libgfortran/ChangeLog

   1 2011-06-18  Janne Blomqvist  <jb@gcc.gnu.org>
   2
   3         PR libfortran/49296
   4         * io/list_read.c (read_character): Accept EOF as a separator when
   5         reading string.
   6
   7 2011-06-17  Daniel Carrera  <dcarrera@gmail.com>
   8
   9         * caf/single.c (_gfortran_caf_register): Store the address
  10         of all static coarrays in a linked list.
  11         (_gfortran_caf_finalize): Free memory of staic coarrays.
  12         * caf/mpi.c (_gfortran_caf_register): Store the address
  13         of all static coarrays in a linked list. Initialize MPI
  14         if necessary.
  15         (_gfortran_caf_finalize): Free memory of staic coarrays.
  16         (_gfortran_caf_init): Check if MPI is already initialized
  17         before initializing again.
  18         * caf/libcaf.h: Add a type to caf_register_t to distinguish
  19         static coarrays and add the type caf_static_t to make the
  20         linked list of static coarrays.
  21
  22 2011-06-11  Janne Blomqvist  <jb@gcc.gnu.org>
  23
  24         * io/unix.c (buf_seek): Return error if file is not seekable.
  25         (buf_tell): Call buf_seek.
  26
  27 2011-06-11  Janne Blomqvist  <jb@gcc.gnu.org>
  28
  29         * io/unix.c (fd_to_stream): Figure out if a fd is seekable by
  30         trying lseek().
  31
  32 2011-06-10  Daniel Carrera  <dcarrera@gmail.com>
  33
  34         * caf/mpi.c (_gfortran_caf_sync_all,
  35         _gfortran_caf_sync_images): Functions have void return type
  36         and move status into parameter list.
  37         * caf/single.c (_gfortran_caf_sync_all,
  38         _gfortran_caf_sync_images): Functions have void return type
  39         and move status into parameter list.
  40         * caf/libcaf.h (_gfortran_caf_sync_all,
  41         _gfortran_caf_sync_images): Functions have void return type
  42         and move status into parameter list.
  43
  44 2011-06-03  Richard Henderson  <rth@redhat.com>
  45             Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
  46
  47         * config/fpu-387.h [__sun__ && __svr4__] (sigill_hdlr): Correct
  48         insn, insn size.
  49         (has_sse) [!__x86_64__ && __sun__ && __svr4__]: Use movaps.
  50
  51 2011-05-29  Janne Blomqvist  <jb@gcc.gnu.org>
  52
  53         PR libfortran/48931
  54         * libgfortran.h (find_addr2line): New prototype.
  55         * runtime/backtrace.c (show_backtrace): Use async-signal-safe
  56         execve and stored path of addr2line.
  57         * runtime/compile_options.c (maybe_find_addr2line): New function.
  58         (set_options): Call maybe_find_addr2line if backtracing is enabled.
  59         * runtime/main.c (find_addr2line): New function.
  60         (init): Call find_addr2line if backtracing is enabled.
  61         (cleanup): Free addr2line_path.
  62
  63 2011-05-29  Janne Blomqvist  <jb@gcc.gnu.org>
  64
  65         PR libfortran/49214
  66         * runtime/backtrace.c (fd_gets): Return NULL if nothing was read.
  67
  68 2011-05-29  Janne Blomqvist  <jb@gcc.gnu.org>
  69
  70         PR libfortran/19155
  71         * io/read.c (convert_real): Check for invalid input by comparing
  72         endptr instead of EINVAL.
  73         (read_f): Fixup floating point input without significand.
  74
  75 2011-05-22  Tobias Burnus  <burnus@net-b.de>
  76
  77         * runtime/stop.c (stop_string,error_stop_string): Minor cleanup.
  78
  79 2011-05-22  Janne Blomqvist  <jb@gcc.gnu.org>
  80
  81         PR libfortran/48931
  82         * configure.ac: Check for backtrace_symbols_fd instead of
  83         backtrace_symbols, check for readlink.
  84         * config.h.in: Regenerated.
  85         * configure: Regenerated.
  86         * runtime/backtrace.c (local_strcasestr): Remove.
  87         (bt_header): New function.
  88         (dump_glibc_backtrace): Remove.
  89         (fd_gets): New function.
  90         (show_backtrace): Rework to use backtrace_symbols_fd and pipes
  91         avoiding functions that are not async-signal-safe, reformat
  92         output.
  93         * runtime/main.c (store_exe_path): Try to check /proc/self/exe
  94         first.
  95
  96 2011-05-20  Janne Blomqvist  <jb@gcc.gnu.org>
  97             Uros Bizjak  <ubizjak@gmail.com>
  98
  99         PR libfortran/48977
 100         * configure.host: Swap order of glibc and x86 tests.
 101
 102 2011-05-20  Janne Blomqvist  <jb@gcc.gnu.org>
 103
 104         * config/fpu-387.h (set_fpu): Use renamed inexact macro.
