OSDN Git Service

2011-05-04 Jerry DeLisle <jvdelisle@gcc.gnu.org>
[pf3gnuchains/gcc-fork.git] / libgfortran / ChangeLog
1 2011-05-04  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
2
3         PR libgfortran/48787
4         * io/write_float.def (output_float): Adjust up and down rounding for
5         cases where 'd' = 0. Gather common code to one location.
6
7 2011-05-01  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
8
9         PR libgfortran/48787
10         * io/write_float.def (output_float): Gather up integer declarations and
11         add new 'p' for scale factor. Use 'p' in place of the 'dtp' reference
12         everywhere. For ROUND_UP scan the digit string and only perform
13         rounding if something other than '0' is found.
14
15 2011-04-29  Janne Blomqvist  <jb@gcc.gnu.org>
16
17         * io/unix.c (min): New macro.
18         (unpack_filename): Return errno number for errors.
19         (regular_file): Use appropriately sized buffer for path.
20         (compare_file_filename): Likewise.
21         (find_file): Likewise.
22         (delete_file): Likewise.
23         (file_exists): Likewise.
24         (file_size): Likewise.
25         (inquire_sequential): Likewise.
26         (inquire_direct): Likewise.
27         (inquire_formatted): Likewise.
28         (inquire_access): Likewise.
29
30 2011-04-29  Janne Blomqvist  <jb@gcc.gnu.org>
31
32         PR libfortran/48488
33         * io/write.c (write_real, write_real_g0): Update comments.
34
35 2011-04-29  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
36             Janne Blomqvist  <jb@gcc.gnu.org>
37
38         PR libgfortran/48488
39         PR libgfortran/48602
40         PR libgfortran/48615
41         PR libgfortran/48684
42         PR libgfortran/48787
43         * io/write.c (write_d, write_e, write_f, write_en,
44         write_es): Add precision compemsation parameter to call.
45         (set_fnode_default): Adjust default widths to assure
46         round trip on write and read. (write_real): Adjust call to write_float.
47         (write_real_g0): Calculate compensation for extra precision and adjust
48         call to write_float.
49         * io/write_float.def (output_float_FMT_G_): Use volatile rather than
50         asm volatile to avoid optimization issue. Correctly calculate the
51         number of blanks (nb) to be appended and simplify calculation logic.
52         (write_float): Increase MIN_FIELD_WIDTH by one to accomodate the new
53         default widths. Eliminate the code that attempted to reduce the
54         the precision used in later sprintf functions.  Add call parameter to
55         compensate for extra precision.
56
57 2011-04-20  Jim Meyering  <meyering@redhat.com>
58
59         * intrinsics/move_alloc.c (move_alloc): Remove useless
60         if-before-free test.
61         * io/fbuf.c (fbuf_destroy): Likewise.
62         * io/format.c (save_parsed_format): Likewise.
63         * io/open.c (already_open): Likewise.
64         * io/unit.c (free_internal_unit, close_unit_1): Likewise.
65         * io/unix.c (mem_close): Likewise.
66
67 2011-04-19  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
68
69         PR libgfortran/48602
70         * io/write_float.def (output_float_FMT_G): Fix reversal in conditional.
71         Use asm volatile to mark temp variable, avoiding optimization errors.
72
73 2011-04-17  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
74
75         PR libgfortran/48602
76         * io/write_float.def (output_float_FMT_G): Use current rounding mode
77         to set the rounding parameters. (output_float): Skip rounding
78         if value is zero.
79         
80 2011-04-16  Janne Blomqvist  <jb@gcc.gnu.org>
81
82         * intrinsics/date_and_time.c (date_and_time): Remove sprintf CPP
83         branch.
84         * io/format.c (format_error): Use snprintf instead of sprintf.
85         * io/list_read.c: Move snprintf fallback macro to libgfortran.h.
86         (convert_integer): Use snprintf instead of sprintf.
87         (parse_repeat): Likewise.
88         (read_logical): Likewise.
89         (read_integer): Likewise.
90         (read_character): Likewise.
