libquadmath/ChangeLog

   1 2012-11-25  Tobias Burnus  <burnus@net-b.de>
   2
   3         PR libquadmath/55462
   4         * strtod/strtod_l.c (round_and_return): Use HAVE_FENV_H
   5         instead of nonexisting HAVE_GET_ROUNDING_MODE.
   6
   7 2012-11-23  H.J. Lu  <hongjiu.lu@intel.com>
   8
   9         PR bootstrap/55455
  10         * quadmath-rounding-mode.h (get_rounding_mode): Don't pass
  11         void to fegetround.
  12
  13 2012-11-23  Tobias Burnus  <burnus@net-b.de>
  14             Joseph Myers  <joseph@codesourcery.com>
  15
  16         * quadmath-rounding-mode.h: New.
  17         * printf/fpioconst.c: Update from GLIBC. Fix strtod rounding.
  18         * printf/fpioconst.h: Ditto.
  19         * printf/printf_fp.c (__quadmath_printf_fp): Update from GLIBC.
  20         Make printf respect the rounding mode for decimal output.
  21         * printf/printf_fphex.c (__quadmath_printf_fphex): Update from
  22         GLIBC.  Make printf respect the rounding mode for hex output.
  23         * strtod/strtod_l.c: Update from GLIBC. Make strtod respect the
  24         rounding mode. Fix strtod handling of underflow.
  25
  26 2012-11-22  David S. Miller  <davem@davemloft.net>
  27             Tobias Burnus  <burnus@net-b.de>
  28             Joseph Myers  <joseph@codesourcery.com>
  29
  30         * math/atanq.c (atanq): Update from GLIBC. Handle tiny and
  31         very large arguments properly.
  32         * math/j0q.c (y0q): Update from GLIBC. Avoid arithmetic
  33         underflow when 'x' is very small.
  34         * math/j1q.c (y1q): Ditto.
  35         * math/log1pq.c (log1pq): Update from GLIBC. Saturate
  36         nonzero exponents with absolute value below 0x1p-128 to
  37         +/- 0x1p-128.
  38         * math/powq.c (powq): Update from GLIBC. If xm1 is
  39         smaller than LDBL_EPSILON/2.0L, just return xm1.
  40
  41 2012-11-21  Tobias Burnus  <burnus@net-b.de>
  42
  43         PR libquadmath/55225
  44         * math/lgammaq.c (lgammaq): Use local variable if
  45         math.h does not provide signgam.
  46         * acinclude.m4 (LIBQUAD_CHECK_MATH_H_SIGNGAM): New check.
  47         * configure.ac: Use it.
  48         * configure: Regenerate.
  49         * config.h.in: Regenerate.
  50
  51 2012-11-15  Tobias Burnus  <burnus@net-b.de>
  52             Joseph Myers  <joseph@codesourcery.com>
  53
  54         * math/fmaq.c (fmaq): Merge from GLIBC. Fix fma
  55         underflows with small x * y; Fix overflow results
  56         outside round-to-nearest mode; make use of Dekker
  57         and Knuth algorithms use round-to-nearest.
  58
  59 2012-11-01  Tobias Burnus  <burnus@net-b.de>
  60
  61         * math/fmaq.c (fmaq): Fix build.
  62
  63 2012-11-01  Tobias Burnus  <burnus@net-b.de>
  64             Joseph Myers  <joseph@codesourcery.com>
  65
  66         * math/fmaq.c (fmaq): Merge from GLIBC. Handle cases
  67         with small x * y using scaling, not as x * y + z.
  68         * math/lgammaq.c (lgammaq): Fix signgam handling.
  69
  70 2012-11-01  Tobias Burnus  <burnus@net-b.de>
  71
  72         * Makefile.am (libquadmath_la_SOURCES): Add new math/* files.
  73         * Makefile.in: Regenerated.
  74         * math/acoshq.c: Update comment.
  75         * math/acosq.c: Ditto.
  76         * math/asinhq.c: Ditto.
  77         * math/asinq.c: Ditto.
  78         * math/atan2q.c: Ditto.
  79         * math/atanhq.c: Ditto.
  80         * math/ceilq.c: Ditto.
  81         * math/copysignq.c: Ditto.
