ChangeLog

   1 2005-12-15  Dmitry V. Kustov  <kustov@telex221.ru>
   2
   3         * src/base/random/cl_random_from.cc: Add support for OpenBSD.
   4
   5 2005-11-23  Richard B. Kreckel  <kreckel@ginac.de>
   6
   7         * Version 1.1.11 released.
   8
   9 2005-11-20  Richard B. Kreckel  <kreckel@ginac.de>
  10
  11         * src/integer/conv/cl_I_cached_power.h: New file.
  12         * src/integer/conv/cl_I_cached_power.cc: New file.
  13         Contains power_table and cached_power_table previously...
  14         * src/integer/conv/cl_I_to_digits.cc: ...here.
  15         * src/integer/conv/cl_I_from_digits.cc: Use cached powers.
  16
  17 2005-11-02  Richard B. Kreckel  <kreckel@ginac.de>
  18
  19         * src/integer/conv/cl_I_from_digits.cc: Made input of all numbers in
  20         non-power-of-two base much faster.
  21         * tests/test_I_io.cc: New file...
  22         * tests/Makefile.in, tests/test_I.cc: ...used here.
  23
  24 2005-10-22  Richard B. Kreckel  <kreckel@ginac.de>
  25
  26         * Version 1.1.10 released.
  27
  28 2005-10-22  Richard B. Kreckel  <kreckel@ginac.de>
  29
  30         * src/Makefile.in: Accept CPPFLAGS from environment.
  31         * examples/Makefile.in: Likewise.
  32         * benchmarks/Makefile.in: Likewise.
  33         * tests/Makefile.in: Likewise.
  34
  35 2005-08-30  Richard B. Kreckel  <kreckel@ginac.de>
  36
  37         * include/cln/modules.h (CL_OUTPUT_LABEL): Work around redundant
  38         duplication of basic blocks on m68k.
  39
  40 2005-08-30  Richard B. Kreckel  <kreckel@ginac.de>
  41
  42         * include/cln/modules.h (CL_JUMP_TO): Fix mips* brokenness.
  43
  44 2005-08-27  Bruno Haible  <bruno@clisp.org>
  45
  46         Split aclocal.m4 into individual files.
  47         * m4/alloca.m4, m4/as-underscore.m4, m4/c++-constructors.m4:
  48         * m4/fpu_control.m4, m4/general.m4, m4/gettimeofday.m4:
  49         * m4/longdouble.m4, m4/longlong.m4, m4/param.m4, m4/perror.m4:
  50         * m4/proto.m4, m4/rusage.m4, m4/times.m4:
  51         New files, extracted from autoconf/aclocal.m4.
  52         * autoconf/aclocal.m4: m4_include them.
  53         * Makefile.devel (AUTOCONF_MACROS): New variable.
  54         (configure): Depend on it.
  55         (CLISP_M4DIR): Remove variable.
  56         (autoconf/aclocal.m4): Remove rule.
  57
  58 2005-08-27  Bruno Haible  <bruno@clisp.org>
  59
  60         * src/integer/bitwise/cl_I_ash_I.cc (ash): Avoid shifting a 32-bit
  61         zero value by more than 31 bits.
  62
  63 2005-08-27  Bruno Haible  <bruno@clisp.org>
  64
  65         Make the long-float overflow check work on 64-bit platforms.
  66         * src/float/lfloat/cl_LF.h (LF_exp_mid, LF_exp_high): Define as
  67         'unsigned int', not 'unsigned long'.
  68
  69 2005-08-27  Bruno Haible  <bruno@clisp.org>
  70
  71         * include/cln/modules.h (CL_OUTPUT_LABEL): Work around redundant
  72         duplication of basic blocks by g++ 4.0.
  73         See <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23345>.
  74
  75 2005-08-27  Bruno Haible  <bruno@clisp.org>
  76
  77         Make it possible to cross-compile CLN.
  78         * m4/intparam.m4: New file.
  79         * m4/floatparam.m4: New file.
  80         * autoconf/aclocal.m4: Include both.
  81         (CL_MACHINE): Add an additional CROSS_MACRO parameter.
  82         * configure.ac (CL_MACHINE): When cross-compiling, use
  83         CL_INTPARAM_CROSS and CL_FLOATPARAM_CROSS.
  84
  85 2005-08-27  Bruno Haible  <bruno@clisp.org>
  86
  87         Define HAVE_LONGLONG and HAVE_LONGDOUBLE when cross-compiling.
  88         * autoconf/aclocal.m4 (CL_LONGLONG, CL_LONGDOUBLE): When cross-
  89         compiling, use the test code from gnulib.
  90
  91 2005-08-27  Bruno Haible  <bruno@clisp.org>
  92
  93         * autoconf/aclocal.m4 (CL_RUSAGE): Fix error when cross-compiling.
  94
  95 2005-08-16  Richard B. Kreckel  <kreckel@ginac.de>
  96
  97         The patch of 2005-05-01 made it impossible to test the type of a cl_UP
  98         by comparing with &cl_class_univpoly_ring. We need an alternative:
  99         * include/cln/object.h (cl_class_flags_modint_ring): New #define...
 100         * src/polynomial/elem/cl_UP.cc (cl_class_univpoly_ring): ...used here.
 101         * src/polynomial/elem/cl_UP_GF2.h: Likewise.
 102         * src/polynomial/elem/cl_UP_MI.h: Likewise.
 103         * src/polynomial/elem/cl_UP_gen.h: Likewise.
 104         * src/polynomial/elem/cl_UP_number.h: Likewise.
 105
 106 2005-08-15  Richard B. Kreckel  <kreckel@ginac.de>
 107
 108         * m4/cc.m4 (CL_AS_NOEXECSTACK): New macro...
 109         * configure.ac: ...used here for setting ASMFLAGS...
 110         * src/Makefile.in: ...which are used here.
 111
 112 2005-08-02  Andreas Jochens  <aj@andaco.de>
 113
 114         * include/cln/config.h.in: Add support for PowerPC 64 CPU.
 115         * include/cln/modules.h: Likewise.
 116         * include/cln/object.h: Likewise.
 117         * include/cln/types.h: Likewise.
 118
 119 2005-07-24  Richard B. Kreckel  <kreckel@ginac.de>
 120
 121         Make out of the box build on x86_64 system with complete 32 bit
 122         userland possible
 123         * include/cln/config.h.in: Don't #define __x86_64__ when
 124         __i386__ is defined.
 125         * src/base/digitseq/cl_asm_x86_64_.cc: New file.
 126         * doc/cln.tex: Revert workaround description introduced 2005-05-02.
 127
 128 2005-06-10  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 129
 130         * Makefile.in: Don't enter nonexisting directories.
 131
 132 2005-05-17  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 133
 134         Speed up the linking step
 135         * src/Makefile.in: Use -objectlist for linking libcln.la.
 136
 137 2005-05-15  Richard B. Kreckel  <kreckel@ginac.de>
 138
 139         * autoconf/ltmain.sh: Upgrade to libtool-1.5.16.
 140         * m4/libtool.m4: Upgrade to libtool-1.5.16 (without Comeau patch).
 141
 142 2005-05-02  Richard B. Kreckel  <kreckel@ginac.de>
 143
 144         * doc/cln.tex: Document what to do on a x86_64 machine with 32-bit
 145         userland.
 146
 147 2005-05-01  Richard B. Kreckel  <kreckel@ginac.de>
 148
 149         Fix crashes in find_univpoly_ring and related functions
 150         * include/cln/modinteger.h: Remove vptr from cl_heap_modint_ring;
 151         remove declaration of cl_class cl_class_modint_ring.
 152         * include/cln/univpoly.h: Remove vptr from cl_heap_univpoly_ring;
 153         remove declaration of cl_class_univpoly_ring.
 154         * include/cln/object.h: cl_class_flags_modint_ring: New #define...
 155         * src/modinteger/cl_MI.cc: ...used in cl_class_modint_ring.
 156         * src/modinteger/cl_MI_fix16.h: No vptr, but static dtor and type flag.
 157         * src/modinteger/cl_MI_fix29.h: Likewise.
 158         * src/modinteger/cl_MI_fix32.h: Likewise.
 159         * src/modinteger/cl_MI_int32.h: Likewise.
 160         * src/modinteger/cl_MI_montgom.h: Likewise.
 161         * src/modinteger/cl_MI_pow2: Likewise.
 162         * src/modinteger/cl_MI_pow2m1.h: Likewise.
 163         * src/modinteger/cl_MI_pow2p1.h: Likewise.
 164         * src/modinteger/cl_MI_std.h: Likewise.
 165         * src/polynomial/elem/cl_UP.cc (cl_make_univpoly_ring): Compare with
 166         cl_class_flags_modint_ring, not with cl_class_modint_ring.
 167         * src/polynomial/elem/cl_UP_GF2.h (cl_class_num_univpoly_ring): New.
 168         * src/polynomial/elem/cl_UP_MI.h (cl_class_modint_univpoly_ring): New.
 169         * src/polynomial/elem/cl_UP_gen.h (cl_class_gen_univpoly_ring): New.
