xref: /netbsd-src/external/gpl3/gcc/dist/libquadmath/ChangeLog (revision 122b5006ee1bd67145794b4cde92f4fe4781a5ec)
12021-04-08  Release Manager
2
3	* GCC 10.3.0 released.
4
52021-01-03  Iain Sandoe  <iain@sandoe.co.uk>
6	    Jakub Jelinek   <jakub@redhat.com>
7
8	PR target/97865
9	* configure: Regenerate.
10
112020-07-23  Release Manager
12
13	* GCC 10.2.0 released.
14
152020-05-07  Release Manager
16
17	* GCC 10.1.0 released.
18
192020-02-12  Sandra Loosemore  <sandra@codesourcery.com>
20
21	PR libstdc++/79193
22	PR libstdc++/88999
23
24	* configure: Regenerated.
25
262020-01-24  Maciej W. Rozycki  <macro@wdc.com>
27
28	* configure.ac: Handle `--with-toolexeclibdir='.
29	* Makefile.in: Regenerate.
30	* aclocal.m4: Regenerate.
31	* configure: Regenerate.
32
332020-01-01  Jakub Jelinek  <jakub@redhat.com>
34
35	Update copyright years.
36
37	* libquadmath.texi: Bump @copying's copyright year.
38
392019-09-27  Maciej W. Rozycki  <macro@wdc.com>
40
41	* configure: Regenerate.
42
432019-08-02  Jakub Jelinek  <jakub@redhat.com>
44
45	* quadmath.h (M_Eq, M_LOG2Eq, M_LOG10Eq, M_LN2q, M_LN10q, M_PIq,
46	M_PI_2q, M_PI_4q, M_1_PIq, M_2_PIq, M_2_SQRTPIq, M_SQRT2q,
47	M_SQRT1_2q): Use two more decimal places.
48
492019-01-01  Jakub Jelinek  <jakub@redhat.com>
50
51	Update copyright years.
52
53	* libquadmath.texi: Bump @copying's copyright year.
54
552018-12-11  Jakub Jelinek  <jakub@redhat.com>
56
57	PR c/88430
58	* quadmath_weak.h (__qmath2): Add __quadmath_throw.
59
602018-11-07  Joseph Myers  <joseph@codesourcery.com>
61
62	* quadmath-imp.h (ieee854_float128): Use mantissa0, mantissa1,
63	mantissa2 and mantissa3 fields instead of mant_high and mant_low.
64	Change nan field to ieee_nan.
65	* update-quadmath.py (update_sources): Also update fmaq.c.
66	* math/nanq.c (nanq): Use ieee_nan field of union.
67	Zero-initialize f.  Set quiet_nan field.
68	* printf/flt1282mpn.c, printf/printf_fphex.c, strtod/mpn2flt128.c,
69	strtod/strtoflt128.c: Use mantissa0, mantissa1, mantissa2 and
70	mantissa3 fields.  Use ieee_nan and quiet_nan field.
71	* math/fmaq.c: Regenerate from glibc sources with
72	update-quadmath.py.
73
742018-11-05  Joseph Myers  <joseph@codesourcery.com>
75
76	PR libquadmath/68686
77	* Makefile.am: (libquadmath_la_SOURCES): Remove math/isinf_nsq.c.
78	Add math/exp2q.c math/issignalingq.c math/lgammaq_neg.c
79	math/lgammaq_product.c math/tanq_kernel.c math/tgammaq_product.c
80	math/casinhq_kernel.c.
81	* Makefile.in: Regenerate.
82	* libquadmath.texi (exp2q, issignalingq): Document.
83	* quadmath-imp.h: Include <errno.h>, <limits.h>, <stdbool.h> and
84	<fenv.h>.
85	(HIGH_ORDER_BIT_IS_SET_FOR_SNAN, FIX_FLT128_LONG_CONVERT_OVERFLOW)
86	(FIX_FLT128_LLONG_CONVERT_OVERFLOW, __quadmath_kernel_tanq)
87	(__quadmath_gamma_productq, __quadmath_gammaq_r)
88	(__quadmath_lgamma_negq, __quadmath_lgamma_productq)
89	(__quadmath_lgammaq_r, __quadmath_kernel_casinhq, mul_splitq)
90	(math_check_force_underflow_complex, __glibc_likely)
91	(__glibc_unlikely, struct rm_ctx, SET_RESTORE_ROUNDF128)
92	(libc_feholdsetround_ctx, libc_feresetround_ctx): New.
93	(feraiseexcept, fenv_t, feholdexcept, fesetround, feupdateenv)
94	(fesetenv, fetestexcept, feclearexcept): Define if not supported
95	through <fenv.h>.
96	(__quadmath_isinf_nsq): Remove.
97	* quadmath.h (exp2q, issignalingq): New.
98	* quadmath.map (QUADMATH_1.2): New.
99	* quadmath_weak.h (exp2q, issignalingq): New.
100	* update-quadmath.py: New file.
101	* math/isinf_nsq.c: Remove file.
102	* math/casinhq_kernel.c, math/exp2q.c, math/expq_table.h,
103	math/issignalingq.c, math/lgammaq_neg.c, math/lgammaq_product.c,
104	math/tanq_kernel.c, math/tgammaq_product.c: New files.  Generated
105	from glibc sources with update-quadmath.py.
