]> www.ginac.de Git - cln.git/blob - ChangeLog
* src/float/output/cl_F_dprint.cc (decode_float_decimal): Fix rare
[cln.git] / ChangeLog
1 2006-06-14  Richard B. Kreckel  <kreckel@ginac.de>
2
3         * src/float/output/cl_F_dprint.cc (decode_float_decimal): Fix rare
4         assertion for quite large numbers (cl_LF binary exponent >~ 4.8E8).
5
6 2006-05-20  Bruno Haible  <bruno@clisp.org>
7
8         * src/base/random/cl_random_from.cc: Treat FreeBSD, NetBSD like other
9         Unix platforms.
10
11 2006-05-07  Richard B. Kreckel  <kreckel@ginac.de>
12
13         * include/cln/modules.h (CL_JUMP_TO): Fix for Intel Mac.
14
15 2006-04-22  Richard B. Kreckel  <kreckel@ginac.de>
16
17         * Apply patch for x86_64 from mainline:
18         2005-12-02  Bruno Haible  <bruno@clisp.org>
19                 * src/base/cl_low.h (mulu64) [x86_64]: Change asm restriction,
20                 since mulq doesn't accept immediate arguments.
21
22 2005-12-17  Richard B. Kreckel  <kreckel@ginac.de>
23
24         * Branched cln_1-1. This is the maintenance branch. It is for
25         low-impact patches. The main branch will eventually become CLN 1.2.0.
26
27 2005-12-15  Dmitry V. Kustov  <kustov@telex221.ru>
28
29         * src/base/random/cl_random_from.cc: Add support for OpenBSD.
30
31 2005-11-23  Richard B. Kreckel  <kreckel@ginac.de>
32
33         * Version 1.1.11 released.
34
35 2005-11-20  Richard B. Kreckel  <kreckel@ginac.de>
36
37         * src/integer/conv/cl_I_cached_power.h: New file.
38         * src/integer/conv/cl_I_cached_power.cc: New file.
39         Contains power_table and cached_power_table previously...
40         * src/integer/conv/cl_I_to_digits.cc: ...here.
41         * src/integer/conv/cl_I_from_digits.cc: Use cached powers.
42
43 2005-11-02  Richard B. Kreckel  <kreckel@ginac.de>
44
45         * src/integer/conv/cl_I_from_digits.cc: Made input of all numbers in
46         non-power-of-two base much faster.
47         * tests/test_I_io.cc: New file...
48         * tests/Makefile.in, tests/test_I.cc: ...used here.
49
50 2005-10-22  Richard B. Kreckel  <kreckel@ginac.de>
51
52         * Version 1.1.10 released.
53
54 2005-10-22  Richard B. Kreckel  <kreckel@ginac.de>
55
56         * src/Makefile.in: Accept CPPFLAGS from environment.
57         * examples/Makefile.in: Likewise.
58         * benchmarks/Makefile.in: Likewise.
59         * tests/Makefile.in: Likewise.
60
61 2005-08-30  Richard B. Kreckel  <kreckel@ginac.de>
62
63         * include/cln/modules.h (CL_OUTPUT_LABEL): Work around redundant
64         duplication of basic blocks on m68k.
65
66 2005-08-30  Richard B. Kreckel  <kreckel@ginac.de>
67
68         * include/cln/modules.h (CL_JUMP_TO): Fix mips* brokenness.
69
70 2005-08-27  Bruno Haible  <bruno@clisp.org>
71
72         Split aclocal.m4 into individual files.
73         * m4/alloca.m4, m4/as-underscore.m4, m4/c++-constructors.m4:
74         * m4/fpu_control.m4, m4/general.m4, m4/gettimeofday.m4:
75         * m4/longdouble.m4, m4/longlong.m4, m4/param.m4, m4/perror.m4:
76         * m4/proto.m4, m4/rusage.m4, m4/times.m4:
77         New files, extracted from autoconf/aclocal.m4.
78         * autoconf/aclocal.m4: m4_include them.
79         * Makefile.devel (AUTOCONF_MACROS): New variable.
80         (configure): Depend on it.
81         (CLISP_M4DIR): Remove variable.
82         (autoconf/aclocal.m4): Remove rule.