 170         * src/polynomial/elem/cl_UP_number.h (cl_class_num_univpoly_ring): New.
 171         Reported by Ralf Goertz <R_Goertz@web.de>.
 172
 173 2005-04-29  Richard B. Kreckel  <kreckel@ginac.de>
 174             Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 175
 176         * m4/cc.m4: Emit a warning if g++ is used and optimization turned off.
 177
 178 2005-04-24  Richard B. Kreckel  <kreckel@ginac.de>
 179
 180         Make GCC compiler flags default to -O
 181         * m4/cc.m4: New file...
 182         * configure.ac: ...used here.
 183         * autoconf/aclocal.m4: Regenerate.
 184
 185 2005-04-18  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 186
 187         * include/cln/string.h: Declare cl_string.
 188
 189 2005-03-17  Richard B. Kreckel  <kreckel@ginac.de>
 190
 191         * autoconf/ltmain.sh: Upgrade to libtool-1.5.14.
 192         * m4/libtool.m4: Upgrade to libtool-1.5.14 with Comeau patch.
 193         * autoconf/aclocal.m4: Regenerate.
 194
 195 2005-03-15  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 196
 197         * src/Makefile.in: Use $CXX instead of $CC when linking.
 198
 199 2005-02-27  Richard B. Kreckel  <kreckel@ginac.de>
 200
 201         * examples/perfnum.cc: update to presumed 42st Mersenne prime.
 202
 203 2004-11-28  Richard B. Kreckel  <kreckel@ginac.de>
 204
 205         Disambiguate binary operators of CLN types with float/double
 206         * include/cln/dfloat.h: Add binary operator overloads for arguments of
 207         type double.
 208         * include/cln/ffloat.h: Likewise, for arguments of type float.
 209         * include/cln/float.h: Likewise, both for arguments of types double and
 210         float.
 211         * include/cln/real.h: Likewise.
 212         Reported by Isidro Cachadiña Gutiérrez <icacha@unex.es>.
 213
 214 2004-11-03  Richard B. Kreckel  <kreckel@ginac.de>
 215
 216         * Version 1.1.9 released.
 217
 218 2004-10-28  Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
 219
 220         * src/Makefile.in: Let config.status set LDFLAGS.
 221
 222 2004-10-27  Peter Breitenlohner  <peb@mppmu.mpg.de>
 223
 224         * cln.pc.in: Fix typo.
 225         * Makefile.in (INSTALL_SCRIPT): Added, to be used for scripts like
 226         cln-config.  Allows us to do special things to binaries, like
 227         installing them with install -s.
 228         * doc/Makefile.in: add/remove the cln.info from the installed
 229         $(infodir)/dir unless this is debian install-info (code copied from
 230         what GNU automake would produce).
 231
 232 2004-10-26  Richard B. Kreckel  <kreckel@ginac.de>
 233
 234         * src/integer/input/cl_I_read_stream.cc (read_integer): Fix a bug
 235         that caused radix specifiers to not work when reading from a stream.
 236         * src/rational/input/cl_RA_read_stream.cc (read_rational): Likewise.
 237         * src/real/input/cl_R_read_stream.cc (read_real): Likewise.
 238         * src/float/input/cl_F_read_stream.cc (read_float): Likewise.
 239         * src/complex/input/cl_N_read_stream.cc (read_complex): Likewise.
 240
 241 2004-10-25  Richard B. Kreckel  <kreckel@ginac.de>
 242
 243         * src/base/cl_low.h: Add mulu64 assembler macro for ia64.
 244
 245 2004-10-24  Richard B. Kreckel  <kreckel@ginac.de>
 246
 247         * src/base/cl_low.h: Add mul and div macros for x86_64.
 248
 249 2004-10-23  Richard B. Kreckel  <kreckel@ginac.de>
 250
 251         * src/integer/conv/cl_I_from_digits.cc (digits_to_I): Fix thinko in
 252         new code for base power of two.
 253
 254 2004-10-22  Richard B. Kreckel  <kreckel@ginac.de>
 255
 256         * src/integer/conv/cl_I_to_digits (I_to_digits): Fix an elusive stack
 257         overwriting problem. That was the real cause for Debian bug#246319.
 258         * src/integer/output/cl_I_print.cc (print_integer): Revert workaround
 259         for the bug fixed above.
 260
 261 2004-10-20  Richard B. Kreckel  <kreckel@ginac.de>
 262
 263         * include/cln/types.h: Use 64 bit digits on x86_64 CPU.
 264
 265 2004-10-12  Richard B. Kreckel  <kreckel@ginac.de>
 266
 267         * src/integer/conv/cl_I_from_digits.cc (digits_to_I): Speedup when
 268         the base is a power of two.
 269
 270 2004-10-05  Richard B. Kreckel  <kreckel@ginac.de>
 271
 272         * src/integer/conv/cl_I_to_digits.cc (I_to_digits): Fix bug in base 32.
 273
 274 2004-09-27  Richard B. Kreckel  <kreckel@ginac.de>
 275
 276         Support for little-endian Mips, second shot
 277         * src/base/digitseq/cl_asm_mipsel_.cc: New file...
 278         * src/base/digitseq/cl_asm_cc: ...used here.
 279         * src/base/digitseq/cl_asm.h: Include cl_asm_mips.h for any endianness.
 280         * include/cln/object.h: Set alignment for mipsel explicitly.
 281
 282 2004-09-05  Richard B. Kreckel  <kreckel@ginac.de>
 283
 284         Support for little-endian Mips
 285         * include/cln/config.h.in: Add __mipsel__.
 286         * include/cln/modules.h: For Mips, this is endianness-agnostic.
 287         * src/base/digitseq/cl_asm_.cc, src/base/digitseq/cl_asm.h:
 288         Mask out assembler for little-endian Mips.
 289
 290 2004-08-30  Bruno Haible  <bruno@clisp.org>
 291
 292         * benchmarks/timebench2.sh: Multiply all repeat counts by 100.
 293         * benchmarks/timebench2.results: Add recent PowerPC G4 results.
 294
 295 2004-08-26  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 296
 297         * examples/e.cc: remove extra semicolon.
 298
 299 2004-08-25  Bruno Haible  <bruno@clisp.org>
 300
 301         * autoconf/ltmain.sh: Upgrade to libtool-1.5.6.
 302         * m4/libtool.m4: New file, from libtool-1.5.6 with modifications:
 303         2004-08-22  Bruno Haible  <bruno@clisp.org>
 304                 * m4/libtool.m4: Add support for Comeau C++ on Linux.
 305                 Reported by Prof. Roberto Bagnara <bagnara@cs.unipr.it>.
 306         * autoconf/aclocal.m4: Regenerate.
 307
 308 2004-08-19  Bruno Haible  <bruno@clisp.org>
 309
 310         * include/cln/modules.h (CL_GLOBALIZE_JUMP_LABEL, CL_JUMP_TO): When
 311         converting a label to a string, use ASM_UNDERSCORE_PREFIX. Needed on
 312         MacOS X.
 313         Reported by Darren Bane <darren.bane@ul.ie>.
 314
 315 2004-07-01  Richard B. Kreckel  <kreckel@ginac.de>
 316
 317         * Version 1.1.8 released.
 318
 319 2004-06-30  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 320
 321         * src/base/cl_macros.h: alloca(3) has size_t argument type.
 322
 323 2004-06-30  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 324
 325         * include/cln/floatformat.h: Do define a type here.
 326
 327 2004-06-27  Richard B. Kreckel  <kreckel@ginac.de>
 328
 329         * include/cln/modules.h (CL_JUMP_TO): Fix AMD64 brokenness.
 330
 331 2004-06-23  Bruno Haible  <bruno@clisp.org>
 332
 333         * configure.ac: Pretend ftime() is not available. Needed by
 334         CL_TIMES_CLOCK.
 335
 336 2004-06-21  Ralf Stephan  <ralf@ark.in-berlin.de>
 337
 338         * doc/cln.tex: Document jacobi, isprobprime and nextprobprime.
 339
 340 2004-06-18  Richard B. Kreckel  <kreckel@ginac.de>
 341
 342         * rational/transcendental/cl_RA_logp.cc: fix bug where base is
 343         reciprocal of an integer.
 344         Reported by Niklas Knutsson <nq@altern.org>.
 345
 346 2004-06-18  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 347
 348         * src/complex/ring/cl_C_ring.cc, src/integer/ring/cl_I_ring.cc,
 349         src/rational/ring/cl_RA_ring.cc, src/real/ring/cl_R_ring.cc:
 350         Make template specializations explicit.
 351
 352 2004-06-17  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 353
 354         * src/complex/input/cl_N_read.cc, src/float/input/cl_F_read.cc,
 355         src/integer/input/cl_I_read.cc, src/rational/input/cl_RA_read.cc,
 356         src/real/input/cl_R_read.cc:  Remove unused labels.