  82         * math/cosq.c: Ditto.
  83         * math/coshq.c: Ditto.
  84         * math/erfq.c: Ditto.
  85         * math/fabsq.c: Ditto.
  86         * math/finiteq.c: Ditto.
  87         * math/floorq.c: Ditto.
  88         * math/fmodq.c: Ditto.
  89         * math/frexpq.c: Ditto.
  90         * math/isnanq.c: Ditto.
  91         * math/j0q.c: Ditto.
  92         * math/j1q.c: Ditto.
  93         * math/ldexpq.c: Ditto.
  94         * math/llroundq.c: Ditto.
  95         * math/log10q.c: Ditto.
  96         * math/log1pq.c: Ditto.
  97         * math/log2q.c: Ditto.
  98         * math/logq.c: Ditto.
  99         * math/lroundq.c: Ditto.
 100         * math/modfq.c: Ditto.
 101         * math/nextafterq.c: Ditto.
 102         * math/powq.c: Ditto.
 103         * math/rem_pio2q.c: Ditto.
 104         * math/remainderq.c: Ditto.
 105         * math/rintq.c: Ditto.
 106         * math/roundq.c: Ditto.
 107         * math/scalblnq.c: Ditto.
 108         * math/scalbnq.c: Ditto.
 109         * math/sincosq_kernel.c: Ditto.
 110         * math/sinq.c: Ditto.
 111         * math/tanq.c: Ditto.
 112         * math/expq.c: Ditto.
 113         (__expq_table, expq): Renamed local array from __expl_table.
 114         * math/cosq_kernel.c (__quadmath_kernel_cosq): Fix sign handling.
 115         * math/cacoshq.c: Changes from GLIBC; fix returned sign.
 116         * math/casinhq.c: Changes from GLIBC to fix special-case.
 117         * math/cbrtq.c: Use modified GLIBC version.
 118         * math/complex.c (ccoshd, cexpq, clog10q, clogq, csinhq, csinq,
 119         ctanhq, ctanq): Moved to separates files.
 120         (mult_c128, div_c128): Removed no longer needed functions.
 121         (cexpiq): Call sincosq instead of sinq and cosq.
 122         (cosq): Call cosh(-re,im) instead of cosq/sinq/sinh/cosh.
 123         * math/ccoshq.c (ccoshq): New file, moved from complex.c and
 124         modified based on GLIBC.
 125         * math/cexpq.c (cexp): Ditto.
 126         * math/clog10q.c (clog10q): Ditto.
 127         * math/clogq.c (clogq): Ditto.
 128         * math/csinhq.c: Ditto.
 129         * math/csinq.c: Ditto.
 130         * math/csqrtq.c: Ditto.
 131         * math/ctanhq.c: Ditto.
 132         * math/ctanq.c: Ditto.
 133         * math/fmaq.c (fmaq): Port TININESS_AFTER_ROUNDING handling
 134         from GLIBC.
 135         * math/ilogbq.c (ilogbq): Add errno = EDOM handling.
 136         * math/isinf_nsq.c (__quadmath_isinf_nsq): New file, ported
 137         from GLIBC.
 138         * math/lgammaq.c (lgammaq): Add signgam handling.
 139         * math/sinhq.c (sinhq): Fix sign handling.
 140         * math/sinq_kernel.c (__quadmath_kernel_sinq): Ditto.
 141         * math/tgammaq.c (tgammaq): Ditto.
 142         * math/x2y2m1q.c: New file.
 143         * quadmath-imp.h (TININESS_AFTER_ROUNDING): New define.
 144         (__quadmath_x2y2m1q, __quadmath_isinf_nsq): New prototypes.
 145
 146 2012-10-31  Tobias Burnus  <burnus@net-b.de>
 147             Joseph Myers <joseph@codesourcery.com>
 148             David S. Miller <davem@davemloft.net>
 149             Ulrich Drepper <drepper@redhat.com>
 150             Marek Polacek <polacek@redhat.com>:
 151             Petr Baudis <pasky@suse.cz>
 152
 153         * math/complex.c (csqrtq): NaN and INF fixes.
 154         * math/sqrtq.c (sqrt): NaN, INF and < 0 fixes.
