ChangeLog

   1 2006-04-25  Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
   2
   3         Prepare for autoconf-2.60.
   4         * Makefile.in (datarootdir): New variable.
   5         * src/Makefile.in (datarootdir): New variable.
   6         * doc/Makefile.in (datarootdir): New variable.
   7
   8 2006-04-25  Bruno Haible  <bruno@clisp.org>
   9             Richard B. Kreckel  <kreckel@ginac.de>
  10
  11         Make it theoretically possible to use bignums and long-floats with
  12         more than 2^32 significant digits or bits.
  13         * doc/cln.tex (logcount): Change return type to uintC.
  14         (struct cl_byte): Change elements to uintC.
  15         (integer_length, ord2, power2p): Change return type to uintC.
  16         (scale_float): Change argument type to sintC.
  17         (float_digits, float_precision): Change return type to uintC.
  18         * examples/atan_recip.cc: Use uintC instead of uintL where appropriate.
  19         * examples/atanh_recip.cc: Likewise.
  20         * include/cln/GV.h: Likewise.
  21         * include/cln/GV_complex.h: Likewise.
  22         * include/cln/GV_integer.h: Likewise.
  23         * include/cln/GV_modinteger.h: Likewise.
  24         * include/cln/GV_number.h: Likewise.
  25         * include/cln/GV_rational.h: Likewise.
  26         * include/cln/GV_real.h: Likewise.
  27         * include/cln/SV.h: Likewise.
  28         * include/cln/SV_complex.h: Likewise.
  29         * include/cln/SV_integer.h: Likewise.
  30         * include/cln/SV_number.h: Likewise.
  31         * include/cln/SV_rational.h: Likewise.
  32         * include/cln/SV_real.h: Likewise.
  33         * include/cln/SV_ringelt.h: Likewise.
  34         * include/cln/dfloat.h: Likewise.
  35         * include/cln/ffloat.h: Likewise.
  36         * include/cln/float.h: Likewise.
  37         * include/cln/integer.h: Likewise.
  38         * include/cln/lfloat.h: Likewise.
  39         * include/cln/modinteger.h: Likewise.
  40         * include/cln/sfloat.h: Likewise.
  41         * src/base/cl_low.h (integerlengthC): New macro.
  42         * src/base/digitseq/cl_2DS_div.cc: Use uintC instead of uintL where
  43         appropriate.
  44         * src/base/digitseq/cl_2DS_recip.cc: Likewise.
  45         * src/base/digitseq/cl_DS.h: Likewise.
  46         * src/base/digitseq/cl_DS_mul.c: Likewise.
  47         * src/base/digitseq/cl_DS_mul_fftc.h: Likewise.
  48         * src/base/digitseq/cl_DS_mul_fftcs.h: Likewise.
  49         * src/base/digitseq/cl_DS_mul_fftm.h: Likewise.
  50         * src/base/digitseq/cl_DS_mul_fftp.h: Likewise.
  51         * src/base/digitseq/cl_DS_mul_fftp3.h: Likewise.
  52         * src/base/digitseq/cl_DS_mul_fftp3m.h: Likewise.
  53         * src/base/digitseq/cl_DS_mul_fftr.h: Likewise.
  54         * src/base/digitseq/cl_DS_mul_kara.h: Likewise.
  55         * src/base/digitseq/cl_DS_mul_nuss.h: Likewise.
  56         * src/base/digitseq/cl_DS_recip.cc: Likewise.
  57         * src/base/digitseq/cl_DS_recipsqrt.cc: Likewise.
  58         * src/base/digitseq/cl_DS_sqrt.cc: Likewise.
  59         * src/base/digitseq/cl_DS_trandom.cc: Likewise.
  60         * src/complex/input/cl_N_read.cc: Likewise.
  61         * src/complex/transcendental/cl_C_asinh_aux.cc: Likewise.
  62         * src/complex/transcendental/cl_C_expt_C.cc: Likewise.
  63         * src/float/cl_F.h: Likewise.
  64         * src/float/conv/cl_F_from_F_f.cc: Likewise.
  65         * src/float/conv/cl_F_from_I_f.cc: Likewise.
  66         * src/float/conv/cl_F_from_RA_f.cc: Likewise.
  67         * src/float/dfloat/conv/cl_I_to_double.cc: Likewise.
  68         * src/float/dfloat/conv/cl_RA_to_double.cc: Likewise.
  69         * src/float/dfloat/elem/cl_DF_from_I.cc: Likewise.
  70         * src/float/dfloat/elem/cl_DF_from_RA.cc: Likewise.
  71         * src/float/dfloat/elem/cl_DF_scale.cc: Likewise.
  72         * src/float/dfloat/misc/cl_DF_digits.cc: Likewise.
  73         * src/float/dfloat/misc/cl_DF_precision.cc: Likewise.
  74         * src/float/elem/cl_F_scale.cc: Likewise.
  75         * src/float/ffloat/conv/cl_I_to_float.cc: Likewise.
  76         * src/float/ffloat/conv/cl_RA_to_float.cc: Likewise.
  77         * src/float/ffloat/elem/cl_FF_from_I.cc: Likewise.
  78         * src/float/ffloat/elem/cl_FF_from_RA.cc: Likewise.
  79         * src/float/ffloat/elem/cl_FF_scale.cc: Likewise.
  80         * src/float/ffloat/misc/cl_FF_digits.cc: Likewise.
  81         * src/float/ffloat/misc/cl_FF_precision.cc: Likewise.
  82         * src/float/input/cl_F_read.cc: Likewise.
  83         * src/float/lfloat/algebraic/cl_LF_sqrt.cc: Likewise.
  84         * src/float/lfloat/elem/cl_LF_1plus.cc: Likewise.
  85         * src/float/lfloat/elem/cl_LF_I_div.cc: Likewise.
  86         * src/float/lfloat/elem/cl_LF_I_mul.cc: Likewise.
  87         * src/float/lfloat/elem/cl_LF_div.cc: Likewise.
  88         * src/float/lfloat/elem/cl_LF_from_I.cc: Likewise.
  89         * src/float/lfloat/elem/cl_LF_from_RA.cc: Likewise.
  90         * src/float/lfloat/elem/cl_LF_fround.cc: Likewise.
  91         * src/float/lfloat/elem/cl_LF_ftrunc.cc: Likewise.
  92         * src/float/lfloat/elem/cl_LF_futrunc.cc: Likewise.
  93         * src/float/lfloat/elem/cl_LF_scale.cc: Likewise.
  94         * src/float/lfloat/elem/cl_LF_to_I.cc: Likewise.
  95         * src/float/lfloat/misc/cl_LF_digits.cc: Likewise.
  96         * src/float/lfloat/misc/cl_LF_idecode.cc: Likewise.
  97         * src/float/lfloat/misc/cl_LF_leninc.cc: Likewise.
  98         * src/float/lfloat/misc/cl_LF_lenincx.cc: Likewise.
  99         * src/float/lfloat/misc/cl_LF_precision.cc: Likewise.
 100         * src/float/lfloat/misc/cl_LF_shortenrel.cc: Likewise.
 101         * src/float/lfloat/misc/cl_LF_shortenwith.cc: Likewise.
 102         * src/float/misc/cl_F_digits.cc: Likewise.
 103         * src/float/misc/cl_F_epsneg.cc: Likewise.
 104         * src/float/misc/cl_F_epspos.cc: Likewise.
 105         * src/float/misc/cl_F_leastneg.cc: Likewise.
 106         * src/float/misc/cl_F_leastpos.cc: Likewise.
 107         * src/float/misc/cl_F_mostneg.cc: Likewise.
 108         * src/float/misc/cl_F_mostpos.cc: Likewise.
 109         * src/float/misc/cl_F_precision.cc: Likewise.
 110         * src/float/misc/cl_F_rational.cc: Likewise.
 111         * src/float/misc/cl_F_shortenrel.cc: Likewise.
 112         * src/float/output/cl_F_dprint.cc: Likewise.
 113         * src/float/random/cl_F_random.cc: Likewise.
 114         * src/float/sfloat/elem/cl_SF_from_I.cc: Likewise.
 115         * src/float/sfloat/elem/cl_SF_from_RA.cc: Likewise.
 116         * src/float/sfloat/elem/cl_SF_scale.cc: Likewise.
 117         * src/float/sfloat/misc/cl_SF_digits.cc: Likewise.
 118         * src/float/sfloat/misc/cl_SF_precision.cc: Likewise.
 119         * src/float/transcendental/cl_F_atanhx.cc: Likewise.
 120         * src/float/transcendental/cl_F_atanx.cc: Likewise.
 121         * src/float/transcendental/cl_F_catalanconst_f.cc: Likewise.
 122         * src/float/transcendental/cl_F_cos.cc: Likewise.
 123         * src/float/transcendental/cl_F_cosh.cc: Likewise.
 124         * src/float/transcendental/cl_F_coshsinh.cc: Likewise.
 125         * src/float/transcendental/cl_F_cossin.cc: Likewise.
 126         * src/float/transcendental/cl_F_eulerconst_f.cc: Likewise.
 127         * src/float/transcendental/cl_F_exp1_f.cc: Likewise.
 128         * src/float/transcendental/cl_F_expx.cc: Likewise.
 129         * src/float/transcendental/cl_F_ln10_f.cc: Likewise.
 130         * src/float/transcendental/cl_F_ln2_f.cc: Likewise.
 131         * src/float/transcendental/cl_F_lnx.cc: Likewise.
 132         * src/float/transcendental/cl_F_pi_f.cc: Likewise.
 133         * src/float/transcendental/cl_F_sin.cc: Likewise.
 134         * src/float/transcendental/cl_F_sinh.cc: Likewise.
 135         * src/float/transcendental/cl_F_sinhx.cc: Likewise.
 136         * src/float/transcendental/cl_F_sinx.cc: Likewise.
 137         * src/float/transcendental/cl_F_tran.h: Likewise.
 138         * src/float/transcendental/cl_F_zeta_int_f.cc: Likewise.
 139         * src/float/transcendental/cl_LF_atan_recip.cc: Likewise.
 140         * src/float/transcendental/cl_LF_atanh_recip.cc: Likewise.
 141         * src/float/transcendental/cl_LF_catalanconst.cc: Likewise.
 142         * src/float/transcendental/cl_LF_coshsinh_aux.cc: Likewise.
 143         * src/float/transcendental/cl_LF_cossin_aux.cc: Likewise.
 144         * src/float/transcendental/cl_LF_eulerconst.cc: Likewise.
 145         * src/float/transcendental/cl_LF_exp1.cc: Likewise.
 146         * src/float/transcendental/cl_LF_exp_aux.cc: Likewise.
 147         * src/float/transcendental/cl_LF_pi.cc: Likewise.
 148         * src/float/transcendental/cl_LF_ratseries.cc: Likewise.
 149         * src/float/transcendental/cl_LF_ratseries_.cc: Likewise.
 150         * src/float/transcendental/cl_LF_ratseries_a.cc: Likewise.
 151         * src/float/transcendental/cl_LF_ratseries_ab.cc: Likewise.
 152         * src/float/transcendental/cl_LF_ratseries_b.cc: Likewise.
 153         * src/float/transcendental/cl_LF_ratseries_p.cc: Likewise.
 154         * src/float/transcendental/cl_LF_ratseries_pa.cc: Likewise.
 155         * src/float/transcendental/cl_LF_ratseries_pab.cc: Likewise.
 156         * src/float/transcendental/cl_LF_ratseries_pb.cc: Likewise.
 157         * src/float/transcendental/cl_LF_ratseries_pq.cc: Likewise.
 158         * src/float/transcendental/cl_LF_ratseries_pqa.cc: Likewise.
 159         * src/float/transcendental/cl_LF_ratseries_pqab.cc: Likewise.
 160         * src/float/transcendental/cl_LF_ratseries_pqb.cc: Likewise.
 161         * src/float/transcendental/cl_LF_ratseries_q.cc: Likewise.
 162         * src/float/transcendental/cl_LF_ratseries_qa.cc: Likewise.
 163         * src/float/transcendental/cl_LF_ratseries_qab.cc: Likewise.
 164         * src/float/transcendental/cl_LF_ratseries_qb.cc: Likewise.
 165         * src/float/transcendental/cl_LF_ratseries_stream_pq.cc: Likewise.
 166         * src/float/transcendental/cl_LF_ratseries_stream_pqa.cc: Likewise.