106	* math/acoshq.c, math/acosq.c, math/asinhq.c, math/asinq.c,
107	math/atan2q.c, math/atanhq.c, math/atanq.c, math/cacoshq.c,
108	math/cacosq.c, math/casinhq.c, math/casinq.c, math/catanhq.c,
109	math/catanq.c, math/cbrtq.c, math/ccoshq.c, math/ceilq.c,
110	math/cexpq.c, math/cimagq.c, math/clog10q.c, math/clogq.c,
111	math/conjq.c, math/copysignq.c, math/coshq.c, math/cosq.c,
112	math/cosq_kernel.c, math/cprojq.c, math/crealq.c, math/csinhq.c,
113	math/csinq.c, math/csqrtq.c, math/ctanhq.c, math/ctanq.c,
114	math/erfq.c, math/expm1q.c, math/expq.c, math/fabsq.c,
115	math/fdimq.c, math/finiteq.c, math/floorq.c, math/fmaxq.c,
116	math/fminq.c, math/fmodq.c, math/frexpq.c, math/hypotq.c,
117	math/ilogbq.c, math/isinfq.c, math/isnanq.c, math/j0q.c,
118	math/j1q.c, math/jnq.c, math/ldexpq.c, math/lgammaq.c,
119	math/llrintq.c, math/llroundq.c, math/log10q.c, math/log1pq.c,
120	math/log2q.c, math/logbq.c, math/logq.c, math/lrintq.c,
121	math/lroundq.c, math/modfq.c, math/nearbyintq.c,
122	math/nextafterq.c, math/powq.c, math/remainderq.c, math/remquoq.c,
123	math/rintq.c, math/roundq.c, math/scalblnq.c, math/scalbnq.c,
124	math/signbitq.c, math/sincos_table.c, math/sincosq.c,
125	math/sincosq_kernel.c, math/sinhq.c, math/sinq.c,
126	math/sinq_kernel.c, math/tanhq.c, math/tanq.c, math/tgammaq.c,
127	math/truncq.c, math/x2y2m1q.c: Regenerate from glibc sources with
128	update-quadmath.py.
129
1302018-10-31  Joseph Myers  <joseph@codesourcery.com>
131
132	PR bootstrap/82856
133	* Makefile.am: Include multilib.am.
134	(AUTOMAKE_OPTIONS): Remove 1.8.  Add info-in-builddir.
135	(all-local): Define outside conditional code.
136	(CLEANFILES): Remove libquadmath.info.
137	* configure.ac: Remove AC_PREREQ.
138	* Makefile.in, aclocal.m4, config.h.in, configure: Regenerate.
139
1402018-04-24  H.J. Lu  <hongjiu.lu@intel.com>
141
142	* configure: Regenerated.
143
1442018-04-19  Jakub Jelinek  <jakub@redhat.com>
145
146	* configure: Regenerated.
147
1482018-04-18  David Malcolm  <dmalcolm@redhat.com>
149
150	PR jit/85384
151	* configure: Regenerate.
152
1532018-02-14  Igor Tsimbalist  <igor.v.tsimbalist@intel.com>
154
155	PR target/84148
156	* configure: Regenerate.
157
1582018-01-03  Jakub Jelinek  <jakub@redhat.com>
159
160	* libquadmath.texi: Bump @copying's copyright year.
161
1622017-11-17  Igor Tsimbalist  <igor.v.tsimbalist@intel.com>
163
164	* Makefile.am: Update AM_CFLAGS.
165	* Makefile.in: Regenerate:
166	* acinclude.m4: Add enable.m4 and cet.m4.
167	* configure: Regenerate.
168	* configure.ac: Set CET_FLAGS. Update XCFLAGS.
169
1702017-11-05  Tom de Vries  <tom@codesourcery.com>
171
172	PR other/82784
173	* printf/gmp-impl.h (MPN_MUL_N_RECURSE): Remove semicolon after
174	"do {} while (0)".
175
1762017-09-01  Michael Meissner  <meissner@linux.vnet.ibm.com>
177
178	PR libquadmath/81848
179	* configure.ac (powerpc*-linux*): Use attribute mode KC to create
180	complex __float128 on PowerPC instead of attribute mode TC.
181	* quadmath.h (__complex128): Likewise.
182	* configure: Regenerate.
183	* math/cbrtq.c (CBRT2): Use __float128 not long double.
184	(CBRT4): Likewise.
185	(CBRT2I): Likewise.
186	(CBRT4I): Likewise.
187	* math/j0q.c (U0): Likewise.
188	* math/sqrtq.c (sqrtq): Don't depend on implicit conversion
189	between __float128, instead explicitly convert the __float128
190	value to long double because the PowerPC does not allow __float128
191	and long double in the same expression.
192
1932017-07-19  Gerald Pfeifer  <gerald@pfeifer.com>
194
195	* math/powq.c (powq): Use uint32_t instead of u_int32_t.
196
1972017-07-19  Jakub Jelinek  <jakub@redhat.com>
198
199	PR libquadmath/65757
200	* quadmath-imp.h (math_opt_barrier, math_force_eval,
201	math_narrow_eval, math_check_force_underflow,
202	math_check_force_underflow_nonneg): Define.
203	* math/ceilq.c: Backport changes from upstream glibc
204	between 2012-11-01 and 2017-07-13.
205	* math/remquoq.c: Likewise.
206	* math/expq.c: Likewise.
207	* math/llroundq.c: Likewise.
208	* math/logq.c: Likewise.
209	* math/atanq.c: Likewise.
210	* math/nearbyintq.c: Likewise.
211	* math/scalblnq.c: Likewise.
212	* math/finiteq.c: Likewise.
213	* math/atanhq.c: Likewise.
214	* math/expm1q.c: Likewise.
215	* math/sinhq.c: Likewise.
216	* math/log10q.c: Likewise.
217	* math/rintq.c: Likewise.
218	* math/roundq.c: Likewise.
219	* math/fmaq.c: Likewise.
220	* math/erfq.c: Likewise.
221	* math/log2q.c: Likewise.
222	* math/lroundq.c: Likewise.
223	* math/j1q.c: Likewise.
224	* math/scalbnq.c: Likewise.
225	* math/truncq.c: Likewise.
226	* math/frexpq.c: Likewise.
227	* math/sincosq.c: Likewise.
228	* math/tanhq.c: Likewise.
229	* math/asinq.c: Likewise.
230	* math/coshq.c: Likewise.
231	* math/j0q.c: Likewise.
232	* math/asinhq.c: Likewise.
233	* math/floorq.c: Likewise.
234	* math/sinq_kernel.c: Likewise.
235	* math/powq.c: Likewise.
236	* math/hypotq.c: Likewise.
237	* math/sincos_table.c: Likewise.
238	* math/rem_pio2q.c: Likewise.
239	* math/nextafterq.c: Likewise.
240	* math/log1pq.c: Likewise.
241	* math/sincosq_kernel.c: Likewise.
242	* math/tanq.c: Likewise.
243	* math/acosq.c: Likewise.