 155         * math/expm1q.c (expm1q): Changes from GLIBC. Use expq for
 156         large parameters. Fix errno for boundary conditions.
 157         * math/finiteq.c (finiteq): Add comment.
 158         * math/fmaq.c (fmaq): Changes from GLIBC. Fix missing underflows
 159         and bad results for some subnormal results. Fix sign of inexact
 160         zero return. Fix sign of exact zero return.
 161         Ensure additions are not scheduled after fetestexcept.
 162         * math/jnq.c (jnq): Changes from GLIBC. Set up errno properly
 163         for ynq. Fix jnq precision.
 164         * math/nearbyintq.c (nearbyintq): Changes from GLIBC. Do not
 165         manipulate bits before adding and subtracting TWO112[sx].
 166         * math/rintq.c (rintq): Ditto.
 167         * math/scalbnq.c (scalbnq): Changes from GLIBC. Fix integer
 168         overflow.
 169
 170 2012-09-14  David Edelsohn  <dje.gcc@gmail.com>
 171
 172         * configure: Regenerated.
 173
 174 2012-05-29  Benjamin Kosnik  <bkoz@redhat.com>
 175
 176         PR libstdc++/51007
 177         * configure.ac: Allow gnu, gnu* variants for --enable-symvers argument.
 178         * configure: Regenerated.
 179
 180 2012-05-16  H.J. Lu  <hongjiu.lu@intel.com>
 181
 182         * configure: Regenerated.
 183
 184 2011-11-21  Andreas Tobler  <andreast@fgznet.ch>
 185
 186         * configure: Regenerate.
 187
 188 2011-11-07  Kai Tietz  <ktietz@redhat.com>
 189
 190         PR target/51007
 191         * quadmath-imp.h (ieee854_float128): Adjust
 192         for ms-bitfield layout.
 193
 194 2011-11-02  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
 195
 196         * printf/gmp-impl.h: Adapt path to longlong.h.
 197
 198 2011-09-21  Joseph Myers  <joseph@codesourcery.com>
 199
 200         * Makefile.am (FLAGS_TO_PASS): Define.
 201         * Makefile.in: Regenerate.
 202
 203 2011-08-01  Jakub Jelinek  <jakub@redhat.com>
 204
 205         * math/rem_pio2q.c (__quadmath_kernel_rem_pio2): Fix up fq to y
 206         conversion for prec 3 and __FLT_EVAL_METHOD__ != 0.
 207
 208 2011-04-03  Tobias Burnus  <burnus@net-b.de>
 209
 210         PR fortran/48311
 211         * libquadmath.texi (strtoflt128): Fix typo.
 212         (quadmath_snprintf): Move note up, make @item.
 213
 214 2011-03-25  Jakub Jelinek  <jakub@redhat.com>
 215
 216         * printf/printf_fp.c (__quadmath_printf_fp): Use memcpy instead of
 217         mempcpy.
 218
 219 2011-03-21  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
 220
 221         PR bootstrap/48135
 222         * configure.ac (quadmath_use_symver): Handle --disable-symvers.
 223         * configure: Regenerate.
 224
 225 2011-02-28  Jakub Jelinek  <jakub@redhat.com>
 226
 227         * printf/quadmath-printf.c (quadmath_snprintf): Make sure
 228         that for size > 0 str is always zero terminated.
 229
 230 2011-02-20  Jakub Jelinek  <jakub@redhat.com>
 231
 232         PR bootstrap/47827
 233         * printf/quadmath-printf.h (NAN): Redefine to __builtin_nanf ("").
 234
 235 2011-02-17  Jakub Jelinek  <jakub@redhat.com>
 236
 237         * libquadmath.texi (FLT128_DIG, FLT128_MIN_10_EXP,
 238         FLT128_MAX_10_EXP): Document.
 239         (strtoflt128): Remove obsolete comment.
 240         * configure.ac (HAVE_STRTOULL): New check.
 241         * printf/gmp-impl.h (mpn_construct_float128): New prototype,
 242         define.
 243         * printf/mul_n.c: Include <config.h>.
 244         * printf/add_n.c: Likewise.
 245         * printf/cmp.c: Likewise.
 246         * printf/fpioconst.c: Likewise.
 247         * printf/mul_1.c: Likewise.
 248         * printf/rshift.c: Likewise.