 167         * src/float/transcendental/cl_LF_ratseries_stream_pqab.cc: Likewise.
 168         * src/float/transcendental/cl_LF_ratseries_stream_pqb.cc: Likewise.
 169         * src/float/transcendental/cl_LF_ratsumseries_pqcd.cc: Likewise.
 170         * src/float/transcendental/cl_LF_ratsumseries_pqcd_aux.cc: Likewise.
 171         * src/float/transcendental/cl_LF_ratsumseries_pqd.cc: Likewise.
 172         * src/float/transcendental/cl_LF_ratsumseries_pqd_aux.cc: Likewise.
 173         * src/float/transcendental/cl_LF_tran.h: Likewise.
 174         * src/float/transcendental/cl_LF_zeta3.cc: Likewise.
 175         * src/float/transcendental/cl_LF_zeta_int.cc: Likewise.
 176         * src/integer/algebraic/cl_I_rootp_I.cc: Likewise.
 177         * src/integer/algebraic/cl_I_rootp_aux.cc: Likewise.
 178         * src/integer/bitwise/cl_I_ash.cc: Likewise.
 179         * src/integer/bitwise/cl_I_ash_I.cc: Likewise.
 180         * src/integer/bitwise/cl_I_byte.h: Likewise.
 181         * src/integer/bitwise/cl_I_fullbyte.cc: Likewise.
 182         * src/integer/bitwise/cl_I_ilength.cc: Likewise.
 183         * src/integer/bitwise/cl_I_ldb.cc: Likewise.
 184         * src/integer/bitwise/cl_I_ldbtest.cc: Likewise.
 185         * src/integer/bitwise/cl_I_ldbx.cc: Likewise.
 186         * src/integer/bitwise/cl_I_ldbxtest.cc: Likewise.
 187         * src/integer/bitwise/cl_I_logbitp.cc: Likewise.
 188         * src/integer/bitwise/cl_I_logbitp_I.cc: Likewise.
 189         * src/integer/bitwise/cl_I_logcount.cc: Likewise.
 190         * src/integer/bitwise/cl_I_mkf.cc: Likewise.
 191         * src/integer/bitwise/cl_I_mkfx.cc: Likewise.
 192         * src/integer/cl_I.h: Likewise.
 193         * src/integer/conv/cl_I_to_digits.cc: Likewise.
 194         * src/integer/conv/cl_I_digits_need.cc: Likewise.
 195         * src/integer/conv/cl_I_from_digits.cc: Likewise.
 196         * src/integer/gcd/cl_I_gcd.cc: Likewise.
 197         * src/integer/gcd/cl_I_xgcd.cc: Likewise.
 198         * src/integer/misc/cl_I_eqhashcode.cc: Likewise.
 199         * src/integer/misc/cl_I_ord2.cc: Likewise.
 200         * src/integer/misc/cl_I_power2p.cc: Likewise.
 201         * src/integer/output/cl_I_cached_power.h (cached_power_table): allow
 202         for 40 elements.
 203         * src/integer/output/cl_I_decstring.cc: Use uintC instead of uintL
 204         where appropriate.
 205         * src/integer/output/cl_I_print.cc: Likewise.
 206         * src/integer/output/cl_I_print_string.cc: Likewise.
 207         * src/modinteger/cl_MI.cc: Likewise.
 208         * src/modinteger/cl_MI_lshift.cc: Likewise.
 209         * src/modinteger/cl_MI_montgom.h: Likewise.
 210         * src/modinteger/cl_MI_pow2.h: Likewise.
 211         * src/modinteger/cl_MI_pow2m1.h: Likewise.
 212         * src/modinteger/cl_MI_pow2p1.h: Likewise.
 213         * src/modinteger/cl_MI_rshift.cc: Likewise.
 214         * src/modinteger/cl_MI_std.h: Likewise.
 215         * src/numtheory/cl_IF_millerrabin.cc: Likewise.
 216         * src/numtheory/cl_nt_isprobprime.cc: Likewise.
 217         * src/numtheory/cl_nt_sqrtmodp.cc: Likewise.
 218         * src/polynomial/elem/cl_UP_GF2.h: Likewise.
 219         * src/real/conv/cl_F_from_R_f.cc: Likewise.
 220         * src/real/format-output/cl_fmt_floatstring.cc: Likewise.
 221         * src/real/input/cl_R_read.cc: Likewise.
 222         * src/vector/cl_GV_I.cc: Likewise.
 223         * src/vector/cl_GV_I_copy.cc: Likewise.
 224         * src/vector/cl_GV_number.cc: Likewise.
 225         * src/vector/cl_GV_number_copy.cc: Likewise.
 226         * src/vector/cl_SV_copy.cc: Likewise.
 227         * src/vector/cl_SV_number.cc: Likewise.
 228         * src/vector/cl_SV_ringelt.cc: Likewise.
 229         * tests/main.cc: Likewise.
 230         * tests/test_I_ilength.cc: Likewise.
 231         * tests/test_I_ord2.cc: Likewise.
 232
 233 2006-04-19  Bruno Haible  <bruno@clisp.org>
 234
 235         Prepare for autoconf-2.60.
 236         * general.m4 (CL_CC_WORKS): Include <stdlib.h>, for exit() declaration.
 237         * longdouble.m4 (CL_LONGDOUBLE): Likewise.
 238         * longlong.m4 (CL_LONGLONG): Likewise.
 239         * times.m4 (CL_TIMES_CLOCK): Likewise.
 240         Reported by Ralf Wildenhues <Ralf.Wildenhues@gmx.de>.
 241
 242 2005-12-04  Bruno Haible  <bruno@clisp.org>
 243
 244         * src/integer/conv/cl_I_to_digits.cc (I_to_digits_noshrink): Set
 245         erg->len at the end.
 246
 247 2005-12-04  Bruno Haible  <bruno@clisp.org>
 248
 249         Extend the fixnum range from 32 bits to 61 bits on 64-bit platforms.
 250         * doc/cln.tex (gcd, jacobi): Take 'long' instead of 'int32' arguments.
 251         * include/cln/object.h (cl_value_shift): Define as 3, not 32, in the
 252         64-bit case.
 253         * include/cln/types.h (intVsize): New macro.
 254         (sintV, uintV): New types.
 255         * include/cln/integer.h (gcd): Take uintV arguments.
 256         * include/cln/numtheory.h (jacobi): Take sintV arguments.
 257         * src/complex/input/cl_N_read.cc (read_complex): Call FN_to_UV instead
 258         of FN_to_UL.
 259         * src/complex/transcendental/cl_C_expt_C.cc: Likewise.
 260         * src/float/dfloat/elem/cl_DF_scale_I.cc: Use uintV instead of uintL
 261         * src/float/dfloat/elem/cl_DF_from_RA.cc: Don't assume values >= 2^53
 262         and < 2^55 are always bignums.
 263         where appropriate.
 264         * src/float/ffloat/conv/cl_RA_to_float.cc: Call FN_to_UV instead of
 265         FN_to_UL.
 266         * src/float/ffloat/elem/cl_FF_from_RA.cc: Likewise.
 267         * src/float/ffloat/elem/cl_FF_scale_I.cc: Call FN_to_V instead of
 268         FN_to_L.
 269         * src/float/lfloat/elem/cl_LF_scale_I.cc: Likewise.
 270         * src/float/output/cl_F_dprint.cc: Likewise.
 271         * src/float/sfloat/elem/cl_SF_from_RA.cc: Call FN_to_UV instead of
 272         FN_to_UL.
 273         * src/float/sfloat/elem/cl_SF_scale_I.cc: Call FN_to_V instead of
 274         FN_to_L.
 275         * src/integer/cl_I.h (FN_to_UV): Renamed from FN_to_UL, change return
 276         type.
 277         (FN_to_V): Renamed from FN_to_L, change return type.
 278         (FN_V_zerop): Renamed from FN_L_zerop.
 279         (FN_V_minusp): Renamed from FN_L_minusp.
 280         (cl_I_constructor_from_L2, cl_I_constructor_from_UL2): Define as an
 281         inline function on 64-bit platforms.
 282         (V_to_I, UV_to_I): New macros.
 283         (pFN_maxlength_digits_at): Return an uintV instead of an uint32.
 284         (set_pFN_maxlength_digits_at): Take an uintV instead of an uint32.
 285         * src/integer/algebraic/cl_I_sqrtp.cc: Call FN_to_UV instead of
 286         FN_to_UL.
 287         * src/integer/bitwise/cl_I_ash_I.cc: Call FN_to_V instead of FN_to_L.
 288         * src/integer/bitwise/cl_I_ilength.cc: Likewise.
 289         * src/integer/bitwise/cl_I_log_aux.cc: Likewise.
 290         * src/integer/bitwise/cl_I_logbitp_I.cc: Likewise.
 291         * src/integer/bitwise/cl_I_logcount.cc: Likewise.
 292         * src/integer/bitwise/cl_I_logtest.cc: Likewise.
 293         * src/integer/conv/cl_I_from_L2.cc: Don't produce code on 64-bit
 294         platforms.
 295         * src/integer/conv/cl_I_from_UL2.cc: Likewise.
 296         * src/integer/conv/cl_I_from_NDS.cc: Update.
 297         * src/integer/conv/cl_I_from_Q.cc: Remove cast to sint32.
 298         * src/integer/conv/cl_I_from_UQ.cc: Remove cast to uint32.
 299         * src/integer/conv/cl_I_to_L.cc: Check again fixnum that needs more
 300         than 32 bits.
 301         * src/integer/conv/cl_I_to_Q.cc: Call FN_to_V instead of FN_to_L.
 302         * src/integer/conv/cl_I_to_UL.cc: Likewise.
 303         * src/integer/conv/cl_I_to_UQ.cc: Likewise.
 304         * src/integer/elem/cl_I_div.cc: Treat fixnums that need more than 32
 305         bits like bignums.
 306         * src/integer/elem/cl_I_minus.cc: Call FN_to_V instead of FN_to_L.
 307         * src/integer/elem/cl_I_mul.cc: Likewise.
 308         * src/integer/elem/cl_I_plus.cc: Likewise.
 309         * src/integer/elem/cl_I_square.cc: Likewise.
 310         * src/integer/elem/cl_I_uminus.cc: Likewise.
 311         * src/integer/gcd/cl_I_gcd.cc: Likewise.
 312         * src/integer/gcd/cl_low_gcd.cc (gcd): Take uintV arguments.
 313         * src/integer/hash/cl_I_hashcode.cc: Call FN_to_V instead of FN_to_L.
 314         * src/integer/input/cl_I_read.cc (read_integer): Call FN_to_UV instead
 315         of FN_to_UL.
 316         * src/integer/misc/cl_I_ord2.cc (ord2): Call ord2_64 on 64-bit
 317         platforms.
 318         * src/integer/misc/cl_I_power2p.cc: Call FN_to_UV instead of FN_to_UL.
 319         * src/integer/misc/combin/cl_I_doublefactorial.cc (doublefakul_table):
 320         Extend table for larger fixnums.
 321         (doublefactorial): Update.
 322         * src/integer/misc/combin/cl_I_factorial.cc (fakul_table): Extend table
 323         for larger fixnums.
 324         (factorial): Update.
 325         * src/modinteger/cl_MI_fix16.h: Call FN_to_UV instead of FN_to_UL.
 326         * src/modinteger/cl_MI_fix29.h: Likewise.
 327         * src/modinteger/cl_MI_fix32.h: Likewise.
 328         * src/modinteger/cl_MI_std.h: Likewise.
 329         * src/numtheory/cl_nt_cornacchia4.cc: Call FN_to_V instead of FN_to_L.
 330         * src/numtheory/cl_nt_jacobi.cc: Likewise.
 331         * src/numtheory/cl_nt_jacobi_low.cc (jacobi_aux): Take uintV arguments
 332         instead of uint32 arguments.
 333         (jacobi): Take sintV argument instead of a sint32 argument.
 334         * src/rational/input/cl_RA_read.cc: Call FN_to_UV instead of FN_to_UL.
 335         * src/real/input/cl_R_read.cc: Likewise.
 336         * src/vector/cl_GV_I.cc: Likewise.
 337         * tests/timefact.cc: Call FN_to_V instead of FN_to_L.
 338
 339 2005-12-04  Bruno Haible  <bruno@clisp.org>
 340
 341         More complete 64-bit division macros.
 342         * src/base/cl_low.h (divu_6432_3232_w): Choose a different macro
 343         expansion on x86_64.