 357
 358 2004-06-17  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 359
 360         * benchmarks/timebench1.cc, benchmarks/timebench2a.LiDIA.cc,
 361         benchmarks/timebench2a.cc, benchmarks/timebench2ap.cc,
 362         benchmarks/timebench2b.LiDIA.cc, benchmarks/timebench2b.cc,
 363         examples/atan_recip.cc, examples/atanh_recip.cc,
 364         examples/contfrac.cc, examples/e.cc, examples/legendre.cc,
 365         examples/lucaslehmer.cc, examples/pi.cc, include/cln/GV.h,
 366         include/cln/SV.h, include/cln/malloc.h, include/cln/modules.h,
 367         include/cln/object.h, include/cln/string.h, src/base/cl_abort.cc,
 368         src/base/cl_alloca.h, src/base/cl_malloc.cc,
 369         src/base/random/cl_random_from.cc,
 370         src/base/string/cl_spushstring_append.cc,
 371         src/base/string/cl_spushstring_push.cc,
 372         src/base/string/cl_st_debug.cc,
 373         src/base/string/input/cl_st_gettoken.cc,
 374         src/complex/input/cl_N_read.cc, src/float/input/cl_F_read.cc,
 375         src/float/output/cl_F_dprint.cc, src/integer/input/cl_I_read.cc,
 376         src/rational/input/cl_RA_read.cc,
 377         src/real/format-output/cl_fmt_integer.cc,
 378         src/real/format-output/cl_fmt_paddedstring.cc,
 379         src/real/input/cl_R_read.cc, src/timing/cl_t_current.cc,
 380         src/timing/cl_t_current2.cc, tests/exam.cc, tests/tests.cc,
 381         tests/timeLFRAmul.cc, tests/timeLFatan-compare.cc,
 382         tests/timeLFatan.cc, tests/timeLFatanh-compare.cc,
 383         tests/timeLFatanh.cc, tests/timeLFcos-compare.cc, tests/timeLFcos.cc,
 384         tests/timeLFcosh.cc, tests/timeLFexp-compare.cc, tests/timeLFexp.cc,
 385         tests/timeLFln-compare.cc, tests/timeLFln.cc,
 386         tests/timeLFsin-compare.cc, tests/timeLFsin.cc, tests/timeLFsinh.cc,
 387         tests/timeLFsqrt.cc, tests/timeMImisc5.cc, tests/timeMIpow2div.cc,
 388         tests/timeMIpow2recip.cc, tests/timeRALFdiv.cc, tests/timeRAtoLF.cc,
 389         tests/timeUPMImul.cc, tests/timecatalan.cc, tests/timediv.cc,
 390         tests/timediv2adic-compare.cc, tests/timediv2adic.cc,
 391         tests/timeeuler.cc, tests/timeexp1.cc, tests/timefact.cc,
 392         tests/timegcd.cc, tests/timemul-compare.cc, tests/timemul.cc,
 393         tests/timepi.cc, tests/timeprint-compare.cc, tests/timeprint.cc,
 394         tests/timerecip2adic-compare.cc, tests/timerecip2adic.cc,
 395         tests/timesqrt.cc, tests/timesqrtmodp.cc, tests/timesquare.cc,
 396         tests/timezeta3.cc: Change all C include headers to ISO style
 397         within C++ code.
 398
 399 2004-06-10  Richard B. Kreckel  <kreckel@ginac.de>
 400
 401         * examples/perfnum.cc: update to presumed 41st Mersenne prime.
 402
 403 2004-05-02  Richard B. Kreckel  <kreckel@ginac.de>
 404
 405         * Version 1.1.7 released.
 406
 407 2004-05-02  Richard B. Kreckel  <kreckel@ginac.de>
 408
 409         * examples/pi.cc and examples/pi.1: New files.
 410         * examples/Makefile.in: Build the pi executable.
 411
 412 2004-05-01  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 413
 414         * src/Makefile.in: Fix for parallel build: wait for subdir objects to
 415         be finished before creating the library.
 416
 417 2004-04-30  Richard B. Kreckel  <kreckel@ginac.de>
 418
 419         * src/integer/output/cl_I_print.cc (print_integer): workaround
 420         GCC compiler bug (cf. Debian bug#246319).
 421
 422 2004-03-20  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 423
 424         * m4/gmp.m4 (CL_GMP_CHECK): Do not lose LIBS setting with config.cache
 425         enabled.
 426
 427 2004-03-08  Bruno Haible  <bruno@clisp.org>
 428
 429         * src/float/lfloat/elem/cl_LF_mul.cc (operator*): Fix the second
 430         underflow condition.
 431         * src/float/lfloat/algebraic/cl_LF_sqrt.cc (sqrt): Fix a bug with large
 432         uexp whereby SQRT of MOST-POSITIVE-LONG-FLOAT was less than 1.
 433
 434 2004-03-04  Richard B. Kreckel  <kreckel@ginac.de>
 435
 436         * Makefile.in (install): Add ${srcdir} for cln.m4.
 437         * m4/gmp.m4: quote macro names.
 438         Reported by Ralf Wildenhues <Ralf.Wildenhues@gmx.de>.
 439
 440 2004-01-01  Richard B. Kreckel  <kreckel@ginac.de>
 441
 442         * Version 1.1.6 released.
 443
 444 2004-01-01  Richard B. Kreckel  <kreckel@ginac.de>
 445
 446         * include/cln/univpoly.h, include/cln/univpoly_complex.h,
 447           include/cln//univpoly_integer.h, include/cln/univpoly_modint.h,
 448           include/cln/univpoly_rational.h, include/cln/univpoly_real.h,
 449           src/polynomial/elem/cl_UP_GF2.h, src/polynomial/elem/cl_UP_MI.h,
 450           src/polynomial/elem/cl_UP_gen.h, src/polynomial/elem/cl_UP_no_ring.cc,
 451           src/polynomial/elem/cl_UP_number.h (ldegree): New function.
 452         * doc/cln.tex: Document `ldegree'.
 453
 454 2003-12-29  Richard B. Kreckel  <kreckel@ginac.de>
 455
 456         Rework of autoconfiscation infrastructure
 457         * autoconf/config.{guess,sub}: Update to GNU version 2003-10-07.
 458         * autoconf/ltmain.sh: Update to GNU version 1.4.3.
 459         * autoconf/autoconf: Remove (from now on we assume autoconf is
 460           installed properly on the sytem).
 461         * autoconf/autoconf.m4: Likewise.
 462         * autoconf/autoconf.m4f: Likewise.
 463         * autoconf/acgeneral.m4: Likewise.
 464         * autoconf/acspecific.m4: Likewise.
 465         * autoconf/aclocal.m4: Regenerate.
 466         * autoconf/acinclude.m4: Remove (while moving the macros...)
 467         * m4/gmp.m4: New file (...to here).
 468         * Makefile.devel: Update to new scheme.
 469         * configure.ac: Likewise.
 470         * include/cln/GV_integer.h: Assume template specializations work.
 471         * include/cln/GV_modinteger.h: Likewise.
 472         * include/cln/config.h.in: Likewise, and drop HAVE_BOOL.
 473         * src/base/cl_base_config.h.in: Drop support for obsolete ftime(3).
 474         * src/base/random/cl_random_from.cc: Likewise.
 475         * src/timing/cl_base_config.h.in: Likewise.
 476         * src/timing/cl_t_current.cc: Likewise.
 477
 478 2003-12-27  Richard B. Kreckel  <kreckel@ginac.de>
 479
 480         * src/polynomial/cl_UP_gen.h (gen_minus): Fix case where first
 481           argument is zero.
 482         * src/polynomial/cl_UP_MI.h (modint_minus): Likewise.
 483         * src/polynomial/cl_UP_number.h (num_minus): Likewise.
 484         Reported by Munagala Ramanath <amberarrow@yahoo.com>.
 485
 486 2003-12-02  Richard B. Kreckel  <kreckel@ginac.de>
 487
 488         * examples/perfnum.cc: update to presumed 40th Mersenne prime.
 489
 490 2003-11-20  Christian Bauer  <cbauer@ginac.de>
 491
 492         Added pkg-config support
 493         * cln.pc.in: New file.
 494         * Makefile.in: Take care of cln.pc.
 495         * configure.ac: Likewise.
 496
 497 2003-08-06  Richard B. Kreckel  <kreckel@ginac.de>
 498
 499         * src/numtheory/cl_nt_sqrtmodp.cc: #undef _R.
 500         Reported by Andrew Rechnitzer <A.Rechnitzer@ms.unimelb.edu.au>.
 501
 502 2003-08-01  Richard Kreckel  <kreckel@ginac.de>
 503
 504         More dependent base resolution issues
 505         * src/base/hash/cl_hash.h (cl_heap_hashtable<T>::iterator()):
 506         portable syntactic simplification.
 507         * src/base/hash/cl_hashset.h: Preceed inherited members with this->.
 508         * src/base/hash/cl_hash1.h: Likewise for member functions.
 509         * src/base/hash/cl_hash2.h: Likewise.
 510         * src/base/hash/cl_hashuniq.h: Likewise.
 511         * src/base/hash/cl_hashuniqweak.h: Likewise.
 512         * src/base/hash/cl_hash.h: Revert explicit static member function
 513         lookup since that was GCC's fault.
 514         * src/base/hash/cl_hash2weak.h: Likewise.
 515         * src/base/hash/cl_hashuniqweak.h: Likewise.