83
84 2005-08-27  Bruno Haible  <bruno@clisp.org>
85
86         * src/integer/bitwise/cl_I_ash_I.cc (ash): Avoid shifting a 32-bit
87         zero value by more than 31 bits.
88
89 2005-08-27  Bruno Haible  <bruno@clisp.org>
90
91         Make the long-float overflow check work on 64-bit platforms.
92         * src/float/lfloat/cl_LF.h (LF_exp_mid, LF_exp_high): Define as
93         'unsigned int', not 'unsigned long'.
94
95 2005-08-27  Bruno Haible  <bruno@clisp.org>
96
97         * include/cln/modules.h (CL_OUTPUT_LABEL): Work around redundant
98         duplication of basic blocks by g++ 4.0.
99         See <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23345>.
100
101 2005-08-27  Bruno Haible  <bruno@clisp.org>
102
103         Make it possible to cross-compile CLN.
104         * m4/intparam.m4: New file.
105         * m4/floatparam.m4: New file.
106         * autoconf/aclocal.m4: Include both.
107         (CL_MACHINE): Add an additional CROSS_MACRO parameter.
108         * configure.ac (CL_MACHINE): When cross-compiling, use
109         CL_INTPARAM_CROSS and CL_FLOATPARAM_CROSS.
110
111 2005-08-27  Bruno Haible  <bruno@clisp.org>
112
113         Define HAVE_LONGLONG and HAVE_LONGDOUBLE when cross-compiling.
114         * autoconf/aclocal.m4 (CL_LONGLONG, CL_LONGDOUBLE): When cross-
115         compiling, use the test code from gnulib.
116
117 2005-08-27  Bruno Haible  <bruno@clisp.org>
118
119         * autoconf/aclocal.m4 (CL_RUSAGE): Fix error when cross-compiling.
120
121 2005-08-16  Richard B. Kreckel  <kreckel@ginac.de>
122
123         The patch of 2005-05-01 made it impossible to test the type of a cl_UP
124         by comparing with &cl_class_univpoly_ring. We need an alternative:
125         * include/cln/object.h (cl_class_flags_modint_ring): New #define...
126         * src/polynomial/elem/cl_UP.cc (cl_class_univpoly_ring): ...used here.
127         * src/polynomial/elem/cl_UP_GF2.h: Likewise.
128         * src/polynomial/elem/cl_UP_MI.h: Likewise.
129         * src/polynomial/elem/cl_UP_gen.h: Likewise.
130         * src/polynomial/elem/cl_UP_number.h: Likewise.
131
132 2005-08-15  Richard B. Kreckel  <kreckel@ginac.de>
133
134         * m4/cc.m4 (CL_AS_NOEXECSTACK): New macro...
135         * configure.ac: ...used here for setting ASMFLAGS...
136         * src/Makefile.in: ...which are used here.
137
138 2005-08-02  Andreas Jochens  <aj@andaco.de>
139
140         * include/cln/config.h.in: Add support for PowerPC 64 CPU.
141         * include/cln/modules.h: Likewise.
142         * include/cln/object.h: Likewise.
143         * include/cln/types.h: Likewise.
144
145 2005-07-24  Richard B. Kreckel  <kreckel@ginac.de>
146
147         Make out of the box build on x86_64 system with complete 32 bit
148         userland possible
149         * include/cln/config.h.in: Don't #define __x86_64__ when
150         __i386__ is defined.
151         * src/base/digitseq/cl_asm_x86_64_.cc: New file.
152         * doc/cln.tex: Revert workaround description introduced 2005-05-02.
153
154 2005-06-10  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
155
156         * Makefile.in: Don't enter nonexisting directories.
157
158 2005-05-17  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
159
160         Speed up the linking step
161         * src/Makefile.in: Use -objectlist for linking libcln.la.
162
163 2005-05-15  Richard B. Kreckel  <kreckel@ginac.de>
164
165         * autoconf/ltmain.sh: Upgrade to libtool-1.5.16.
166         * m4/libtool.m4: Upgrade to libtool-1.5.16 (without Comeau patch).
167
168 2005-05-02  Richard B. Kreckel  <kreckel@ginac.de>
169
170         * doc/cln.tex: Document what to do on a x86_64 machine with 32-bit
171         userland.
