libgfortran/ChangeLog

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