 516
 517 2003-06-29  Richard Kreckel  <kreckel@ginac.de>
 518
 519         Dependent base resolution needed for GCC-3.4
 520         * include/cln/GV.h: Preceed inherited members with this->.
 521         * include/cln/SV.h: Likewise.
 522         * include/cln/object.h: Likewise.
 523         * src/base/hash/cl_hash1.h: Likewise.
 524         * src/base/hash/cl_hash1weak.h: Likewise.
 525         * src/base/hash/cl_hash2.h: Likewise.
 526         * src/base/hash/cl_hashuniq.h: Likewise.
 527         * src/base/hash/cl_hash.h: Make lookup of static member function
 528         explicit.
 529         * src/base/hash/cl_hash2weak.h: Likewise.
 530         * src/base/hash/cl_hashuniqweak.h: Likewise.
 531         * src/modinteger/cl_MI.cc: Make shell-comment a C-comment. Geez.
 532
 533 2003-02-24  Bruno Haible  <bruno@clisp.org>
 534
 535         * src/base/random/cl_random_from.cc (random_state::random_state): Add
 536         support for MacOS X.
 537         * src/modinteger/cl_MI.cc: #undef _R.
 538         Reported by Erann Gat <gat@jpl.nasa.gov>.
 539
 540 2002-08-03  Joerg Arndt  <jj@suse.de>
 541
 542         * include/cln/config.h.in: Add support for x86_64 CPU.
 543         * include/cln/modules.h: Likewise.
 544         * include/cln/types.h: Likewise.
 545         * include/cln/object.h: Likewise.
 546
 547 2002-06-08  Richard Kreckel  <kreckel@ginac.de>
 548
 549         * src/base/digitseq/cl_asm.h: ensure intCsize==16 when including
 550         m68k Assembler routines.
 551         * src/base/digitseq/cl_asm_.cc: Likewise.
 552
 553 2002-05-28  Richard Kreckel  <kreckel@ginac.de>
 554
 555         * Version 1.1.5 released.
 556
 557 2002-05-27  Richard Kreckel  <kreckel@ginac.de>
 558
 559         * include/cln/modules.h (CL_CONCATENATE): New macro.
 560
 561 2002-05-10  Richard Kreckel  <kreckel@ginac.de>
 562
 563         * doc/cln.tex (Building the library): Update recommendations for
 564         compiling on Tru64 using g++ 3.0 and 3.1.
 565         * README: Update homepage.
 566
 567 2002-05-05  Bruno Haible  <bruno@clisp.org>
 568
 569         * doc/cln.tex (Building the library): Give some recommendations for
 570         g++ 3.0 and 3.1.
 571
 572         Force link-time references despite optimizations done by g++ 2.95
 573         and newer.
 574         * include/cln/modules.h (CL_FORCE_LINK): New macro.
 575         * Use CL_FORCE_LINK.
 576         * include/cln/GV_integer.h (cl_GV_I_debug_dummy): Likewise.
 577         * include/cln/GV_number.h (cl_GV_number_debug_dummy): Likewise.
 578         * include/cln/SV_number.h (cl_SV_number_debug_dummy): Likewise.
 579         * include/cln/SV_ringelt.h (cl_SV_ringelt_debug_dummy): Likewise.
 580         * include/cln/dfloat.h (cl_DF_debug_dummy): Likewise.
 581         * include/cln/ffloat.h (cl_FF_classes_dummy, cl_FF_debug_dummy):
 582         Likewise.
 583         * include/cln/integer.h (cl_I_classes_dummy, cl_I_debug_dummy):
 584         Likewise.
 585         * include/cln/lfloat.h (cl_LF_debug_dummy): Likewise.
 586         * include/cln/modinteger.h (cl_MI_debug_dummy): Likewise.
 587         * include/cln/rational.h (cl_RA_debug_dummy): Likewise.
 588         * include/cln/real.h (cl_R_debug_dummy): Likewise.
 589         * include/cln/ring.h (cl_ring_debug_dummy): Likewise.
 590         * include/cln/sfloat.h (cl_SF_classes_dummy, cl_SF_classes_dummy):
 591         Likewise.
 592         * include/cln/string.h (cl_string_debug_dummy): Likewise.
 593         * include/cln/univpoly.h (cl_UP_debug_dummy): Likewise.
 594         * src/float/base/cl_ieee.h (cl_ieee_dummy_NNN): Likewise.
 595
 596         Avoid g++ 3.1 warnings.
 597         * src/base/cl_offsetof.h (offsetof): Redefine each time.
 598         * src/base/digitseq/cl_DS.h: Include "cl_offsetof.h" after <gmp.h>.
 599
 600         Avoid g++ 3.1 warnings.
 601         * src/base/hash/cl_hash.h (struct cl_heap_hashtable): Use typename
 602         where needed.
 603         * src/base/hash/cl_hash1.h (struct cl_heap_hashtable_1): typedef
 604         htxentry as a shortcut.
 605         * src/base/hash/cl_hash2.h (struct cl_heap_hashtable_2): Likewise.
 606         * src/base/hash/cl_hashset.h (struct cl_heap_hashtable_set): Likewise.
 607         * src/base/hash/cl_hashuniq.h (struct cl_heap_hashtable_uniq):
 608         Likewise.
 609
 610 2002-03-15  Bruno Haible  <bruno@clisp.org>
 611
 612         * cln.tex: Document problem with GNU make 3.77.
 613         Reported by Michael Somos <somos@grail.cba.csuohio.edu>.
 614
 615 2002-02-16  Richard Kreckel  <kreckel@ginac.de>
 616
 617         * cln.m4: quote macro name.
 618           Pointed out by Roberto Bagnara.
 619
 620 2002-01-20  Richard Kreckel  <kreckel@ginac.de>
 621
 622         * autoconf/config.{guess,sub}: Update to GNU version 2002-01-02.
 623           (the old one was broken on Linux/Mips.)
 624
 625 2002-01-04  Richard Kreckel  <kreckel@ginac.de>
 626
 627         * autoconf/autoconf.m4f: get brutal in order to adhere to FHS.
 628         * Version 1.1.4 released.
 629
 630 2002-01-03  Richard Kreckel  <kreckel@ginac.de>
 631
 632         * autoconf/acinclude.m4: revamp MPN-matcher.
 633         * autoconf/aclocal.m4: upgrade to autoconf-2.52 infrastructure,
 634           sync with CLisp from CVS.
 635         * autoconf/autoconf: Likewise.
 636         * autoconf/autoconf.m4f: Likewise (new file).
 637         * configure.ac: Likewise (new file, replaces configure.in).
 638         * configure.in: Likewise (deleted, replaced by configure.ac).
 639         * autoconf/config.{guess,sub}: Update to GNU version 2001-12-13.
 640         * src/Makefile.in: made VPATH safe for autoconf-2.52.
 641         * include/cln/config.h.in: Add __s390__.
 642
 643 2001-12-31  Richard Kreckel  <kreckel@ginac.de>
 644
 645         * src/base/digitseq/cl_DS.h: <gmp.h> is not included extern "C"
 646           any more since GMP4 has some C++ support in it.
 647
 648 2001-12-14  Richard Kreckel  <kreckel@ginac.de>
 649
 650         * include/cln/modules.h, include/cln/object.h: add support for
 651           s390.
 652         * src/numtheory/cl_nt_sqrtmodp.cc: workaround for GCC2.x compiler-bug
 653           on s390, provided by Gerhard Tonn.
 654
 655 2001-11-05  Richard Kreckel  <kreckel@ginac.de>
 656
 657         * autoconf/ltmain.sh: Upgrade to libtool-1.4.2.
 658         * autoconf/config.{guess,sub}: Update to GNU version 2001-09-07.
 659         * Version 1.1.3 released.
 660
 661 2001-11-04  Bruno Haible  <haible@clisp.cons.org>
 662
 663         Interoperability with gcc-3.0 -fuse-cxa-atexit.
 664         * autoconf/aclocal.m4 (CL_GLOBAL_CONSTRUCTORS): Add test whether
 665           global destructors actually exist.
 666         * include/cln/modules.h (CL_PROVIDE, CL_PROVIDE_END, CL_PROVIDE_END):
 667           Don't hack the global destructors if there is no global destructors
 668           function.
 669
 670 2001-11-03  Richard Kreckel  <kreckel@ginac.de>
 671
 672         * src/float/transcendental/cl_F_sinx.cc (sinx_naive): For small
 673           values of x, return square(x) instead of x.
 674         * src/float/transcendental/cl_F_sinhx.cc (sinhx_naive): Likewise.
 675
 676 2001-07-25  Richard Kreckel  <kreckel@ginac.de>
 677
 678         * Version 1.1.2 released.
 679
 680 2001-07-24  Richard Kreckel  <kreckel@ginac.de>
 681
 682         * src/base/hash/cl_hash.h: declare _cl_hashtable_iterator<htentry> a
 683           friend of cl_heap_hashtable<htentry>.
 684
 685 2001-07-22  Richard Kreckel  <kreckel@ginac.de>
 686
 687         * src/float/base/cl_ieee.cc: try to do magic to the FPU only if
 688           _FPU_IEEE is really defined.