 249         * printf/lshift.c: Likewise.
 250         * printf/submul_1.c: Likewise.
 251         * printf/sub_n.c: Likewise.
 252         * printf/divrem.c: Likewise.
 253         * printf/addmul_1.c: Likewise.
 254         * printf/mul.c: Likewise.
 255         * printf/quadmath-printf.h (isupper, isdigit, tolower): Change
 256         to avoid evaluating argument multiple times.
 257         (isxdigit): Redefine.
 258         * strtod/strtoflt128.c: New file.
 259         * strtod/strtod_l.c: New file.
 260         * strtod/mpn2flt128.c: New file.
 261         * strtod/grouping.h: New file.
 262         * strtod/tens_in_limb.c: New file.
 263         * gdtoa/arith.h: Removed.
 264         * gdtoa/gd_qnan.h: Removed.
 265         * gdtoa/gdtoa_fltrnds.h: Removed.
 266         * gdtoa/gdtoa.h: Removed.
 267         * gdtoa/gdtoaimp.h: Removed.
 268         * gdtoa/gethex.c: Removed.
 269         * gdtoa/gmisc.c: Removed.
 270         * gdtoa/hd_init.c: Removed.
 271         * gdtoa/hexnan.c: Removed.
 272         * gdtoa/makefile: Removed.
 273         * gdtoa/misc.c: Removed.
 274         * gdtoa/README.gdtoa: Removed.
 275         * gdtoa/smisc.c: Removed.
 276         * gdtoa/strtodg.c: Removed.
 277         * gdtoa/strtopQ.c: Removed.
 278         * gdtoa/sum.c: Removed.
 279         * quadmath.h (FLT128_DIG, FLT128_MIN_10_EXP, FLT128_MAX_10_EXP):
 280         Define.
 281         * Makefile.am (libquadmath_la_SOURCES): Remove gdtoa/*, add
 282         strtod/strtoflt128.c, strtod/mpn2flt128.c and strtod/tens_in_limb.c.
 283         * config.h.in: Regenerated.
 284         * configure: Regenerated.
 285         * Makefile.in: Regenerated.
 286
 287 2011-02-16  Jakub Jelinek  <jakub@redhat.com>
 288
 289         * printf/quadmath-printf.c: Also check __GLIBC__ when checking
 290         whether workarounds for printf hook handling should be added.
 291
 292         * configure.ac: Check for locale.h too.
 293         (USE_LOCALE_SUPPORT): Remove check.
 294         (USE_NL_LANGINFO, USE_NL_LANGINFO_WC, USE_LOCALECONV): New checks.
 295         (USE_I18_NUMBER_H): Check also for _NL_CTYPE_MB_CUR_MAX.
 296         * printf/printf_fphex.c (__quadmath_printf_fphex): Use nl_langinfo
 297         or localeconv for narrow version and nl_langinfo if USE_NL_LANGINFO_WC
 298         for wide version.
 299         * printf/quadmath-printf.h: Include locale.h if HAVE_LOCALE_H.
 300         * printf/printf_fp.c (USE_I18N_NUMBER_H): Don't define to 0.
 301         (__quadmath_printf_fp): Use nl_langinfo or localeconv for narrow
 302         version and nl_langinfo if USE_NL_LANGINFO_WC for wide version.
 303         Guard nl_langinfo (_NL_CTYPE_MB_CUR_MAX) use with
 304         USE_I18N_NUMBER_H #ifdef.
 305         * configure: Regenerated.
 306         * config.h.in: Regenerated.
 307
 308 2011-02-14  Jakub Jelinek  <jakub@redhat.com>
 309
 310         PR bootstrap/47736
 311         * configure.ac (HAVE_PRINTF_HOOKS): Test if printf_info struct has
 312         user field.
 313         * printf/quadmath-printf.c (quadmath_snprintf): Clear whole info
 314         field instead of setting individual fields to 0.  Don't set info.user
 315         to -1.
 316         * configure: Regenerated.
 317
 318         * configure.ac (HAVE_HIDDEN_VISIBILITY): Test with -Werror in CFLAGS.
 319         * printf/printf_fp.c: Don't include <alloca.h>.
 320         * printf/quadmath-printf.h (_itoa): Redefine to __quadmath_itoa.