91         (parse_real): Likewise.
92         (read_complex): Likewise.
93         (read_real): Likewise.
94         (check_type): Likewise.
95         (nml_parse_qualifier): Add string length argument, use snprintf
96         instead of sprintf.
97         (nml_get_obj_data): Use snprintf instead of sprintf.
98         * io/open.c (new_unit): Remove sprintf CPP branch, use snprintf
99         instead of sprintf.
100         * io/transfer.c (require_type): Use snprintf instead of sprintf.
101         * io/unix.c (tempfile): Likewise.
102         * io/write.c (nml_write_obj): Likewise.
103         * io/write_float.def (output_float): Remove sprintf CPP branch,
104         use snprintf instead of sprintf.
105         * libgfortran.h: Add fallback snprintf macro from io/list_read.c.
106         * runtime/backtrace.c (show_backtrace): Remove sprintf CPP branch.
107         * runtime/main.c (store_exe_path): Use snprintf instead of
108         sprintf.
109
110 2011-04-15  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
111
112         PR libgfortran/48589
113         * io/write_float.def (write_infnan): Set width properly for G0.
114
115 2011-04-15  Tobias Burnus  <burnus@net-b.de>
116
117         PR fortran/18918
118         * Makefile.am: Build and install caf/single.c as
119         libcaf_single.a.
120         * Makefile.in: Regenerate.
121
122 2011-04-15  Janne Blomqvist  <jb@gcc.gnu.org>
123
124         PR libfortran/47571
125         * configure: Regenerated.
126         * config.h.in: Regenerated.
127         * acinclude.m4: Add alpha*-dec-osf* to gthread blacklist.
128         * configure.ac: Use separate symbol for clock_gettime in librt.
129         * intrinsics/system_clock.c: Use weakrefs only when needed and
130         supported.
131
132 2011-04-12  Janne Blomqvist  <jb@gcc.gnu.org>
133
134         * configure.ac: Use AC_TYPE_* to make sure we have (u)intptr_t,
135         check for ptrdiff_t.
136         * libgfortran.h: Remove (u)intptr_t definitions, use ptrdiff_t for
137         index_type, change cshift0* prototypes.
138         * config.h.in: Regenerated.
139         * configure: Regenerated.
140         * intrinsics/cshift0.c (cshift0): Use ptrdiff_t instead of ssize_t.
141         * io/io.h (array_loop_spec): Use index_type instead of ssize_t.
142         * io/list_read.c (nml_parse_qualifier): Likewise.
143         * io/write.c (nml_write_obj): Likewise.
144         * m4/cshift0.c (cshift0_'rtype_code`): Likewise.
145         * generated/cshift0_*.c: Regenerated.
146
147 2011-04-12  Janne Blomqvist  <jb@gcc.gnu.org>
148
149         * Makefile.am: Remove _GNU_SOURCE from AM_CPPFLAGS.
150         * Makefile.in: Regenerated.
151
152 2011-04-06  Tobias Burnus  <burnus@net-b.de>
153
154         PR fortran/18918
155         * caf/mpi.c (_gfortran_caf_init, _gfortran_caf_finalize):
156         Add global variable caf_mpi_initialized and use it for when
157         finalizing.
158
159 2011-04-04  Tobias Burnus  <burnus@net-b.de>
160
161         * unix.c: Adapt stat DEFINEs since MinGW64 supports LFS.
162         (fallback_access, open_internal4, compare_file_filename,
163         find_file, file_size, inquire_sequential, inquire_direct,
164         inquire_formatted): Use "struct stat" instead of gfstat_t.
165
166 2011-03-27  Tobias Burnus  <burnus@net-b.de>
167
168         PR fortran/18918
169         * caf/libcaf.h: New - coarray communication library.
170         * caf/mpi.c: New.
171         * caf/single.c: New.
172
173 2011-03-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
174
175         PR libgfortran/48030
176         * io/read.c (read_x): Re-implement using fbuf_getc.
177
178 2011-03-21  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
179
180         PR bootstrap/48135
181         * configure.ac (gfortran_use_symver): Handle --disable-symvers.