 105         * config/fpu-aix.h (set_fpu): Clarify error messages, use renamed
 106         inexact macro, set TRP_INEXACT for inexact exception instead of
 107         TRP_UNDERFLOW.
 108         * config/fpu-generic.h (set_fpu): Clarify error messages, use
 109         renamed inexact macro.
 110         * config/fpu-glibc.h (set_fpu): Likewise.
 111         * config/fpu-sysv.h (set_fpu): Likewise.
 112
 113 2011-05-14  Tobias Burnus  <burnus@net-b.de>
 114
 115         * runtime/stop.c (error_stop_string, error_stop_numeric):
 116         Do not backtrace for ERROR STOP.
 117
 118 2011-05-14  Janne Blomqvist  <jb@gcc.gnu.org>
 119
 120         PR libfortran/48915
 121         * intrinsics/abort.c (abort): Call sys_abort().
 122         * io/unix.c (st_vprintf): Call sys_abort().
 123         * libgfortran.h (options_t): Remove dump_core member.
 124         (handler): Rename to backtrace_handler.
 125         (compile_options_t): Remove dump_core member.
 126         (sys_exit): Remove.
 127         (sys_abort): New function.
 128         * runtime/backtrace.c (show_backtrace): Print a message if
 129         backtrace is not available on target, update filter symbol name
 130         due to backtrace_handler renaming.
 131         * runtime/compile_options.c (backtrace_handler): New
 132         implementation based on example in Glibc manual.
 133         (set_options): Remove dump_core handling, make signal handler
 134         handle more signals.
 135         (init_compile_options): Remove dump_core.
 136         * runtime/environ.c (variable_table[]): Remove
 137         GFORTRAN_ERROR_DUMPCORE element.
 138         (show_variables): Update name, call exit().
 139         * runtime/error.c (sys_exit): Remove.
 140         (sys_abort): New function.
 141         (recursion_check): Call sys_abort().
 142         (os_error): Call exit().
 143         (runtime_error): Likewise.
 144         (runtime_error_at): Likewise.
 145         (internal_error): Likewise.
 146         (generate_error): Likewise.
 147         (notify_std): Likewise.
 148         * runtime/stop.c (stop_numeric): Call exit().
 149         (stop_numeric_f08): Likewise.
 150         (stop_string): Likewise.
 151         (error_stop_string): Call sys_abort().
 152         (error_stop_numeric): Print backtrace, exit with provided code.
 153
 154 2011-05-14  Janne Blomqvist  <jb@gcc.gnu.org>
 155
 156         * io/unix.c (st_vprintf,st_printf): Move to runtime/error.c.
 157         * libgfortran.h (struct options_t): Remove use_stderr field.
 158         (st_vprintf,st_printf): Move prototypes.
 159         (estr_write): New prototype.
 160         * runtime/error.c (sys_exit): Use estr_write instead of st_printf.
 161         (estr_write): New function.
 162         (st_vprintf): Move from io/unix.c, use stack allocated buffer,
 163         always output to stderr.
 164         (st_printf): Move from io/unix.c.
 165         (show_locus): Use a local variable instead of static.
 166         (os_error): Use estr_write instead of st_printf.
 167         (runtime_error): Likewise.
 168         (runtime_error_at): Likewise.
 169         (runtime_warning_at): Likewise.
 170         (internal_error): Likewise.
 171         (generate_error): Likewise.
 172         (generate_warning): Likewise.
 173         (notify_std): Likewise.
 174         * runtime/pause.c (do_pause): Likewise.
 175         (pause_string): Likewise.
 176         * runtime/stop.c (stop_string): Likewise.
 177         (error_stop_string): Likewise.
 178         * config/fpu_aix.h (set_fpu): Likewise.
 179         * config/fpu_generic.h (set_fpu): Likewise.
 180         * config/fpu_glibc.h (set_fpu): Likewise.
 181         * config/fpu-sysv.h (set_fpu): Likewise.
 182         * runtime/backtrace.c (dump_glibc_backtrace): Likewise.
 183         (show_backtrace): Likewise.
 184         * runtime/environ.c (print_spaces): Likewise.
 185         (show_string): Likewise.
 186         (show_variables): Likewise.
 187         (variable_table[]): Remove GFORTRAN_USE_STDERR entry.
 188
 189 2011-05-14  Tobias Burnus  <burnus@net-b.de>
 190
 191         PR fortran/48961
 192         * intrinsics/execute_command_line.c (set_cmdstat): Don't abort if
 193         synchronously executing with WAIT=.false.
 194         (execute_command_line): Fix setting of cmdstat and exitstat.
 195
 196 2011-05-06  Tobias Burnus  <burnus@net-b.de>
 197
 198         PR fortran/18918
 199         * caf/libcaf.h: Cleanup headers.
 200         (_gfortran_caf_critical, _gfortran_caf_end_critical): Make stub.
 201         (caf_register_t): New enum.