244	* math/lrintq.c: Likewise.
245	* math/llrintq.c: Likewise.
246
2472017-02-09  Gerald Pfeifer  <gerald@pfeifer.com>
248
249	* configure.ac (ACX_BUGURL): Update.
250	* configure: Regenerate.
251
2522017-01-17  Jakub Jelinek  <jakub@redhat.com>
253
254	PR other/79046
255	* configure.ac: Add GCC_BASE_VER.
256	* Makefile.am (gcc_version): Use @get_gcc_base_ver@ instead of cat to
257	get version from BASE-VER file.
258	* configure: Regenerated.
259	* Makefile.in: Regenerated.
260
2612017-01-01  Jakub Jelinek  <jakub@redhat.com>
262
263	* libquadmath.texi: Bump @copying's copyright year.
264
2652016-11-15  Matthias Klose  <doko@ubuntu.com>
266
267	* configure: Regenerate.
268
2692016-01-04  Jakub Jelinek  <jakub@redhat.com>
270
271	* libquadmath.texi: Bump @copying's copyright year.
272
2732015-08-09  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
274
275	* Makefile.am (libquadmath_la_SOURCES): Add matha/logbq.c.
276	* Makefile.in: Regenerate.
277	* libquadmath.texi: Document logbq.
278	* quadmath.h: Add logbq prototype.
279	* quadmath.map: Add logbq.
280	* quadmath_weak.h: Add logbq prototype.
281	* math/logbq.c: New file
282
2832015-05-13  Michael Haubenwallner  <michael.haubenwallner@ssi-schaefer.com>
284
285	* Makefile.in: Regenerated with automake-1.11.6.
286	* aclocal.m4: Likewise.
287	* configure: Likewise.
288
2892015-01-26  Matthias Klose  <doko@ubuntu.com>
290
291	* configure.ac: Move AM_ENABLE_MULTILIB before AC_PROG_CC.
292	* configure: Regenerate.
293
2942015-01-05  Jakub Jelinek  <jakub@redhat.com>
295
296	* libquadmath.texi: Bump @copying's copyright year.
297
2982014-11-21  H.J. Lu  <hongjiu.lu@intel.com>
299
300	PR bootstrap/63784
301	* configure: Regenerated.
302
3032014-11-11  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
304
305	PR target/63610
306	* configure: Regenerate.
307
3082014-10-21  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
309
310	PR libquadmath/55821
311	* Makefile.am: Unconditionally define libquadmath_TEXINFOS.
312	* Makefile.in: Regenerate.
313
3142014-10-09  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
315
316	PR libquadmath/55821
317	* Makefile.am: Conditionally define info_TEXINFOS and
318	libquadmath_TEXINFOS.
319	* Makefile.in: Regenerate.
320
3212014-10-08  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
322
323	PR libquadmath/63487
324	* libquadmath.texi (sincosq): Fix typo.
325
3262014-01-02  Tobias Burnus  <burnus@net-b.de>
327
328	* libquadmath.texi: Bump @copying's copyright year.
329
3302013-12-04  Richard Sandiford  <rdsandiford@googlemail.com>
331
332	* Makefile.am (AM_CPPFLAGS): Define.
333	* Makefile.in: Regenerate.
334	* printf/gmp-impl.h: Remove path from longlong.h include.
335
3362013-09-20  Alan Modra  <amodra@gmail.com>
337
338	* configure: Regenerate.
339
3402013-03-06  Shakthi Kannan  <shakthimaan@gmail.com>
341
342	PR libquadmath/55473
343	* quadmath.h: Add 'extern "C"' block for C++ use.
344
3452013-02-19  Jakub Jelinek  <jakub@redhat.com>
346
347	PR libquadmath/56379
348	* strtod/strtod_l.c (mpn_lshift_1): Rewritten as function-like
349	macro.
350
3512013-02-17  Tobias Burnus  <burnus@net-b.de>
352
353	* math/cacoshq.c (cacoshq): Call signbitq instead of signbit.
354
3552013-02-06  Richard Sandiford  <rdsandiford@googlemail.com>
356
357	Revert previous patch.
358
3592013-02-03  Richard Sandiford  <rdsandiford@googlemail.com>
360
361	Update copyright years.
362
3632013-01-22  Jakub Jelinek  <jakub@redhat.com>
364
365	PR libquadmath/56072
366	* libquadmath.texi (M_PI_2q, M_PI_4q): Fix up description.
367
3682012-12-13  Jakub Jelinek  <jakub@redhat.com>
369
370	* math/cbrtq.c (cbrtq): Use Q suffixed floating point constants
371	instead of L suffixed ones.
372	* math/fmaq.c (fmaq): Likewise.
373	* math/rintq.c (TWO112): Likewise.
374
3752012-12-03  Tobias Burnus  <burnus@net-b.de>
376
377	* strtod/strtod_l.c (___STRTOF_INTERNAL): Fix exponent
378	reading.
379
3802012-11-25  Tobias Burnus  <burnus@net-b.de>
381
382	PR libquadmath/55462
383	* strtod/strtod_l.c (round_and_return): Use HAVE_FENV_H
384	instead of nonexisting HAVE_GET_ROUNDING_MODE.
385
3862012-11-23  H.J. Lu  <hongjiu.lu@intel.com>
387
388	PR bootstrap/55455
389	* quadmath-rounding-mode.h (get_rounding_mode): Don't pass
390	void to fegetround.
391
3922012-11-23  Tobias Burnus  <burnus@net-b.de>
393	    Joseph Myers  <joseph@codesourcery.com>
394
395	* quadmath-rounding-mode.h: New.
396	* printf/fpioconst.c: Update from GLIBC. Fix strtod rounding.
397	* printf/fpioconst.h: Ditto.
398	* printf/printf_fp.c (__quadmath_printf_fp): Update from GLIBC.
399	Make printf respect the rounding mode for decimal output.
400	* printf/printf_fphex.c (__quadmath_printf_fphex): Update from
401	GLIBC.  Make printf respect the rounding mode for hex output.
402	* strtod/strtod_l.c: Update from GLIBC. Make strtod respect the
403	rounding mode. Fix strtod handling of underflow.