172
173 2005-05-01  Richard B. Kreckel  <kreckel@ginac.de>
174
175         Fix crashes in find_univpoly_ring and related functions
176         * include/cln/modinteger.h: Remove vptr from cl_heap_modint_ring;
177         remove declaration of cl_class cl_class_modint_ring.
178         * include/cln/univpoly.h: Remove vptr from cl_heap_univpoly_ring;
179         remove declaration of cl_class_univpoly_ring.
180         * include/cln/object.h: cl_class_flags_modint_ring: New #define...
181         * src/modinteger/cl_MI.cc: ...used in cl_class_modint_ring.
182         * src/modinteger/cl_MI_fix16.h: No vptr, but static dtor and type flag.
183         * src/modinteger/cl_MI_fix29.h: Likewise.
184         * src/modinteger/cl_MI_fix32.h: Likewise.
185         * src/modinteger/cl_MI_int32.h: Likewise.
186         * src/modinteger/cl_MI_montgom.h: Likewise.
187         * src/modinteger/cl_MI_pow2: Likewise.
188         * src/modinteger/cl_MI_pow2m1.h: Likewise.
189         * src/modinteger/cl_MI_pow2p1.h: Likewise.
190         * src/modinteger/cl_MI_std.h: Likewise.
191         * src/polynomial/elem/cl_UP.cc (cl_make_univpoly_ring): Compare with
192         cl_class_flags_modint_ring, not with cl_class_modint_ring.
193         * src/polynomial/elem/cl_UP_GF2.h (cl_class_num_univpoly_ring): New.
194         * src/polynomial/elem/cl_UP_MI.h (cl_class_modint_univpoly_ring): New.
195         * src/polynomial/elem/cl_UP_gen.h (cl_class_gen_univpoly_ring): New.
196         * src/polynomial/elem/cl_UP_number.h (cl_class_num_univpoly_ring): New.
197         Reported by Ralf Goertz <R_Goertz@web.de>.
198
199 2005-04-29  Richard B. Kreckel  <kreckel@ginac.de>
200             Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
201
202         * m4/cc.m4: Emit a warning if g++ is used and optimization turned off.
203
204 2005-04-24  Richard B. Kreckel  <kreckel@ginac.de>
205
206         Make GCC compiler flags default to -O
207         * m4/cc.m4: New file...
208         * configure.ac: ...used here.
209         * autoconf/aclocal.m4: Regenerate.
210
211 2005-04-18  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
212
213         * include/cln/string.h: Declare cl_string.
214
215 2005-03-17  Richard B. Kreckel  <kreckel@ginac.de>
216
217         * autoconf/ltmain.sh: Upgrade to libtool-1.5.14.
218         * m4/libtool.m4: Upgrade to libtool-1.5.14 with Comeau patch.
219         * autoconf/aclocal.m4: Regenerate.
220
221 2005-03-15  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
222
223         * src/Makefile.in: Use $CXX instead of $CC when linking.
224
225 2005-02-27  Richard B. Kreckel  <kreckel@ginac.de>
226         
227         * examples/perfnum.cc: update to presumed 42st Mersenne prime.
228
229 2004-11-28  Richard B. Kreckel  <kreckel@ginac.de>
230
231         Disambiguate binary operators of CLN types with float/double
232         * include/cln/dfloat.h: Add binary operator overloads for arguments of
233         type double.
234         * include/cln/ffloat.h: Likewise, for arguments of type float.
235         * include/cln/float.h: Likewise, both for arguments of types double and
236         float.
237         * include/cln/real.h: Likewise.
238         Reported by Isidro Cachadiña Gutiérrez <icacha@unex.es>.
239
240 2004-11-03  Richard B. Kreckel  <kreckel@ginac.de>
241
242         * Version 1.1.9 released.
243
244 2004-10-28  Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
245
246         * src/Makefile.in: Let config.status set LDFLAGS.
247
248 2004-10-27  Peter Breitenlohner  <peb@mppmu.mpg.de>
249
250         * cln.pc.in: Fix typo.
251         * Makefile.in (INSTALL_SCRIPT): Added, to be used for scripts like
252         cln-config.  Allows us to do special things to binaries, like
253         installing them with install -s.
254         * doc/Makefile.in: add/remove the cln.info from the installed
255         $(infodir)/dir unless this is debian install-info (code copied from
256         what GNU automake would produce).