 344         (divu_6432_6432): New macro.
 345         (divu_6464_6464): Choose a different macro expansion for all CPUs
 346         except sparc64 and x86_64.
 347         (divu_12864_6464): Define NEED_VAR_divu_64_rest here.
 348         * src/base/low/cl_low_div.cc (divu_64_rest): Avoid defining it twice.
 349         (divu_6464_6464_): New function.
 350
 351 2005-12-04  Bruno Haible  <bruno@clisp.org>
 352
 353         * src/base/cl_low.h (ord2_64): New macro.
 354
 355 2005-12-02  Bruno Haible  <bruno@clisp.org>
 356
 357         * src/base/cl_low.h (mulu64) [x86_64]: Change asm restriction, since
 358         mulq doesn't accept immediate arguments.
 359
 360 2005-11-26  Bruno Haible  <bruno@clisp.org>
 361
 362         * src/base/cl_low.h (GENERIC_INTEGERLENGTH32): New macro.
 363         (integerlength64): Define using integerlength32 if integerlength32
 364         is not defined generically.
 365
 366 2005-11-26  Bruno Haible  <bruno@clisp.org>
 367
 368         * src/base/cl_low.h (mulu32) [SPARC64]: Remove rd instruction, since
 369         umul returns the complete 64-bit product in a register.
 370         (mulu32_w) [SPARC64]: Prefer umul over mulx instruction.
 371         (divu_6432_3232_w) [SPARC64]: Prefer umul/udiv over mulx/udivx
 372         instructions.
 373
 374 2005-11-26  Bruno Haible  <bruno@clisp.org>
 375
 376         * src/base/cl_low.h (divu_3216_1616): Prepend underscore to local
 377         variables.
 378
 379 2005-11-26  Bruno Haible  <bruno@clisp.org>
 380
 381         * src/base/cl_low.h (ord2_32): Parenthesize macro argument.
 382
 383 2005-12-17  Richard B. Kreckel  <kreckel@ginac.de>
 384
 385         * Created branch cln_1-1 for maintenance patches.
 386         This is the main branch, which will eventually become CLN 1.2.0.
 387
 388 2005-12-15  Dmitry V. Kustov  <kustov@telex221.ru>
 389
 390         * src/base/random/cl_random_from.cc: Add support for OpenBSD.
 391
 392 2005-11-23  Richard B. Kreckel  <kreckel@ginac.de>
 393
 394         * Version 1.1.11 released.
 395
 396 2005-11-20  Richard B. Kreckel  <kreckel@ginac.de>
 397
 398         * src/integer/conv/cl_I_cached_power.h: New file.
 399         * src/integer/conv/cl_I_cached_power.cc: New file.
 400         Contains power_table and cached_power_table previously...
 401         * src/integer/conv/cl_I_to_digits.cc: ...here.
 402         * src/integer/conv/cl_I_from_digits.cc: Use cached powers.
 403
 404 2005-11-02  Richard B. Kreckel  <kreckel@ginac.de>
 405
 406         * src/integer/conv/cl_I_from_digits.cc: Made input of all numbers in
 407         non-power-of-two base much faster.
 408         * tests/test_I_io.cc: New file...
 409         * tests/Makefile.in, tests/test_I.cc: ...used here.
 410
 411 2005-10-22  Richard B. Kreckel  <kreckel@ginac.de>
 412
 413         * Version 1.1.10 released.
 414
 415 2005-10-22  Richard B. Kreckel  <kreckel@ginac.de>
 416
 417         * src/Makefile.in: Accept CPPFLAGS from environment.
 418         * examples/Makefile.in: Likewise.
 419         * benchmarks/Makefile.in: Likewise.
 420         * tests/Makefile.in: Likewise.
 421
 422 2005-08-30  Richard B. Kreckel  <kreckel@ginac.de>
 423
 424         * include/cln/modules.h (CL_OUTPUT_LABEL): Work around redundant
 425         duplication of basic blocks on m68k.
 426
 427 2005-08-30  Richard B. Kreckel  <kreckel@ginac.de>
 428
 429         * include/cln/modules.h (CL_JUMP_TO): Fix mips* brokenness.
 430
 431 2005-08-27  Bruno Haible  <bruno@clisp.org>
 432
 433         Split aclocal.m4 into individual files.
 434         * m4/alloca.m4, m4/as-underscore.m4, m4/c++-constructors.m4:
 435         * m4/fpu_control.m4, m4/general.m4, m4/gettimeofday.m4:
 436         * m4/longdouble.m4, m4/longlong.m4, m4/param.m4, m4/perror.m4:
 437         * m4/proto.m4, m4/rusage.m4, m4/times.m4:
 438         New files, extracted from autoconf/aclocal.m4.
 439         * autoconf/aclocal.m4: m4_include them.
 440         * Makefile.devel (AUTOCONF_MACROS): New variable.
 441         (configure): Depend on it.
 442         (CLISP_M4DIR): Remove variable.
 443         (autoconf/aclocal.m4): Remove rule.
 444
 445 2005-08-27  Bruno Haible  <bruno@clisp.org>
 446
 447         * src/integer/bitwise/cl_I_ash_I.cc (ash): Avoid shifting a 32-bit
 448         zero value by more than 31 bits.
 449
 450 2005-08-27  Bruno Haible  <bruno@clisp.org>
 451
 452         Make the long-float overflow check work on 64-bit platforms.
 453         * src/float/lfloat/cl_LF.h (LF_exp_mid, LF_exp_high): Define as
 454         'unsigned int', not 'unsigned long'.
 455
 456 2005-08-27  Bruno Haible  <bruno@clisp.org>
 457
 458         * include/cln/modules.h (CL_OUTPUT_LABEL): Work around redundant
 459         duplication of basic blocks by g++ 4.0.
 460         See <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23345>.
 461
 462 2005-08-27  Bruno Haible  <bruno@clisp.org>
 463
 464         Make it possible to cross-compile CLN.
 465         * m4/intparam.m4: New file.
 466         * m4/floatparam.m4: New file.
 467         * autoconf/aclocal.m4: Include both.
 468         (CL_MACHINE): Add an additional CROSS_MACRO parameter.
 469         * configure.ac (CL_MACHINE): When cross-compiling, use
 470         CL_INTPARAM_CROSS and CL_FLOATPARAM_CROSS.
 471
 472 2005-08-27  Bruno Haible  <bruno@clisp.org>
 473
 474         Define HAVE_LONGLONG and HAVE_LONGDOUBLE when cross-compiling.
 475         * autoconf/aclocal.m4 (CL_LONGLONG, CL_LONGDOUBLE): When cross-
 476         compiling, use the test code from gnulib.
 477
 478 2005-08-27  Bruno Haible  <bruno@clisp.org>
 479
 480         * autoconf/aclocal.m4 (CL_RUSAGE): Fix error when cross-compiling.
 481
 482 2005-08-16  Richard B. Kreckel  <kreckel@ginac.de>
 483
 484         The patch of 2005-05-01 made it impossible to test the type of a cl_UP
 485         by comparing with &cl_class_univpoly_ring. We need an alternative:
 486         * include/cln/object.h (cl_class_flags_modint_ring): New #define...
 487         * src/polynomial/elem/cl_UP.cc (cl_class_univpoly_ring): ...used here.
 488         * src/polynomial/elem/cl_UP_GF2.h: Likewise.
 489         * src/polynomial/elem/cl_UP_MI.h: Likewise.
 490         * src/polynomial/elem/cl_UP_gen.h: Likewise.
 491         * src/polynomial/elem/cl_UP_number.h: Likewise.
 492
 493 2005-08-15  Richard B. Kreckel  <kreckel@ginac.de>
 494
 495         * m4/cc.m4 (CL_AS_NOEXECSTACK): New macro...
 496         * configure.ac: ...used here for setting ASMFLAGS...
 497         * src/Makefile.in: ...which are used here.
 498
 499 2005-08-02  Andreas Jochens  <aj@andaco.de>
 500
 501         * include/cln/config.h.in: Add support for PowerPC 64 CPU.
 502         * include/cln/modules.h: Likewise.
 503         * include/cln/object.h: Likewise.
 504         * include/cln/types.h: Likewise.
 505
 506 2005-07-24  Richard B. Kreckel  <kreckel@ginac.de>
 507
 508         Make out of the box build on x86_64 system with complete 32 bit
 509         userland possible
 510         * include/cln/config.h.in: Don't #define __x86_64__ when
 511         __i386__ is defined.
 512         * src/base/digitseq/cl_asm_x86_64_.cc: New file.
 513         * doc/cln.tex: Revert workaround description introduced 2005-05-02.
 514
 515 2005-06-10  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 516
 517         * Makefile.in: Don't enter nonexisting directories.
 518
 519 2005-05-17  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 520
 521         Speed up the linking step
 522         * src/Makefile.in: Use -objectlist for linking libcln.la.
 523
 524 2005-05-15  Richard B. Kreckel  <kreckel@ginac.de>
 525
 526         * autoconf/ltmain.sh: Upgrade to libtool-1.5.16.
 527         * m4/libtool.m4: Upgrade to libtool-1.5.16 (without Comeau patch).
 528
 529 2005-05-02  Richard B. Kreckel  <kreckel@ginac.de>
 530
 531         * doc/cln.tex: Document what to do on a x86_64 machine with 32-bit
 532         userland.
 533
 534 2005-05-01  Richard B. Kreckel  <kreckel@ginac.de>
 535
 536         Fix crashes in find_univpoly_ring and related functions
 537         * include/cln/modinteger.h: Remove vptr from cl_heap_modint_ring;
 538         remove declaration of cl_class cl_class_modint_ring.
 539         * include/cln/univpoly.h: Remove vptr from cl_heap_univpoly_ring;
 540         remove declaration of cl_class_univpoly_ring.
 541         * include/cln/object.h: cl_class_flags_modint_ring: New #define...
 542         * src/modinteger/cl_MI.cc: ...used in cl_class_modint_ring.
 543         * src/modinteger/cl_MI_fix16.h: No vptr, but static dtor and type flag.
 544         * src/modinteger/cl_MI_fix29.h: Likewise.
 545         * src/modinteger/cl_MI_fix32.h: Likewise.
 546         * src/modinteger/cl_MI_int32.h: Likewise.
 547         * src/modinteger/cl_MI_montgom.h: Likewise.
 548         * src/modinteger/cl_MI_pow2: Likewise.
 549         * src/modinteger/cl_MI_pow2m1.h: Likewise.
 550         * src/modinteger/cl_MI_pow2p1.h: Likewise.
 551         * src/modinteger/cl_MI_std.h: Likewise.
 552         * src/polynomial/elem/cl_UP.cc (cl_make_univpoly_ring): Compare with
 553         cl_class_flags_modint_ring, not with cl_class_modint_ring.
 554         * src/polynomial/elem/cl_UP_GF2.h (cl_class_num_univpoly_ring): New.
 555         * src/polynomial/elem/cl_UP_MI.h (cl_class_modint_univpoly_ring): New.
 556         * src/polynomial/elem/cl_UP_gen.h (cl_class_gen_univpoly_ring): New.
 557         * src/polynomial/elem/cl_UP_number.h (cl_class_num_univpoly_ring): New.
 558         Reported by Ralf Goertz <R_Goertz@web.de>.
 559
 560 2005-04-29  Richard B. Kreckel  <kreckel@ginac.de>
 561             Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 562
 563         * m4/cc.m4: Emit a warning if g++ is used and optimization turned off.
 564
 565 2005-04-24  Richard B. Kreckel  <kreckel@ginac.de>
 566
 567         Make GCC compiler flags default to -O
 568         * m4/cc.m4: New file...
 569         * configure.ac: ...used here.
 570         * autoconf/aclocal.m4: Regenerate.
 571
 572 2005-04-18  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 573
 574         * include/cln/string.h: Declare cl_string.
 575
 576 2005-03-17  Richard B. Kreckel  <kreckel@ginac.de>
 577
 578         * autoconf/ltmain.sh: Upgrade to libtool-1.5.14.
 579         * m4/libtool.m4: Upgrade to libtool-1.5.14 with Comeau patch.
 580         * autoconf/aclocal.m4: Regenerate.
 581
 582 2005-03-15  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 583
 584         * src/Makefile.in: Use $CXX instead of $CC when linking.
 585
 586 2005-02-27  Richard B. Kreckel  <kreckel@ginac.de>
 587
 588         * examples/perfnum.cc: update to presumed 42st Mersenne prime.
