 184         * generated/all_l8.c: Regenerated.
 185         * generated/any_l16.c: Regenerated.
 186         * generated/any_l1.c: Regenerated.
 187         * generated/any_l2.c: Regenerated.
 188         * generated/any_l4.c: Regenerated.
 189         * generated/any_l8.c: Regenerated.
 190         * generated/count_16_l.c: Regenerated.
 191         * generated/count_1_l.c: Regenerated.
 192         * generated/count_2_l.c: Regenerated.
 193         * generated/count_4_l.c: Regenerated.
 194         * generated/count_8_l.c: Regenerated.
 195         * generated/maxloc1_16_i16.c: Regenerated.
 196         * generated/maxloc1_16_i1.c: Regenerated.
 197         * generated/maxloc1_16_i2.c: Regenerated.
 198         * generated/maxloc1_16_i4.c: Regenerated.
 199         * generated/maxloc1_16_i8.c: Regenerated.
 200         * generated/maxloc1_16_r10.c: Regenerated.
 201         * generated/maxloc1_16_r16.c: Regenerated.
 202         * generated/maxloc1_16_r4.c: Regenerated.
 203         * generated/maxloc1_16_r8.c: Regenerated.
 204         * generated/maxloc1_4_i16.c: Regenerated.
 205         * generated/maxloc1_4_i1.c: Regenerated.
 206         * generated/maxloc1_4_i2.c: Regenerated.
 207         * generated/maxloc1_4_i4.c: Regenerated.
 208         * generated/maxloc1_4_i8.c: Regenerated.
 209         * generated/maxloc1_4_r10.c: Regenerated.
 210         * generated/maxloc1_4_r16.c: Regenerated.
 211         * generated/maxloc1_4_r4.c: Regenerated.
 212         * generated/maxloc1_4_r8.c: Regenerated.
 213         * generated/maxloc1_8_i16.c: Regenerated.
 214         * generated/maxloc1_8_i1.c: Regenerated.
 215         * generated/maxloc1_8_i2.c: Regenerated.
 216         * generated/maxloc1_8_i4.c: Regenerated.
 217         * generated/maxloc1_8_i8.c: Regenerated.
 218         * generated/maxloc1_8_r10.c: Regenerated.
 219         * generated/maxloc1_8_r16.c: Regenerated.
 220         * generated/maxloc1_8_r4.c: Regenerated.
 221         * generated/maxloc1_8_r8.c: Regenerated.
 222         * generated/maxval_i16.c: Regenerated.
 223         * generated/maxval_i1.c: Regenerated.
 224         * generated/maxval_i2.c: Regenerated.
 225         * generated/maxval_i4.c: Regenerated.
 226         * generated/maxval_i8.c: Regenerated.
 227         * generated/maxval_r10.c: Regenerated.
 228         * generated/maxval_r16.c: Regenerated.
 229         * generated/maxval_r4.c: Regenerated.
 230         * generated/maxval_r8.c: Regenerated.
 231         * generated/minloc1_16_i16.c: Regenerated.
 232         * generated/minloc1_16_i1.c: Regenerated.
 233         * generated/minloc1_16_i2.c: Regenerated.
 234         * generated/minloc1_16_i4.c: Regenerated.
 235         * generated/minloc1_16_i8.c: Regenerated.
 236         * generated/minloc1_16_r10.c: Regenerated.
 237         * generated/minloc1_16_r16.c: Regenerated.
 238         * generated/minloc1_16_r4.c: Regenerated.
 239         * generated/minloc1_16_r8.c: Regenerated.
 240         * generated/minloc1_4_i16.c: Regenerated.
 241         * generated/minloc1_4_i1.c: Regenerated.
 242         * generated/minloc1_4_i2.c: Regenerated.
 243         * generated/minloc1_4_i4.c: Regenerated.
 244         * generated/minloc1_4_i8.c: Regenerated.
 245         * generated/minloc1_4_r10.c: Regenerated.