 689         * include/cln/modules.h: change assembler labels from `label' to
 690           `label:' on hppa, needed by Linux (see comment).
 691         * autoconf/acinclude.m4: new file (for storing CLN-specific macros).
 692         * Makefile.devel: adjusted.
 693         * autoconf/aclocal.m4: regenerate.
 694         * src/base/low/cl_low_mul.cc: moved POD variables that are declared
 695           extern "C" elsewhere out of the namespace.
 696         * src/base/low/cl_low_div.cc: Likewise.
 697
 698 2001-06-08  Bruno Haible  <haible@clisp.cons.org>
 699
 700         * autoconf/config.{guess,sub}: Update to GNU version 2001-05-11.
 701         * autoconf/aclocal.m4: Upgrade to libtool-1.4.
 702         * autoconf/ltmain.sh: Likewise.
 703         * autoconf/ltconfig: Remove file.
 704         * autoconf/install-sh: New file.
 705         * configure.in: Add AC_CONFIG_AUX_DIR call.
 706
 707 2001-06-05  Richard Kreckel  <kreckel@ginac.de>
 708
 709         * tests/tests.cc: resolve namespace ambiguity about strcmp().
 710
 711 2001-05-31  Richard Kreckel  <kreckel@ginac.de>
 712
 713         * Version 1.1.1 released.
 714
 715 2001-05-28  Richard Kreckel  <kreckel@ginac.de>
 716
 717         * cln/cln.tex: documented problems with shared library on Sparc
 718           using gcc older than 2.95.3.
 719         * configure.in: Fixed typos in versioning docu.
 720
 721 2001-05-25  Bruno Haible  <haible@clisp.cons.org>
 722
 723         * src/base/digitseq/cl_asm_arm_.cc: Use #0x instead of #& to designate
 724           hexadecimal constants.
 725
 726 2001-05-25  Richard Kreckel  <kreckel@ginac.de>
 727
 728         * autoconf/floatparam.c (double_wordorder_bigendian_p): new symbol.
 729         * src/float/dfloat/cl_DF.h: Check for double_wordorder_bigendian_p.
 730         * Removed LiDIA interface since that is now outdated (namespace cln)
 731           and maintained elsewhere.
 732         * Adjusted dates and final touches for 1.1.1.
 733
 734 2001-05-19  Richard Kreckel  <kreckel@ginac.de>
 735
 736         * INSTALL: Update toolchain info: no egcs, some more platforms.
 737         * doc/cln.tex: Likewise.
 738
 739 2001-05-18  Richard Kreckel  <kreckel@ginac.de>
 740
 741         * src/base/cl_low.h: prepended variables declared inside macros
 742           with underscore.  Fixes equal_hashcode() on various platforms.
 743
 744 2001-04-25  Richard Kreckel  <kreckel@ginac.de>
 745
 746         * src/base/cl_low.h: Added several checks if NO_ASM is defined, so
 747           this definition becomes actually useful.  This is needed for
 748           compilation on Arm until somebody fixes the assembler files for Arm.
 749         * src/base/digitseq/cl_asm.h: Likewise.
 750         * src/base/digitseq/cl_asm_.cc: Likewise.
 751         * */Makefile.in: Added `override' in front of `CPPFLAGS +=' so
 752            one can say `make CPPFLAGS=-DFOOBAR'.
 753
 754 2001-03-26  Arvid Norberg <c99ang@cs.umu.se>
 755
 756         * src/base/random/cl_random_from.cc: ported to beos.
 757
 758 2001-03-05  Richard Kreckel  <kreckel@ginac.de>
 759
 760         * include/cln/modules.h (CL_JUMP_TO): Fix IA64 brokenness.
 761
 762 2001-01-28  Richard Kreckel  <kreckel@ginac.de>
 763
 764         * include/cln/number.h (cl_as_N): Remove bogus comment.
 765
 766 2001-01-22  Richard Kreckel  <kreckel@ginac.de>
 767
 768         * configure.in: Make build in separate builddir possible (again).
 769
 770 2001-01-22  Richard Kreckel  <kreckel@ginac.de>
 771
 772         * include/cln/*.h: Change signatures of all classes' methods
 773           cln::cl_foo::operator new(size_t, cl_foo*) to
 774           cln::cl_foo::operator new(size_t, void*) so one can declare
 775           std::vector<cln::cl_foo>, std::list<cln::cl_foo> etc. for
 776           certain STL implementations (like libstdc++-v3).
 777
 778 2000-12-14  Richard Kreckel  <kreckel@ginac.de>
 779
 780         * Version 1.1 released.
 781
 782 2000-12-13  Richard Kreckel  <kreckel@ginac.de>
 783
 784         * */*: cl_istream -> std::istream, cl_ostream -> std::ostream.
 785
 786 2000-12-05  Richard Kreckel  <kreckel@ginac.de>
 787
 788         * Makefile.in, src/Makefile.in, doc/Makefile.in: Use mkdir -p.
 789         * include/cln/version.h.in, src/base/verion.cc: New files.
 790         * configure.in: Generate include/cln/version.h.
 791         * cln.m4: Rewrote it. Check result of cln-config without compiling.
 792           Do cross-check library version and header version information.
 793
 794 2000-12-02  Christian Bauer  <cbauer@ginac.de>
 795
 796         * Makefile.in, src/Makefile.in, doc/Makefile.in: Added $DESTDIR.
 797         * cln.m4, cln.spec.in: some minor fixes with respect to RPM package
 798           building.
 799
 800 2000-11-24  Richard Kreckel  <kreckel@ginac.de>
 801
 802         * */*: Removed problematic stdin, stdout and stderr definitions.
 803           Use std::cin, std::cout, std::cerr instead (obsoletes 2000-10-29).
 804
 805 2000-11-20  Bruno Haible
 806
 807         * cln-config.1: change title.
 808
 809 2000-11-18  Richard Kreckel  <kreckel@ginac.de>
 810
 811         * cln.m4: New file.
 812         * doc/cln.tex: Document package tools cln-config and cln.m4.
 813         * Makefile.in: Care about cln.m4.
 814
 815 2000-11-17  Richard Kreckel  <kreckel@ginac.de>
 816
 817         * cln-config.1: added manpage, as required by a couple of distros.
 818         * Makefile.in, doc/Makefile.in: target install depends on installdirs.
 819
 820 2000-11-16  Richard Kreckel  <kreckel@ginac.de>
 821
 822         * autoconf/aclocal.m4 (CL_GMP_SET_UINTD): New macro...
 823         * configure.in: ...used here.
 824         * include/cln/config.h.in: Put in macros defined by CL_GMP_SET_UINTD...
 825         * include/cln/types.h: ...used here.
 826         * autoconf/acgeneral.m4, autoconf/aclocal.m4: Adhere to FHS.
 827
 828 2000-11-13  Richard Kreckel  <kreckel@ginac.de>
 829
 830         * src/base/digitseq/cl_asm.h: Test if (intDsize==32) for MIPS and HPPA,
 831           in order to guard against an accidented configuration.
 832         * src/integer/conv/cl_I_to_digits.cc (table): member b_hoch_k of
 833           struct power_table_entry initialized as ULL instead of as UL, if
 834           intDsize==64 (caused misprinting on MIPS w/ GMP).
 835         * src/base/cl_macros.h (minus_bitm, bitc): implemented ULL, if
 836           HAVE_FAST_LONGLONG.
 837         * src/integer/bitwise/cl_I_mkfx.cc (mkf_extract): Likewise.
 838         * src/integer/conv/cl_I_from_L.cc (cl_I_from_L): Added trival
 839           generation of Bignum for intDsize==64 and a notreached-check at end.
 840         * autoconf/config.guess, autoconf/config.sub: updated from FSF.
 841         * include/cln/config.h.in: Prepared support for IA64.
 842         * include/cln/types.h: Likewise.
 843         * include/cln/object.h: Likewise.
 844         * include/cln/modules.h: Likewise.
 845         * src/base/cl_macros.h (nonreturning_function): Likewise (NUE's
 846           compiler claims __GNUC_MINOR__==9).
 847
 848 2000-11-03  Richard Kreckel  <kreckel@ginac.de>
 849
 850         * src/base/cl_macros.h (bit, minus_bit): changed criterion for ULL from
 851           HAVE_DD to HAVE_FAST_LONGLONG.
 852         * src/base/cl_macros.h (bitm): implemented ULL, if HAVE_FAST_LONGLONG.
 853         * src/base/cl_low.h: actually no need to include "cln/types.h" here.
 854         * src/base/cl_low.h (logcount_64): always ULL, independent of HAVE_DD.
 855         * src/base/random/cl_UL_random.cc (random32): a is always ULL.
 856
 857 2000-11-01  Richard Kreckel  <kreckel@ginac.de>
 858
 859         * include/cln/object.h (cl_combine): define additional signatures, if
 860           HAVE_LONGLONG is defined, in order to keep the compiler happy.
 861         * src/base/cl_macros.h: include "cln/types.h", since we need HAVE_DD...
 862         * src/base/cl_macros.h (bit): ...for this macro...
