1 2020-12-05 Iain Sandoe <iain@sandoe.co.uk>
4 * configure: Regenerate.
6 2020-11-29 John David Anglin <danglin@gcc.gnu.org>
8 * configure: Regenerate.
10 2020-11-26 Maciej W. Rozycki <macro@linux-mips.org>
12 * io/io.h [HAVE_NEWLOCALE]: Also check for HAVE_FREELOCALE and
14 [HAVE_FREELOCALE && HAVE_NEWLOCALE && HAVE_USELOCALE]
15 (HAVE_POSIX_2008_LOCALE): New macro.
16 (st_parameter_dt) [HAVE_NEWLOCALE]: Check for
17 HAVE_POSIX_2008_LOCALE instead.
18 * io/transfer.c (data_transfer_init_worker, finalize_transfer)
19 [HAVE_USELOCALE]: Check for HAVE_POSIX_2008_LOCALE instead.
20 * io/unit.c [HAVE_NEWLOCALE]: Likewise.
21 (init_units) [HAVE_NEWLOCALE]: Likewise.
22 (close_units) [HAVE_FREELOCALE]: Likewise.
23 * runtime/error.c (gf_strerror) [HAVE_USELOCALE]: Likewise.
25 2020-11-26 Maciej W. Rozycki <macro@linux-mips.org>
27 * libgfortran.h: Use #if rather than #ifdef with
28 __FLT_HAS_INFINITY__, __DBL_HAS_INFINITY__,
29 __LDBL_HAS_INFINITY__, __FLT_HAS_QUIET_NAN__,
30 __DBL_HAS_QUIET_NAN__, and __LDBL_HAS_QUIET_NAN__.
32 2020-11-21 Iain Sandoe <iain@sandoe.co.uk>
34 * intrinsics/execute_command_line.c (environ): Use
35 _NSGetEnviron to get the environment pointer on Darwin.
37 2020-10-30 Harald Anlauf <anlauf@gmx.de>
39 * intrinsics/random.c (SZ_IN_INT_4): Define size of state in int32_t.
40 (SZ_IN_INT_8): Define size of state in int64_t.
42 (random_seed_i4): Use size SZ_IN_INT_4 instead of SZ.
43 (random_seed_i8): Use size SZ_IN_INT_8 instead of SZ.
45 2020-10-18 Harald Anlauf <anlauf@gmx.de>
47 * m4/matmul_internal.m4: Move check for rank-1 times rank-2 before
48 checks on strides for rank-2 times rank-2.
49 * generated/matmul_c10.c: Regenerated.
50 * generated/matmul_c16.c: Likewise.
51 * generated/matmul_c4.c: Likewise.
52 * generated/matmul_c8.c: Likewise.
53 * generated/matmul_i1.c: Likewise.
54 * generated/matmul_i16.c: Likewise.
55 * generated/matmul_i2.c: Likewise.
56 * generated/matmul_i4.c: Likewise.
57 * generated/matmul_i8.c: Likewise.
58 * generated/matmul_r10.c: Likewise.
59 * generated/matmul_r16.c: Likewise.
60 * generated/matmul_r4.c: Likewise.
61 * generated/matmul_r8.c: Likewise.
62 * generated/matmulavx128_c10.c: Likewise.
63 * generated/matmulavx128_c16.c: Likewise.
64 * generated/matmulavx128_c4.c: Likewise.
65 * generated/matmulavx128_c8.c: Likewise.
66 * generated/matmulavx128_i1.c: Likewise.
67 * generated/matmulavx128_i16.c: Likewise.
68 * generated/matmulavx128_i2.c: Likewise.
69 * generated/matmulavx128_i4.c: Likewise.
70 * generated/matmulavx128_i8.c: Likewise.
71 * generated/matmulavx128_r10.c: Likewise.
72 * generated/matmulavx128_r16.c: Likewise.
73 * generated/matmulavx128_r4.c: Likewise.
74 * generated/matmulavx128_r8.c: Likewise.
76 2020-09-28 Tobias Burnus <tobias@codesourcery.com>
78 * m4/unpack.m4 (unpack0_'rtype_code`,
79 unpack1_'rtype_code`): Move 'rstride[0]' initialization outside
80 conditional branch to silence -Wmaybe-uninitialized.