 202         (_gfortran_caf_register, _gfortran_caf_deregister): New prototype.
 203         * caf/single.c (_gfortran_caf_critical,
 204         _gfortran_caf_end_critical): Remove.
 205         (_gfortran_caf_register, _gfortran_caf_deregister): New functions.
 206         * caf/mpi.c (_gfortran_caf_critical,
 207         _gfortran_caf_end_critical): Remove.
 208         (_gfortran_caf_register, _gfortran_caf_deregister): New functions.
 209         (caf_world_window): Remove global variable.
 210         (_gfortran_caf_init): Fix off-by-one error of this_image.
 211
 212 2011-05-04  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 213
 214         PR libgfortran/48787
 215         * io/write_float.def (output_float): Adjust up and down rounding for
 216         cases where 'd' = 0. Gather common code to one location.
 217
 218 2011-05-01  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 219
 220         PR libgfortran/48787
 221         * io/write_float.def (output_float): Gather up integer declarations and
 222         add new 'p' for scale factor. Use 'p' in place of the 'dtp' reference
 223         everywhere. For ROUND_UP scan the digit string and only perform
 224         rounding if something other than '0' is found.
 225
 226 2011-04-29  Janne Blomqvist  <jb@gcc.gnu.org>
 227
 228         * io/unix.c (min): New macro.
 229         (unpack_filename): Return errno number for errors.
 230         (regular_file): Use appropriately sized buffer for path.
 231         (compare_file_filename): Likewise.
 232         (find_file): Likewise.
 233         (delete_file): Likewise.
 234         (file_exists): Likewise.
 235         (file_size): Likewise.
 236         (inquire_sequential): Likewise.
 237         (inquire_direct): Likewise.
 238         (inquire_formatted): Likewise.
 239         (inquire_access): Likewise.
 240
 241 2011-04-29  Janne Blomqvist  <jb@gcc.gnu.org>
 242
 243         PR libfortran/48488
 244         * io/write.c (write_real, write_real_g0): Update comments.
 245
 246 2011-04-29  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 247             Janne Blomqvist  <jb@gcc.gnu.org>
 248
 249         PR libgfortran/48488
 250         PR libgfortran/48602
 251         PR libgfortran/48615
 252         PR libgfortran/48684
 253         PR libgfortran/48787
 254         * io/write.c (write_d, write_e, write_f, write_en,
 255         write_es): Add precision compemsation parameter to call.
 256         (set_fnode_default): Adjust default widths to assure
 257         round trip on write and read. (write_real): Adjust call to write_float.
 258         (write_real_g0): Calculate compensation for extra precision and adjust
 259         call to write_float.
 260         * io/write_float.def (output_float_FMT_G_): Use volatile rather than
 261         asm volatile to avoid optimization issue. Correctly calculate the
 262         number of blanks (nb) to be appended and simplify calculation logic.
 263         (write_float): Increase MIN_FIELD_WIDTH by one to accomodate the new
 264         default widths. Eliminate the code that attempted to reduce the
 265         the precision used in later sprintf functions.  Add call parameter to
 266         compensate for extra precision.
 267
 268 2011-04-20  Jim Meyering  <meyering@redhat.com>
 269
 270         * intrinsics/move_alloc.c (move_alloc): Remove useless
 271         if-before-free test.
 272         * io/fbuf.c (fbuf_destroy): Likewise.
 273         * io/format.c (save_parsed_format): Likewise.
 274         * io/open.c (already_open): Likewise.
 275         * io/unit.c (free_internal_unit, close_unit_1): Likewise.
 276         * io/unix.c (mem_close): Likewise.
 277
 278 2011-04-19  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 279
 280         PR libgfortran/48602
 281         * io/write_float.def (output_float_FMT_G): Fix reversal in conditional.
 282         Use asm volatile to mark temp variable, avoiding optimization errors.
 283
 284 2011-04-17  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 285
 286         PR libgfortran/48602
 287         * io/write_float.def (output_float_FMT_G): Use current rounding mode
 288         to set the rounding parameters. (output_float): Skip rounding
 289         if value is zero.
 290
 291 2011-04-16  Janne Blomqvist  <jb@gcc.gnu.org>
 292
 293         * intrinsics/date_and_time.c (date_and_time): Remove sprintf CPP
 294         branch.
 295         * io/format.c (format_error): Use snprintf instead of sprintf.
 296         * io/list_read.c: Move snprintf fallback macro to libgfortran.h.
 297         (convert_integer): Use snprintf instead of sprintf.
 298         (parse_repeat): Likewise.
 299         (read_logical): Likewise.
 300         (read_integer): Likewise.
 301         (read_character): Likewise.
 302         (parse_real): Likewise.
 303         (read_complex): Likewise.
 304         (read_real): Likewise.
 305         (check_type): Likewise.
 306         (nml_parse_qualifier): Add string length argument, use snprintf
 307         instead of sprintf.
 308         (nml_get_obj_data): Use snprintf instead of sprintf.