 246         * generated/minloc1_4_r16.c: Regenerated.
 247         * generated/minloc1_4_r4.c: Regenerated.
 248         * generated/minloc1_4_r8.c: Regenerated.
 249         * generated/minloc1_8_i16.c: Regenerated.
 250         * generated/minloc1_8_i1.c: Regenerated.
 251         * generated/minloc1_8_i2.c: Regenerated.
 252         * generated/minloc1_8_i4.c: Regenerated.
 253         * generated/minloc1_8_i8.c: Regenerated.
 254         * generated/minloc1_8_r10.c: Regenerated.
 255         * generated/minloc1_8_r16.c: Regenerated.
 256         * generated/minloc1_8_r4.c: Regenerated.
 257         * generated/minloc1_8_r8.c: Regenerated.
 258         * generated/minval_i16.c: Regenerated.
 259         * generated/minval_i1.c: Regenerated.
 260         * generated/minval_i2.c: Regenerated.
 261         * generated/minval_i4.c: Regenerated.
 262         * generated/minval_i8.c: Regenerated.
 263         * generated/minval_r10.c: Regenerated.
 264         * generated/minval_r16.c: Regenerated.
 265         * generated/minval_r4.c: Regenerated.
 266         * generated/minval_r8.c: Regenerated.
 267         * generated/product_c10.c: Regenerated.
 268         * generated/product_c16.c: Regenerated.
 269         * generated/product_c4.c: Regenerated.
 270         * generated/product_c8.c: Regenerated.
 271         * generated/product_i16.c: Regenerated.
 272         * generated/product_i1.c: Regenerated.
 273         * generated/product_i2.c: Regenerated.
 274         * generated/product_i4.c: Regenerated.
 275         * generated/product_i8.c: Regenerated.
 276         * generated/product_r10.c: Regenerated.
 277         * generated/product_r16.c: Regenerated.
 278         * generated/product_r4.c: Regenerated.
 279         * generated/product_r8.c: Regenerated.
 280         * generated/sum_c10.c: Regenerated.
 281         * generated/sum_c16.c: Regenerated.
 282         * generated/sum_c4.c: Regenerated.
 283         * generated/sum_c8.c: Regenerated.
 284         * generated/sum_i16.c: Regenerated.
 285         * generated/sum_i1.c: Regenerated.
 286         * generated/sum_i2.c: Regenerated.
 287         * generated/sum_i4.c: Regenerated.
 288         * generated/sum_i8.c: Regenerated.
 289         * generated/sum_r10.c: Regenerated.
 290         * generated/sum_r16.c: Regenerated.
 291         * generated/sum_r4.c: Regenerated.
 292         * generated/sum_r8.c: Regenerated.
 293
 294 2011-03-11  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 295
 296         PR libgfortran/48047
 297         * io/write_float.def (write_float): Change MIN_FIELD_WIDTH to 48.
 298
 299 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 300
 301         PR libfortran/47802
 302         * intrinsics/ctime.c (strctime): Use builtins to check localtime_r
 303         return type.
 304
 305 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 306
 307         PR libfortran/47802
 308         * intrinsics/ctime.c (strctime): Don't use return value of
 309         localtime_r.
 310
 311 2011-02-28  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 312
 313         PR libgfortran/47567
 314         * io/write_float.def (output_float): Move handling of w = 0 to after
 315         output rounding. Check for zero and set zero_flag accordingly. Set
 316         width according to zero_flag. Add better comments.
 317
 318 2011-02-27  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 319
 320         PR libgfortran/47778
 321         * io/list_read.c (namelist_read): Intialize the error string buffere.
 322         If pprev_nl was used during the previous namelist read and the rank
 323         was zero, reset the pointer to NULL for the next namelist read.
 324
 325 2011-02-26  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 326
 327         PR libfortran/45165
 328         * unix.c (fallback_access): Fix file descriptor leaks.