182         * configure: Regenerate.
183
184 2011-03-19  John David Anglin  <dave.anglin@nrc-cnrc.gc.ca>
185
186         PR fortran/35667
187         * io/io.h (convert_infnan): Declare.
188         * io/read.c (convert_infnan): New.
189         (read_f): Use convert_infnan to convert INFs and NANs.
190         * list_read.c (parse_real, read_real): Likewise.
191
192 2011-03-19  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
193
194         PR libfortran/47439
195         * io/unix.c (tempfile): Work around poor mktemp() implementations.
196
197 2011-03-16  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
198
199         PR libfortran/47883
200         * acinclude.m4: Don't use AC_TRY_ macros.
201         * configure.ac: Likewise.
202         * configure: Regenerate.
203
204 2011-03-12  Thomas Koenig  <tkoenig@gcc.gnu.org>
205
206         PR libfortran/48066
207         * m4/ifunction.m4:  If return array is empty, return.
208         * m4/ifunction_logical.m4:  Likewise.
209         * generated/all_l16.c: Regenerated.
210         * generated/all_l1.c: Regenerated.
211         * generated/all_l2.c: Regenerated.
212         * generated/all_l4.c: Regenerated.
213         * generated/all_l8.c: Regenerated.
214         * generated/any_l16.c: Regenerated.
215         * generated/any_l1.c: Regenerated.
216         * generated/any_l2.c: Regenerated.
217         * generated/any_l4.c: Regenerated.
218         * generated/any_l8.c: Regenerated.
219         * generated/count_16_l.c: Regenerated.
220         * generated/count_1_l.c: Regenerated.
221         * generated/count_2_l.c: Regenerated.
222         * generated/count_4_l.c: Regenerated.
223         * generated/count_8_l.c: Regenerated.
224         * generated/maxloc1_16_i16.c: Regenerated.
225         * generated/maxloc1_16_i1.c: Regenerated.
226         * generated/maxloc1_16_i2.c: Regenerated.
227         * generated/maxloc1_16_i4.c: Regenerated.
228         * generated/maxloc1_16_i8.c: Regenerated.
229         * generated/maxloc1_16_r10.c: Regenerated.
230         * generated/maxloc1_16_r16.c: Regenerated.
231         * generated/maxloc1_16_r4.c: Regenerated.
232         * generated/maxloc1_16_r8.c: Regenerated.
233         * generated/maxloc1_4_i16.c: Regenerated.
234         * generated/maxloc1_4_i1.c: Regenerated.
235         * generated/maxloc1_4_i2.c: Regenerated.
236         * generated/maxloc1_4_i4.c: Regenerated.
237         * generated/maxloc1_4_i8.c: Regenerated.
238         * generated/maxloc1_4_r10.c: Regenerated.
239         * generated/maxloc1_4_r16.c: Regenerated.
240         * generated/maxloc1_4_r4.c: Regenerated.
241         * generated/maxloc1_4_r8.c: Regenerated.
242         * generated/maxloc1_8_i16.c: Regenerated.
243         * generated/maxloc1_8_i1.c: Regenerated.
244         * generated/maxloc1_8_i2.c: Regenerated.
245         * generated/maxloc1_8_i4.c: Regenerated.
246         * generated/maxloc1_8_i8.c: Regenerated.
247         * generated/maxloc1_8_r10.c: Regenerated.
248         * generated/maxloc1_8_r16.c: Regenerated.
249         * generated/maxloc1_8_r4.c: Regenerated.
250         * generated/maxloc1_8_r8.c: Regenerated.
251         * generated/maxval_i16.c: Regenerated.
252         * generated/maxval_i1.c: Regenerated.
253         * generated/maxval_i2.c: Regenerated.
254         * generated/maxval_i4.c: Regenerated.
255         * generated/maxval_i8.c: Regenerated.
256         * generated/maxval_r10.c: Regenerated.
257         * generated/maxval_r16.c: Regenerated.
258         * generated/maxval_r4.c: Regenerated.