 321         * configure: Regenerated.
 322
 323         PR fortran/47642
 324         * libquadmath.texi (quadmath_snprintf): Document.
 325         (quadmath_flt128tostr): Remove.
 326         * Makefile.am (libquadmath_la_SOURCES): Add printf/*.c.
 327         Remove  quadmath_io.c, gdtoa/gdtoa.c, gdtoa/g__fmt.c,
 328         gdtoa/g_Qfmt.c, gdtoa/dmisc.c and gdtoa/ulp.c.
 329         * quadmath.h (quadmath_snprintf): New prototype.
 330         (quadmath_flt128tostr): Remove.
 331         * quadmath_weak.h (quadmath_snprintf): Add.
 332         (quadmath_flt128tostr): Remove.
 333         * configure.ac: New AC_CHECK_HEADERS headers: langinfo.h, wchar.h,
 334         wctype.h, limits.h, ctype.h, printf.h, errno.h.
 335         (AC_USE_SYSTEM_EXTENSIONS): Add.
 336         (HAVE_HIDDEN_VISIBILITY, HAVE_PRINTF_HOOKS,
 337         USE_LOCALE_SUPPORT, USE_I18N_NUMBER_H): New checks.
 338         * quadmath.map (QUADMATH_1.0): Add quadmath_snprintf.  Remove
 339         quadmath_flt128tostr.
 340         * printf/printf_fphex.c: New file.
 341         * printf/_itowa.h: New file.
 342         * printf/mul_n.c: New file.
 343         * printf/quadmath-printf.h: New file.
 344         * printf/submul_1.c: New file.
 345         * printf/quadmath-printf.c: New file.
 346         * printf/gmp-impl.h: New file.
 347         * printf/lshift.c: New file.
 348         * printf/fpioconst.h: New file.
 349         * printf/add_n.c: New file.
 350         * printf/cmp.c: New file.
 351         * printf/sub_n.c: New file.
 352         * printf/mul.c: New file.
 353         * printf/divrem.c: New file.
 354         * printf/addmul_1.c: New file.
 355         * printf/printf_fp.c: New file.
 356         * printf/_itoa.h: New file.
 357         * printf/fpioconst.c: New file.
 358         * printf/_i18n_number.h: New file.
 359         * printf/flt1282mpn.c: New file.
 360         * printf/rshift.c: New file.
 361         * printf/mul_1.c: New file.
 362         * quadmath_io.c: Removed.
 363         * gdtoa/gdtoa.c: Removed.
 364         * gdtoa/g__fmt.c: Removed.
 365         * gdtoa/g_Qfmt.c: Removed.
 366         * gdtoa/dmisc.c: Removed.
 367         * gdtoa/ulp.c: Removed.
 368         * config.h.in: Regenerated.
 369         * configure: Regenerated.
 370         * Makefile.in: Regenerated.
 371
 372 2011-02-13  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 373
 374         * Makefile.in: Regenerate.
 375         * aclocal.m4: Likewise.
 376         * configure: Likewise.
 377
 378 2011-01-31  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
 379
 380         PR libquadmath/47293
 381         * gdtoa/qd_qnan.h: Fix NAN bit patterns.
 382         * gdtoa/strtopQ.c (strtoflt128): Handle endianess with NAN.
 383
 384 2011-01-17  Tobias Burnus  <burnus@net-b.de>
 385
 386         PR fortran/47295
 387         * libquadmath.text: Document typedef and constants.
 388
 389 2011-01-17  Tobias Burnus  <burnus@net-b.de>
 390
 391         PR fortran/46817
 392         * quadmath-imp.h: Refer to libquadmath not ot libiberty,
 393         bump copyright year and use /**/ instead of // comments.
 394         * quadmath.h: Ditto.
 395         * quadmath-weak.h: Ditto.
 396         * quadmath_io.c: Ditto.
 397
 398 2011-01-16  Jakub Jelinek  <jakub@redhat.com>
 399
 400         PR libfortran/47322
 401         * math/remquoq.c (remquoq): Use uint64_t type instead of u_int64_t.
 402
 403         PR fortran/46625
 404         * quadmath.map (QUADMATH_1.0): Remove quadmath_strtopQ
 405         and quadmath_dtoaq.  Add strtoflt128 and quadmath_flt128tostr.