257
258 2004-10-26  Richard B. Kreckel  <kreckel@ginac.de>
259
260         * src/integer/input/cl_I_read_stream.cc (read_integer): Fix a bug
261         that caused radix specifiers to not work when reading from a stream.
262         * src/rational/input/cl_RA_read_stream.cc (read_rational): Likewise.
263         * src/real/input/cl_R_read_stream.cc (read_real): Likewise.
264         * src/float/input/cl_F_read_stream.cc (read_float): Likewise.
265         * src/complex/input/cl_N_read_stream.cc (read_complex): Likewise.
266
267 2004-10-25  Richard B. Kreckel  <kreckel@ginac.de>
268
269         * src/base/cl_low.h: Add mulu64 assembler macro for ia64.
270
271 2004-10-24  Richard B. Kreckel  <kreckel@ginac.de>
272
273         * src/base/cl_low.h: Add mul and div macros for x86_64.
274
275 2004-10-23  Richard B. Kreckel  <kreckel@ginac.de>
276
277         * src/integer/conv/cl_I_from_digits.cc (digits_to_I): Fix thinko in
278         new code for base power of two.
279
280 2004-10-22  Richard B. Kreckel  <kreckel@ginac.de>
281
282         * src/integer/conv/cl_I_to_digits (I_to_digits): Fix an elusive stack
283         overwriting problem. That was the real cause for Debian bug#246319.
284         * src/integer/output/cl_I_print.cc (print_integer): Revert workaround
285         for the bug fixed above.
286
287 2004-10-20  Richard B. Kreckel  <kreckel@ginac.de>
288
289         * include/cln/types.h: Use 64 bit digits on x86_64 CPU.
290
291 2004-10-12  Richard B. Kreckel  <kreckel@ginac.de>
292
293         * src/integer/conv/cl_I_from_digits.cc (digits_to_I): Speedup when
294         the base is a power of two.
295
296 2004-10-05  Richard B. Kreckel  <kreckel@ginac.de>
297
298         * src/integer/conv/cl_I_to_digits.cc (I_to_digits): Fix bug in base 32.
299
300 2004-09-27  Richard B. Kreckel  <kreckel@ginac.de>
301
302         Support for little-endian Mips, second shot
303         * src/base/digitseq/cl_asm_mipsel_.cc: New file...
304         * src/base/digitseq/cl_asm_cc: ...used here.
305         * src/base/digitseq/cl_asm.h: Include cl_asm_mips.h for any endianness.
306         * include/cln/object.h: Set alignment for mipsel explicitly.
307
308 2004-09-05  Richard B. Kreckel  <kreckel@ginac.de>
309
310         Support for little-endian Mips
311         * include/cln/config.h.in: Add __mipsel__.
312         * include/cln/modules.h: For Mips, this is endianness-agnostic.
313         * src/base/digitseq/cl_asm_.cc, src/base/digitseq/cl_asm.h:
314         Mask out assembler for little-endian Mips.
315
316 2004-08-30  Bruno Haible  <bruno@clisp.org>
317
318         * benchmarks/timebench2.sh: Multiply all repeat counts by 100.
319         * benchmarks/timebench2.results: Add recent PowerPC G4 results.
320
321 2004-08-26  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
322
323         * examples/e.cc: remove extra semicolon.
324
325 2004-08-25  Bruno Haible  <bruno@clisp.org>
326
327         * autoconf/ltmain.sh: Upgrade to libtool-1.5.6.
328         * m4/libtool.m4: New file, from libtool-1.5.6 with modifications:
329         2004-08-22  Bruno Haible  <bruno@clisp.org>
330                 * m4/libtool.m4: Add support for Comeau C++ on Linux.
331                 Reported by Prof. Roberto Bagnara <bagnara@cs.unipr.it>.
332         * autoconf/aclocal.m4: Regenerate.
333
334 2004-08-19  Bruno Haible  <bruno@clisp.org>
335
336         * include/cln/modules.h (CL_GLOBALIZE_JUMP_LABEL, CL_JUMP_TO): When
337         converting a label to a string, use ASM_UNDERSCORE_PREFIX. Needed on
338         MacOS X.
339         Reported by Darren Bane <darren.bane@ul.ie>.
340
341 2004-07-01  Richard B. Kreckel  <kreckel@ginac.de>
342
343         * Version 1.1.8 released.