259         * generated/maxval_r8.c: Regenerated.
260         * generated/minloc1_16_i16.c: Regenerated.
261         * generated/minloc1_16_i1.c: Regenerated.
262         * generated/minloc1_16_i2.c: Regenerated.
263         * generated/minloc1_16_i4.c: Regenerated.
264         * generated/minloc1_16_i8.c: Regenerated.
265         * generated/minloc1_16_r10.c: Regenerated.
266         * generated/minloc1_16_r16.c: Regenerated.
267         * generated/minloc1_16_r4.c: Regenerated.
268         * generated/minloc1_16_r8.c: Regenerated.
269         * generated/minloc1_4_i16.c: Regenerated.
270         * generated/minloc1_4_i1.c: Regenerated.
271         * generated/minloc1_4_i2.c: Regenerated.
272         * generated/minloc1_4_i4.c: Regenerated.
273         * generated/minloc1_4_i8.c: Regenerated.
274         * generated/minloc1_4_r10.c: Regenerated.
275         * generated/minloc1_4_r16.c: Regenerated.
276         * generated/minloc1_4_r4.c: Regenerated.
277         * generated/minloc1_4_r8.c: Regenerated.
278         * generated/minloc1_8_i16.c: Regenerated.
279         * generated/minloc1_8_i1.c: Regenerated.
280         * generated/minloc1_8_i2.c: Regenerated.
281         * generated/minloc1_8_i4.c: Regenerated.
282         * generated/minloc1_8_i8.c: Regenerated.
283         * generated/minloc1_8_r10.c: Regenerated.
284         * generated/minloc1_8_r16.c: Regenerated.
285         * generated/minloc1_8_r4.c: Regenerated.
286         * generated/minloc1_8_r8.c: Regenerated.
287         * generated/minval_i16.c: Regenerated.
288         * generated/minval_i1.c: Regenerated.
289         * generated/minval_i2.c: Regenerated.
290         * generated/minval_i4.c: Regenerated.
291         * generated/minval_i8.c: Regenerated.
292         * generated/minval_r10.c: Regenerated.
293         * generated/minval_r16.c: Regenerated.
294         * generated/minval_r4.c: Regenerated.
295         * generated/minval_r8.c: Regenerated.
296         * generated/product_c10.c: Regenerated.
297         * generated/product_c16.c: Regenerated.
298         * generated/product_c4.c: Regenerated.
299         * generated/product_c8.c: Regenerated.
300         * generated/product_i16.c: Regenerated.
301         * generated/product_i1.c: Regenerated.
302         * generated/product_i2.c: Regenerated.
303         * generated/product_i4.c: Regenerated.
304         * generated/product_i8.c: Regenerated.
305         * generated/product_r10.c: Regenerated.
306         * generated/product_r16.c: Regenerated.
307         * generated/product_r4.c: Regenerated.
308         * generated/product_r8.c: Regenerated.
309         * generated/sum_c10.c: Regenerated.
310         * generated/sum_c16.c: Regenerated.
311         * generated/sum_c4.c: Regenerated.
312         * generated/sum_c8.c: Regenerated.
313         * generated/sum_i16.c: Regenerated.
314         * generated/sum_i1.c: Regenerated.
315         * generated/sum_i2.c: Regenerated.
316         * generated/sum_i4.c: Regenerated.
317         * generated/sum_i8.c: Regenerated.
318         * generated/sum_r10.c: Regenerated.
319         * generated/sum_r16.c: Regenerated.
320         * generated/sum_r4.c: Regenerated.
321         * generated/sum_r8.c: Regenerated.
322
323 2011-03-11  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
324
325         PR libgfortran/48047
326         * io/write_float.def (write_float): Change MIN_FIELD_WIDTH to 48.
327
328 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
329
330         PR libfortran/47802
331         * intrinsics/ctime.c (strctime): Use builtins to check localtime_r
332         return type.
333
334 2011-03-04  Janne Blomqvist  <jb@gcc.gnu.org>
335
336         PR libfortran/47802
337         * intrinsics/ctime.c (strctime): Don't use return value of
338         localtime_r.