 309         * io/open.c (new_unit): Remove sprintf CPP branch, use snprintf
 310         instead of sprintf.
 311         * io/transfer.c (require_type): Use snprintf instead of sprintf.
 312         * io/unix.c (tempfile): Likewise.
 313         * io/write.c (nml_write_obj): Likewise.
 314         * io/write_float.def (output_float): Remove sprintf CPP branch,
 315         use snprintf instead of sprintf.
 316         * libgfortran.h: Add fallback snprintf macro from io/list_read.c.
 317         * runtime/backtrace.c (show_backtrace): Remove sprintf CPP branch.
 318         * runtime/main.c (store_exe_path): Use snprintf instead of
 319         sprintf.
 320
 321 2011-04-15  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 322
 323         PR libgfortran/48589
 324         * io/write_float.def (write_infnan): Set width properly for G0.
 325
 326 2011-04-15  Tobias Burnus  <burnus@net-b.de>
 327
 328         PR fortran/18918
 329         * Makefile.am: Build and install caf/single.c as
 330         libcaf_single.a.
 331         * Makefile.in: Regenerate.
 332
 333 2011-04-15  Janne Blomqvist  <jb@gcc.gnu.org>
 334
 335         PR libfortran/47571
 336         * configure: Regenerated.
 337         * config.h.in: Regenerated.
 338         * acinclude.m4: Add alpha*-dec-osf* to gthread blacklist.
 339         * configure.ac: Use separate symbol for clock_gettime in librt.
 340         * intrinsics/system_clock.c: Use weakrefs only when needed and
 341         supported.
 342
 343 2011-04-12  Janne Blomqvist  <jb@gcc.gnu.org>
 344
 345         * configure.ac: Use AC_TYPE_* to make sure we have (u)intptr_t,
 346         check for ptrdiff_t.
 347         * libgfortran.h: Remove (u)intptr_t definitions, use ptrdiff_t for
 348         index_type, change cshift0* prototypes.
 349         * config.h.in: Regenerated.
 350         * configure: Regenerated.
 351         * intrinsics/cshift0.c (cshift0): Use ptrdiff_t instead of ssize_t.
 352         * io/io.h (array_loop_spec): Use index_type instead of ssize_t.
 353         * io/list_read.c (nml_parse_qualifier): Likewise.
 354         * io/write.c (nml_write_obj): Likewise.
 355         * m4/cshift0.c (cshift0_'rtype_code`): Likewise.
 356         * generated/cshift0_*.c: Regenerated.
 357
 358 2011-04-12  Janne Blomqvist  <jb@gcc.gnu.org>
 359
 360         * Makefile.am: Remove _GNU_SOURCE from AM_CPPFLAGS.
 361         * Makefile.in: Regenerated.
 362
 363 2011-04-06  Tobias Burnus  <burnus@net-b.de>
 364
 365         PR fortran/18918
 366         * caf/mpi.c (_gfortran_caf_init, _gfortran_caf_finalize):
 367         Add global variable caf_mpi_initialized and use it for when
 368         finalizing.
 369
 370 2011-04-04  Tobias Burnus  <burnus@net-b.de>
 371
 372         * unix.c: Adapt stat DEFINEs since MinGW64 supports LFS.
 373         (fallback_access, open_internal4, compare_file_filename,
 374         find_file, file_size, inquire_sequential, inquire_direct,
 375         inquire_formatted): Use "struct stat" instead of gfstat_t.
 376
 377 2011-03-27  Tobias Burnus  <burnus@net-b.de>
 378
 379         PR fortran/18918
 380         * caf/libcaf.h: New - coarray communication library.
 381         * caf/mpi.c: New.
 382         * caf/single.c: New.
 383
 384 2011-03-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 385
 386         PR libgfortran/48030
 387         * io/read.c (read_x): Re-implement using fbuf_getc.
 388
 389 2011-03-21  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
 390
 391         PR bootstrap/48135
 392         * configure.ac (gfortran_use_symver): Handle --disable-symvers.
 393         * configure: Regenerate.
 394
 395 2011-03-19  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
 396
 397         PR fortran/35667
 398         * io/io.h (convert_infnan): Declare.
 399         * io/read.c (convert_infnan): New.
 400         (read_f): Use convert_infnan to convert INFs and NANs.
 401         * list_read.c (parse_real, read_real): Likewise.
 402
 403 2011-03-19  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 404
 405         PR libfortran/47439
 406         * io/unix.c (tempfile): Work around poor mktemp() implementations.
 407
 408 2011-03-16  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 409
 410         PR libfortran/47883
 411         * acinclude.m4: Don't use AC_TRY_ macros.
 412         * configure.ac: Likewise.
 413         * configure: Regenerate.
 414
 415 2011-03-12  Thomas Koenig  <tkoenig@gcc.gnu.org>
 416
 417         PR libfortran/48066
 418         * m4/ifunction.m4:  If return array is empty, return.
 419         * m4/ifunction_logical.m4:  Likewise.