81 * generated/unpack_c10.c: Regenerate.
82 * generated/unpack_c16.c: Regenerate.
83 * generated/unpack_c4.c: Regenerate.
84 * generated/unpack_c8.c: Regenerate.
85 * generated/unpack_i1.c: Regenerate.
86 * generated/unpack_i16.c: Regenerate.
87 * generated/unpack_i2.c: Regenerate.
88 * generated/unpack_i4.c: Regenerate.
89 * generated/unpack_i8.c: Regenerate.
90 * generated/unpack_r10.c: Regenerate.
91 * generated/unpack_r16.c: Regenerate.
92 * generated/unpack_r4.c: Regenerate.
93 * generated/unpack_r8.c: Regenerate.
95 2020-09-27 Clement Chigot <clement.chigot@atos.net>
97 * config/t-aix: Use $(AR) without -X32_64.
99 2020-09-03 Harald Anlauf <anlauf@gmx.de>
101 * m4/iall.m4: Initial value for result should be -1.
102 * generated/iall_i1.c (miall_i1): Generated.
103 * generated/iall_i16.c (miall_i16): Likewise.
104 * generated/iall_i2.c (miall_i2): Likewise.
105 * generated/iall_i4.c (miall_i4): Likewise.
106 * generated/iall_i8.c (miall_i8): Likewise.
108 2020-08-24 Mark Eggleston <markeggleston@gcc.gnu.org>
111 * intrinsics/env.c: If value_len is > 0 blank the string.
112 Copy the result only if its length is > 0.
114 2020-08-20 Mark Eggleston <markeggleston@gcc.gnu.org>
117 * io/format.c (parse_format_list): Add new local variable
118 "standard" to hold the required standard to check. If the
119 format width is zero select standard depending on descriptor.
120 Call notification_std using the new standard variable.
122 2020-08-17 David Edelsohn <dje.gcc@gmail.com>
123 Clement Chigot <clement.chigot@atos.net>
125 * config/t-aix: Add complementary mode object file to libcaf_single.a.
127 2020-08-03 Tobias Burnus <tobias@codesourcery.com>
129 * caf/single.c (_gfortran_caf_lock): Fix typo.
131 2020-07-26 Tobias Burnus <tobias@codesourcery.com>
134 * configure.ac: Add LIBGOMP_CHECKED_INT_KINDS and
135 LIBGOMP_CHECKED_REAL_KINDS and use it for to skip
136 integer kind=16 checks for amdgcn.
137 * Makefile.am (kinds.h, selected_int_kind.inc,
138 selected_real_kind.inc): Pass them on.
139 * mk-kinds-h.sh: Takes to-be-check kinds as argument.
140 * mk-sik-inc.sh: Likewise.
141 * mk-srk-inc.sh: Likewise.
142 * Makefile.in: Regenerate.
143 * configure: Regenerate.
145 2020-07-24 Dominique d'Humieres <dhumieres.dominique@free.fr>
148 * io/write_float.def (determine_en_precision): Fix switch between
151 2020-07-24 Dominique d'Humieres <dhumieres.dominique@free.fr>
154 * io/write_float.def (build_float_string): Do not reset
155 nbefore for FMT_F and FMT_EN.
157 2020-07-14 David Edelsohn <dje.gcc@gmail.com>
159 * config/t-aix: Set BITS from compiler cpp macro.
161 2020-06-22 David Edelsohn <dje.gcc@gmail.com>
163 * Makefile.am: Use -include.
164 * Makefile.in: Regenerate.
166 2020-06-21 David Edelsohn <dje.gcc@gmail.com>
168 * Makefile.am (tmake_file): Build and install AIX-style FAT libraries.
169 * Makefile.in: Regenerate.
170 * configure.ac (tmake_file): Substitute.
171 * configure: Regenerate.
172 * configure.host: Add system configury stanza. Define tmake_file.
173 * config/t-aix: New file.
175 2020-06-13 Thomas Koenig <tkoenig@gcc.gnu.org>
178 * io/write.c (ztoa_big): Disable -Wstringop-overflow for one
181 2020-06-08 Harald Anlauf <anlauf@gmx.de>
184 * io/transfer.c (finalize_transfer): Fix type in error message.