339
340 2011-02-28  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
341
342         PR libgfortran/47567
343         * io/write_float.def (output_float): Move handling of w = 0 to after
344         output rounding. Check for zero and set zero_flag accordingly. Set
345         width according to zero_flag. Add better comments.
346
347 2011-02-27  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
348
349         PR libgfortran/47778
350         * io/list_read.c (namelist_read): Intialize the error string buffere.
351         If pprev_nl was used during the previous namelist read and the rank
352         was zero, reset the pointer to NULL for the next namelist read.
353
354 2011-02-26  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
355
356         PR libfortran/45165
357         * unix.c (fallback_access): Fix file descriptor leaks.
358
359 2011-02-25  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
360
361         * acinclude.m4 (LIBGFOR_CHECK_FPSETMASK): Set shell variable
362         tested in configure.host.
363         * configure: Regenerate.
364
365 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
366
367         PR libfortran/47802
368         * config.h.in: Regenerated.
369         * configure: Regenerated.
370         * configure.ac: Remove checks for ctime and ctime_r, add check for
371         strftime.
372         * intrinsics/date_and_time.c (localtime_r): Move fallback
373         implementation to time_1.h.
374         * intrinsics/time_1.h (localtime_r): Fallback implementation.
375         * intrinsics/ctime.c: Include time_1.h.
376         (ctime_r): Remove fallback implementation.
377         (strctime): New function.
378         (fdate): Use strctime instead of ctime_r.
379         (fdate_sub): Likewise.
380         (ctime): Likewise.
381         (ctime_sub): Likewise.
382
383 2011-02-24  Jakub Jelinek  <jakub@redhat.com>
384
385         PR fortran/47878
386         * io/transfer.c (read_sf): Call fbuf_getptr only at the end,
387         and subtract n, dtp->u.p.sf_seen_eor and seen_comma from it.
388
389 2011-02-24  Janne Blomqvist  <jb@gcc.gnu.org>
390
391         PR libfortran/47802
392         * configure.ac: Add test for POSIX getpwuid_r.
393         * intrinsics/getlog.c (getlog): CPP test for
394         HAVE_POSIX_GETPWUID_R.
395         * config.h.in: Regenerated.
396         * configure: Regenerated.
397
398 2011-02-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
399
400         PR libgfortran/47567
401         * io/write_float.def (output_float): Remove special case handling of
402         zero with width 1.
403
404 2011-02-23  Janne Blomqvist  <jb@gcc.gnu.org>
405             Jerry DeLisle    <jvdelisle@gcc.gnu.org>
406
407         PR libfortran/47694
408         * io/fbuf.h (fbuf_getptr): New inline function.
409         * io/transfer.c (read_sf): Use fbuf_getptr and fbuf_getc to scan
410         through the string instead of fbuf_read.
411
412 2011-02-22  Tobias Burnus  <burnus@net-b.de>
413             Kai-Uwe Eckhardt  <kuehro@gmx.de>
414
415         PR libfortran/47830
416         * intrinsics/c99_functions.c (roundl): Make C valid for
417         HAVE_NEXTAFTERL.
418
419 2011-02-19  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
420
421         PR libgfortran/47567
422         * io/write_float.def (output_float): Adjust width for F0.d to
423         allow space for negative signs on zero.
424
425 2011-02-16  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
426
427         PR libgfortran/47667
428         * io/list_read.c (read_logical): Check for end of line before calling
429         eat_line. (read_integer): Likewise. (parse_real): Don't unget the
430         separator. Check for end of line before calling eat_line.
431         (read_complex): Allow line-end before and after parenthesis and comma.
432         Check for end of line before calling eat_line. (read_real): Check for
433         end of line before calling eat_line.
434
435 2011-02-16  Jakub Jelinek  <jakub@redhat.com>
436
437         PR libfortran/47757
438         * gfortran.map (GFORTRAN_1.4): Export
439         _gfortran_{m,s}i{all,any,parity}_i{1,2,4,8,16} and
440         _gfortran_{cshift0,eoshift{0,2}}_16_char4.