404
4052012-11-22  David S. Miller  <davem@davemloft.net>
406	    Tobias Burnus  <burnus@net-b.de>
407	    Joseph Myers  <joseph@codesourcery.com>
408
409	* math/atanq.c (atanq): Update from GLIBC. Handle tiny and
410	very large arguments properly.
411	* math/j0q.c (y0q): Update from GLIBC. Avoid arithmetic
412	underflow when 'x' is very small.
413	* math/j1q.c (y1q): Ditto.
414	* math/log1pq.c (log1pq): Update from GLIBC. Saturate
415	nonzero exponents with absolute value below 0x1p-128 to
416	+/- 0x1p-128.
417	* math/powq.c (powq): Update from GLIBC. If xm1 is
418	smaller than LDBL_EPSILON/2.0L, just return xm1.
419
4202012-11-21  Tobias Burnus  <burnus@net-b.de>
421
422	PR libquadmath/55225
423	* math/lgammaq.c (lgammaq): Use local variable if
424	math.h does not provide signgam.
425	* acinclude.m4 (LIBQUAD_CHECK_MATH_H_SIGNGAM): New check.
426	* configure.ac: Use it.
427	* configure: Regenerate.
428	* config.h.in: Regenerate.
429
4302012-11-15  Tobias Burnus  <burnus@net-b.de>
431	    Joseph Myers  <joseph@codesourcery.com>
432
433	* math/fmaq.c (fmaq): Merge from GLIBC. Fix fma
434	underflows with small x * y; Fix overflow results
435	outside round-to-nearest mode; make use of Dekker
436	and Knuth algorithms use round-to-nearest.
437
4382012-11-01  Tobias Burnus  <burnus@net-b.de>
439
440	* math/fmaq.c (fmaq): Fix build.
441
4422012-11-01  Tobias Burnus  <burnus@net-b.de>
443	    Joseph Myers  <joseph@codesourcery.com>
444
445	* math/fmaq.c (fmaq): Merge from GLIBC. Handle cases
446	with small x * y using scaling, not as x * y + z.
447	* math/lgammaq.c (lgammaq): Fix signgam handling.
448
4492012-11-01  Tobias Burnus  <burnus@net-b.de>
450
451	* Makefile.am (libquadmath_la_SOURCES): Add new math/* files.
452	* Makefile.in: Regenerated.
453	* math/acoshq.c: Update comment.
454	* math/acosq.c: Ditto.
455	* math/asinhq.c: Ditto.
456	* math/asinq.c: Ditto.
457	* math/atan2q.c: Ditto.
458	* math/atanhq.c: Ditto.
459	* math/ceilq.c: Ditto.
460	* math/copysignq.c: Ditto.
461	* math/cosq.c: Ditto.
462	* math/coshq.c: Ditto.
463	* math/erfq.c: Ditto.
464	* math/fabsq.c: Ditto.
465	* math/finiteq.c: Ditto.
466	* math/floorq.c: Ditto.
467	* math/fmodq.c: Ditto.
468	* math/frexpq.c: Ditto.
469	* math/isnanq.c: Ditto.
470	* math/j0q.c: Ditto.
471	* math/j1q.c: Ditto.
472	* math/ldexpq.c: Ditto.
473	* math/llroundq.c: Ditto.
474	* math/log10q.c: Ditto.
475	* math/log1pq.c: Ditto.
476	* math/log2q.c: Ditto.
477	* math/logq.c: Ditto.
478	* math/lroundq.c: Ditto.
479	* math/modfq.c: Ditto.
480	* math/nextafterq.c: Ditto.
481	* math/powq.c: Ditto.
482	* math/rem_pio2q.c: Ditto.
483	* math/remainderq.c: Ditto.
484	* math/rintq.c: Ditto.
485	* math/roundq.c: Ditto.
486	* math/scalblnq.c: Ditto.
487	* math/scalbnq.c: Ditto.
488	* math/sincosq_kernel.c: Ditto.
489	* math/sinq.c: Ditto.
490	* math/tanq.c: Ditto.
491	* math/expq.c: Ditto.
492	(__expq_table, expq): Renamed local array from __expl_table.
493	* math/cosq_kernel.c (__quadmath_kernel_cosq): Fix sign handling.
494	* math/cacoshq.c: Changes from GLIBC; fix returned sign.
495	* math/casinhq.c: Changes from GLIBC to fix special-case.
496	* math/cbrtq.c: Use modified GLIBC version.
497	* math/complex.c (ccoshd, cexpq, clog10q, clogq, csinhq, csinq,
498	ctanhq, ctanq): Moved to separates files.
499	(mult_c128, div_c128): Removed no longer needed functions.
500	(cexpiq): Call sincosq instead of sinq and cosq.
501	(cosq): Call cosh(-re,im) instead of cosq/sinq/sinh/cosh.
502	* math/ccoshq.c (ccoshq): New file, moved from complex.c and
503	modified based on GLIBC.
504	* math/cexpq.c (cexp): Ditto.
505	* math/clog10q.c (clog10q): Ditto.
506	* math/clogq.c (clogq): Ditto.
507	* math/csinhq.c: Ditto.
508	* math/csinq.c: Ditto.
509	* math/csqrtq.c: Ditto.
510	* math/ctanhq.c: Ditto.
511	* math/ctanq.c: Ditto.
512	* math/fmaq.c (fmaq): Port TININESS_AFTER_ROUNDING handling
513	from GLIBC.
514	* math/ilogbq.c (ilogbq): Add errno = EDOM handling.
515	* math/isinf_nsq.c (__quadmath_isinf_nsq): New file, ported
516	from GLIBC.
517	* math/lgammaq.c (lgammaq): Add signgam handling.
518	* math/sinhq.c (sinhq): Fix sign handling.
519	* math/sinq_kernel.c (__quadmath_kernel_sinq): Ditto.
520	* math/tgammaq.c (tgammaq): Ditto.
521	* math/x2y2m1q.c: New file.
522	* quadmath-imp.h (TININESS_AFTER_ROUNDING): New define.
523	(__quadmath_x2y2m1q, __quadmath_isinf_nsq): New prototypes.