 420         * generated/all_l16.c: Regenerated.
 421         * generated/all_l1.c: Regenerated.
 422         * generated/all_l2.c: Regenerated.
 423         * generated/all_l4.c: Regenerated.
 424         * generated/all_l8.c: Regenerated.
 425         * generated/any_l16.c: Regenerated.
 426         * generated/any_l1.c: Regenerated.
 427         * generated/any_l2.c: Regenerated.
 428         * generated/any_l4.c: Regenerated.
 429         * generated/any_l8.c: Regenerated.
 430         * generated/count_16_l.c: Regenerated.
 431         * generated/count_1_l.c: Regenerated.
 432         * generated/count_2_l.c: Regenerated.
 433         * generated/count_4_l.c: Regenerated.
 434         * generated/count_8_l.c: Regenerated.
 435         * generated/maxloc1_16_i16.c: Regenerated.
 436         * generated/maxloc1_16_i1.c: Regenerated.
 437         * generated/maxloc1_16_i2.c: Regenerated.
 438         * generated/maxloc1_16_i4.c: Regenerated.
 439         * generated/maxloc1_16_i8.c: Regenerated.
 440         * generated/maxloc1_16_r10.c: Regenerated.
 441         * generated/maxloc1_16_r16.c: Regenerated.
 442         * generated/maxloc1_16_r4.c: Regenerated.
 443         * generated/maxloc1_16_r8.c: Regenerated.
 444         * generated/maxloc1_4_i16.c: Regenerated.
 445         * generated/maxloc1_4_i1.c: Regenerated.
 446         * generated/maxloc1_4_i2.c: Regenerated.
 447         * generated/maxloc1_4_i4.c: Regenerated.
 448         * generated/maxloc1_4_i8.c: Regenerated.
 449         * generated/maxloc1_4_r10.c: Regenerated.
 450         * generated/maxloc1_4_r16.c: Regenerated.
 451         * generated/maxloc1_4_r4.c: Regenerated.
 452         * generated/maxloc1_4_r8.c: Regenerated.
 453         * generated/maxloc1_8_i16.c: Regenerated.
 454         * generated/maxloc1_8_i1.c: Regenerated.
 455         * generated/maxloc1_8_i2.c: Regenerated.
 456         * generated/maxloc1_8_i4.c: Regenerated.
 457         * generated/maxloc1_8_i8.c: Regenerated.
 458         * generated/maxloc1_8_r10.c: Regenerated.
 459         * generated/maxloc1_8_r16.c: Regenerated.
 460         * generated/maxloc1_8_r4.c: Regenerated.
 461         * generated/maxloc1_8_r8.c: Regenerated.
 462         * generated/maxval_i16.c: Regenerated.
 463         * generated/maxval_i1.c: Regenerated.
 464         * generated/maxval_i2.c: Regenerated.
 465         * generated/maxval_i4.c: Regenerated.
 466         * generated/maxval_i8.c: Regenerated.
 467         * generated/maxval_r10.c: Regenerated.
 468         * generated/maxval_r16.c: Regenerated.
 469         * generated/maxval_r4.c: Regenerated.
 470         * generated/maxval_r8.c: Regenerated.
 471         * generated/minloc1_16_i16.c: Regenerated.
 472         * generated/minloc1_16_i1.c: Regenerated.
 473         * generated/minloc1_16_i2.c: Regenerated.
 474         * generated/minloc1_16_i4.c: Regenerated.
 475         * generated/minloc1_16_i8.c: Regenerated.
 476         * generated/minloc1_16_r10.c: Regenerated.
 477         * generated/minloc1_16_r16.c: Regenerated.
 478         * generated/minloc1_16_r4.c: Regenerated.
 479         * generated/minloc1_16_r8.c: Regenerated.
 480         * generated/minloc1_4_i16.c: Regenerated.
 481         * generated/minloc1_4_i1.c: Regenerated.
 482         * generated/minloc1_4_i2.c: Regenerated.
 483         * generated/minloc1_4_i4.c: Regenerated.
 484         * generated/minloc1_4_i8.c: Regenerated.
 485         * generated/minloc1_4_r10.c: Regenerated.
 486         * generated/minloc1_4_r16.c: Regenerated.
 487         * generated/minloc1_4_r4.c: Regenerated.
 488         * generated/minloc1_4_r8.c: Regenerated.
 489         * generated/minloc1_8_i16.c: Regenerated.
 490         * generated/minloc1_8_i1.c: Regenerated.
 491         * generated/minloc1_8_i2.c: Regenerated.
 492         * generated/minloc1_8_i4.c: Regenerated.
 493         * generated/minloc1_8_i8.c: Regenerated.
 494         * generated/minloc1_8_r10.c: Regenerated.
 495         * generated/minloc1_8_r16.c: Regenerated.
 496         * generated/minloc1_8_r4.c: Regenerated.
 497         * generated/minloc1_8_r8.c: Regenerated.