 406         * quadmath_weak.h (quadmath_strtopQ, quadmath_dtoaq): Remove.
 407         (strtoflt128, quadmath_flt128tostr): Add.
 408         * gdtoa/strtopQ.c (quadmath_strtopQ): Rename to...
 409         (strtoflt128): ... this.  Return __float128, instead of writing
 410         to memory pointed by last argument.
 411         * quadmath.h: Use C style comments instead of C++ style.
 412         (quadmath_strtopQ, quadmath_dtoaq): Remove prototypes.
 413         (strtoflt128, quadmath_flt128tostr): Add prototypes.
 414         * libquadmath.texi (quadmath_dtoaq): Rename to quadmath_flt128tostr.
 415         (quadmath_strtopQ): Rename to strtoflt128.  Adjust prototype,
 416         adjust examples.
 417         * quadmath_io.c (quadmath_dtoaq): Rename to...
 418         (quadmath_flt128tostr): ... this.
 419
 420         * quadmath.h (__quadmath_throw, __quadmath_nth): Define.
 421         Use it for all prototypes.
 422
 423         PR fortran/46416
 424         * quadmath.h (cbrtq, finiteq, isnanq, signbitq, sqrtq): Remove
 425         const from prototype argument.
 426         (cimagq, conjq, cprojq, crealq, fdimq, fmaxq, fminq, ilogbq,
 427         llrintq, log2q, lrintq, nearbyintq, remquoq): New prototypes.
 428         (__quadmath_extern_inline): Define.
 429         (cimagq, conjq, crealq): New inlines.
 430         * Makefile.am (libquadmath_la_SOURCES): Add math/cimagq.c,
 431         math/conjq.c, math/cprojq.c, math/crealq.c, math/fdimq.c,
 432         math/fmaxq.c, math/fminq.c, math/ilogbq.c, math/llrintq.c,
 433         math/log2q.c, math/lrintq.c, math/nearbyintq.c and math/remquoq.c.
 434         * Makefile.in: Regenerated.
 435         * quadmath_weak.h (cimagq, conjq, cprojq, crealq, fdimq, fmaxq,
 436         fminq, ilogbq, llrintq, log2q, lrintq, nearbyintq, remquoq): Add.
 437         * quadmath-imp.h (__LITTLE_ENDIAN__): Don't define.
 438         (ieee854_float128): Use __BYTE_ORDER == __ORDER_BIG_ENDIAN__
 439         tests instead of __BIG_ENDIAN__.
 440         * quadmath.map (QUADMATH_1.0): Add cimagq, conjq, cprojq, crealq,
 441         fdimq, fmaxq, fminq, ilogbq, llrintq, log2q, lrintq, nearbyintq
 442         and remquoq.
 443         * libquadmath.texi (cimagq, conjq, cprojq, crealq, fdimq, fmaxq,
 444         fminq, ilogbq, llrintq, log2q, lrintq, nearbyintq, remquoq): Add.
 445         * math/cprojq.c: New file.
 446         * math/ilogbq.c: New file.
 447         * math/fminq.c: New file.
 448         * math/llrintq.c: New file.
 449         * math/log2q.c: New file.
 450         * math/lrintq.c: New file.
 451         * math/crealq.c: New file.
 452         * math/nearbyintq.c: New file.
 453         * math/fmaxq.c: New file.
 454         * math/conjq.c: New file.
 455         * math/remquoq.c: New file.
 456         * math/cimagq.c: New file.
 457         * math/fdimq.c: New file.
 458
 459         * math/ldexpq.c: Include errno.h.  Set errno to ERANGE if needed.
 460
 461         PR fortran/46416
 462         * quadmath.h (cacosq, cacoshq, casinq, casinhq, catanq, catanhq):
 463         New prototypes.
 464         (M_Eq, M_LOG2Eq, M_LOG10Eq, M_LN2q, M_LN10q, M_PIq, M_PI_2q, M_PI_4q,
 465         M_1_PIq, M_2_PIq, M_2_SQRTPIq, M_SQRT2q, M_SQRT1_2q): Define.
 466         * quadmath_weak.h (cacosq, cacoshq, casinq, casinhq, catanq,
 467         catanhq): Add.