524
5252012-10-31  Tobias Burnus  <burnus@net-b.de>
526	    Joseph Myers <joseph@codesourcery.com>
527	    David S. Miller <davem@davemloft.net>
528	    Ulrich Drepper <drepper@redhat.com>
529	    Marek Polacek <polacek@redhat.com>:
530	    Petr Baudis <pasky@suse.cz>
531
532	* math/complex.c (csqrtq): NaN and INF fixes.
533	* math/sqrtq.c (sqrt): NaN, INF and < 0 fixes.
534	* math/expm1q.c (expm1q): Changes from GLIBC. Use expq for
535	large parameters. Fix errno for boundary conditions.
536	* math/finiteq.c (finiteq): Add comment.
537	* math/fmaq.c (fmaq): Changes from GLIBC. Fix missing underflows
538	and bad results for some subnormal results. Fix sign of inexact
539	zero return. Fix sign of exact zero return.
540	Ensure additions are not scheduled after fetestexcept.
541	* math/jnq.c (jnq): Changes from GLIBC. Set up errno properly
542	for ynq. Fix jnq precision.
543	* math/nearbyintq.c (nearbyintq): Changes from GLIBC. Do not
544	manipulate bits before adding and subtracting TWO112[sx].
545	* math/rintq.c (rintq): Ditto.
546	* math/scalbnq.c (scalbnq): Changes from GLIBC. Fix integer
547	overflow.
548
5492012-09-14  David Edelsohn  <dje.gcc@gmail.com>
550
551	* configure: Regenerated.
552
5532012-05-29  Benjamin Kosnik  <bkoz@redhat.com>
554
555	PR libstdc++/51007
556	* configure.ac: Allow gnu, gnu* variants for --enable-symvers argument.
557	* configure: Regenerated.
558
5592012-05-16  H.J. Lu  <hongjiu.lu@intel.com>
560
561	* configure: Regenerated.
562
5632011-11-21  Andreas Tobler  <andreast@fgznet.ch>
564
565	* configure: Regenerate.
566
5672011-11-07  Kai Tietz  <ktietz@redhat.com>
568
569	PR target/51007
570	* quadmath-imp.h (ieee854_float128): Adjust
571	for ms-bitfield layout.
572
5732011-11-02  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
574
575	* printf/gmp-impl.h: Adapt path to longlong.h.
576
5772011-09-21  Joseph Myers  <joseph@codesourcery.com>
578
579	* Makefile.am (FLAGS_TO_PASS): Define.
580	* Makefile.in: Regenerate.
581
5822011-08-01  Jakub Jelinek  <jakub@redhat.com>
583
584	* math/rem_pio2q.c (__quadmath_kernel_rem_pio2): Fix up fq to y
585	conversion for prec 3 and __FLT_EVAL_METHOD__ != 0.
586
5872011-04-03  Tobias Burnus  <burnus@net-b.de>
588
589	PR fortran/48311
590	* libquadmath.texi (strtoflt128): Fix typo.
591	(quadmath_snprintf): Move note up, make @item.
592
5932011-03-25  Jakub Jelinek  <jakub@redhat.com>
594
595	* printf/printf_fp.c (__quadmath_printf_fp): Use memcpy instead of
596	mempcpy.
597
5982011-03-21  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
599
600	PR bootstrap/48135
601	* configure.ac (quadmath_use_symver): Handle --disable-symvers.
602	* configure: Regenerate.
603
6042011-02-28  Jakub Jelinek  <jakub@redhat.com>
605
606	* printf/quadmath-printf.c (quadmath_snprintf): Make sure
607	that for size > 0 str is always zero terminated.
608
6092011-02-20  Jakub Jelinek  <jakub@redhat.com>
610
611	PR bootstrap/47827
612	* printf/quadmath-printf.h (NAN): Redefine to __builtin_nanf ("").
613
6142011-02-17  Jakub Jelinek  <jakub@redhat.com>
615
616	* libquadmath.texi (FLT128_DIG, FLT128_MIN_10_EXP,
617	FLT128_MAX_10_EXP): Document.
618	(strtoflt128): Remove obsolete comment.
619	* configure.ac (HAVE_STRTOULL): New check.
620	* printf/gmp-impl.h (mpn_construct_float128): New prototype,
621	define.
622	* printf/mul_n.c: Include <config.h>.
623	* printf/add_n.c: Likewise.
624	* printf/cmp.c: Likewise.
625	* printf/fpioconst.c: Likewise.
626	* printf/mul_1.c: Likewise.
627	* printf/rshift.c: Likewise.
628	* printf/lshift.c: Likewise.
629	* printf/submul_1.c: Likewise.
630	* printf/sub_n.c: Likewise.
631	* printf/divrem.c: Likewise.
632	* printf/addmul_1.c: Likewise.
633	* printf/mul.c: Likewise.
634	* printf/quadmath-printf.h (isupper, isdigit, tolower): Change
635	to avoid evaluating argument multiple times.
636	(isxdigit): Redefine.
637	* strtod/strtoflt128.c: New file.
638	* strtod/strtod_l.c: New file.
639	* strtod/mpn2flt128.c: New file.
640	* strtod/grouping.h: New file.
641	* strtod/tens_in_limb.c: New file.
642	* gdtoa/arith.h: Removed.
643	* gdtoa/gd_qnan.h: Removed.
644	* gdtoa/gdtoa_fltrnds.h: Removed.
645	* gdtoa/gdtoa.h: Removed.
646	* gdtoa/gdtoaimp.h: Removed.
647	* gdtoa/gethex.c: Removed.
648	* gdtoa/gmisc.c: Removed.
649	* gdtoa/hd_init.c: Removed.
650	* gdtoa/hexnan.c: Removed.
651	* gdtoa/makefile: Removed.
652	* gdtoa/misc.c: Removed.
653	* gdtoa/README.gdtoa: Removed.
654	* gdtoa/smisc.c: Removed.
655	* gdtoa/strtodg.c: Removed.
656	* gdtoa/strtopQ.c: Removed.
657	* gdtoa/sum.c: Removed.
658	* quadmath.h (FLT128_DIG, FLT128_MIN_10_EXP, FLT128_MAX_10_EXP):
659	Define.
660	* Makefile.am (libquadmath_la_SOURCES): Remove gdtoa/*, add
661	strtod/strtoflt128.c, strtod/mpn2flt128.c and strtod/tens_in_limb.c.