 863         * src/base/cl_macros.h (minus_bit): ...and this one.
 864         * src/base/cl_low.h: include "cln/types.h", since we need HAVE_DD...
 865         * src/base/cl_low.h (logcount_64): ...for this macro.
 866         * src/base/random/cl_UL_random.cc (random32): if HAVE_DD a is an ULL.
 867         * src/integer/gcd/cl_I_gcd_aux2.cc (floorDD): fixed algorithmic bug.
 868           that turned up when intDsize==32 and cl_word_size==64.
 869         * src/float/dfloat/elem/cl_DF_div.cc (operator/): fixed a missing cast
 870           to uint64 that turned up when intDsize==32 and cl_word_size==64.
 871
 872 2000-10-29  Richard Kreckel  <kreckel@ginac.de>
 873
 874         * src/real/input/cl_R_read.cc, src/complex/input/cl_N_read.cc:
 875           #undef stderr, if it's defined so cln::stderr isn't confused.
 876         * src/base/input/cl_read_globals.cc: stdin should not be extern.
 877
 878 2000-09-05  Richard Kreckel  <kreckel@ginac.de>
 879
 880         * include/cln/number.h (As): Fix it in namespace by suffixing `_As'
 881           to the appropiate method instead of prefixing `as_'.
 882         * src/complex/misc/cl_N_as.cc (cl_N_As): Likewise.
 883         * src/real/misc/cl_R_as.cc (cl_R_As): Likewise.
 884         * src/rational/misc/cl_RA_as.cc (cl_RA_As): Likewise.
 885         * src/integer/misc/cl_I_as.cc (cl_I_As): Likewise.
 886         * src/float/misc/cl_F_as.cc (cl_F_As): Likewise.
 887         * src/float/sfloat/misc/cl_SF_as.cc (cl_SF_As): Likewise.
 888         * src/float/lfloat/misc/cl_LF_as.cc (cl_LF_As): Likewise.
 889         * src/float/ffloat/misc/cl_FF_as.cc (cl_FF_As): Likewise.
 890         * src/float/dfloat/misc/cl_DF_as.cc (cl_DF_As): Likewise.
 891
 892 2000-09-05  Richard Kreckel  <kreckel@ginac.de>
 893
 894         * src/complex/transcendental/cl_C_expt_C.cc (expt): fix logic for
 895           the 0^y cases.
 896
 897 2000-08-30  Richard Kreckel  <kreckel@ginac.de>
 898
 899         * include/cln/number.h, cl_number::_as_cl_private_thing(): removed.
 900           Rearranged for a clearer distinction between macros and remaining
 901           identifiers, so Cint can parse it smoothly.
 902
 903 2000-08-29  Richard Kreckel  <kreckel@ginac.de>
 904
 905         * include/cln/number.h, the(const cl_number& x): New template
 906           function.
 907
 908 2000-08-29  Richard Kreckel  <kreckel@ginac.de>
 909
 910         * */*: Pushed CLN into a namespace `cln'.  While doing so, the
 911           following identifiers got their poor-man's namespace (i.e. the
 912           prefix `cl_') stripped off:
 913           cl_catalanconst()        ->  catalanconst()
 914           cl_compare()             ->  compare()
 915           cl_cos_sin()             ->  cos_sin()
 916           cl_cos_sin_t             ->  cos_sin_t
 917           cl_cosh_sinh()           ->  cosh_sinh()
 918           cl_cosh_sinh_t           ->  cosh_sinh_t
 919           cl_decoded_dfloat        ->  decoded_dfloat
 920           cl_decoded_ffloat        ->  decoded_ffloat
 921           cl_decoded_float         ->  decoded_float
 922           cl_decoded_lfloat        ->  decoded_lfloat
 923           cl_decoded_sfloat        ->  decoded_sfloat
 924           cl_default_float_format  ->  default_float_format
 925           cl_default_print_flags   ->  default_print_flags
 926           cl_default_random_state  ->  default_random_state
 927           cl_double_approx()       ->  double_approx()
 928           cl_equal()               ->  equal()
 929           cl_equal_hashcode()      ->  equal_hashcode()
 930           cl_eulerconst()          ->  eulerconst()
 931           cl_find_modint_ring()    ->  find_modint_ring()
 932           cl_find_univpoly_ring()  ->  find_univ_poly_ring()
 933           cl_float_approx()        ->  float_approx
 934           cl_float_format()        ->  float_format()
 935           cl_float_format_t        ->  float_format_t
 936           cl_free_hook()           ->  free_hook()
 937           cl_hermite()             ->  hermite()
 938           cl_laguerre()            ->  laguerre()
 939           cl_legendre()            ->  legandre()
 940           cl_malloc_hook()         ->  malloc_hook()
 941           cl_pi()                  ->  pi()
 942           cl_tschebychev()         ->  tschebychev()
 943           cl_zeta()                ->  zeta()
 944           NB: For functions these changes includes all signatures.
 945         * include/*: moved to include/cln/*, stripped `cl_' off filenames.
 946         * cln-config.in: new file.
 947
 948 2000-08-26  Bruno Haible  <haible@clisp.cons.org>
 949
 950         * autoconf/acgeneral.m4 (AC_OUTPUT): Use braces in exec_prefix default
 951           value, not parens.
 952
 953 2000-08-18  Bruno Haible  <haible@clisp.cons.org>
 954
 955         * include/cl_univpoly_modint.h: Fix typo.
 956
 957 2000-07-13  Bruno Haible  <haible@clisp.cons.org>
 958
 959         * src/float/input/cl_F_read_stream.cc (number_char_p): Accept '_',
 960           used as precision marker for floats.
 961           Reported by Keith Briggs (in 1998) and Thomas Roessler.
 962         * src/integer/input/cl_I_read_stream.cc (number_char_p): Likewise.
 963         * src/rational/input/cl_RA_read_stream.cc (number_char_p): Likewise.
 964         * src/real/input/cl_R_read_stream.cc (number_char_p): Likewise.
 965         * src/complex/input/cl_N_read_stream.cc (number_char_p): Likewise.
 966
 967 2000-06-22  Bruno Haible  <haible@clisp.cons.org>
 968
 969         * include/cl_object.h: Rename cl_word_size to cl_pointer_size.
 970         * include/cl_types.h (HAVE_FAST_LONGLONG): Also define on Irix6
 971           with N32 ABI.
 972           (cl_word_size): New macro.
 973         * src/float/sfloat/cl_SF.h: Use cl_pointer_size instead of
 974           cl_word_size.
 975
 976 2000-05-31  Bruno Haible  <haible@clisp.cons.org>
 977
 978         * tests/exam_I_floor.cc (integer_floor_tests): Add one more entry.
 979           From a sample that fails with gcc-2.95.2 on Sparc.
 980         * tests/exam_I_gcd.cc (integer_gcd_tests): Likewise.
 981
 982 2000-05-30  Richard Kreckel  <kreckel@ginac.de>
 983
 984         * configure.in, autoconf/aclocal.m4 (CL_GMP_H_VERSION, CL_GMP_CHECK):
 985           New macros.
 986         * configure.in, include/cl_config.h.in (CL_VERSION, CL_VERSION_MINOR,
 987           CL_VERSION_PATCHLEVEL): New definitions.
 988         * autoconf/config.guess, autoconf/config.sub, autoconf/ltconfig,
 989           autoconf/ltmain.sh: updated from FSF (libtool 1.3.5, etc).
 990         * src/Makefile.in, configure.in: release-variables renamed from
 991           CLN_* to CL_*.
 992         * configure.in: default to build both shared and static library
 993           (i.e. default to the most common behaviour).
 994
 995 2000-05-29  Richard Kreckel  <kreckel@ginac.de>
 996
 997         * autoconf/aclocal.m4 (CL_CANONICAL_HOST): Added missing changequote
 998           environment around the patch of 2000-05-23.
 999
1000 2000-05-29  Bruno Haible  <haible@clisp.cons.org>
1001
1002         * autoconf/aclocal.m4 (CL_PROG_INSTALL): Fix typo.
1003           Reported by Thomas Klausner <wiz@danbala.ifoer.tuwien.ac.at>.
1004
1005 2000-05-27  Richard Kreckel  <kreckel@ginac.de>
1006
1007         * src/float/lfloat/algebraic/cl_LF_sqrt.cc,
1008           src/base/digitseq/cl_DS_sqrt.cc: Readjusted break-even points.
1009
1010 2000-05-24  Richard Kreckel  <kreckel@ginac.de>
1011
1012         * autoconf/config.*: Updated to new version from FSF
1013           (the new libtool wants this).
1014         * src/Makefile.in: added $(LDFLAGS) to link step.
1015         * src/base/digitseq/cl_2DS_div.cc, cl_2DS_recip.cc: Readjusted
1016           break-even points.
1017
1018 2000-05-23  Bruno Haible  <haible@clisp.cons.org>
1019
1020         * autoconf/aclocal.m4 (CL_CANONICAL_HOST): Determine host_cpu,
1021           host_vendor, host_os correctly if $host has more than two hyphens.