 498         * generated/minval_i16.c: Regenerated.
 499         * generated/minval_i1.c: Regenerated.
 500         * generated/minval_i2.c: Regenerated.
 501         * generated/minval_i4.c: Regenerated.
 502         * generated/minval_i8.c: Regenerated.
 503         * generated/minval_r10.c: Regenerated.
 504         * generated/minval_r16.c: Regenerated.
 505         * generated/minval_r4.c: Regenerated.
 506         * generated/minval_r8.c: Regenerated.
 507         * generated/product_c10.c: Regenerated.
 508         * generated/product_c16.c: Regenerated.
 509         * generated/product_c4.c: Regenerated.
 510         * generated/product_c8.c: Regenerated.
 511         * generated/product_i16.c: Regenerated.
 512         * generated/product_i1.c: Regenerated.
 513         * generated/product_i2.c: Regenerated.
 514         * generated/product_i4.c: Regenerated.
 515         * generated/product_i8.c: Regenerated.
 516         * generated/product_r10.c: Regenerated.
 517         * generated/product_r16.c: Regenerated.
 518         * generated/product_r4.c: Regenerated.
 519         * generated/product_r8.c: Regenerated.
 520         * generated/sum_c10.c: Regenerated.
 521         * generated/sum_c16.c: Regenerated.
 522         * generated/sum_c4.c: Regenerated.
 523         * generated/sum_c8.c: Regenerated.
 524         * generated/sum_i16.c: Regenerated.
 525         * generated/sum_i1.c: Regenerated.
 526         * generated/sum_i2.c: Regenerated.
 527         * generated/sum_i4.c: Regenerated.
 528         * generated/sum_i8.c: Regenerated.
 529         * generated/sum_r10.c: Regenerated.
 530         * generated/sum_r16.c: Regenerated.
 531         * generated/sum_r4.c: Regenerated.
 532         * generated/sum_r8.c: Regenerated.
 533
 534 2011-03-11  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 535
 536         PR libgfortran/48047
 537         * io/write_float.def (write_float): Change MIN_FIELD_WIDTH to 48.
 538
 539 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 540
 541         PR libfortran/47802
 542         * intrinsics/ctime.c (strctime): Use builtins to check localtime_r
 543         return type.
 544
 545 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
 546
 547         PR libfortran/47802
 548         * intrinsics/ctime.c (strctime): Don't use return value of
 549         localtime_r.
 550
 551 2011-02-28  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 552
 553         PR libgfortran/47567
 554         * io/write_float.def (output_float): Move handling of w = 0 to after
 555         output rounding. Check for zero and set zero_flag accordingly. Set
 556         width according to zero_flag. Add better comments.
 557
 558 2011-02-27  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 559
 560         PR libgfortran/47778
 561         * io/list_read.c (namelist_read): Intialize the error string buffere.
 562         If pprev_nl was used during the previous namelist read and the rank
 563         was zero, reset the pointer to NULL for the next namelist read.
 564
 565 2011-02-26  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 566
 567         PR libfortran/45165
 568         * unix.c (fallback_access): Fix file descriptor leaks.
 569
 570 2011-02-25  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 571
 572         * acinclude.m4 (LIBGFOR_CHECK_FPSETMASK): Set shell variable
 573         tested in configure.host.
 574         * configure: Regenerate.
 575
 576 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 577
 578         PR libfortran/47802
 579         * config.h.in: Regenerated.
 580         * configure: Regenerated.
 581         * configure.ac: Remove checks for ctime and ctime_r, add check for
 582         strftime.
 583         * intrinsics/date_and_time.c (localtime_r): Move fallback
 584         implementation to time_1.h.
 585         * intrinsics/time_1.h (localtime_r): Fallback implementation.
 586         * intrinsics/ctime.c: Include time_1.h.
 587         (ctime_r): Remove fallback implementation.
 588         (strctime): New function.
 589         (fdate): Use strctime instead of ctime_r.
 590         (fdate_sub): Likewise.
 591         (ctime): Likewise.
 592         (ctime_sub): Likewise.
 593
 594 2011-02-24  Jakub Jelinek  <jakub@redhat.com>
 595
 596         PR fortran/47878
 597         * io/transfer.c (read_sf): Call fbuf_getptr only at the end,
 598         and subtract n, dtp->u.p.sf_seen_eor and seen_comma from it.
 599
 600 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
 601
 602         PR libfortran/47802
 603         * configure.ac: Add test for POSIX getpwuid_r.
 604         * intrinsics/getlog.c (getlog): CPP test for
 605         HAVE_POSIX_GETPWUID_R.
 606         * config.h.in: Regenerated.
 607         * configure: Regenerated.
 608
 609 2011-02-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 610
 611         PR libgfortran/47567
 612         * io/write_float.def (output_float): Remove special case handling of
 613         zero with width 1.