662	* config.h.in: Regenerated.
663	* configure: Regenerated.
664	* Makefile.in: Regenerated.
665
6662011-02-16  Jakub Jelinek  <jakub@redhat.com>
667
668	* printf/quadmath-printf.c: Also check __GLIBC__ when checking
669	whether workarounds for printf hook handling should be added.
670
671	* configure.ac: Check for locale.h too.
672	(USE_LOCALE_SUPPORT): Remove check.
673	(USE_NL_LANGINFO, USE_NL_LANGINFO_WC, USE_LOCALECONV): New checks.
674	(USE_I18_NUMBER_H): Check also for _NL_CTYPE_MB_CUR_MAX.
675	* printf/printf_fphex.c (__quadmath_printf_fphex): Use nl_langinfo
676	or localeconv for narrow version and nl_langinfo if USE_NL_LANGINFO_WC
677	for wide version.
678	* printf/quadmath-printf.h: Include locale.h if HAVE_LOCALE_H.
679	* printf/printf_fp.c (USE_I18N_NUMBER_H): Don't define to 0.
680	(__quadmath_printf_fp): Use nl_langinfo or localeconv for narrow
681	version and nl_langinfo if USE_NL_LANGINFO_WC for wide version.
682	Guard nl_langinfo (_NL_CTYPE_MB_CUR_MAX) use with
683	USE_I18N_NUMBER_H #ifdef.
684	* configure: Regenerated.
685	* config.h.in: Regenerated.
686
6872011-02-14  Jakub Jelinek  <jakub@redhat.com>
688
689	PR bootstrap/47736
690	* configure.ac (HAVE_PRINTF_HOOKS): Test if printf_info struct has
691	user field.
692	* printf/quadmath-printf.c (quadmath_snprintf): Clear whole info
693	field instead of setting individual fields to 0.  Don't set info.user
694	to -1.
695	* configure: Regenerated.
696
697	* configure.ac (HAVE_HIDDEN_VISIBILITY): Test with -Werror in CFLAGS.
698	* printf/printf_fp.c: Don't include <alloca.h>.
699	* printf/quadmath-printf.h (_itoa): Redefine to __quadmath_itoa.
700	* configure: Regenerated.
701
702	PR fortran/47642
703	* libquadmath.texi (quadmath_snprintf): Document.
704	(quadmath_flt128tostr): Remove.
705	* Makefile.am (libquadmath_la_SOURCES): Add printf/*.c.
706	Remove  quadmath_io.c, gdtoa/gdtoa.c, gdtoa/g__fmt.c,
707	gdtoa/g_Qfmt.c, gdtoa/dmisc.c and gdtoa/ulp.c.
708	* quadmath.h (quadmath_snprintf): New prototype.
709	(quadmath_flt128tostr): Remove.
710	* quadmath_weak.h (quadmath_snprintf): Add.
711	(quadmath_flt128tostr): Remove.
712	* configure.ac: New AC_CHECK_HEADERS headers: langinfo.h, wchar.h,
713	wctype.h, limits.h, ctype.h, printf.h, errno.h.
714	(AC_USE_SYSTEM_EXTENSIONS): Add.
715	(HAVE_HIDDEN_VISIBILITY, HAVE_PRINTF_HOOKS,
716	USE_LOCALE_SUPPORT, USE_I18N_NUMBER_H): New checks.
717	* quadmath.map (QUADMATH_1.0): Add quadmath_snprintf.  Remove
718	quadmath_flt128tostr.
719	* printf/printf_fphex.c: New file.
720	* printf/_itowa.h: New file.
721	* printf/mul_n.c: New file.
722	* printf/quadmath-printf.h: New file.
723	* printf/submul_1.c: New file.
724	* printf/quadmath-printf.c: New file.
725	* printf/gmp-impl.h: New file.
726	* printf/lshift.c: New file.
727	* printf/fpioconst.h: New file.
728	* printf/add_n.c: New file.
729	* printf/cmp.c: New file.
730	* printf/sub_n.c: New file.
731	* printf/mul.c: New file.
732	* printf/divrem.c: New file.
733	* printf/addmul_1.c: New file.
734	* printf/printf_fp.c: New file.
735	* printf/_itoa.h: New file.
736	* printf/fpioconst.c: New file.
737	* printf/_i18n_number.h: New file.
738	* printf/flt1282mpn.c: New file.
739	* printf/rshift.c: New file.
740	* printf/mul_1.c: New file.
741	* quadmath_io.c: Removed.
742	* gdtoa/gdtoa.c: Removed.
743	* gdtoa/g__fmt.c: Removed.
744	* gdtoa/g_Qfmt.c: Removed.
745	* gdtoa/dmisc.c: Removed.
746	* gdtoa/ulp.c: Removed.
747	* config.h.in: Regenerated.
748	* configure: Regenerated.
749	* Makefile.in: Regenerated.
750
7512011-02-13  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
752
753	* Makefile.in: Regenerate.
754	* aclocal.m4: Likewise.
755	* configure: Likewise.
756
7572011-01-31  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
758
759	PR libquadmath/47293
760	* gdtoa/qd_qnan.h: Fix NAN bit patterns.
761	* gdtoa/strtopQ.c (strtoflt128): Handle endianess with NAN.
762
7632011-01-17  Tobias Burnus  <burnus@net-b.de>
764
765	PR fortran/47295
766	* libquadmath.text: Document typedef and constants.
767
7682011-01-17  Tobias Burnus  <burnus@net-b.de>
769
770	PR fortran/46817
771	* quadmath-imp.h: Refer to libquadmath not ot libiberty,
772	bump copyright year and use /**/ instead of // comments.
773	* quadmath.h: Ditto.
774	* quadmath-weak.h: Ditto.
775	* quadmath_io.c: Ditto.
776
7772011-01-16  Jakub Jelinek  <jakub@redhat.com>
778
779	PR libfortran/47322
780	* math/remquoq.c (remquoq): Use uint64_t type instead of u_int64_t.