344
345 2004-06-30  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
346
347         * src/base/cl_macros.h: alloca(3) has size_t argument type.
348
349 2004-06-30  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
350
351         * include/cln/floatformat.h: Do define a type here.
352
353 2004-06-27  Richard B. Kreckel  <kreckel@ginac.de>
354
355         * include/cln/modules.h (CL_JUMP_TO): Fix AMD64 brokenness.
356
357 2004-06-23  Bruno Haible  <bruno@clisp.org>
358
359         * configure.ac: Pretend ftime() is not available. Needed by
360         CL_TIMES_CLOCK.
361
362 2004-06-21  Ralf Stephan  <ralf@ark.in-berlin.de>
363
364         * doc/cln.tex: Document jacobi, isprobprime and nextprobprime.
365
366 2004-06-18  Richard B. Kreckel  <kreckel@ginac.de>
367
368         * rational/transcendental/cl_RA_logp.cc: fix bug where base is
369         reciprocal of an integer.
370         Reported by Niklas Knutsson <nq@altern.org>.
371
372 2004-06-18  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
373
374         * src/complex/ring/cl_C_ring.cc, src/integer/ring/cl_I_ring.cc,
375         src/rational/ring/cl_RA_ring.cc, src/real/ring/cl_R_ring.cc:
376         Make template specializations explicit.
377
378 2004-06-17  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
379
380         * src/complex/input/cl_N_read.cc, src/float/input/cl_F_read.cc,
381         src/integer/input/cl_I_read.cc, src/rational/input/cl_RA_read.cc,
382         src/real/input/cl_R_read.cc:  Remove unused labels.
383
384 2004-06-17  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
385
386         * benchmarks/timebench1.cc, benchmarks/timebench2a.LiDIA.cc,
387         benchmarks/timebench2a.cc, benchmarks/timebench2ap.cc,
388         benchmarks/timebench2b.LiDIA.cc, benchmarks/timebench2b.cc,
389         examples/atan_recip.cc, examples/atanh_recip.cc,
390         examples/contfrac.cc, examples/e.cc, examples/legendre.cc,
391         examples/lucaslehmer.cc, examples/pi.cc, include/cln/GV.h,
392         include/cln/SV.h, include/cln/malloc.h, include/cln/modules.h,
393         include/cln/object.h, include/cln/string.h, src/base/cl_abort.cc,
394         src/base/cl_alloca.h, src/base/cl_malloc.cc,
395         src/base/random/cl_random_from.cc,
396         src/base/string/cl_spushstring_append.cc,
397         src/base/string/cl_spushstring_push.cc,
398         src/base/string/cl_st_debug.cc,
399         src/base/string/input/cl_st_gettoken.cc,
400         src/complex/input/cl_N_read.cc, src/float/input/cl_F_read.cc,
401         src/float/output/cl_F_dprint.cc, src/integer/input/cl_I_read.cc,
402         src/rational/input/cl_RA_read.cc,
403         src/real/format-output/cl_fmt_integer.cc,
404         src/real/format-output/cl_fmt_paddedstring.cc,
405         src/real/input/cl_R_read.cc, src/timing/cl_t_current.cc,
406         src/timing/cl_t_current2.cc, tests/exam.cc, tests/tests.cc,
407         tests/timeLFRAmul.cc, tests/timeLFatan-compare.cc,
408         tests/timeLFatan.cc, tests/timeLFatanh-compare.cc,
409         tests/timeLFatanh.cc, tests/timeLFcos-compare.cc, tests/timeLFcos.cc,
410         tests/timeLFcosh.cc, tests/timeLFexp-compare.cc, tests/timeLFexp.cc,
411         tests/timeLFln-compare.cc, tests/timeLFln.cc,
412         tests/timeLFsin-compare.cc, tests/timeLFsin.cc, tests/timeLFsinh.cc,
413         tests/timeLFsqrt.cc, tests/timeMImisc5.cc, tests/timeMIpow2div.cc,
414         tests/timeMIpow2recip.cc, tests/timeRALFdiv.cc, tests/timeRAtoLF.cc,
415         tests/timeUPMImul.cc, tests/timecatalan.cc, tests/timediv.cc,
416         tests/timediv2adic-compare.cc, tests/timediv2adic.cc,
417         tests/timeeuler.cc, tests/timeexp1.cc, tests/timefact.cc,
418         tests/timegcd.cc, tests/timemul-compare.cc, tests/timemul.cc,
419         tests/timepi.cc, tests/timeprint-compare.cc, tests/timeprint.cc,
420         tests/timerecip2adic-compare.cc, tests/timerecip2adic.cc,
421         tests/timesqrt.cc, tests/timesqrtmodp.cc, tests/timesquare.cc,
422         tests/timezeta3.cc: Change all C include headers to ISO style
423         within C++ code.