 589
 590 2004-11-28  Richard B. Kreckel  <kreckel@ginac.de>
 591
 592         Disambiguate binary operators of CLN types with float/double
 593         * include/cln/dfloat.h: Add binary operator overloads for arguments of
 594         type double.
 595         * include/cln/ffloat.h: Likewise, for arguments of type float.
 596         * include/cln/float.h: Likewise, both for arguments of types double and
 597         float.
 598         * include/cln/real.h: Likewise.
 599         Reported by Isidro Cachadiña Gutiérrez <icacha@unex.es>.
 600
 601 2004-11-03  Richard B. Kreckel  <kreckel@ginac.de>
 602
 603         * Version 1.1.9 released.
 604
 605 2004-10-28  Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
 606
 607         * src/Makefile.in: Let config.status set LDFLAGS.
 608
 609 2004-10-27  Peter Breitenlohner  <peb@mppmu.mpg.de>
 610
 611         * cln.pc.in: Fix typo.
 612         * Makefile.in (INSTALL_SCRIPT): Added, to be used for scripts like
 613         cln-config.  Allows us to do special things to binaries, like
 614         installing them with install -s.
 615         * doc/Makefile.in: add/remove the cln.info from the installed
 616         $(infodir)/dir unless this is debian install-info (code copied from
 617         what GNU automake would produce).
 618
 619 2004-10-26  Richard B. Kreckel  <kreckel@ginac.de>
 620
 621         * src/integer/input/cl_I_read_stream.cc (read_integer): Fix a bug
 622         that caused radix specifiers to not work when reading from a stream.
 623         * src/rational/input/cl_RA_read_stream.cc (read_rational): Likewise.
 624         * src/real/input/cl_R_read_stream.cc (read_real): Likewise.
 625         * src/float/input/cl_F_read_stream.cc (read_float): Likewise.
 626         * src/complex/input/cl_N_read_stream.cc (read_complex): Likewise.
 627
 628 2004-10-25  Richard B. Kreckel  <kreckel@ginac.de>
 629
 630         * src/base/cl_low.h: Add mulu64 assembler macro for ia64.
 631
 632 2004-10-24  Richard B. Kreckel  <kreckel@ginac.de>
 633
 634         * src/base/cl_low.h: Add mul and div macros for x86_64.
 635
 636 2004-10-23  Richard B. Kreckel  <kreckel@ginac.de>
 637
 638         * src/integer/conv/cl_I_from_digits.cc (digits_to_I): Fix thinko in
 639         new code for base power of two.
 640
 641 2004-10-22  Richard B. Kreckel  <kreckel@ginac.de>
 642
 643         * src/integer/conv/cl_I_to_digits (I_to_digits): Fix an elusive stack
 644         overwriting problem. That was the real cause for Debian bug#246319.
 645         * src/integer/output/cl_I_print.cc (print_integer): Revert workaround
 646         for the bug fixed above.
 647
 648 2004-10-20  Richard B. Kreckel  <kreckel@ginac.de>
 649
 650         * include/cln/types.h: Use 64 bit digits on x86_64 CPU.
 651
 652 2004-10-12  Richard B. Kreckel  <kreckel@ginac.de>
 653
 654         * src/integer/conv/cl_I_from_digits.cc (digits_to_I): Speedup when
 655         the base is a power of two.
 656
 657 2004-10-05  Richard B. Kreckel  <kreckel@ginac.de>
 658
 659         * src/integer/conv/cl_I_to_digits.cc (I_to_digits): Fix bug in base 32.
 660
 661 2004-09-27  Richard B. Kreckel  <kreckel@ginac.de>
 662
 663         Support for little-endian Mips, second shot
 664         * src/base/digitseq/cl_asm_mipsel_.cc: New file...
 665         * src/base/digitseq/cl_asm_cc: ...used here.
 666         * src/base/digitseq/cl_asm.h: Include cl_asm_mips.h for any endianness.
 667         * include/cln/object.h: Set alignment for mipsel explicitly.
 668
 669 2004-09-05  Richard B. Kreckel  <kreckel@ginac.de>
 670
 671         Support for little-endian Mips
 672         * include/cln/config.h.in: Add __mipsel__.
 673         * include/cln/modules.h: For Mips, this is endianness-agnostic.
 674         * src/base/digitseq/cl_asm_.cc, src/base/digitseq/cl_asm.h:
 675         Mask out assembler for little-endian Mips.
 676
 677 2004-08-30  Bruno Haible  <bruno@clisp.org>
 678
 679         * benchmarks/timebench2.sh: Multiply all repeat counts by 100.
 680         * benchmarks/timebench2.results: Add recent PowerPC G4 results.
 681
 682 2004-08-26  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 683
 684         * examples/e.cc: remove extra semicolon.
 685
 686 2004-08-25  Bruno Haible  <bruno@clisp.org>
 687
 688         * autoconf/ltmain.sh: Upgrade to libtool-1.5.6.
 689         * m4/libtool.m4: New file, from libtool-1.5.6 with modifications:
 690         2004-08-22  Bruno Haible  <bruno@clisp.org>
 691           * m4/libtool.m4: Add support for Comeau C++ on Linux.
 692           Reported by Prof. Roberto Bagnara <bagnara@cs.unipr.it>.
 693         * autoconf/aclocal.m4: Regenerate.
 694
 695 2004-08-19  Bruno Haible  <bruno@clisp.org>
 696
 697         * include/cln/modules.h (CL_GLOBALIZE_JUMP_LABEL, CL_JUMP_TO): When
 698         converting a label to a string, use ASM_UNDERSCORE_PREFIX. Needed on
 699         MacOS X.
 700         Reported by Darren Bane <darren.bane@ul.ie>.
 701
 702 2004-07-01  Richard B. Kreckel  <kreckel@ginac.de>
 703
 704         * Version 1.1.8 released.
 705
 706 2004-06-30  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 707
 708         * src/base/cl_macros.h: alloca(3) has size_t argument type.
 709
 710 2004-06-30  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 711
 712         * include/cln/floatformat.h: Do define a type here.
 713
 714 2004-06-27  Richard B. Kreckel  <kreckel@ginac.de>
 715
 716         * include/cln/modules.h (CL_JUMP_TO): Fix AMD64 brokenness.
 717
 718 2004-06-23  Bruno Haible  <bruno@clisp.org>
 719
 720         * configure.ac: Pretend ftime() is not available. Needed by
 721         CL_TIMES_CLOCK.
 722
 723 2004-06-21  Ralf Stephan  <ralf@ark.in-berlin.de>
 724
 725         * doc/cln.tex: Document jacobi, isprobprime and nextprobprime.
 726
 727 2004-06-18  Richard B. Kreckel  <kreckel@ginac.de>
 728
 729         * rational/transcendental/cl_RA_logp.cc: fix bug where base is
 730         reciprocal of an integer.
 731         Reported by Niklas Knutsson <nq@altern.org>.
 732
 733 2004-06-18  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 734
 735         * src/complex/ring/cl_C_ring.cc, src/integer/ring/cl_I_ring.cc,
 736         src/rational/ring/cl_RA_ring.cc, src/real/ring/cl_R_ring.cc:
 737         Make template specializations explicit.
 738
 739 2004-06-17  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 740
 741         * src/complex/input/cl_N_read.cc, src/float/input/cl_F_read.cc,
 742         src/integer/input/cl_I_read.cc, src/rational/input/cl_RA_read.cc,
 743         src/real/input/cl_R_read.cc:  Remove unused labels.
 744
 745 2004-06-17  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 746
 747         * benchmarks/timebench1.cc, benchmarks/timebench2a.LiDIA.cc,
 748         benchmarks/timebench2a.cc, benchmarks/timebench2ap.cc,
 749         benchmarks/timebench2b.LiDIA.cc, benchmarks/timebench2b.cc,
 750         examples/atan_recip.cc, examples/atanh_recip.cc,
 751         examples/contfrac.cc, examples/e.cc, examples/legendre.cc,
 752         examples/lucaslehmer.cc, examples/pi.cc, include/cln/GV.h,
 753         include/cln/SV.h, include/cln/malloc.h, include/cln/modules.h,
 754         include/cln/object.h, include/cln/string.h, src/base/cl_abort.cc,
 755         src/base/cl_alloca.h, src/base/cl_malloc.cc,
 756         src/base/random/cl_random_from.cc,
 757         src/base/string/cl_spushstring_append.cc,
 758         src/base/string/cl_spushstring_push.cc,
 759         src/base/string/cl_st_debug.cc,
 760         src/base/string/input/cl_st_gettoken.cc,
 761         src/complex/input/cl_N_read.cc, src/float/input/cl_F_read.cc,
 762         src/float/output/cl_F_dprint.cc, src/integer/input/cl_I_read.cc,
 763         src/rational/input/cl_RA_read.cc,
 764         src/real/format-output/cl_fmt_integer.cc,
 765         src/real/format-output/cl_fmt_paddedstring.cc,
 766         src/real/input/cl_R_read.cc, src/timing/cl_t_current.cc,
 767         src/timing/cl_t_current2.cc, tests/exam.cc, tests/tests.cc,
 768         tests/timeLFRAmul.cc, tests/timeLFatan-compare.cc,
 769         tests/timeLFatan.cc, tests/timeLFatanh-compare.cc,
 770         tests/timeLFatanh.cc, tests/timeLFcos-compare.cc, tests/timeLFcos.cc,
 771         tests/timeLFcosh.cc, tests/timeLFexp-compare.cc, tests/timeLFexp.cc,
 772         tests/timeLFln-compare.cc, tests/timeLFln.cc,
 773         tests/timeLFsin-compare.cc, tests/timeLFsin.cc, tests/timeLFsinh.cc,
 774         tests/timeLFsqrt.cc, tests/timeMImisc5.cc, tests/timeMIpow2div.cc,
 775         tests/timeMIpow2recip.cc, tests/timeRALFdiv.cc, tests/timeRAtoLF.cc,
 776         tests/timeUPMImul.cc, tests/timecatalan.cc, tests/timediv.cc,
 777         tests/timediv2adic-compare.cc, tests/timediv2adic.cc,
 778         tests/timeeuler.cc, tests/timeexp1.cc, tests/timefact.cc,
 779         tests/timegcd.cc, tests/timemul-compare.cc, tests/timemul.cc,
 780         tests/timepi.cc, tests/timeprint-compare.cc, tests/timeprint.cc,
 781         tests/timerecip2adic-compare.cc, tests/timerecip2adic.cc,
 782         tests/timesqrt.cc, tests/timesqrtmodp.cc, tests/timesquare.cc,
 783         tests/timezeta3.cc: Change all C include headers to ISO style
 784         within C++ code.
 785
 786 2004-06-10  Richard B. Kreckel  <kreckel@ginac.de>
 787
 788         * examples/perfnum.cc: update to presumed 41st Mersenne prime.
 789
 790 2004-05-02  Richard B. Kreckel  <kreckel@ginac.de>
 791
 792         * Version 1.1.7 released.
 793
 794 2004-05-02  Richard B. Kreckel  <kreckel@ginac.de>
 795
 796         * examples/pi.cc and examples/pi.1: New files.
 797         * examples/Makefile.in: Build the pi executable.
 798
 799 2004-05-01  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 800
 801         * src/Makefile.in: Fix for parallel build: wait for subdir objects to
 802         be finished before creating the library.
 803
 804 2004-04-30  Richard B. Kreckel  <kreckel@ginac.de>
 805
 806         * src/integer/output/cl_I_print.cc (print_integer): workaround
 807         GCC compiler bug (cf. Debian bug#246319).
 808
 809 2004-03-20  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
 810
 811         * m4/gmp.m4 (CL_GMP_CHECK): Do not lose LIBS setting with config.cache
 812         enabled.
 813
 814 2004-03-08  Bruno Haible  <bruno@clisp.org>
 815
 816         * src/float/lfloat/elem/cl_LF_mul.cc (operator*): Fix the second
 817         underflow condition.
 818         * src/float/lfloat/algebraic/cl_LF_sqrt.cc (sqrt): Fix a bug with large
 819         uexp whereby SQRT of MOST-POSITIVE-LONG-FLOAT was less than 1.
 820
 821 2004-03-04  Richard B. Kreckel  <kreckel@ginac.de>
 822
 823         * Makefile.in (install): Add ${srcdir} for cln.m4.
 824         * m4/gmp.m4: quote macro names.