 329
 330 2011-02-25  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 331
 332         * acinclude.m4 (LIBGFOR_CHECK_FPSETMASK): Set shell variable
 333         tested in configure.host.
 334         * configure: Regenerate.
 335
 336 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 337
 338         PR libfortran/47802
 339         * config.h.in: Regenerated.
 340         * configure: Regenerated.
 341         * configure.ac: Remove checks for ctime and ctime_r, add check for
 342         strftime.
 343         * intrinsics/date_and_time.c (localtime_r): Move fallback
 344         implementation to time_1.h.
 345         * intrinsics/time_1.h (localtime_r): Fallback implementation.
 346         * intrinsics/ctime.c: Include time_1.h.
 347         (ctime_r): Remove fallback implementation.
 348         (strctime): New function.
 349         (fdate): Use strctime instead of ctime_r.
 350         (fdate_sub): Likewise.
 351         (ctime): Likewise.
 352         (ctime_sub): Likewise.
 353
 354 2011-02-24  Jakub Jelinek  <jakub@redhat.com>
 355
 356         PR fortran/47878
 357         * io/transfer.c (read_sf): Call fbuf_getptr only at the end,
 358         and subtract n, dtp->u.p.sf_seen_eor and seen_comma from it.
 359
 360 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 361
 362         PR libfortran/47802
 363         * configure.ac: Add test for POSIX getpwuid_r.
 364         * intrinsics/getlog.c (getlog): CPP test for
 365         HAVE_POSIX_GETPWUID_R.
 366         * config.h.in: Regenerated.
 367         * configure: Regenerated.
 368
 369 2011-02-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 370
 371         PR libgfortran/47567
 372         * io/write_float.def (output_float): Remove special case handling of
 373         zero with width 1.
 374
 375 2011-02-23  Janne Blomqvist  <jb@gcc.gnu.org>
 376             Jerry DeLisle    <jvdelisle@gcc.gnu.org>
 377
 378         PR libfortran/47694
 379         * io/fbuf.h (fbuf_getptr): New inline function.
 380         * io/transfer.c (read_sf): Use fbuf_getptr and fbuf_getc to scan
 381         through the string instead of fbuf_read.
 382
 383 2011-02-22  Tobias Burnus  <burnus@net-b.de>
 384             Kai-Uwe Eckhardt  <kuehro@gmx.de>
 385
 386         PR libfortran/47830
 387         * intrinsics/c99_functions.c (roundl): Make C valid for
 388         HAVE_NEXTAFTERL.
 389
 390 2011-02-19  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 391
 392         PR libgfortran/47567
 393         * io/write_float.def (output_float): Adjust width for F0.d to
 394         allow space for negative signs on zero.
 395
 396 2011-02-16  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 397
 398         PR libgfortran/47667
 399         * io/list_read.c (read_logical): Check for end of line before calling
 400         eat_line. (read_integer): Likewise. (parse_real): Don't unget the
 401         separator. Check for end of line before calling eat_line.
 402         (read_complex): Allow line-end before and after parenthesis and comma.
 403         Check for end of line before calling eat_line. (read_real): Check for
 404         end of line before calling eat_line.
 405
 406 2011-02-16  Jakub Jelinek  <jakub@redhat.com>
 407
 408         PR libfortran/47757
 409         * gfortran.map (GFORTRAN_1.4): Export
 410         _gfortran_{m,s}i{all,any,parity}_i{1,2,4,8,16} and
 411         _gfortran_{cshift0,eoshift{0,2}}_16_char4.
 412
 413 2011-02-15  Tobias Burnus  <burnus@net-b.de>
 414
 415         PR fortran/47716
 416         PR fortran/47648
 417         * acinclude.m4 (LIBGFOR_CHECK_FLOAT128): Use check from
 418         libquadmath, which uses more features.
 419         * configure: Regenerate.