441
442 2011-02-15  Tobias Burnus  <burnus@net-b.de>
443
444         PR fortran/47716
445         PR fortran/47648
446         * acinclude.m4 (LIBGFOR_CHECK_FLOAT128): Use check from
447         libquadmath, which uses more features.
448         * configure: Regenerate.
449
450 2011-02-14  Jakub Jelinek  <jakub@redhat.com>
451
452         PR fortran/47642
453         * io/write_float.def (DTOAQ): Use quadmath_snprintf instead of
454         quadmath_flt128tostr.
455         * io/transfer128.c (tmp2): Initialize to quadmath_snprintf instead
456         of quadmath_flt128tostr.
457
458 2011-02-13  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
459
460         * Makefile.in: Regenerate.
461         * aclocal.m4: Likewise.
462         * configure: Likewise.
463
464 2011-02-05  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
465
466         PR libgfortran/47567
467         * io/write_float.def (output_float): Eliminate some redundant code.
468         Adjust width for case of F0.X for values of zero and all other values.
469         Expand cases where '*' is set to give cleaner results.
470
471 2011-02-05  Janne Blomqvist  <jb@gcc.gnu.org>
472
473         PR libfortran/47571
474         * intrinsics/time_1.h (GF_CLOCK_MONOTONIC): Move to system_clock.c.
475         (weak_gettime): Likewise.
476         (gf_gettime): Change API, move weak_gettime() usage to
477         system_clock.c
478         * intrinsics/system_clock.c (GTHREAD_USE_WEAK): Define.
479         (gf_gettime_mono): New function.
480         (system_clock_4): Use gf_gettime_mono().
481         (system_clock_8): Likewise.
482         * intrinsics/date_and_time.c (date_and_time): Update gf_gettime()
483         usage.
484
485 2011-02-02  Janne Blomqvist  <jb@gcc.gnu.org>
486
487         PR libfortran/47571
488         * configure: Regenerated.
489         * configure.ac: Don't add librt to LIBS.
490         * intrinsics/time_1.h (weak_gettime): Weakref trickery for
491         clock_gettime().
492         (gf_gettime): Use weak_gettime() instead of clock_gettime().
493
494 2011-02-01  Janne Blomqvist  <jb@gcc.gnu.org>
495
496         * intrinsics/time_1.h: Include errno.h needed by fallbacks.
497
498 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
499
500         * intrinsics/time_1.h: Fix definition of GF_CLOCK_MONOTONIC macro.
501
502 2011-01-31  Janne Blomqvist  <jb@gcc.gnu.org>
503
504         * configure.ac: Check for clock_gettime().
505         * configure: Regenerated.
506         * config.h.in: Regenerated.
507         * intrinsics/time_1.h (__time_1): Rename to gf_cputime, add
508         times() fallback.
509         (gf_gettime): New function.
510         * intrinsics/cpu_time.c (__cpu_time_1): Update to call gf_cputime.
511         * intrinsics/date_and_time.c (date_and_time): Use gf_gettime.
512         * intrinsics/dtime.c (dtime_sub): Use gf_cputime.
513         * intrinsics/etime.c (etime_sub): Use gf_cputime.
514         * intrinsics/system_clock.c (system_clock_4): Use gf_gettime.
515         (system_clock_8): Use gf_gettime, increase count rate to allow
516         nanosecond precision, remove overflow prone branch.
517
518 2011-01-29  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
519
520         PR libgfortran/47434
521         * io/write_float.def (write_infnan): Use calculate_sign to determine
522         if the sign should be given and check field widths accordingly.
523
524 2011-01-29  Kai Tietz  <kai.tietz@onevision.com>
525
526         * intrinsics/ctime.c (ctime_r): Improve implementation.
527
528 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
529
530         PR libfortran/47431
531         * config.h.in: Regenerated.
532         * configure: Regenerated.
533         * configure.ac: Add check for ctime_r().
534         * intrinsics/ctime.c (ctime_r): Fallback implementation.