1022
1023 2000-05-19  Richard Kreckel  <kreckel@ginac.de>
1024
1025         * src/base/digitseq/cl_DS_mul.cc: Rearranged break-even points to
1026           better match present-day CPUs whenever GMP3 is used.
1027         * src/base/digitseq/cl_DS_div.cc: dto.
1028         * src/TUNING: Added comment about order of tuning.
1029         * configure, configure.in: Safer GMP3-detection.
1030         * INSTALL.generic: Clarified behaviour of --with-gmp.
1031         * autoconf/config.guess: updated from Clisp-2000-03-06 sources.
1032
1033 2000-05-04  Richard Kreckel  <kreckel@ginac.de>
1034
1035         * gmp/: removed completely.  From now on we are going to link
1036           externally against libgmp3.0 or above!
1037         * configure, configure.in, Makefile.in, */Makeflags: removed support
1038           of internal gmp 2.0.2, like $GMP_INCLUDES, which should be done by
1039           setting $CPPFLAGS instead.
1040         * Makefile.in: Added libtool inter-library dependency for -lgmp and
1041           conforming interface versioning (-version-info).
1042         * autoconf/ltconfig, autoconf/ltmain.sh: Updated to newer versions
1043           from libtool 1.3.4.
1044
1045 2000-02-22  Bruno Haible  <haible@clisp.cons.org>
1046
1047         * src/base/digitseq/cl_asm_mips_.cc: Starting at argument 5 the
1048           parameter passing was changed for the MIPS n32 ABI.
1049
1050 2000-01-24  Richard Kreckel  <kreckel@ginac.de>
1051
1052         * gmp/*: Replaced the complete mpn sources with the ones from
1053           Debian since they are maintained while the ones from FSF
1054           aren't and there were problems on some architectures, PowerPC
1055           in particular.  See the file gmp/README.CLN.  This way the
1056           hard links in this directory have vanished, they were causing
1057           trouble for people working in AFS.  This became necessary for
1058           Debian, because there it woudn't compile on PPC.
1059
1060 2000-01-13  Richard Kreckel  <kreckel@ginac.de>
1061
1062         * Version 1.0.3 released.
1063
1064 2000-01-13  Richard Kreckel  <kreckel@ginac.de>
1065
1066         * src/base/cl_macros.h (nonreturning_function): For egcs and newer
1067           use __attribute__ ((__noreturn__)) instead of the __volatile__
1068           storage class.
1069           (nonreturning): Remove macro.
1070         * include/*: Minor fixes to stop -ansi -pedantic from complaining.
1071         * include/cl_integer.h: (doublefactorial): New declaration.
1072         * src/integer/misc/combin/cl_I_doublefactorial.cc: New file.
1073
1074 1999-12-18  Bruno Haible  <haible@clisp.cons.org>
1075
1076         * autoconf/acgeneral.m4 (AC_ARG_PROGRAM): Create conftestsed using
1077           "cat", not "echo".
1078         * autoconf/ltconfig: Improve support for recent FreeBSD 3.
1079         * include/cl_GV.h (cl_GV_vectorops): Change return type from 'T' to
1080           'const T', to match definition in src/vector/cl_GV_number.cc.
1081           Reported by Duncan Simpson <dps@io.stargate.co.uk>.
1082         * gmp/mpn/Makefile.in (.S.lo): Use 'if', not '&&', because '&&' may
1083           yield return code 1.
1084
1085 1999-09-07  Bruno Haible  <haible@clisp.cons.org>
1086
1087         * Version 1.0.2 released.
1088
1089 1999-09-06  Bruno Haible  <haible@clisp.cons.org>
1090
1091         * src/rational/cl_RA.h (integerp, ratiop): Instead of returning a
1092           boolean expression, write alternatives ending with either
1093           "return cl_true;" or "return cl_false;". This way, g++ does a
1094           better job inlining it.
1095         * src/float/cl_F.h (longfloatp): Likewise.
1096         * src/real/cl_R.h (rationalp, integerp, floatp): Likewise.
1097         * src/complex/cl_C.h (realp, complexp): Likewise.
1098
1099 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1100
1101         * include/cl_integer.h (cl_equal_hashcode): New declaration.
1102         * include/cl_rational.h (cl_equal_hashcode): New declaration.
1103         * include/cl_sfloat.h (cl_equal_hashcode): New declaration.
1104         * include/cl_ffloat.h (cl_equal_hashcode): New declaration.
1105         * include/cl_dfloat.h (cl_equal_hashcode): New declaration.
1106         * include/cl_lfloat.h (cl_equal_hashcode): New declaration.
1107         * include/cl_float.h (cl_equal_hashcode): New declaration.
1108         * include/cl_real.h (cl_equal_hashcode): New declaration.
1109         * include/cl_complex.h (cl_equal_hashcode): New declaration.
1110         * src/base/cl_N.h (equal_hashcode_low, equal_hashcode_one): New macros.
1111         * src/integer/misc/cl_I_eqhashcode.cc: New file.
1112         * src/rational/misc/cl_RA_eqhashcode.cc: New file.
1113         * src/float/sfloat/misc/cl_SF_eqhashcode.cc: New file.
1114         * src/float/ffloat/misc/cl_FF_eqhashcode.cc: New file.
1115         * src/float/dfloat/misc/cl_DF_eqhashcode.cc: New file.
1116         * src/float/lfloat/misc/cl_LF_eqhashcode.cc: New file.
1117         * src/float/misc/cl_F_eqhashcode.cc: New file.
1118         * src/real/misc/cl_R_eqhashcode.cc: New file.
1119         * src/complex/misc/cl_C_eqhashcode.cc: New file.
1120         * doc/cln.tex: Document `cl_equal_hashcode'.
1121
1122 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1123
1124         * include/cl_ring.h (cl_number_ring_ops): Add `contains' member.
1125           (cl_number_ring): New class.
1126           (cl_specialized_number_ring<T>): Inherit from cl_number_ring.
1127           (instanceof): New function.
1128         * src/integer/ring/cl_I_ring.cc (cl_I_p): New function.
1129         * src/integer/misc/cl_I_as.cc (cl_I_p): Add comment.
1130         * src/rational/ring/cl_RA_ring.cc (cl_RA_p): New function.
1131         * src/rational/misc/cl_RA_as.cc (cl_RA_p): Add comment.
1132         * src/real/ring/cl_R_ring.cc (cl_R_p): New function.
1133         * src/real/misc/cl_R_as.cc (cl_R_p): Add comment.
1134         * src/complex/ring/cl_C_ring.cc (cl_N_p): New function.
1135         * src/complex/misc/cl_N_as.cc (cl_N_p): Add comment.
1136         * doc/cln.tex: Document `instanceof'.
1137
1138 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1139
1140         * include/cl_rational.h (numerator, denominator): New declarations.
1141         * src/rational/elem/cl_RA_numerator.cc: New file.
1142         * src/rational/elem/cl_RA_denominator.cc: New file.
1143         * include/cl_integer.h (numerator, denominator): New inline functions.
1144         * doc/cln.tex: Document `numerator' and `denominator'.
1145
1146 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1147
1148         * src/rational/algebraic/cl_RA_rootp.cc (rootp): Fix endless loop
1149           in the integer case.
1150         * src/rational/algebraic/cl_RA_rootp_I.cc (rootp): Likewise.
1151
1152 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1153
1154         * include/cl_config.h.in: Support for sparc64 CPU.
1155         * include/cl_modules.h: Likewise.
1156         * include/cl_types.h: Likewise.
1157         * include/cl_object.h: Likewise.
1158         * include/cl_GV.h: Likewise.
1159         * src/Makefile.in: Likewise.
1160         * src/base/cl_alloca.h: Likewise.
1161         * src/base/cl_macros.h: Likewise.
1162         * src/base/cl_sysdep.h: Likewise.
1163         * src/base/cl_low.h: Likewise.
1164         * src/base/digitseq/cl_asm.h: Likewise.
1165         * src/base/digitseq/cl_asm_.cc: Likewise.
1166         * src/base/digitseq/cl_asm_sparc64.h: New file.
1167         * src/base/digitseq/cl_asm_sparc64_.cc: New file.
1168         * src/modinteger/cl_MI_int.h: Support for sparc64 CPU.
1169         * src/polynomial/elem/cl_UP_no_ring.cc: Likewise.
1170         * src/polynomial/elem/cl_UP_GF2.h: Likewise.
1171         * src/polynomial/elem/cl_asm_GF2.cc: Likewise.
1172
1173 1999-09-04  Bruno Haible  <haible@clisp.cons.org>
1174
1175         * src/base/digitseq/cl_asm_sparc_.cc (orc2_loop_up, orc2_loop_down):
1176           Use the `orn' instruction.
1177
1178 1999-08-14  Bruno Haible  <haible@clisp.cons.org>
1179
1180         Assume all platforms have <stdlib.h> and clock_t.
1181         * configure.in: Don't call CL_STDLIB_H and CL_CLOCK_T.
1182         * src/base/cl_base_config.h.in (CLOCK_T): Remove definition.
1183         * src/base/random/cl_random_from.cc: Use clock_t instead of CLOCK_T.
1184         * src/timing/cl_t_config.h.in (CLOCK_T): Remove definition.