 825         Reported by Ralf Wildenhues <Ralf.Wildenhues@gmx.de>.
 826
 827 2004-01-01  Richard B. Kreckel  <kreckel@ginac.de>
 828
 829         * Version 1.1.6 released.
 830
 831 2004-01-01  Richard B. Kreckel  <kreckel@ginac.de>
 832
 833         * include/cln/univpoly.h, include/cln/univpoly_complex.h,
 834           include/cln//univpoly_integer.h, include/cln/univpoly_modint.h,
 835           include/cln/univpoly_rational.h, include/cln/univpoly_real.h,
 836           src/polynomial/elem/cl_UP_GF2.h, src/polynomial/elem/cl_UP_MI.h,
 837           src/polynomial/elem/cl_UP_gen.h, src/polynomial/elem/cl_UP_no_ring.cc,
 838           src/polynomial/elem/cl_UP_number.h (ldegree): New function.
 839         * doc/cln.tex: Document `ldegree'.
 840
 841 2003-12-29  Richard B. Kreckel  <kreckel@ginac.de>
 842
 843         Rework of autoconfiscation infrastructure
 844         * autoconf/config.{guess,sub}: Update to GNU version 2003-10-07.
 845         * autoconf/ltmain.sh: Update to GNU version 1.4.3.
 846         * autoconf/autoconf: Remove (from now on we assume autoconf is
 847           installed properly on the sytem).
 848         * autoconf/autoconf.m4: Likewise.
 849         * autoconf/autoconf.m4f: Likewise.
 850         * autoconf/acgeneral.m4: Likewise.
 851         * autoconf/acspecific.m4: Likewise.
 852         * autoconf/aclocal.m4: Regenerate.
 853         * autoconf/acinclude.m4: Remove (while moving the macros...)
 854         * m4/gmp.m4: New file (...to here).
 855         * Makefile.devel: Update to new scheme.
 856         * configure.ac: Likewise.
 857         * include/cln/GV_integer.h: Assume template specializations work.
 858         * include/cln/GV_modinteger.h: Likewise.
 859         * include/cln/config.h.in: Likewise, and drop HAVE_BOOL.
 860         * src/base/cl_base_config.h.in: Drop support for obsolete ftime(3).
 861         * src/base/random/cl_random_from.cc: Likewise.
 862         * src/timing/cl_base_config.h.in: Likewise.
 863         * src/timing/cl_t_current.cc: Likewise.
 864
 865 2003-12-27  Richard B. Kreckel  <kreckel@ginac.de>
 866
 867         * src/polynomial/cl_UP_gen.h (gen_minus): Fix case where first
 868           argument is zero.
 869         * src/polynomial/cl_UP_MI.h (modint_minus): Likewise.
 870         * src/polynomial/cl_UP_number.h (num_minus): Likewise.
 871         Reported by Munagala Ramanath <amberarrow@yahoo.com>.
 872
 873 2003-12-02  Richard B. Kreckel  <kreckel@ginac.de>
 874
 875         * examples/perfnum.cc: update to presumed 40th Mersenne prime.
 876
 877 2003-11-20  Christian Bauer  <cbauer@ginac.de>
 878
 879         Added pkg-config support
 880         * cln.pc.in: New file.
 881         * Makefile.in: Take care of cln.pc.
 882         * configure.ac: Likewise.
 883
 884 2003-08-06  Richard B. Kreckel  <kreckel@ginac.de>
 885
 886         * src/numtheory/cl_nt_sqrtmodp.cc: #undef _R.
 887         Reported by Andrew Rechnitzer <A.Rechnitzer@ms.unimelb.edu.au>.
 888
 889 2003-08-01  Richard Kreckel  <kreckel@ginac.de>
 890
 891         More dependent base resolution issues
 892         * src/base/hash/cl_hash.h (cl_heap_hashtable<T>::iterator()):
 893         portable syntactic simplification.
 894         * src/base/hash/cl_hashset.h: Preceed inherited members with this->.
 895         * src/base/hash/cl_hash1.h: Likewise for member functions.
 896         * src/base/hash/cl_hash2.h: Likewise.
 897         * src/base/hash/cl_hashuniq.h: Likewise.
 898         * src/base/hash/cl_hashuniqweak.h: Likewise.
 899         * src/base/hash/cl_hash.h: Revert explicit static member function
 900         lookup since that was GCC's fault.
 901         * src/base/hash/cl_hash2weak.h: Likewise.
 902         * src/base/hash/cl_hashuniqweak.h: Likewise.
 903
 904 2003-06-29  Richard Kreckel  <kreckel@ginac.de>
 905
 906         Dependent base resolution needed for GCC-3.4
 907         * include/cln/GV.h: Preceed inherited members with this->.
 908         * include/cln/SV.h: Likewise.
 909         * include/cln/object.h: Likewise.
 910         * src/base/hash/cl_hash1.h: Likewise.
 911         * src/base/hash/cl_hash1weak.h: Likewise.
 912         * src/base/hash/cl_hash2.h: Likewise.
 913         * src/base/hash/cl_hashuniq.h: Likewise.
 914         * src/base/hash/cl_hash.h: Make lookup of static member function
 915         explicit.
 916         * src/base/hash/cl_hash2weak.h: Likewise.
 917         * src/base/hash/cl_hashuniqweak.h: Likewise.
 918         * src/modinteger/cl_MI.cc: Make shell-comment a C-comment. Geez.
 919
 920 2003-02-24  Bruno Haible  <bruno@clisp.org>
 921
 922         * src/base/random/cl_random_from.cc (random_state::random_state): Add
 923         support for MacOS X.
 924         * src/modinteger/cl_MI.cc: #undef _R.
 925         Reported by Erann Gat <gat@jpl.nasa.gov>.
 926
 927 2002-08-03  Joerg Arndt  <jj@suse.de>
 928
 929         * include/cln/config.h.in: Add support for x86_64 CPU.
 930         * include/cln/modules.h: Likewise.
 931         * include/cln/types.h: Likewise.
 932         * include/cln/object.h: Likewise.
 933
 934 2002-06-08  Richard Kreckel  <kreckel@ginac.de>
 935
 936         * src/base/digitseq/cl_asm.h: ensure intCsize==16 when including
 937         m68k Assembler routines.
 938         * src/base/digitseq/cl_asm_.cc: Likewise.
 939
 940 2002-05-28  Richard Kreckel  <kreckel@ginac.de>
 941
 942         * Version 1.1.5 released.
 943
 944 2002-05-27  Richard Kreckel  <kreckel@ginac.de>
 945
 946         * include/cln/modules.h (CL_CONCATENATE): New macro.
 947
 948 2002-05-10  Richard Kreckel  <kreckel@ginac.de>
 949
 950         * doc/cln.tex (Building the library): Update recommendations for
 951         compiling on Tru64 using g++ 3.0 and 3.1.
 952         * README: Update homepage.
 953
 954 2002-05-05  Bruno Haible  <bruno@clisp.org>
 955
 956         * doc/cln.tex (Building the library): Give some recommendations for
 957         g++ 3.0 and 3.1.
 958
 959         Force link-time references despite optimizations done by g++ 2.95
 960         and newer.
 961         * include/cln/modules.h (CL_FORCE_LINK): New macro.
 962         * Use CL_FORCE_LINK.
 963         * include/cln/GV_integer.h (cl_GV_I_debug_dummy): Likewise.
 964         * include/cln/GV_number.h (cl_GV_number_debug_dummy): Likewise.
 965         * include/cln/SV_number.h (cl_SV_number_debug_dummy): Likewise.
 966         * include/cln/SV_ringelt.h (cl_SV_ringelt_debug_dummy): Likewise.
 967         * include/cln/dfloat.h (cl_DF_debug_dummy): Likewise.
 968         * include/cln/ffloat.h (cl_FF_classes_dummy, cl_FF_debug_dummy):
 969         Likewise.
 970         * include/cln/integer.h (cl_I_classes_dummy, cl_I_debug_dummy):
 971         Likewise.
 972         * include/cln/lfloat.h (cl_LF_debug_dummy): Likewise.
 973         * include/cln/modinteger.h (cl_MI_debug_dummy): Likewise.
 974         * include/cln/rational.h (cl_RA_debug_dummy): Likewise.
 975         * include/cln/real.h (cl_R_debug_dummy): Likewise.
 976         * include/cln/ring.h (cl_ring_debug_dummy): Likewise.
 977         * include/cln/sfloat.h (cl_SF_classes_dummy, cl_SF_classes_dummy):
 978         Likewise.
 979         * include/cln/string.h (cl_string_debug_dummy): Likewise.
 980         * include/cln/univpoly.h (cl_UP_debug_dummy): Likewise.
 981         * src/float/base/cl_ieee.h (cl_ieee_dummy_NNN): Likewise.
 982
 983         Avoid g++ 3.1 warnings.
 984         * src/base/cl_offsetof.h (offsetof): Redefine each time.
 985         * src/base/digitseq/cl_DS.h: Include "cl_offsetof.h" after <gmp.h>.
 986
 987         Avoid g++ 3.1 warnings.
 988         * src/base/hash/cl_hash.h (struct cl_heap_hashtable): Use typename
 989         where needed.
 990         * src/base/hash/cl_hash1.h (struct cl_heap_hashtable_1): typedef
 991         htxentry as a shortcut.
 992         * src/base/hash/cl_hash2.h (struct cl_heap_hashtable_2): Likewise.
 993         * src/base/hash/cl_hashset.h (struct cl_heap_hashtable_set): Likewise.
 994         * src/base/hash/cl_hashuniq.h (struct cl_heap_hashtable_uniq):
 995         Likewise.
 996
 997 2002-03-15  Bruno Haible  <bruno@clisp.org>
 998
 999         * cln.tex: Document problem with GNU make 3.77.
1000         Reported by Michael Somos <somos@grail.cba.csuohio.edu>.
1001
1002 2002-02-16  Richard Kreckel  <kreckel@ginac.de>
1003
1004         * cln.m4: quote macro name.
1005           Pointed out by Roberto Bagnara.
1006
1007 2002-01-20  Richard Kreckel  <kreckel@ginac.de>
1008
1009         * autoconf/config.{guess,sub}: Update to GNU version 2002-01-02.
1010           (the old one was broken on Linux/Mips.)
1011
1012 2002-01-04  Richard Kreckel  <kreckel@ginac.de>
1013
1014         * autoconf/autoconf.m4f: get brutal in order to adhere to FHS.
1015         * Version 1.1.4 released.
1016
1017 2002-01-03  Richard Kreckel  <kreckel@ginac.de>
1018
1019         * autoconf/acinclude.m4: revamp MPN-matcher.
1020         * autoconf/aclocal.m4: upgrade to autoconf-2.52 infrastructure,
1021           sync with CLisp from CVS.
1022         * autoconf/autoconf: Likewise.
1023         * autoconf/autoconf.m4f: Likewise (new file).
1024         * configure.ac: Likewise (new file, replaces configure.in).
1025         * configure.in: Likewise (deleted, replaced by configure.ac).
1026         * autoconf/config.{guess,sub}: Update to GNU version 2001-12-13.
1027         * src/Makefile.in: made VPATH safe for autoconf-2.52.
1028         * include/cln/config.h.in: Add __s390__.
1029
1030 2001-12-31  Richard Kreckel  <kreckel@ginac.de>
1031
1032         * src/base/digitseq/cl_DS.h: <gmp.h> is not included extern "C"
1033           any more since GMP4 has some C++ support in it.
1034
1035 2001-12-14  Richard Kreckel  <kreckel@ginac.de>
1036
1037         * include/cln/modules.h, include/cln/object.h: add support for
1038           s390.
1039         * src/numtheory/cl_nt_sqrtmodp.cc: workaround for GCC2.x compiler-bug
1040           on s390, provided by Gerhard Tonn.
1041
1042 2001-11-05  Richard Kreckel  <kreckel@ginac.de>
1043
1044         * autoconf/ltmain.sh: Upgrade to libtool-1.4.2.
1045         * autoconf/config.{guess,sub}: Update to GNU version 2001-09-07.
1046         * Version 1.1.3 released.
1047
1048 2001-11-04  Bruno Haible  <haible@clisp.cons.org>
1049
1050         Interoperability with gcc-3.0 -fuse-cxa-atexit.
1051         * autoconf/aclocal.m4 (CL_GLOBAL_CONSTRUCTORS): Add test whether
1052           global destructors actually exist.