535         (fdate): Use ctime_r() instead of ctime().
536         (fdate_sub): Likewise.
537         (ctime): Likewise.
538         (ctime_sub): Likewise.
539
540 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
541
542         PR libfortran/47432
543         * config.h.in: Regenerated.
544         * configure: Regenerated.
545         * configure.ac: Add check for ttyname_r.
546         * io/unix.h: Add TTY_NAME_MAX, change stream_ttyname prototype.
547         * io/unix.c (stream_ttyname): Use ttyname_r if available, conform
548         to new prototype.
549         * io/inquire.c (inquire_via_unit): Use changed stream_ttyname.
550         * io/intrinsics.c (ttynam_sub): Likewise.
551         (ttynam): Likewise.
552
553 2011-01-27  Janne Blomqvist  <jb@gcc.gnu.org>
554
555         PR libfortran/47491
556         * configure.ac: Call AC_USE_SYSTEM_EXTENSIONS to enable common
557         extensions.
558         * config.h.in: Regenerate.
559         * configure: Regenerate.
560
561 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
562
563         PR libgfortran/47285
564         * io/write_float.def (write_infnan): Adjust processor selected width
565         to 3 if NaN.
566
567 2011-01-26  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
568
569         PR libgfortran/47285
570         * io/write_float.def (output_float): Return SUCCESS or FAILURE and use
571         the result to set the padding.
572
573 2011-01-26  Kai Tietz  <kai.tietz@onevision.com>
574
575         * intrinsics/getlog.c (getlog): Fix label/statement issue.
576
577 2011-01-25  Janne Blomqvist  <jb@gcc.gnu.org>
578
579         PR libfortran/47375
580         * config.h.in: Regenerated.
581         * configure: Regenerated.
582         * configure.ac: Add check for getpwuid_r.
583         * intrinsics/getlog.c (getlog): Use getpwuid_r() if available.
584
585 2011-01-22  Janne Blomqvist  <jb@gcc.gnu.org>
586
587         PR libfortran/46267
588         * config.h.in: Regenerated.
589         * configure: Regenerated.
590         * configure.ac: Check presence of strerror_r.
591         * intrinsics/gerror.c (gerror): Use gf_strerror, modify logic.
592         * io/unix.c (get_oserror): Remove.
593         * libgfortran.h (gf_strerror): Add prototype.
594         (get_oserror): Remove prototype.
595         * runtime/error.c (gf_strerror): New function.
596         (os_error): Use gf_strerror instead of get_oserror.
597         (generate_errror): Likewise.
598
599 2011-01-17  Janne Blomqvist  <jb@gcc.gnu.org>
600
601         PR libfortran/47296
602         * io/unix.c (tempfile): Set opp->file and opp->file_len also if an
603         error occurs.
604
605 2011-01-16  Jakub Jelinek  <jakub@redhat.com>
606
607         PR fortran/46625
608         * io/write_float.def (DTOAQ): Use quadmath_flt128tostr
609         instead of quadmath_dtoa.
610         * io/transfer128.c (tmp1, tmp2): New variables, bring in
611         strtoflt128 and quadmath_flt128tostr.
612         (transfer_real128, transfer_real128_write, transfer_complex128,
613         transfer_complex128_write): Remove tmp1/tmp2 variables.
614         * io/read.c (convert_real): Use strtoflt128 instead of
615         quadmath_strtopQ, adjust for the changed arguments and return
616         value.
617
618 2011-01-14  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
619
620         PR libgfortran/47296
621         * io/unix.c (unpack_filename): Return non-zero if the filename passed
622         in is NULL.
623
624 2011-01-04  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
625
626         PR libgfortran/47154
627         * io/list_read.c (namelist_read): Remove calls to hit_eof to avoid the
628         duplicate calls via next_record.
629
630 \f
631 Copyright (C) 2011 Free Software Foundation, Inc.
632
633 Copying and distribution of this file, with or without modification,
634 are permitted in any medium without royalty provided the copyright
635 notice and this notice are preserved.