186 2020-06-01 Uroš Bizjak <ubizjak@gmail.com>
189 * config/fpu-387.h (struct fenv): Add __attribute__ ((gcc_struct)).
191 2020-05-29 H.J. Lu <hjl.tools@gmail.com>
194 * configure: Regenerated.
196 2020-05-29 Jakub Jelinek <jakub@redhat.com>
199 * Makefile.am (i_findloc0_c): Add findloc0_i10.c.
200 (i_findloc1_c): Add findloc1_i10.c.
201 * gfortran.map (GFORTRAN_10.2): New symbol version, export
202 _gfortran_{,m,s}findloc{0,1}_c10 symbols.
203 * Makefile.in: Regenerated.
204 * generated/findloc0_c10.c: Generated.
205 * generated/findloc1_c10.c: Generated.
207 2020-05-28 Harald Anlauf <anlauf@gmx.de>
210 * io/unit.c (unlock_unit): Guard by check for NULL pointer.
212 2020-05-26 Harald Anlauf <anlauf@gmx.de>
213 Steven G. Kargl <kargl@gcc.gnu.org>
216 * io/transfer.c (st_wait_async): Do not dereference NULL pointer.
218 2020-05-26 Harald Anlauf <anlauf@gmx.de>
221 * io/transfer.c (finalize_transfer): Generate runtime error for
222 namelist input/output to unformatted file.
224 2020-05-23 Thomas Koenig <tkoenig@gcc.gnu.org>
227 * io/async.c (async_wait_id): Generate error if ID is higher
228 than the highest current ID.
229 * runtime/error.c (translate_error): Handle LIBERROR_BAD_WAIT_ID.
231 2020-05-21 H.J. Lu <hongjiu.lu@intel.com>
233 * m4/matmul.m4: Don't include <config/i386/cpuinfo.h>. Use
234 __builtin_cpu_is/__builtin_cpu_supports
235 * generated/matmul_c10.c: Regenerated.
236 * generated/matmul_c16.c: Likewise.
237 * generated/matmul_c4.c: Likewise.
238 * generated/matmul_c8.c: Likewise.
239 * generated/matmul_i1.c: Likewise.
240 * generated/matmul_i16.c: Likewise.
241 * generated/matmul_i2.c: Likewise.
242 * generated/matmul_i4.c: Likewise.
243 * generated/matmul_i8.c: Likewise.
244 * generated/matmul_r10.c: Likewise.
245 * generated/matmul_r16.c: Likewise.
246 * generated/matmul_r4.c: Likewise.
247 * generated/matmul_r8.c: Likewise.
249 2020-05-15 H.J. Lu <hongjiu.lu@intel.com>
252 * configure: Regenerated.
254 2020-05-14 Thomas Koenig <tkoenig@gcc.gnu.org>
257 * io/close.c (close_status): Add CLOSE_INVALID.
258 (st_close): Return early on invalid STATUS parameter.
260 2020-05-14 H.J. Lu <hongjiu.lu@intel.com>
262 * configure: Regenerated.
264 2020-05-06 Uroš Bizjak <ubizjak@gmail.com>
266 * config/fpu-387.h (__math_force_eval): Remove.
267 (__math_force_eval_div): New define.
268 (local_feraiseexcept): Use __math_force_eval_div to use
269 generic division to generate INVALID, DIVZERO and INEXACT
271 (struct fenv): Define named struct instead of typedef.
273 2020-05-01 Uroš Bizjak <ubizjak@gmail.com>
275 * config/fpu-387.h (__math_force_eval): New define.
276 (local_feraiseexcept): Use __math_force_eval to evaluate
277 generic division to generate INVALID and DIVZERO exceptions.
279 2020-04-22 Fritz Reese <foreese@gcc.gnu.org>
281 * intrinsics/trigd.c, intrinsics/trigd_lib.inc, intrinsics/trigd.inc:
282 Guard against unavailable math functions.
283 Use suffixes from kinds.h based on the REAL kind.
285 2020-04-22 Jakub Jelinek <jakub@redhat.com>
289 * configure.ac: Add math func checks for fmaf, fma and fmal. Add
290 HAVE_INLINE_BUILTIN_COPYSIGN check.
291 * c99_protos.h (copysign, fmaf, fma, fmal): Provide fallback
293 (HAVE_COPYSIGN, HAVE_FMAF, HAVE_FMA, HAVE_FMAL): Define if not
294 defined and fallback version is provided.