1053         * include/cln/modules.h (CL_PROVIDE, CL_PROVIDE_END, CL_PROVIDE_END):
1054           Don't hack the global destructors if there is no global destructors
1055           function.
1056
1057 2001-11-03  Richard Kreckel  <kreckel@ginac.de>
1058
1059         * src/float/transcendental/cl_F_sinx.cc (sinx_naive): For small
1060           values of x, return square(x) instead of x.
1061         * src/float/transcendental/cl_F_sinhx.cc (sinhx_naive): Likewise.
1062
1063 2001-07-25  Richard Kreckel  <kreckel@ginac.de>
1064
1065         * Version 1.1.2 released.
1066
1067 2001-07-24  Richard Kreckel  <kreckel@ginac.de>
1068
1069         * src/base/hash/cl_hash.h: declare _cl_hashtable_iterator<htentry> a
1070           friend of cl_heap_hashtable<htentry>.
1071
1072 2001-07-22  Richard Kreckel  <kreckel@ginac.de>
1073
1074         * src/float/base/cl_ieee.cc: try to do magic to the FPU only if
1075           _FPU_IEEE is really defined.
1076         * include/cln/modules.h: change assembler labels from `label' to
1077           `label:' on hppa, needed by Linux (see comment).
1078         * autoconf/acinclude.m4: new file (for storing CLN-specific macros).
1079         * Makefile.devel: adjusted.
1080         * autoconf/aclocal.m4: regenerate.
1081         * src/base/low/cl_low_mul.cc: moved POD variables that are declared
1082           extern "C" elsewhere out of the namespace.
1083         * src/base/low/cl_low_div.cc: Likewise.
1084
1085 2001-06-08  Bruno Haible  <haible@clisp.cons.org>
1086
1087         * autoconf/config.{guess,sub}: Update to GNU version 2001-05-11.
1088         * autoconf/aclocal.m4: Upgrade to libtool-1.4.
1089         * autoconf/ltmain.sh: Likewise.
1090         * autoconf/ltconfig: Remove file.
1091         * autoconf/install-sh: New file.
1092         * configure.in: Add AC_CONFIG_AUX_DIR call.
1093
1094 2001-06-05  Richard Kreckel  <kreckel@ginac.de>
1095
1096         * tests/tests.cc: resolve namespace ambiguity about strcmp().
1097
1098 2001-05-31  Richard Kreckel  <kreckel@ginac.de>
1099
1100         * Version 1.1.1 released.
1101
1102 2001-05-28  Richard Kreckel  <kreckel@ginac.de>
1103
1104         * cln/cln.tex: documented problems with shared library on Sparc
1105           using gcc older than 2.95.3.
1106         * configure.in: Fixed typos in versioning docu.
1107
1108 2001-05-25  Bruno Haible  <haible@clisp.cons.org>
1109
1110         * src/base/digitseq/cl_asm_arm_.cc: Use #0x instead of #& to designate
1111           hexadecimal constants.
1112
1113 2001-05-25  Richard Kreckel  <kreckel@ginac.de>
1114
1115         * autoconf/floatparam.c (double_wordorder_bigendian_p): new symbol.
1116         * src/float/dfloat/cl_DF.h: Check for double_wordorder_bigendian_p.
1117         * Removed LiDIA interface since that is now outdated (namespace cln)
1118           and maintained elsewhere.
1119         * Adjusted dates and final touches for 1.1.1.
1120
1121 2001-05-19  Richard Kreckel  <kreckel@ginac.de>
1122
1123         * INSTALL: Update toolchain info: no egcs, some more platforms.
1124         * doc/cln.tex: Likewise.
1125
1126 2001-05-18  Richard Kreckel  <kreckel@ginac.de>
1127
1128         * src/base/cl_low.h: prepended variables declared inside macros
1129           with underscore.  Fixes equal_hashcode() on various platforms.
1130
1131 2001-04-25  Richard Kreckel  <kreckel@ginac.de>
1132
1133         * src/base/cl_low.h: Added several checks if NO_ASM is defined, so
1134           this definition becomes actually useful.  This is needed for
1135           compilation on Arm until somebody fixes the assembler files for Arm.
1136         * src/base/digitseq/cl_asm.h: Likewise.
1137         * src/base/digitseq/cl_asm_.cc: Likewise.
1138         * */Makefile.in: Added `override' in front of `CPPFLAGS +=' so
1139            one can say `make CPPFLAGS=-DFOOBAR'.
1140
1141 2001-03-26  Arvid Norberg <c99ang@cs.umu.se>
1142
1143         * src/base/random/cl_random_from.cc: ported to beos.
1144
1145 2001-03-05  Richard Kreckel  <kreckel@ginac.de>
1146
1147         * include/cln/modules.h (CL_JUMP_TO): Fix IA64 brokenness.
1148
1149 2001-01-28  Richard Kreckel  <kreckel@ginac.de>
1150
1151         * include/cln/number.h (cl_as_N): Remove bogus comment.
1152
1153 2001-01-22  Richard Kreckel  <kreckel@ginac.de>
1154
1155         * configure.in: Make build in separate builddir possible (again).
1156
1157 2001-01-22  Richard Kreckel  <kreckel@ginac.de>
1158
1159         * include/cln/*.h: Change signatures of all classes' methods
1160           cln::cl_foo::operator new(size_t, cl_foo*) to
1161           cln::cl_foo::operator new(size_t, void*) so one can declare
1162           std::vector<cln::cl_foo>, std::list<cln::cl_foo> etc. for
1163           certain STL implementations (like libstdc++-v3).
1164
1165 2000-12-14  Richard Kreckel  <kreckel@ginac.de>
1166
1167         * Version 1.1 released.
1168
1169 2000-12-13  Richard Kreckel  <kreckel@ginac.de>
1170
1171         * */*: cl_istream -> std::istream, cl_ostream -> std::ostream.
1172
1173 2000-12-05  Richard Kreckel  <kreckel@ginac.de>
1174
1175         * Makefile.in, src/Makefile.in, doc/Makefile.in: Use mkdir -p.
1176         * include/cln/version.h.in, src/base/verion.cc: New files.
1177         * configure.in: Generate include/cln/version.h.
1178         * cln.m4: Rewrote it. Check result of cln-config without compiling.
1179           Do cross-check library version and header version information.
1180
1181 2000-12-02  Christian Bauer  <cbauer@ginac.de>
1182
1183         * Makefile.in, src/Makefile.in, doc/Makefile.in: Added $DESTDIR.
1184         * cln.m4, cln.spec.in: some minor fixes with respect to RPM package
1185           building.
1186
1187 2000-11-24  Richard Kreckel  <kreckel@ginac.de>
1188
1189         * */*: Removed problematic stdin, stdout and stderr definitions.
1190           Use std::cin, std::cout, std::cerr instead (obsoletes 2000-10-29).
1191
1192 2000-11-20  Bruno Haible
1193
1194         * cln-config.1: change title.
1195
1196 2000-11-18  Richard Kreckel  <kreckel@ginac.de>
1197
1198         * cln.m4: New file.
1199         * doc/cln.tex: Document package tools cln-config and cln.m4.
1200         * Makefile.in: Care about cln.m4.
1201
1202 2000-11-17  Richard Kreckel  <kreckel@ginac.de>
1203
1204         * cln-config.1: added manpage, as required by a couple of distros.
1205         * Makefile.in, doc/Makefile.in: target install depends on installdirs.
1206
1207 2000-11-16  Richard Kreckel  <kreckel@ginac.de>
1208
1209         * autoconf/aclocal.m4 (CL_GMP_SET_UINTD): New macro...
1210         * configure.in: ...used here.
1211         * include/cln/config.h.in: Put in macros defined by CL_GMP_SET_UINTD...
1212         * include/cln/types.h: ...used here.
1213         * autoconf/acgeneral.m4, autoconf/aclocal.m4: Adhere to FHS.
1214
1215 2000-11-13  Richard Kreckel  <kreckel@ginac.de>
1216
1217         * src/base/digitseq/cl_asm.h: Test if (intDsize==32) for MIPS and HPPA,
1218           in order to guard against an accidented configuration.
1219         * src/integer/conv/cl_I_to_digits.cc (table): member b_hoch_k of
1220           struct power_table_entry initialized as ULL instead of as UL, if
1221           intDsize==64 (caused misprinting on MIPS w/ GMP).
1222         * src/base/cl_macros.h (minus_bitm, bitc): implemented ULL, if
1223           HAVE_FAST_LONGLONG.
1224         * src/integer/bitwise/cl_I_mkfx.cc (mkf_extract): Likewise.
1225         * src/integer/conv/cl_I_from_L.cc (cl_I_from_L): Added trival
1226           generation of Bignum for intDsize==64 and a notreached-check at end.
1227         * autoconf/config.guess, autoconf/config.sub: updated from FSF.
1228         * include/cln/config.h.in: Prepared support for IA64.
1229         * include/cln/types.h: Likewise.
1230         * include/cln/object.h: Likewise.
1231         * include/cln/modules.h: Likewise.
1232         * src/base/cl_macros.h (nonreturning_function): Likewise (NUE's
1233           compiler claims __GNUC_MINOR__==9).
1234
1235 2000-11-03  Richard Kreckel  <kreckel@ginac.de>
1236
1237         * src/base/cl_macros.h (bit, minus_bit): changed criterion for ULL from
1238           HAVE_DD to HAVE_FAST_LONGLONG.
1239         * src/base/cl_macros.h (bitm): implemented ULL, if HAVE_FAST_LONGLONG.
1240         * src/base/cl_low.h: actually no need to include "cln/types.h" here.
1241         * src/base/cl_low.h (logcount_64): always ULL, independent of HAVE_DD.
1242         * src/base/random/cl_UL_random.cc (random32): a is always ULL.
1243
1244 2000-11-01  Richard Kreckel  <kreckel@ginac.de>
1245
1246         * include/cln/object.h (cl_combine): define additional signatures, if
1247           HAVE_LONGLONG is defined, in order to keep the compiler happy.
1248         * src/base/cl_macros.h: include "cln/types.h", since we need HAVE_DD...
1249         * src/base/cl_macros.h (bit): ...for this macro...
1250         * src/base/cl_macros.h (minus_bit): ...and this one.
1251         * src/base/cl_low.h: include "cln/types.h", since we need HAVE_DD...
1252         * src/base/cl_low.h (logcount_64): ...for this macro.
1253         * src/base/random/cl_UL_random.cc (random32): if HAVE_DD a is an ULL.
1254         * src/integer/gcd/cl_I_gcd_aux2.cc (floorDD): fixed algorithmic bug.
1255           that turned up when intDsize==32 and cl_word_size==64.
1256         * src/float/dfloat/elem/cl_DF_div.cc (operator/): fixed a missing cast
1257           to uint64 that turned up when intDsize==32 and cl_word_size==64.
1258
1259 2000-10-29  Richard Kreckel  <kreckel@ginac.de>
1260
1261         * src/real/input/cl_R_read.cc, src/complex/input/cl_N_read.cc:
1262           #undef stderr, if it's defined so cln::stderr isn't confused.
1263         * src/base/input/cl_read_globals.cc: stdin should not be extern.
1264
1265 2000-09-05  Richard Kreckel  <kreckel@ginac.de>
1266
1267         * include/cln/number.h (As): Fix it in namespace by suffixing `_As'
1268           to the appropiate method instead of prefixing `as_'.
1269         * src/complex/misc/cl_N_as.cc (cl_N_As): Likewise.
1270         * src/real/misc/cl_R_as.cc (cl_R_As): Likewise.
1271         * src/rational/misc/cl_RA_as.cc (cl_RA_As): Likewise.
1272         * src/integer/misc/cl_I_as.cc (cl_I_As): Likewise.
1273         * src/float/misc/cl_F_as.cc (cl_F_As): Likewise.
1274         * src/float/sfloat/misc/cl_SF_as.cc (cl_SF_As): Likewise.
1275         * src/float/lfloat/misc/cl_LF_as.cc (cl_LF_As): Likewise.
1276         * src/float/ffloat/misc/cl_FF_as.cc (cl_FF_As): Likewise.
1277         * src/float/dfloat/misc/cl_DF_as.cc (cl_DF_As): Likewise.
1278
1279 2000-09-05  Richard Kreckel  <kreckel@ginac.de>
1280
1281         * src/complex/transcendental/cl_C_expt_C.cc (expt): fix logic for
1282           the 0^y cases.