781
782	PR fortran/46625
783	* quadmath.map (QUADMATH_1.0): Remove quadmath_strtopQ
784	and quadmath_dtoaq.  Add strtoflt128 and quadmath_flt128tostr.
785	* quadmath_weak.h (quadmath_strtopQ, quadmath_dtoaq): Remove.
786	(strtoflt128, quadmath_flt128tostr): Add.
787	* gdtoa/strtopQ.c (quadmath_strtopQ): Rename to...
788	(strtoflt128): ... this.  Return __float128, instead of writing
789	to memory pointed by last argument.
790	* quadmath.h: Use C style comments instead of C++ style.
791	(quadmath_strtopQ, quadmath_dtoaq): Remove prototypes.
792	(strtoflt128, quadmath_flt128tostr): Add prototypes.
793	* libquadmath.texi (quadmath_dtoaq): Rename to quadmath_flt128tostr.
794	(quadmath_strtopQ): Rename to strtoflt128.  Adjust prototype,
795	adjust examples.
796	* quadmath_io.c (quadmath_dtoaq): Rename to...
797	(quadmath_flt128tostr): ... this.
798
799	* quadmath.h (__quadmath_throw, __quadmath_nth): Define.
800	Use it for all prototypes.
801
802	PR fortran/46416
803	* quadmath.h (cbrtq, finiteq, isnanq, signbitq, sqrtq): Remove
804	const from prototype argument.
805	(cimagq, conjq, cprojq, crealq, fdimq, fmaxq, fminq, ilogbq,
806	llrintq, log2q, lrintq, nearbyintq, remquoq): New prototypes.
807	(__quadmath_extern_inline): Define.
808	(cimagq, conjq, crealq): New inlines.
809	* Makefile.am (libquadmath_la_SOURCES): Add math/cimagq.c,
810	math/conjq.c, math/cprojq.c, math/crealq.c, math/fdimq.c,
811	math/fmaxq.c, math/fminq.c, math/ilogbq.c, math/llrintq.c,
812	math/log2q.c, math/lrintq.c, math/nearbyintq.c and math/remquoq.c.
813	* Makefile.in: Regenerated.
814	* quadmath_weak.h (cimagq, conjq, cprojq, crealq, fdimq, fmaxq,
815	fminq, ilogbq, llrintq, log2q, lrintq, nearbyintq, remquoq): Add.
816	* quadmath-imp.h (__LITTLE_ENDIAN__): Don't define.
817	(ieee854_float128): Use __BYTE_ORDER == __ORDER_BIG_ENDIAN__
818	tests instead of __BIG_ENDIAN__.
819	* quadmath.map (QUADMATH_1.0): Add cimagq, conjq, cprojq, crealq,
820	fdimq, fmaxq, fminq, ilogbq, llrintq, log2q, lrintq, nearbyintq
821	and remquoq.
822	* libquadmath.texi (cimagq, conjq, cprojq, crealq, fdimq, fmaxq,
823	fminq, ilogbq, llrintq, log2q, lrintq, nearbyintq, remquoq): Add.
824	* math/cprojq.c: New file.
825	* math/ilogbq.c: New file.
826	* math/fminq.c: New file.
827	* math/llrintq.c: New file.
828	* math/log2q.c: New file.
829	* math/lrintq.c: New file.
830	* math/crealq.c: New file.
831	* math/nearbyintq.c: New file.
832	* math/fmaxq.c: New file.
833	* math/conjq.c: New file.
834	* math/remquoq.c: New file.
835	* math/cimagq.c: New file.
836	* math/fdimq.c: New file.
837
838	* math/ldexpq.c: Include errno.h.  Set errno to ERANGE if needed.
839
840	PR fortran/46416
841	* quadmath.h (cacosq, cacoshq, casinq, casinhq, catanq, catanhq):
842	New prototypes.
843	(M_Eq, M_LOG2Eq, M_LOG10Eq, M_LN2q, M_LN10q, M_PIq, M_PI_2q, M_PI_4q,
844	M_1_PIq, M_2_PIq, M_2_SQRTPIq, M_SQRT2q, M_SQRT1_2q): Define.
845	* quadmath_weak.h (cacosq, cacoshq, casinq, casinhq, catanq,
846	catanhq): Add.
847	* quadmath-imp.h (fpclassifyq, QUADFP_NAN, QUADFP_INFINITE,
848	QUADFP_ZERO, QUADFP_SUBNORMAL, QUADFP_NORMAL): Define.
849	* quadmath.map (QUADMATH_1.0): Add cacosq, cacoshq, casinq, casinhq,
850	catanq and catanhq.
851	* Makefile.am (libquadmath_la_SOURCES): Add math/cacosq.c,
852	math/cacoshq.c, math/casinq.c, math/casinhq.c, math/catanq.c
853	and math/catanhq.c.
854	* Makefile.in: Regenerated.
855	* libquadmath.texi (cacosq, cacoshq, casinq, casinhq,
856	catanq, catanhq): Add.
857	* math/cacoshq.c: New file.
858	* math/cacosq.c: New file.
859	* math/catanq.c: New file.
860	* math/catanhq.c: New file.
861	* math/casinq.c: New file.
862	* math/casinhq.c: New file.
863
864	* math/hypotq.c (hypotq): Use Q suffix instead of L.
865	* math/atan2q.c (tiny, pi_o_4, pi_o_2, pi, pi_lo, atan2q): Likewise.
866	* math/cosq.c (cosq): Likewise.
867
868	PR fortran/46402
869	* quadmath.map (QUADMATH_1.0): Add fmaq.
870	* configure.ac: Check for fenv.h, feholdexcept, fesetround,
871	feupdateenv, fesetenv and fetestexcept.
872	* configure: Regenerated.
873	* config.h.in: Regenerated.
874	* quadmath.h (fmaq): New prototype.
875	* quadmath_weak.h (fmaq): Add.
876	* Makefile.am (libquadmath_la_SOURCES): Add math/fmaq.c.
877	* Makefile.in: Regenerated.
878	* quadmath-imp.h: Include config.h.
879	* math/expq.c: Include fenv.h.
880	(USE_FENV_H): Define if libm support for fe* is there.