295 * intrinsics/c99_functions.c (copysign, fmaf, fma, fmal): Provide
296 fallback implementations if possible
297 * configure: Regenerated.
298 * config.h.in: Regenerated.
300 2020-04-19 Uroš Bizjak <ubizjak@gmail.com>
302 * config/fpu-387.h (local_feraiseexcept) [__SSE_MATH__]:
303 Remove unneeded assignments to volatile memory.
305 2020-04-01 Fritz Reese <foreese@gcc.gnu.org>
306 Steven G. Kargl <kargl@gcc.gnu.org>
309 * Makefile.am, Makefile.in: New make rule for intrinsics/trigd.c.
310 * gfortran.map: New routines for {sind, cosd, tand}X{r4, r8, r10, r16}.
311 * intrinsics/trigd.c, intrinsics/trigd_lib.inc, intrinsics/trigd.inc:
312 New files. Defines native degree-valued trig functions.
314 2020-02-18 Thomas Koenig <tkoenig@gcc.gnu.org>
317 * io/async.c (destroy_adv_cond): Do not destroy lock.
318 (async_io): Make sure au->lock is locked for finishing of thread.
319 Do not lock/unlock around signalling emptysignal. Unlock au->lock
321 (init_adv_cond): Do not initialize lock.
322 (enqueue_transfer): Unlock after signal.
323 (enqueue_done_id): Likewise.
324 (enqueue_done): Likewise.
325 (enqueue_close): Likewise.
326 (enqueue_data_transfer): Likewise.
327 (async_wait_id): Do not lock/unlock around signalling au->work.
328 (async_wait): Unlock after signal.
329 * io/async.h (SIGNAL): Add comment about needed au->lock.
330 Remove locking/unlocking of advcond->lock.
331 (WAIT_SIGNAL_MUTEX): Add comment. Remove locking/unlocking of
332 advcond->lock. Unlock mutex only at the end. Loop on
333 __ghread_cond_wait returning zero.
334 (REVOKE_SIGNAL): Add comment. Remove locking/unlocking of
336 (struct adv_cond): Remove mutex from struct.
338 2020-02-12 Sandra Loosemore <sandra@codesourcery.com>
343 * configure: Regenerated.
345 2020-01-24 Maciej W. Rozycki <macro@wdc.com>
347 * configure.ac: Handle `--with-toolexeclibdir='.
348 * Makefile.in: Regenerate.
349 * aclocal.m4: Regenerate.
350 * configure: Regenerate.
352 2020-01-17 Jerry DeLisle <jvdelisle@gcc.gnu.org>
355 * io/unit.c (set_internal_unit): Set round and sign flags
358 2020-01-17 Jerry DeLisle <jvdelisle@gcc.gnu.org>
361 * io/format.c (parse_format_list): Zero width not allowed with
363 * io/write_float.def (build_float_string): Include range of
364 higher exponent values that require wider width.
366 2020-01-01 Jerry DeLisle <jvdelisle@gcc.gnu.org>
369 * io/format.c (parse_format_list): Implement the E0 exponent
370 width to provide smallest possible width for exponent fields.
371 Refactor code for correct parsing and better readability of the
373 * io/io.h (write_real_w0): Change interface to pass in pointer
375 * io/transfer.c: Update all calls to write_real_w0 to use the
377 * io/write.c ((write_real_w0): Use the new interface with fnode
378 to access both the decimal precision and exponent widths used in
380 * io/write_float.def (build_float_string): Use the passed in
381 exponent width to calculate the used width in the case of E0.
383 2020-01-01 Jakub Jelinek <jakub@redhat.com>
385 Update copyright years.
387 Copyright (C) 2020 Free Software Foundation, Inc.
389 Copying and distribution of this file, with or without modification,
390 are permitted in any medium without royalty provided the copyright
391 notice and this notice are preserved.