1283
1284 2000-08-30  Richard Kreckel  <kreckel@ginac.de>
1285
1286         * include/cln/number.h, cl_number::_as_cl_private_thing(): removed.
1287           Rearranged for a clearer distinction between macros and remaining
1288           identifiers, so Cint can parse it smoothly.
1289
1290 2000-08-29  Richard Kreckel  <kreckel@ginac.de>
1291
1292         * include/cln/number.h, the(const cl_number& x): New template
1293           function.
1294
1295 2000-08-29  Richard Kreckel  <kreckel@ginac.de>
1296
1297         * */*: Pushed CLN into a namespace `cln'.  While doing so, the
1298           following identifiers got their poor-man's namespace (i.e. the
1299           prefix `cl_') stripped off:
1300           cl_catalanconst()        ->  catalanconst()
1301           cl_compare()             ->  compare()
1302           cl_cos_sin()             ->  cos_sin()
1303           cl_cos_sin_t             ->  cos_sin_t
1304           cl_cosh_sinh()           ->  cosh_sinh()
1305           cl_cosh_sinh_t           ->  cosh_sinh_t
1306           cl_decoded_dfloat        ->  decoded_dfloat
1307           cl_decoded_ffloat        ->  decoded_ffloat
1308           cl_decoded_float         ->  decoded_float
1309           cl_decoded_lfloat        ->  decoded_lfloat
1310           cl_decoded_sfloat        ->  decoded_sfloat
1311           cl_default_float_format  ->  default_float_format
1312           cl_default_print_flags   ->  default_print_flags
1313           cl_default_random_state  ->  default_random_state
1314           cl_double_approx()       ->  double_approx()
1315           cl_equal()               ->  equal()
1316           cl_equal_hashcode()      ->  equal_hashcode()
1317           cl_eulerconst()          ->  eulerconst()
1318           cl_find_modint_ring()    ->  find_modint_ring()
1319           cl_find_univpoly_ring()  ->  find_univ_poly_ring()
1320           cl_float_approx()        ->  float_approx
1321           cl_float_format()        ->  float_format()
1322           cl_float_format_t        ->  float_format_t
1323           cl_free_hook()           ->  free_hook()
1324           cl_hermite()             ->  hermite()
1325           cl_laguerre()            ->  laguerre()
1326           cl_legendre()            ->  legandre()
1327           cl_malloc_hook()         ->  malloc_hook()
1328           cl_pi()                  ->  pi()
1329           cl_tschebychev()         ->  tschebychev()
1330           cl_zeta()                ->  zeta()
1331           NB: For functions these changes includes all signatures.
1332         * include/*: moved to include/cln/*, stripped `cl_' off filenames.
1333         * cln-config.in: new file.
1334
1335 2000-08-26  Bruno Haible  <haible@clisp.cons.org>
1336
1337         * autoconf/acgeneral.m4 (AC_OUTPUT): Use braces in exec_prefix default
1338           value, not parens.
1339
1340 2000-08-18  Bruno Haible  <haible@clisp.cons.org>
1341
1342         * include/cl_univpoly_modint.h: Fix typo.
1343
1344 2000-07-13  Bruno Haible  <haible@clisp.cons.org>
1345
1346         * src/float/input/cl_F_read_stream.cc (number_char_p): Accept '_',
1347           used as precision marker for floats.
1348           Reported by Keith Briggs (in 1998) and Thomas Roessler.
1349         * src/integer/input/cl_I_read_stream.cc (number_char_p): Likewise.
1350         * src/rational/input/cl_RA_read_stream.cc (number_char_p): Likewise.
1351         * src/real/input/cl_R_read_stream.cc (number_char_p): Likewise.
1352         * src/complex/input/cl_N_read_stream.cc (number_char_p): Likewise.
1353
1354 2000-06-22  Bruno Haible  <haible@clisp.cons.org>
1355
1356         * include/cl_object.h: Rename cl_word_size to cl_pointer_size.
1357         * include/cl_types.h (HAVE_FAST_LONGLONG): Also define on Irix6
1358           with N32 ABI.
1359           (cl_word_size): New macro.
1360         * src/float/sfloat/cl_SF.h: Use cl_pointer_size instead of
1361           cl_word_size.
1362
1363 2000-05-31  Bruno Haible  <haible@clisp.cons.org>
1364
1365         * tests/exam_I_floor.cc (integer_floor_tests): Add one more entry.
1366           From a sample that fails with gcc-2.95.2 on Sparc.
1367         * tests/exam_I_gcd.cc (integer_gcd_tests): Likewise.
1368
1369 2000-05-30  Richard Kreckel  <kreckel@ginac.de>
1370
1371         * configure.in, autoconf/aclocal.m4 (CL_GMP_H_VERSION, CL_GMP_CHECK):
1372           New macros.
1373         * configure.in, include/cl_config.h.in (CL_VERSION, CL_VERSION_MINOR,
1374           CL_VERSION_PATCHLEVEL): New definitions.
1375         * autoconf/config.guess, autoconf/config.sub, autoconf/ltconfig,
1376           autoconf/ltmain.sh: updated from FSF (libtool 1.3.5, etc).
1377         * src/Makefile.in, configure.in: release-variables renamed from
1378           CLN_* to CL_*.
1379         * configure.in: default to build both shared and static library
1380           (i.e. default to the most common behaviour).
1381
1382 2000-05-29  Richard Kreckel  <kreckel@ginac.de>
1383
1384         * autoconf/aclocal.m4 (CL_CANONICAL_HOST): Added missing changequote
1385           environment around the patch of 2000-05-23.
1386
1387 2000-05-29  Bruno Haible  <haible@clisp.cons.org>
1388
1389         * autoconf/aclocal.m4 (CL_PROG_INSTALL): Fix typo.
1390           Reported by Thomas Klausner <wiz@danbala.ifoer.tuwien.ac.at>.
1391
1392 2000-05-27  Richard Kreckel  <kreckel@ginac.de>
1393
1394         * src/float/lfloat/algebraic/cl_LF_sqrt.cc,
1395           src/base/digitseq/cl_DS_sqrt.cc: Readjusted break-even points.
1396
1397 2000-05-24  Richard Kreckel  <kreckel@ginac.de>
1398
1399         * autoconf/config.*: Updated to new version from FSF
1400           (the new libtool wants this).
1401         * src/Makefile.in: added $(LDFLAGS) to link step.
1402         * src/base/digitseq/cl_2DS_div.cc, cl_2DS_recip.cc: Readjusted
1403           break-even points.
1404
1405 2000-05-23  Bruno Haible  <haible@clisp.cons.org>
1406
1407         * autoconf/aclocal.m4 (CL_CANONICAL_HOST): Determine host_cpu,
1408           host_vendor, host_os correctly if $host has more than two hyphens.
1409
1410 2000-05-19  Richard Kreckel  <kreckel@ginac.de>
1411
1412         * src/base/digitseq/cl_DS_mul.cc: Rearranged break-even points to
1413           better match present-day CPUs whenever GMP3 is used.
1414         * src/base/digitseq/cl_DS_div.cc: dto.
1415         * src/TUNING: Added comment about order of tuning.
1416         * configure, configure.in: Safer GMP3-detection.
1417         * INSTALL.generic: Clarified behaviour of --with-gmp.
1418         * autoconf/config.guess: updated from Clisp-2000-03-06 sources.
1419
1420 2000-05-04  Richard Kreckel  <kreckel@ginac.de>
1421
1422         * gmp/: removed completely.  From now on we are going to link
1423           externally against libgmp3.0 or above!
1424         * configure, configure.in, Makefile.in, */Makeflags: removed support
1425           of internal gmp 2.0.2, like $GMP_INCLUDES, which should be done by
1426           setting $CPPFLAGS instead.
1427         * Makefile.in: Added libtool inter-library dependency for -lgmp and
1428           conforming interface versioning (-version-info).
1429         * autoconf/ltconfig, autoconf/ltmain.sh: Updated to newer versions
1430           from libtool 1.3.4.
1431
1432 2000-02-22  Bruno Haible  <haible@clisp.cons.org>
1433
1434         * src/base/digitseq/cl_asm_mips_.cc: Starting at argument 5 the
1435           parameter passing was changed for the MIPS n32 ABI.
1436
1437 2000-01-24  Richard Kreckel  <kreckel@ginac.de>
1438
1439         * gmp/*: Replaced the complete mpn sources with the ones from
1440           Debian since they are maintained while the ones from FSF
1441           aren't and there were problems on some architectures, PowerPC
1442           in particular.  See the file gmp/README.CLN.  This way the
1443           hard links in this directory have vanished, they were causing
1444           trouble for people working in AFS.  This became necessary for
1445           Debian, because there it woudn't compile on PPC.
1446
1447 2000-01-13  Richard Kreckel  <kreckel@ginac.de>
1448
1449         * Version 1.0.3 released.
1450
1451 2000-01-13  Richard Kreckel  <kreckel@ginac.de>
1452
1453         * src/base/cl_macros.h (nonreturning_function): For egcs and newer
1454           use __attribute__ ((__noreturn__)) instead of the __volatile__
1455           storage class.
1456           (nonreturning): Remove macro.
1457         * include/*: Minor fixes to stop -ansi -pedantic from complaining.
1458         * include/cl_integer.h: (doublefactorial): New declaration.
1459         * src/integer/misc/combin/cl_I_doublefactorial.cc: New file.
1460
1461 1999-12-18  Bruno Haible  <haible@clisp.cons.org>
1462
1463         * autoconf/acgeneral.m4 (AC_ARG_PROGRAM): Create conftestsed using
1464           "cat", not "echo".
1465         * autoconf/ltconfig: Improve support for recent FreeBSD 3.
1466         * include/cl_GV.h (cl_GV_vectorops): Change return type from 'T' to
1467           'const T', to match definition in src/vector/cl_GV_number.cc.
1468           Reported by Duncan Simpson <dps@io.stargate.co.uk>.
1469         * gmp/mpn/Makefile.in (.S.lo): Use 'if', not '&&', because '&&' may
1470           yield return code 1.
1471
1472 1999-09-07  Bruno Haible  <haible@clisp.cons.org>
1473
1474         * Version 1.0.2 released.
1475
1476 1999-09-06  Bruno Haible  <haible@clisp.cons.org>
1477
1478         * src/rational/cl_RA.h (integerp, ratiop): Instead of returning a
1479           boolean expression, write alternatives ending with either
1480           "return cl_true;" or "return cl_false;". This way, g++ does a
1481           better job inlining it.
1482         * src/float/cl_F.h (longfloatp): Likewise.
1483         * src/real/cl_R.h (rationalp, integerp, floatp): Likewise.
1484         * src/complex/cl_C.h (realp, complexp): Likewise.
1485
1486 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1487
1488         * include/cl_integer.h (cl_equal_hashcode): New declaration.
1489         * include/cl_rational.h (cl_equal_hashcode): New declaration.
1490         * include/cl_sfloat.h (cl_equal_hashcode): New declaration.
1491         * include/cl_ffloat.h (cl_equal_hashcode): New declaration.
1492         * include/cl_dfloat.h (cl_equal_hashcode): New declaration.
1493         * include/cl_lfloat.h (cl_equal_hashcode): New declaration.
1494         * include/cl_float.h (cl_equal_hashcode): New declaration.
1495         * include/cl_real.h (cl_equal_hashcode): New declaration.
1496         * include/cl_complex.h (cl_equal_hashcode): New declaration.
1497         * src/base/cl_N.h (equal_hashcode_low, equal_hashcode_one): New macros.
1498         * src/integer/misc/cl_I_eqhashcode.cc: New file.
1499         * src/rational/misc/cl_RA_eqhashcode.cc: New file.
1500         * src/float/sfloat/misc/cl_SF_eqhashcode.cc: New file.
1501         * src/float/ffloat/misc/cl_FF_eqhashcode.cc: New file.
1502         * src/float/dfloat/misc/cl_DF_eqhashcode.cc: New file.
1503         * src/float/lfloat/misc/cl_LF_eqhashcode.cc: New file.
1504         * src/float/misc/cl_F_eqhashcode.cc: New file.
1505         * src/real/misc/cl_R_eqhashcode.cc: New file.
1506         * src/complex/misc/cl_C_eqhashcode.cc: New file.
1507         * doc/cln.tex: Document `cl_equal_hashcode'.
1508
1509 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1510
1511         * include/cl_ring.h (cl_number_ring_ops): Add `contains' member.
1512           (cl_number_ring): New class.
1513           (cl_specialized_number_ring<T>): Inherit from cl_number_ring.