 468         * quadmath-imp.h (fpclassifyq, QUADFP_NAN, QUADFP_INFINITE,
 469         QUADFP_ZERO, QUADFP_SUBNORMAL, QUADFP_NORMAL): Define.
 470         * quadmath.map (QUADMATH_1.0): Add cacosq, cacoshq, casinq, casinhq,
 471         catanq and catanhq.
 472         * Makefile.am (libquadmath_la_SOURCES): Add math/cacosq.c,
 473         math/cacoshq.c, math/casinq.c, math/casinhq.c, math/catanq.c
 474         and math/catanhq.c.
 475         * Makefile.in: Regenerated.
 476         * libquadmath.texi (cacosq, cacoshq, casinq, casinhq,
 477         catanq, catanhq): Add.
 478         * math/cacoshq.c: New file.
 479         * math/cacosq.c: New file.
 480         * math/catanq.c: New file.
 481         * math/catanhq.c: New file.
 482         * math/casinq.c: New file.
 483         * math/casinhq.c: New file.
 484
 485         * math/hypotq.c (hypotq): Use Q suffix instead of L.
 486         * math/atan2q.c (tiny, pi_o_4, pi_o_2, pi, pi_lo, atan2q): Likewise.
 487         * math/cosq.c (cosq): Likewise.
 488
 489         PR fortran/46402
 490         * quadmath.map (QUADMATH_1.0): Add fmaq.
 491         * configure.ac: Check for fenv.h, feholdexcept, fesetround,
 492         feupdateenv, fesetenv and fetestexcept.
 493         * configure: Regenerated.
 494         * config.h.in: Regenerated.
 495         * quadmath.h (fmaq): New prototype.
 496         * quadmath_weak.h (fmaq): Add.
 497         * Makefile.am (libquadmath_la_SOURCES): Add math/fmaq.c.
 498         * Makefile.in: Regenerated.
 499         * quadmath-imp.h: Include config.h.
 500         * math/expq.c: Include fenv.h.
 501         (USE_FENV_H): Define if libm support for fe* is there.
 502         (expq): Add fesetround etc. support if USE_FENV_H is defined.
 503         * math/fmaq.c: New file.
 504         * libquadmath.texi (fmaq): Add.
 505
 506 2011-01-14  Tobias Burnus  <burnus@net-b.de>
 507
 508         PR fortran/47182
 509         * configure.ac: Use ACX_BUGURL.
 510         * libquadmath.texi: Include libquadmath-vers.texi for BUGURL.
 511         * Makefile.am: Create libquadmath-vers.texi.
 512         * configure.in: Regenerate.
 513         * Makefile.in: Regenerate.
 514
 515 2011-01-06  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 516
 517         PR fortran/47174
 518         * Makefile.am (libquadmath.info): Unconditionally override
 519         target, not only if BUILD_LIBQUADMATH.
 520         * Makefile.in: Regenerate.
 521         * configure.ac (AM_INIT_AUTOMAKE): Add -Wno-override option to
 522         avoid warning from automake.
 523
 524 2010-12-19  Tobias Burnus  <burnus@net-b.de>
 525
 526         PR fortran/46520
 527         * configure.ac: Do not call AC_CHECK_LIB for gcc_no_link.
 528         * configure: Regenerate
 529
 530 2010-12-13  Tobias Burnus  <burnus@net-b.de>
 531
 532         PR fortran/46625
 533         * gdtoa/gdtoaimp.h: Mangle internal functions by
 534         prefixing them with __quadmath. Don't use gdtoa's strcp(y).
 535         * gdtoa/g_Qfmt.c (g_Qfmt): Use strcpy instead of strcp.
 536         * gdtoa/misc.c (strcpy): Renamed from strcp and only use
 537         if NO_STRING_H is set.
 538         * quadmath-imp.h (__quadmath_rem_pio2q, __quadmath_kernel_sincosq
 539         __quadmath_kernel_sinq, __quadmath_kernel_cosq): Added
 540         __quadmath prefix to internal functions.
 541         * math/cosq.c (cosq): Ditto.
 542         * math/sinq.c (cosq): Ditto.
 543         * math/tanq.c (tanq,__quadmath_kernel_tanq): Ditto.