 614
 615 2011-02-23  Janne Blomqvist  <jb@gcc.gnu.org>
 616             Jerry DeLisle    <jvdelisle@gcc.gnu.org>
 617
 618         PR libfortran/47694
 619         * io/fbuf.h (fbuf_getptr): New inline function.
 620         * io/transfer.c (read_sf): Use fbuf_getptr and fbuf_getc to scan
 621         through the string instead of fbuf_read.
 622
 623 2011-02-22  Tobias Burnus  <burnus@net-b.de>
 624             Kai-Uwe Eckhardt  <kuehro@gmx.de>
 625
 626         PR libfortran/47830
 627         * intrinsics/c99_functions.c (roundl): Make C valid for
 628         HAVE_NEXTAFTERL.
 629
 630 2011-02-19  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 631
 632         PR libgfortran/47567
 633         * io/write_float.def (output_float): Adjust width for F0.d to
 634         allow space for negative signs on zero.
 635
 636 2011-02-16  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 637
 638         PR libgfortran/47667
 639         * io/list_read.c (read_logical): Check for end of line before calling
 640         eat_line. (read_integer): Likewise. (parse_real): Don't unget the
 641         separator. Check for end of line before calling eat_line.
 642         (read_complex): Allow line-end before and after parenthesis and comma.
 643         Check for end of line before calling eat_line. (read_real): Check for
 644         end of line before calling eat_line.
 645
 646 2011-02-16  Jakub Jelinek  <jakub@redhat.com>
 647
 648         PR libfortran/47757
 649         * gfortran.map (GFORTRAN_1.4): Export
 650         _gfortran_{m,s}i{all,any,parity}_i{1,2,4,8,16} and
 651         _gfortran_{cshift0,eoshift{0,2}}_16_char4.
 652
 653 2011-02-15  Tobias Burnus  <burnus@net-b.de>
 654
 655         PR fortran/47716
 656         PR fortran/47648
 657         * acinclude.m4 (LIBGFOR_CHECK_FLOAT128): Use check from
 658         libquadmath, which uses more features.
 659         * configure: Regenerate.
 660
 661 2011-02-14  Jakub Jelinek  <jakub@redhat.com>
 662
 663         PR fortran/47642
 664         * io/write_float.def (DTOAQ): Use quadmath_snprintf instead of
 665         quadmath_flt128tostr.
 666         * io/transfer128.c (tmp2): Initialize to quadmath_snprintf instead
 667         of quadmath_flt128tostr.
 668
 669 2011-02-13  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 670
 671         * Makefile.in: Regenerate.
 672         * aclocal.m4: Likewise.
 673         * configure: Likewise.
 674
 675 2011-02-05  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 676
 677         PR libgfortran/47567
 678         * io/write_float.def (output_float): Eliminate some redundant code.
 679         Adjust width for case of F0.X for values of zero and all other values.
 680         Expand cases where '*' is set to give cleaner results.
 681
 682 2011-02-05  Janne Blomqvist  <jb@gcc.gnu.org>
 683
 684         PR libfortran/47571
 685         * intrinsics/time_1.h (GF_CLOCK_MONOTONIC): Move to system_clock.c.
 686         (weak_gettime): Likewise.
 687         (gf_gettime): Change API, move weak_gettime() usage to
 688         system_clock.c
 689         * intrinsics/system_clock.c (GTHREAD_USE_WEAK): Define.
 690         (gf_gettime_mono): New function.
 691         (system_clock_4): Use gf_gettime_mono().
 692         (system_clock_8): Likewise.
 693         * intrinsics/date_and_time.c (date_and_time): Update gf_gettime()
 694         usage.
 695
 696 2011-02-02  Janne Blomqvist  <jb@gcc.gnu.org>
 697
 698         PR libfortran/47571
 699         * configure: Regenerated.
 700         * configure.ac: Don't add librt to LIBS.
 701         * intrinsics/time_1.h (weak_gettime): Weakref trickery for
 702         clock_gettime().
 703         (gf_gettime): Use weak_gettime() instead of clock_gettime().
 704
 705 2011-02-01  Janne Blomqvist  <jb@gcc.gnu.org>
 706
 707         * intrinsics/time_1.h: Include errno.h needed by fallbacks.
 708
 709 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 710
 711         * intrinsics/time_1.h: Fix definition of GF_CLOCK_MONOTONIC macro.
 712
 713 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
 714
 715         * configure.ac: Check for clock_gettime().
 716         * configure: Regenerated.
 717         * config.h.in: Regenerated.
 718         * intrinsics/time_1.h (__time_1): Rename to gf_cputime, add
 719         times() fallback.
 720         (gf_gettime): New function.
 721         * intrinsics/cpu_time.c (__cpu_time_1): Update to call gf_cputime.
 722         * intrinsics/date_and_time.c (date_and_time): Use gf_gettime.
 723         * intrinsics/dtime.c (dtime_sub): Use gf_cputime.
 724         * intrinsics/etime.c (etime_sub): Use gf_cputime.
 725         * intrinsics/system_clock.c (system_clock_4): Use gf_gettime.