1514           (instanceof): New function.
1515         * src/integer/ring/cl_I_ring.cc (cl_I_p): New function.
1516         * src/integer/misc/cl_I_as.cc (cl_I_p): Add comment.
1517         * src/rational/ring/cl_RA_ring.cc (cl_RA_p): New function.
1518         * src/rational/misc/cl_RA_as.cc (cl_RA_p): Add comment.
1519         * src/real/ring/cl_R_ring.cc (cl_R_p): New function.
1520         * src/real/misc/cl_R_as.cc (cl_R_p): Add comment.
1521         * src/complex/ring/cl_C_ring.cc (cl_N_p): New function.
1522         * src/complex/misc/cl_N_as.cc (cl_N_p): Add comment.
1523         * doc/cln.tex: Document `instanceof'.
1524
1525 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1526
1527         * include/cl_rational.h (numerator, denominator): New declarations.
1528         * src/rational/elem/cl_RA_numerator.cc: New file.
1529         * src/rational/elem/cl_RA_denominator.cc: New file.
1530         * include/cl_integer.h (numerator, denominator): New inline functions.
1531         * doc/cln.tex: Document `numerator' and `denominator'.
1532
1533 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1534
1535         * src/rational/algebraic/cl_RA_rootp.cc (rootp): Fix endless loop
1536           in the integer case.
1537         * src/rational/algebraic/cl_RA_rootp_I.cc (rootp): Likewise.
1538
1539 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1540
1541         * include/cl_config.h.in: Support for sparc64 CPU.
1542         * include/cl_modules.h: Likewise.
1543         * include/cl_types.h: Likewise.
1544         * include/cl_object.h: Likewise.
1545         * include/cl_GV.h: Likewise.
1546         * src/Makefile.in: Likewise.
1547         * src/base/cl_alloca.h: Likewise.
1548         * src/base/cl_macros.h: Likewise.
1549         * src/base/cl_sysdep.h: Likewise.
1550         * src/base/cl_low.h: Likewise.
1551         * src/base/digitseq/cl_asm.h: Likewise.
1552         * src/base/digitseq/cl_asm_.cc: Likewise.
1553         * src/base/digitseq/cl_asm_sparc64.h: New file.
1554         * src/base/digitseq/cl_asm_sparc64_.cc: New file.
1555         * src/modinteger/cl_MI_int.h: Support for sparc64 CPU.
1556         * src/polynomial/elem/cl_UP_no_ring.cc: Likewise.
1557         * src/polynomial/elem/cl_UP_GF2.h: Likewise.
1558         * src/polynomial/elem/cl_asm_GF2.cc: Likewise.
1559
1560 1999-09-04  Bruno Haible  <haible@clisp.cons.org>
1561
1562         * src/base/digitseq/cl_asm_sparc_.cc (orc2_loop_up, orc2_loop_down):
1563           Use the `orn' instruction.
1564
1565 1999-08-14  Bruno Haible  <haible@clisp.cons.org>
1566
1567         Assume all platforms have <stdlib.h> and clock_t.
1568         * configure.in: Don't call CL_STDLIB_H and CL_CLOCK_T.
1569         * src/base/cl_base_config.h.in (CLOCK_T): Remove definition.
1570         * src/base/random/cl_random_from.cc: Use clock_t instead of CLOCK_T.
1571         * src/timing/cl_t_config.h.in (CLOCK_T): Remove definition.
1572         * src/timing/cl_t_current2.cc: Use clock_t instead of CLOCK_T.
1573
1574 1999-07-18  Bruno Haible  <haible@clisp.cons.org>
1575
1576         * gmp/config.guess: Link to autoconf/config.guess.
1577           gmp/config.sub: Link to autoconf/config.sub.
1578           Needed for Win32 platforms.
1579
1580 1999-07-17  Bruno Haible  <haible@clisp.cons.org>
1581
1582         * autoconf/aclocal.m4 (CL_CANONICAL_HOST_CPU): Distinguish "sparc" and
1583           "sparc64" according to the C compiler, not the uname result.
1584
1585 1999-06-17  Bruno Haible  <haible@clisp.cons.org>
1586
1587         * src/base/digitseq/cl_asm_sparc_.cc (compare_loop_up): Fix
1588           COUNTER_LOOPS version, fortunately not used yet.
1589
1590         * include/cl_modules.h: Prepare for gcc version 3.
1591
1592 1999-06-12  Bruno Haible  <haible@clisp.cons.org>
1593
1594         * src/rational/elem/cl_RA_plus.cc, src/rational/elem/cl_RA_minus.cc:
1595           Change the last call from I_I_to_RT to I_I_to_RA.
1596
1597 1999-06-09  Bruno Haible  <haible@clisp.cons.org>
1598
1599         * Version 1.0.1 released.
1600
1601 1999-06-09  Bruno Haible  <haible@clisp.cons.org>
1602
1603         * src/integer/cl_I.h (pFN_maxlength_digits_at): Define also when
1604           intDsize==64.
1605
1606 1999-06-08  Bruno Haible  <haible@clisp.cons.org>
1607
1608         * autoconf/intparam.c (printf_underscored): Change argument type to
1609           `const char*'.
1610         * include/cl_modules.h (CL_OUTPUT_LABEL, CL_JUMP_TO): New macros.
1611           (CL_PROVIDE, CL_PROVIDE_END): Use them.
1612         * include/cl_string.h (cl_heap_string::operator new): Return 1, not 0.
1613         * include/cl_GV.h (cl_GV_inner<T>::operator new): Likewise.
1614         * src/base/ring/cl_no_ring.cc (dummy_canonhom, dummy_expt_pos): Don't
1615           cast a cl_I to void here. Works around a bug in g++-2.95.
1616         * src/complex/misc/cl_C_class.cc: Include "cl_C.h".
1617         * src/polynomial/elem/cl_UP_no_ring.cc (dummy_canonhom,
1618            dummy_expt_pos): Don't cast a cl_I to void here. Works around a bug
1619           in g++-2.95.
1620         * src/polynomial/elem/cl_asm_sparc_GF2.cc (DECLARE_FUNCTION): New
1621           macro.
1622         * src/rational/misc/cl_RA_class.cc: Include "cl_RA.h".
1623         * src/vector/cl_GV_I.cc (cl_heap_GV_I_general::operator new,
1624            DEFINE_cl_heap_GV_I_bits): Return 1, not 0.
1625         * src/vector/cl_GV_number.cc (cl_heap_GV_number_general::operator new):
1626           Likewise.
1627
1628 1999-06-01  Bruno Haible  <haible@clisp.cons.org>
1629
1630         * autoconf/aclocal.m4 (CL_CANONICAL_HOST_CPU): Canonicalize alpha
1631           variants to alpha.
1632
1633 1999-05-29  Bruno Haible  <haible@clisp.cons.org>
1634
1635         * src/base/digitseq/cl_asm_i386_.cc (DECLARE_FUNCTION): Treat
1636           OpenBSD like NetBSD.
1637         * src/base/digitseq/cl_asm_sparc_.cc (DECLARE_FUNCTION): Likewise.
1638         * src/base/digitseq/cl_asm_m68k_.cc (DECLARE_FUNCTION): Treat
1639           OpenBSD like NetBSD, and Linux/ELF like SVR4.
1640
1641 1999-05-16  Bruno Haible  <haible@clisp.cons.org>
1642
1643         * src/base/cl_low.h (integerlength32) [__rs6000__]: Use old assembler
1644           syntax on AIX systems and new assembler syntax on non-AIX systems.
1645
1646 1999-05-01  Bruno Haible  <haible@clisp.cons.org>
1647
1648         * autoconf/config.guess, autoconf/config.sub: Upgrade to newest
1649           version from GNU CVS.
1650
1651 1999-04-24  Bruno Haible  <haible@clisp.cons.org>
1652
1653         * src/integer/bitwise/cl_I_logand.cc (logand): Optimize the case when
1654           either operand is a positive fixnum, O(1) instead of O(N).
1655         * src/integer/bitwise/cl_I_lognand.cc (lognand): Likewise.
1656         * src/integer/bitwise/cl_I_logandc2.cc (logandc2): Likewise for the
1657           first operand.
1658
1659 1999-04-14  Bruno Haible  <haible@clisp.cons.org>
1660
1661         * autoconf/aclocal.m4 (CL_GLOBAL_CONSTRUCTORS): Add check whether
1662           ctor/dtor needs to be exported.
1663         * include/cl_config.h.in (CL_NEED_GLOBALIZE_CTORDTOR): New macro.
1664         * include/cl_modules.h (CL_GLOBALIZE_JUMP_LABEL): Renamed from
1665           CL_GLOBALIZE_LABEL.
1666           (CL_GLOBALIZE_LABEL): New macro.
1667           (CL_GLOBALIZE_CTORDTOR_LABEL): Renamed from CL_GLOBALIZE_ASM_LABEL.
1668           (CL_PROVIDE): Update.
1669
1670 1999-04-12  Bruno Haible  <haible@clisp.cons.org>
1671
1672         * src/Makefile.in ($(ASMFILES_S)): On HPPA, ignore preprocessing
1673           errors ("unterminated string or character constant").
1674           ($(ASMFILES_LO)): On HPPA, try with various settings of
1675           COMPILER_PATH, in order to try /usr/ccs/bin/as and /bin/as.
1676
1677 1999-04-11  Bruno Haible  <haible@clisp.cons.org>
1678
1679         * INSTALL: Mention gmp problems on MIPS.
1680         * doc/cln.tex: Likewise.
1681
1682 1999-03-24  Mumit Khan  <khan@xraylith.wisc.edu>
1683
1684         * src/Makefile.in (SUBDIRS): Filter out CVS and RCS directories from
1685           the source tree.
1686         * include/cl_modules.h (CL_GLOBALIZE_LABEL): Define for Win32.
1687           (CL_GLOBALIZE_ASM_LABEL): New macro.
1688           (CL_PROVIDE): Use it.
1689         * src/base/random/cl_random_from.cc: Handle WIN32.
1690         * src/timing/cl_t_current.cc: Likewise.
1691
1692 1999-03-15  Bruno Haible  <haible@clisp.cons.org>
1693
1694         * autoconf/intparam.c (main7): Use %lX instead of %X for a `long'.
1695           (main8): Adapt for C++.
1696
1697 1999-03-09  Bruno Haible  <haible@clisp.cons.org>
1698
1699         * INSTALL: Mention egcs-1.1 problems on Sparc.
1700         * doc/cln.tex: Likewise.
1701
1702 1999-03-08  Bruno Haible  <haible@clisp.cons.org>
1703
1704         * autoconf/aclocal.m4 (CL_FPU_CONTROL): Fix the "checking for"
1705           messages.
1706
1707 1999-02-25  Bruno Haible  <haible@clisp.cons.org>
1708
1709         * autoconf/aclocal.m4: In test programs, declare `int main()', not
1710           `main()'.
1711
1712         * lidia-interface/src/interfaces/integers/cln/bigint.c
1713           (bigint_to_string): Fix for negative arguments.
1714
1715         * src/base/cl_low.h: Check for `__sparc__', not `SPARC'.
1716         * src/base/cl_alloca.h: Likewise.
1717
1718         * src/base/cl_low.h: Eliminate CLISP style "# " comments.
1719         * src/base/digitseq/cl_asm_arm_.cc,
1720           src/base/digitseq/cl_asm_mips_.cc,
1721           src/float/dfloat/elem/cl_DF_mul.cc: Likewise.
1722
1723         * src/modinteger/cl_MI_pow2.h,
1724           src/modinteger/cl_MI_pow2m1.h,
1725           src/modinteger/cl_MI_pow2p1.h: Workaround g++-2.7.2 inlining bug.
1726
1727 1999-01-18  Bruno Haible  <haible@clisp.cons.org>
1728
1729         * autoconf/acgeneral.m4,
1730           autoconf/acspecific.m4: Upgrade to autoconf-2.13.
1731         * autoconf/config.guess, autoconf/config.sub: Likewise.
1732         * autoconf/aclocal.m4 (CL_ALLOCA): Test for _MSC_VER and alloca being
1733           a macro. Use ${ac_objext}.
1734         * src/base/cl_macros.h (alloca): Put _MSC_VER test before the others,
1735           conforming with CL_ALLOCA.
1736
1737 1999-01-12  Bruno Haible  <haible@clisp.cons.org>
1738
1739         * Version 1.0 released.
1740