 420
 421 2011-02-14  Jakub Jelinek  <jakub@redhat.com>
 422
 423         PR fortran/47642
 424         * io/write_float.def (DTOAQ): Use quadmath_snprintf instead of
 425         quadmath_flt128tostr.
 426         * io/transfer128.c (tmp2): Initialize to quadmath_snprintf instead
 427         of quadmath_flt128tostr.
 428
 429 2011-02-13  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 430
 431         * Makefile.in: Regenerate.
 432         * aclocal.m4: Likewise.
 433         * configure: Likewise.
 434
 435 2011-02-05  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 436
 437         PR libgfortran/47567
 438         * io/write_float.def (output_float): Eliminate some redundant code.
 439         Adjust width for case of F0.X for values of zero and all other values.
 440         Expand cases where '*' is set to give cleaner results.
 441
 442 2011-02-05  Janne Blomqvist  <jb@gcc.gnu.org>
 443
 444         PR libfortran/47571
 445         * intrinsics/time_1.h (GF_CLOCK_MONOTONIC): Move to system_clock.c.
 446         (weak_gettime): Likewise.
 447         (gf_gettime): Change API, move weak_gettime() usage to
 448         system_clock.c
 449         * intrinsics/system_clock.c (GTHREAD_USE_WEAK): Define.
 450         (gf_gettime_mono): New function.
 451         (system_clock_4): Use gf_gettime_mono().
 452         (system_clock_8): Likewise.
 453         * intrinsics/date_and_time.c (date_and_time): Update gf_gettime()
 454         usage.
 455
 456 2011-02-02  Janne Blomqvist  <jb@gcc.gnu.org>
 457
 458         PR libfortran/47571
 459         * configure: Regenerated.
 460         * configure.ac: Don't add librt to LIBS.
 461         * intrinsics/time_1.h (weak_gettime): Weakref trickery for
 462         clock_gettime().
 463         (gf_gettime): Use weak_gettime() instead of clock_gettime().
 464
 465 2011-02-01  Janne Blomqvist  <jb@gcc.gnu.org>
 466
 467         * intrinsics/time_1.h: Include errno.h needed by fallbacks.
 468
 469 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 470
 471         * intrinsics/time_1.h: Fix definition of GF_CLOCK_MONOTONIC macro.
 472
 473 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 474
 475         * configure.ac: Check for clock_gettime().
 476         * configure: Regenerated.
 477         * config.h.in: Regenerated.
 478         * intrinsics/time_1.h (__time_1): Rename to gf_cputime, add
 479         times() fallback.
 480         (gf_gettime): New function.
 481         * intrinsics/cpu_time.c (__cpu_time_1): Update to call gf_cputime.
 482         * intrinsics/date_and_time.c (date_and_time): Use gf_gettime.
 483         * intrinsics/dtime.c (dtime_sub): Use gf_cputime.
 484         * intrinsics/etime.c (etime_sub): Use gf_cputime.
 485         * intrinsics/system_clock.c (system_clock_4): Use gf_gettime.
 486         (system_clock_8): Use gf_gettime, increase count rate to allow
 487         nanosecond precision, remove overflow prone branch.
 488
 489 2011-01-29  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 490
 491         PR libgfortran/47434
 492         * io/write_float.def (write_infnan): Use calculate_sign to determine
 493         if the sign should be given and check field widths accordingly.
 494
 495 2011-01-29  Kai Tietz  <kai.tietz@onevision.com>
 496
 497         * intrinsics/ctime.c (ctime_r): Improve implementation.
 498
 499 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 500
 501         PR libfortran/47431
 502         * config.h.in: Regenerated.
 503         * configure: Regenerated.
 504         * configure.ac: Add check for ctime_r().
 505         * intrinsics/ctime.c (ctime_r): Fallback implementation.
 506         (fdate): Use ctime_r() instead of ctime().
 507         (fdate_sub): Likewise.
 508         (ctime): Likewise.
 509         (ctime_sub): Likewise.