424
425 2004-06-10  Richard B. Kreckel  <kreckel@ginac.de>
426
427         * examples/perfnum.cc: update to presumed 41st Mersenne prime.
428
429 2004-05-02  Richard B. Kreckel  <kreckel@ginac.de>
430
431         * Version 1.1.7 released.
432
433 2004-05-02  Richard B. Kreckel  <kreckel@ginac.de>
434
435         * examples/pi.cc and examples/pi.1: New files.
436         * examples/Makefile.in: Build the pi executable.
437
438 2004-05-01  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
439
440         * src/Makefile.in: Fix for parallel build: wait for subdir objects to
441         be finished before creating the library.
442
443 2004-04-30  Richard B. Kreckel  <kreckel@ginac.de>
444
445         * src/integer/output/cl_I_print.cc (print_integer): workaround
446         GCC compiler bug (cf. Debian bug#246319).
447
448 2004-03-20  Ralf Wildenhues  <Ralf.Wildenhues@gmx.de>
449
450         * m4/gmp.m4 (CL_GMP_CHECK): Do not lose LIBS setting with config.cache
451         enabled.
452
453 2004-03-08  Bruno Haible  <bruno@clisp.org>
454
455         * src/float/lfloat/elem/cl_LF_mul.cc (operator*): Fix the second
456         underflow condition.
457         * src/float/lfloat/algebraic/cl_LF_sqrt.cc (sqrt): Fix a bug with large
458         uexp whereby SQRT of MOST-POSITIVE-LONG-FLOAT was less than 1.
459
460 2004-03-04  Richard B. Kreckel  <kreckel@ginac.de>
461
462         * Makefile.in (install): Add ${srcdir} for cln.m4.
463         * m4/gmp.m4: quote macro names.
464         Reported by Ralf Wildenhues <Ralf.Wildenhues@gmx.de>.
465
466 2004-01-01  Richard B. Kreckel  <kreckel@ginac.de>
467
468         * Version 1.1.6 released.
469
470 2004-01-01  Richard B. Kreckel  <kreckel@ginac.de>
471
472         * include/cln/univpoly.h, include/cln/univpoly_complex.h,
473           include/cln//univpoly_integer.h, include/cln/univpoly_modint.h,
474           include/cln/univpoly_rational.h, include/cln/univpoly_real.h,
475           src/polynomial/elem/cl_UP_GF2.h, src/polynomial/elem/cl_UP_MI.h,
476           src/polynomial/elem/cl_UP_gen.h, src/polynomial/elem/cl_UP_no_ring.cc,
477           src/polynomial/elem/cl_UP_number.h (ldegree): New function.
478         * doc/cln.tex: Document `ldegree'.
479
480 2003-12-29  Richard B. Kreckel  <kreckel@ginac.de>
481
482         Rework of autoconfiscation infrastructure
483         * autoconf/config.{guess,sub}: Update to GNU version 2003-10-07.
484         * autoconf/ltmain.sh: Update to GNU version 1.4.3.
485         * autoconf/autoconf: Remove (from now on we assume autoconf is
486           installed properly on the sytem).
487         * autoconf/autoconf.m4: Likewise.
488         * autoconf/autoconf.m4f: Likewise.
489         * autoconf/acgeneral.m4: Likewise.
490         * autoconf/acspecific.m4: Likewise.
491         * autoconf/aclocal.m4: Regenerate.
492         * autoconf/acinclude.m4: Remove (while moving the macros...)
493         * m4/gmp.m4: New file (...to here).
494         * Makefile.devel: Update to new scheme.
495         * configure.ac: Likewise.
496         * include/cln/GV_integer.h: Assume template specializations work.
497         * include/cln/GV_modinteger.h: Likewise.
