libgfortran/ChangeLog

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