 510
 511 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 512
 513         PR libfortran/47432
 514         * config.h.in: Regenerated.
 515         * configure: Regenerated.
 516         * configure.ac: Add check for ttyname_r.
 517         * io/unix.h: Add TTY_NAME_MAX, change stream_ttyname prototype.
 518         * io/unix.c (stream_ttyname): Use ttyname_r if available, conform
 519         to new prototype.
 520         * io/inquire.c (inquire_via_unit): Use changed stream_ttyname.
 521         * io/intrinsics.c (ttynam_sub): Likewise.
 522         (ttynam): Likewise.
 523
 524 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 525
 526         PR libfortran/47491
 527         * configure.ac: Call AC_USE_SYSTEM_EXTENSIONS to enable common
 528         extensions.
 529         * config.h.in: Regenerate.
 530         * configure: Regenerate.
 531
 532 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 533
 534         PR libgfortran/47285
 535         * io/write_float.def (write_infnan): Adjust processor selected width
 536         to 3 if NaN.
 537
 538 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 539
 540         PR libgfortran/47285
 541         * io/write_float.def (output_float): Return SUCCESS or FAILURE and use
 542         the result to set the padding.
 543
 544 2011-01-26  Kai Tietz  <kai.tietz@onevision.com>
 545
 546         * intrinsics/getlog.c (getlog): Fix label/statement issue.
 547
 548 2011-01-25  Janne Blomqvist  <jb@gcc.gnu.org>
 549
 550         PR libfortran/47375
 551         * config.h.in: Regenerated.
 552         * configure: Regenerated.
 553         * configure.ac: Add check for getpwuid_r.
 554         * intrinsics/getlog.c (getlog): Use getpwuid_r() if available.
 555
 556 2011-01-22  Janne Blomqvist  <jb@gcc.gnu.org>
 557
 558         PR libfortran/46267
 559         * config.h.in: Regenerated.
 560         * configure: Regenerated.
 561         * configure.ac: Check presence of strerror_r.
 562         * intrinsics/gerror.c (gerror): Use gf_strerror, modify logic.
 563         * io/unix.c (get_oserror): Remove.
 564         * libgfortran.h (gf_strerror): Add prototype.
 565         (get_oserror): Remove prototype.
 566         * runtime/error.c (gf_strerror): New function.
 567         (os_error): Use gf_strerror instead of get_oserror.
 568         (generate_errror): Likewise.
 569
 570 2011-01-17  Janne Blomqvist  <jb@gcc.gnu.org>
 571
 572         PR libfortran/47296
 573         * io/unix.c (tempfile): Set opp->file and opp->file_len also if an
 574         error occurs.
 575
 576 2011-01-16  Jakub Jelinek  <jakub@redhat.com>
 577
 578         PR fortran/46625
 579         * io/write_float.def (DTOAQ): Use quadmath_flt128tostr
 580         instead of quadmath_dtoa.
 581         * io/transfer128.c (tmp1, tmp2): New variables, bring in
 582         strtoflt128 and quadmath_flt128tostr.
 583         (transfer_real128, transfer_real128_write, transfer_complex128,
 584         transfer_complex128_write): Remove tmp1/tmp2 variables.
 585         * io/read.c (convert_real): Use strtoflt128 instead of
 586         quadmath_strtopQ, adjust for the changed arguments and return
 587         value.
 588
 589 2011-01-14  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 590
 591         PR libgfortran/47296
 592         * io/unix.c (unpack_filename): Return non-zero if the filename passed
 593         in is NULL.
 594
 595 2011-01-04  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 596
 597         PR libgfortran/47154
 598         * io/list_read.c (namelist_read): Remove calls to hit_eof to avoid the
 599         duplicate calls via next_record.
 600
 601 \f
 602 Copyright (C) 2011 Free Software Foundation, Inc.
 603
 604 Copying and distribution of this file, with or without modification,
 605 are permitted in any medium without royalty provided the copyright
 606 notice and this notice are preserved.