 544         * math/rem_pio2q.c (rem_pio2, __quadmath_kernel_rem_pio2): Ditto.
 545         * math/sinq_kernel.c (__quadmath_kernel_sinq): Ditto.
 546         * math/cosq_kernel.c (__quadmath_kernel_cosq): Ditto.
 547
 548 2010-12-08  Tobias Burnus  <burnus@net-b.de>
 549
 550         PR fortran/46772
 551         * gdtoa/gdtoa.h: Include stdlib.h instead of declare strtod.
 552         * gdtoa/gdtoaimp.h: Don't declare strtod.
 553
 554 2010-12-08  Tobias Burnus  <burnus@net-b.de>
 555
 556         PR fortran/46520
 557         * configure.ac: Use GCC_TRY_COMPILE_OR_LINK instead of
 558         AC_TRY_LINK and cache libquad_have_float128.
 559         * configure: Regenerate.
 560
 561 2010-12-08  Tobias Burnus  <burnus@net-b.de>
 562
 563         PR fortran/46543
 564         * configure.ac: Add texinfo checks.
 565         * Makefile.am: Handle .texi documentation.
 566         * libquadmath.texi: New.
 567         * configure: Regenerated.
 568         * Makefile.in: Regenerated.
 569
 570 2010-12-06  Tobias Burnus  <burnus@net-b.de>
 571
 572         PR fortran/46817
 573         * quadmath-imp.h: Add LGPL header.
 574         * quadmath.h: Add LGPL header.
 575         * quadmath_io.c: Add LGPL header.
 576         * quadmath_weak.h: Add LGPL header.
 577
 578 2010-12-06  Dave Korn  <dave.korn.cygwin@gmail.com>
 579
 580         PR target/40125
 581         PR lto/46695
 582         * configure.ac: Invoke ACX_LT_HOST_FLAGS.
 583         * Makefile.am (libquadmath_la_LDFLAGS): Use lt_host_flags.
 584         * aclocal.m4: Regenerate.
 585         * configure: Regenerate.
 586         * Makefile.in: Regenerate.
 587
 588 2010-11-30  Tobias Burnus  <burnus@net-b.de>
 589
 590         PR fortran/46594
 591         * Makefile.am: Install include files in
 592         target/version specific directory.
 593         * Makefile.in: Regenerate.
 594
 595 2010-11-30  Tobias Burnus  <burnus@net-b.de>
 596
 597         * acinclude.m4 (LIBGOMP_CHECK_SYNC_BUILTINS,
 598         LIBGOMP_CHECK_ATTRIBUTE_VISIBILITY,
 599         LIBGOMP_CHECK_ATTRIBUTE_DLLEXPORT): Remove.
 600
 601 2010-11-29  Kai Tietz  <kai.tietz@onevision.com>
 602
 603         * configure: Regenerated.
 604         * Makefile.in: Regenerated.
 605         * configure.ac (ACX_NONCANONICAL_TARGET): Call it.
 606         (target_alias): Use AC_SUBST for it.
 607         (GCC_NO_EXECUTABLES): Call it.
 608         (AC_PROG_CC): Improve multilib handling.
 609         Check that gcc is used for compilation.
 610         Use pattern from libssp for symbol versioning checking.
 611         * Makefile.am (LTLDFLAGS): Removed.
 612         (libquadmath_la_LDFLAGS): Add -no-undefined.
 613         (libquadmath_la_LINK): Commented out.
 614         (libquadmath_la_DEPENDENCIES): Add libquadmath_la_LIBADD.
 615         (AM_MAKEFLAGS): Add hack for argument passing by top-level.
 616         (MAKEOVERRIDES): New.
 617         * aclocal.m4: Regenerated.
 618
 619 2010-11-19  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
 620
 621         * Makefile.am (libquadmath_la_LIBADD): Move -lm ...
 622         (libquadmath_la_LDFLAGS): ... here.
 623         (quadmath.map-sun): Tabify.
 624         Fix sed expression.
 625         * Makefile.in: Regenerate.
 626
 627 2010-11-16  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>
 628             Tobias Burnus  <burnus@net-b.de>
 629
 630         PR fortran/32049
 631         Initial implementation and checkin.