1185         * src/timing/cl_t_current2.cc: Use clock_t instead of CLOCK_T.
1186
1187 1999-07-18  Bruno Haible  <haible@clisp.cons.org>
1188
1189         * gmp/config.guess: Link to autoconf/config.guess.
1190           gmp/config.sub: Link to autoconf/config.sub.
1191           Needed for Win32 platforms.
1192
1193 1999-07-17  Bruno Haible  <haible@clisp.cons.org>
1194
1195         * autoconf/aclocal.m4 (CL_CANONICAL_HOST_CPU): Distinguish "sparc" and
1196           "sparc64" according to the C compiler, not the uname result.
1197
1198 1999-06-17  Bruno Haible  <haible@clisp.cons.org>
1199
1200         * src/base/digitseq/cl_asm_sparc_.cc (compare_loop_up): Fix
1201           COUNTER_LOOPS version, fortunately not used yet.
1202
1203         * include/cl_modules.h: Prepare for gcc version 3.
1204
1205 1999-06-12  Bruno Haible  <haible@clisp.cons.org>
1206
1207         * src/rational/elem/cl_RA_plus.cc, src/rational/elem/cl_RA_minus.cc:
1208           Change the last call from I_I_to_RT to I_I_to_RA.
1209
1210 1999-06-09  Bruno Haible  <haible@clisp.cons.org>
1211
1212         * Version 1.0.1 released.
1213
1214 1999-06-09  Bruno Haible  <haible@clisp.cons.org>
1215
1216         * src/integer/cl_I.h (pFN_maxlength_digits_at): Define also when
1217           intDsize==64.
1218
1219 1999-06-08  Bruno Haible  <haible@clisp.cons.org>
1220
1221         * autoconf/intparam.c (printf_underscored): Change argument type to
1222           `const char*'.
1223         * include/cl_modules.h (CL_OUTPUT_LABEL, CL_JUMP_TO): New macros.
1224           (CL_PROVIDE, CL_PROVIDE_END): Use them.
1225         * include/cl_string.h (cl_heap_string::operator new): Return 1, not 0.
1226         * include/cl_GV.h (cl_GV_inner<T>::operator new): Likewise.
1227         * src/base/ring/cl_no_ring.cc (dummy_canonhom, dummy_expt_pos): Don't
1228           cast a cl_I to void here. Works around a bug in g++-2.95.
1229         * src/complex/misc/cl_C_class.cc: Include "cl_C.h".
1230         * src/polynomial/elem/cl_UP_no_ring.cc (dummy_canonhom,
1231            dummy_expt_pos): Don't cast a cl_I to void here. Works around a bug
1232           in g++-2.95.
1233         * src/polynomial/elem/cl_asm_sparc_GF2.cc (DECLARE_FUNCTION): New
1234           macro.
1235         * src/rational/misc/cl_RA_class.cc: Include "cl_RA.h".
1236         * src/vector/cl_GV_I.cc (cl_heap_GV_I_general::operator new,
1237            DEFINE_cl_heap_GV_I_bits): Return 1, not 0.
1238         * src/vector/cl_GV_number.cc (cl_heap_GV_number_general::operator new):
1239           Likewise.
1240
1241 1999-06-01  Bruno Haible  <haible@clisp.cons.org>
1242
1243         * autoconf/aclocal.m4 (CL_CANONICAL_HOST_CPU): Canonicalize alpha
1244           variants to alpha.
1245
1246 1999-05-29  Bruno Haible  <haible@clisp.cons.org>
1247
1248         * src/base/digitseq/cl_asm_i386_.cc (DECLARE_FUNCTION): Treat
1249           OpenBSD like NetBSD.
1250         * src/base/digitseq/cl_asm_sparc_.cc (DECLARE_FUNCTION): Likewise.
1251         * src/base/digitseq/cl_asm_m68k_.cc (DECLARE_FUNCTION): Treat
1252           OpenBSD like NetBSD, and Linux/ELF like SVR4.
1253
1254 1999-05-16  Bruno Haible  <haible@clisp.cons.org>
1255
1256         * src/base/cl_low.h (integerlength32) [__rs6000__]: Use old assembler
1257           syntax on AIX systems and new assembler syntax on non-AIX systems.
1258
1259 1999-05-01  Bruno Haible  <haible@clisp.cons.org>
1260
1261         * autoconf/config.guess, autoconf/config.sub: Upgrade to newest
1262           version from GNU CVS.
1263
1264 1999-04-24  Bruno Haible  <haible@clisp.cons.org>
1265
1266         * src/integer/bitwise/cl_I_logand.cc (logand): Optimize the case when
1267           either operand is a positive fixnum, O(1) instead of O(N).
1268         * src/integer/bitwise/cl_I_lognand.cc (lognand): Likewise.
1269         * src/integer/bitwise/cl_I_logandc2.cc (logandc2): Likewise for the
1270           first operand.
1271
1272 1999-04-14  Bruno Haible  <haible@clisp.cons.org>
1273
1274         * autoconf/aclocal.m4 (CL_GLOBAL_CONSTRUCTORS): Add check whether
1275           ctor/dtor needs to be exported.
1276         * include/cl_config.h.in (CL_NEED_GLOBALIZE_CTORDTOR): New macro.
1277         * include/cl_modules.h (CL_GLOBALIZE_JUMP_LABEL): Renamed from
1278           CL_GLOBALIZE_LABEL.
1279           (CL_GLOBALIZE_LABEL): New macro.
1280           (CL_GLOBALIZE_CTORDTOR_LABEL): Renamed from CL_GLOBALIZE_ASM_LABEL.
1281           (CL_PROVIDE): Update.
1282
1283 1999-04-12  Bruno Haible  <haible@clisp.cons.org>
1284
1285         * src/Makefile.in ($(ASMFILES_S)): On HPPA, ignore preprocessing
1286           errors ("unterminated string or character constant").
1287           ($(ASMFILES_LO)): On HPPA, try with various settings of
1288           COMPILER_PATH, in order to try /usr/ccs/bin/as and /bin/as.
1289
1290 1999-04-11  Bruno Haible  <haible@clisp.cons.org>
1291
1292         * INSTALL: Mention gmp problems on MIPS.
1293         * doc/cln.tex: Likewise.
1294
1295 1999-03-24  Mumit Khan  <khan@xraylith.wisc.edu>
1296
1297         * src/Makefile.in (SUBDIRS): Filter out CVS and RCS directories from
1298           the source tree.
1299         * include/cl_modules.h (CL_GLOBALIZE_LABEL): Define for Win32.
1300           (CL_GLOBALIZE_ASM_LABEL): New macro.
1301           (CL_PROVIDE): Use it.
1302         * src/base/random/cl_random_from.cc: Handle WIN32.
1303         * src/timing/cl_t_current.cc: Likewise.
1304
1305 1999-03-15  Bruno Haible  <haible@clisp.cons.org>
1306
1307         * autoconf/intparam.c (main7): Use %lX instead of %X for a `long'.
1308           (main8): Adapt for C++.
1309
1310 1999-03-09  Bruno Haible  <haible@clisp.cons.org>
1311
1312         * INSTALL: Mention egcs-1.1 problems on Sparc.
1313         * doc/cln.tex: Likewise.
1314
1315 1999-03-08  Bruno Haible  <haible@clisp.cons.org>
1316
1317         * autoconf/aclocal.m4 (CL_FPU_CONTROL): Fix the "checking for"
1318           messages.
1319
1320 1999-02-25  Bruno Haible  <haible@clisp.cons.org>
1321
1322         * autoconf/aclocal.m4: In test programs, declare `int main()', not
1323           `main()'.
1324
1325         * lidia-interface/src/interfaces/integers/cln/bigint.c
1326           (bigint_to_string): Fix for negative arguments.
1327
1328         * src/base/cl_low.h: Check for `__sparc__', not `SPARC'.
1329         * src/base/cl_alloca.h: Likewise.
1330
1331         * src/base/cl_low.h: Eliminate CLISP style "# " comments.
1332         * src/base/digitseq/cl_asm_arm_.cc,
1333           src/base/digitseq/cl_asm_mips_.cc,
1334           src/float/dfloat/elem/cl_DF_mul.cc: Likewise.
1335
1336         * src/modinteger/cl_MI_pow2.h,
1337           src/modinteger/cl_MI_pow2m1.h,
1338           src/modinteger/cl_MI_pow2p1.h: Workaround g++-2.7.2 inlining bug.
1339
1340 1999-01-18  Bruno Haible  <haible@clisp.cons.org>
1341
1342         * autoconf/acgeneral.m4,
1343           autoconf/acspecific.m4: Upgrade to autoconf-2.13.
1344         * autoconf/config.guess, autoconf/config.sub: Likewise.
1345         * autoconf/aclocal.m4 (CL_ALLOCA): Test for _MSC_VER and alloca being
1346           a macro. Use ${ac_objext}.
1347         * src/base/cl_macros.h (alloca): Put _MSC_VER test before the others,
1348           conforming with CL_ALLOCA.
1349
1350 1999-01-12  Bruno Haible  <haible@clisp.cons.org>
1351
1352         * Version 1.0 released.
1353