 726         (system_clock_8): Use gf_gettime, increase count rate to allow
 727         nanosecond precision, remove overflow prone branch.
 728
 729 2011-01-29  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 730
 731         PR libgfortran/47434
 732         * io/write_float.def (write_infnan): Use calculate_sign to determine
 733         if the sign should be given and check field widths accordingly.
 734
 735 2011-01-29  Kai Tietz  <kai.tietz@onevision.com>
 736
 737         * intrinsics/ctime.c (ctime_r): Improve implementation.
 738
 739 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 740
 741         PR libfortran/47431
 742         * config.h.in: Regenerated.
 743         * configure: Regenerated.
 744         * configure.ac: Add check for ctime_r().
 745         * intrinsics/ctime.c (ctime_r): Fallback implementation.
 746         (fdate): Use ctime_r() instead of ctime().
 747         (fdate_sub): Likewise.
 748         (ctime): Likewise.
 749         (ctime_sub): Likewise.
 750
 751 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 752
 753         PR libfortran/47432
 754         * config.h.in: Regenerated.
 755         * configure: Regenerated.
 756         * configure.ac: Add check for ttyname_r.
 757         * io/unix.h: Add TTY_NAME_MAX, change stream_ttyname prototype.
 758         * io/unix.c (stream_ttyname): Use ttyname_r if available, conform
 759         to new prototype.
 760         * io/inquire.c (inquire_via_unit): Use changed stream_ttyname.
 761         * io/intrinsics.c (ttynam_sub): Likewise.
 762         (ttynam): Likewise.
 763
 764 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
 765
 766         PR libfortran/47491
 767         * configure.ac: Call AC_USE_SYSTEM_EXTENSIONS to enable common
 768         extensions.
 769         * config.h.in: Regenerate.
 770         * configure: Regenerate.
 771
 772 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 773
 774         PR libgfortran/47285
 775         * io/write_float.def (write_infnan): Adjust processor selected width
 776         to 3 if NaN.
 777
 778 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 779
 780         PR libgfortran/47285
 781         * io/write_float.def (output_float): Return SUCCESS or FAILURE and use
 782         the result to set the padding.
 783
 784 2011-01-26  Kai Tietz  <kai.tietz@onevision.com>
 785
 786         * intrinsics/getlog.c (getlog): Fix label/statement issue.
 787
 788 2011-01-25  Janne Blomqvist  <jb@gcc.gnu.org>
 789
 790         PR libfortran/47375
 791         * config.h.in: Regenerated.
 792         * configure: Regenerated.
 793         * configure.ac: Add check for getpwuid_r.
 794         * intrinsics/getlog.c (getlog): Use getpwuid_r() if available.
 795
 796 2011-01-22  Janne Blomqvist  <jb@gcc.gnu.org>
 797
 798         PR libfortran/46267
 799         * config.h.in: Regenerated.
 800         * configure: Regenerated.
 801         * configure.ac: Check presence of strerror_r.
 802         * intrinsics/gerror.c (gerror): Use gf_strerror, modify logic.
 803         * io/unix.c (get_oserror): Remove.
 804         * libgfortran.h (gf_strerror): Add prototype.
 805         (get_oserror): Remove prototype.
 806         * runtime/error.c (gf_strerror): New function.
 807         (os_error): Use gf_strerror instead of get_oserror.
 808         (generate_errror): Likewise.
 809
 810 2011-01-17  Janne Blomqvist  <jb@gcc.gnu.org>
 811
 812         PR libfortran/47296
 813         * io/unix.c (tempfile): Set opp->file and opp->file_len also if an
 814         error occurs.
 815
 816 2011-01-16  Jakub Jelinek  <jakub@redhat.com>
 817
 818         PR fortran/46625
 819         * io/write_float.def (DTOAQ): Use quadmath_flt128tostr
 820         instead of quadmath_dtoa.
 821         * io/transfer128.c (tmp1, tmp2): New variables, bring in
 822         strtoflt128 and quadmath_flt128tostr.
 823         (transfer_real128, transfer_real128_write, transfer_complex128,
 824         transfer_complex128_write): Remove tmp1/tmp2 variables.
 825         * io/read.c (convert_real): Use strtoflt128 instead of
 826         quadmath_strtopQ, adjust for the changed arguments and return
 827         value.
 828
 829 2011-01-14  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 830
 831         PR libgfortran/47296
 832         * io/unix.c (unpack_filename): Return non-zero if the filename passed
 833         in is NULL.
 834
 835 2011-01-04  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 836
 837         PR libgfortran/47154
 838         * io/list_read.c (namelist_read): Remove calls to hit_eof to avoid the
 839         duplicate calls via next_record.
 840
 841 \f
 842 Copyright (C) 2011 Free Software Foundation, Inc.
 843
 844 Copying and distribution of this file, with or without modification,
 845 are permitted in any medium without royalty provided the copyright
 846 notice and this notice are preserved.