881	(expq): Add fesetround etc. support if USE_FENV_H is defined.
882	* math/fmaq.c: New file.
883	* libquadmath.texi (fmaq): Add.
884
8852011-01-14  Tobias Burnus  <burnus@net-b.de>
886
887	PR fortran/47182
888	* configure.ac: Use ACX_BUGURL.
889	* libquadmath.texi: Include libquadmath-vers.texi for BUGURL.
890	* Makefile.am: Create libquadmath-vers.texi.
891	* configure.in: Regenerate.
892	* Makefile.in: Regenerate.
893
8942011-01-06  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
895
896	PR fortran/47174
897	* Makefile.am (libquadmath.info): Unconditionally override
898	target, not only if BUILD_LIBQUADMATH.
899	* Makefile.in: Regenerate.
900	* configure.ac (AM_INIT_AUTOMAKE): Add -Wno-override option to
901	avoid warning from automake.
902
9032010-12-19  Tobias Burnus  <burnus@net-b.de>
904
905	PR fortran/46520
906	* configure.ac: Do not call AC_CHECK_LIB for gcc_no_link.
907	* configure: Regenerate
908
9092010-12-13  Tobias Burnus  <burnus@net-b.de>
910
911	PR fortran/46625
912	* gdtoa/gdtoaimp.h: Mangle internal functions by
913	prefixing them with __quadmath. Don't use gdtoa's strcp(y).
914	* gdtoa/g_Qfmt.c (g_Qfmt): Use strcpy instead of strcp.
915	* gdtoa/misc.c (strcpy): Renamed from strcp and only use
916	if NO_STRING_H is set.
917	* quadmath-imp.h (__quadmath_rem_pio2q, __quadmath_kernel_sincosq
918	__quadmath_kernel_sinq, __quadmath_kernel_cosq): Added
919	__quadmath prefix to internal functions.
920	* math/cosq.c (cosq): Ditto.
921	* math/sinq.c (cosq): Ditto.
922	* math/tanq.c (tanq,__quadmath_kernel_tanq): Ditto.
923	* math/rem_pio2q.c (rem_pio2, __quadmath_kernel_rem_pio2): Ditto.
924	* math/sinq_kernel.c (__quadmath_kernel_sinq): Ditto.
925	* math/cosq_kernel.c (__quadmath_kernel_cosq): Ditto.
926
9272010-12-08  Tobias Burnus  <burnus@net-b.de>
928
929	PR fortran/46772
930	* gdtoa/gdtoa.h: Include stdlib.h instead of declare strtod.
931	* gdtoa/gdtoaimp.h: Don't declare strtod.
932
9332010-12-08  Tobias Burnus  <burnus@net-b.de>
934
935	PR fortran/46520
936	* configure.ac: Use GCC_TRY_COMPILE_OR_LINK instead of
937	AC_TRY_LINK and cache libquad_have_float128.
938	* configure: Regenerate.
939
9402010-12-08  Tobias Burnus  <burnus@net-b.de>
941
942	PR fortran/46543
943	* configure.ac: Add texinfo checks.
944	* Makefile.am: Handle .texi documentation.
945	* libquadmath.texi: New.
946	* configure: Regenerated.
947	* Makefile.in: Regenerated.
948
9492010-12-06  Tobias Burnus  <burnus@net-b.de>
950
951	PR fortran/46817
952	* quadmath-imp.h: Add LGPL header.
953	* quadmath.h: Add LGPL header.
954	* quadmath_io.c: Add LGPL header.
955	* quadmath_weak.h: Add LGPL header.
956
9572010-12-06  Dave Korn  <dave.korn.cygwin@gmail.com>
958
959	PR target/40125
960	PR lto/46695
961	* configure.ac: Invoke ACX_LT_HOST_FLAGS.
962	* Makefile.am (libquadmath_la_LDFLAGS): Use lt_host_flags.
963	* aclocal.m4: Regenerate.
964	* configure: Regenerate.
965	* Makefile.in: Regenerate.
966
9672010-11-30  Tobias Burnus  <burnus@net-b.de>
968
969	PR fortran/46594
970	* Makefile.am: Install include files in
971	target/version specific directory.
972	* Makefile.in: Regenerate.
973
9742010-11-30  Tobias Burnus  <burnus@net-b.de>
975
976	* acinclude.m4 (LIBGOMP_CHECK_SYNC_BUILTINS,
977	LIBGOMP_CHECK_ATTRIBUTE_VISIBILITY,
978	LIBGOMP_CHECK_ATTRIBUTE_DLLEXPORT): Remove.
979
9802010-11-29  Kai Tietz  <kai.tietz@onevision.com>
981
982	* configure: Regenerated.
983	* Makefile.in: Regenerated.
984	* configure.ac (ACX_NONCANONICAL_TARGET): Call it.
985	(target_alias): Use AC_SUBST for it.
986	(GCC_NO_EXECUTABLES): Call it.
987	(AC_PROG_CC): Improve multilib handling.
988	Check that gcc is used for compilation.
989	Use pattern from libssp for symbol versioning checking.
990	* Makefile.am (LTLDFLAGS): Removed.
991	(libquadmath_la_LDFLAGS): Add -no-undefined.
992	(libquadmath_la_LINK): Commented out.
993	(libquadmath_la_DEPENDENCIES): Add libquadmath_la_LIBADD.
994	(AM_MAKEFLAGS): Add hack for argument passing by top-level.
995	(MAKEOVERRIDES): New.
996	* aclocal.m4: Regenerated.
997
9982010-11-19  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
999
1000	* Makefile.am (libquadmath_la_LIBADD): Move -lm ...
1001	(libquadmath_la_LDFLAGS): ... here.
1002	(quadmath.map-sun): Tabify.
1003	Fix sed expression.
1004	* Makefile.in: Regenerate.
1005
10062010-11-16  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
1007	    Tobias Burnus  <burnus@net-b.de>
1008
1009	PR fortran/32049
1010	Initial implementation and checkin.
1011
1012Copyright (C) 2010-2018 Free Software Foundation, Inc.
1013
1014Copying and distribution of this file, with or without modification,
1015are permitted in any medium without royalty provided the copyright
1016notice and this notice are preserved.
1017