498         * include/cln/config.h.in: Likewise, and drop HAVE_BOOL.
499         * src/base/cl_base_config.h.in: Drop support for obsolete ftime(3).
500         * src/base/random/cl_random_from.cc: Likewise.
501         * src/timing/cl_base_config.h.in: Likewise.
502         * src/timing/cl_t_current.cc: Likewise.
503
504 2003-12-27  Richard B. Kreckel  <kreckel@ginac.de>
505
506         * src/polynomial/cl_UP_gen.h (gen_minus): Fix case where first
507           argument is zero.
508         * src/polynomial/cl_UP_MI.h (modint_minus): Likewise.
509         * src/polynomial/cl_UP_number.h (num_minus): Likewise.
510         Reported by Munagala Ramanath <amberarrow@yahoo.com>.
511
512 2003-12-02  Richard B. Kreckel  <kreckel@ginac.de>
513
514         * examples/perfnum.cc: update to presumed 40th Mersenne prime.
515
516 2003-11-20  Christian Bauer  <cbauer@ginac.de>
517
518         Added pkg-config support
519         * cln.pc.in: New file.
520         * Makefile.in: Take care of cln.pc.
521         * configure.ac: Likewise.
522
523 2003-08-06  Richard B. Kreckel  <kreckel@ginac.de>
524
525         * src/numtheory/cl_nt_sqrtmodp.cc: #undef _R.
526         Reported by Andrew Rechnitzer <A.Rechnitzer@ms.unimelb.edu.au>.
527
528 2003-08-01  Richard Kreckel  <kreckel@ginac.de>
529
530         More dependent base resolution issues
531         * src/base/hash/cl_hash.h (cl_heap_hashtable<T>::iterator()): 
532         portable syntactic simplification.
533         * src/base/hash/cl_hashset.h: Preceed inherited members with this->.
534         * src/base/hash/cl_hash1.h: Likewise for member functions.
535         * src/base/hash/cl_hash2.h: Likewise.
536         * src/base/hash/cl_hashuniq.h: Likewise.
537         * src/base/hash/cl_hashuniqweak.h: Likewise.
538         * src/base/hash/cl_hash.h: Revert explicit static member function
539         lookup since that was GCC's fault.
540         * src/base/hash/cl_hash2weak.h: Likewise.
541         * src/base/hash/cl_hashuniqweak.h: Likewise.
542
543 2003-06-29  Richard Kreckel  <kreckel@ginac.de>
544
545         Dependent base resolution needed for GCC-3.4
546         * include/cln/GV.h: Preceed inherited members with this->.
547         * include/cln/SV.h: Likewise.
548         * include/cln/object.h: Likewise.
549         * src/base/hash/cl_hash1.h: Likewise.
550         * src/base/hash/cl_hash1weak.h: Likewise.
551         * src/base/hash/cl_hash2.h: Likewise.
552         * src/base/hash/cl_hashuniq.h: Likewise.
553         * src/base/hash/cl_hash.h: Make lookup of static member function 
554         explicit.
555         * src/base/hash/cl_hash2weak.h: Likewise.
556         * src/base/hash/cl_hashuniqweak.h: Likewise.
557         * src/modinteger/cl_MI.cc: Make shell-comment a C-comment. Geez.
558
559 2003-02-24  Bruno Haible  <bruno@clisp.org>
560
561         * src/base/random/cl_random_from.cc (random_state::random_state): Add
562         support for MacOS X.
563         * src/modinteger/cl_MI.cc: #undef _R.
564         Reported by Erann Gat <gat@jpl.nasa.gov>.
565
566 2002-08-03  Joerg Arndt  <jj@suse.de>
567
568         * include/cln/config.h.in: Add support for x86_64 CPU.
569         * include/cln/modules.h: Likewise.
570         * include/cln/types.h: Likewise.
571         * include/cln/object.h: Likewise.
572
573 2002-06-08  Richard Kreckel  <kreckel@ginac.de>
574
575         * src/base/digitseq/cl_asm.h: ensure intCsize==16 when including
576         m68k Assembler routines.
577         * src/base/digitseq/cl_asm_.cc: Likewise.
578
579 2002-05-28  Richard Kreckel  <kreckel@ginac.de>
580
581         * Version 1.1.5 released.
582
583 2002-05-27  Richard Kreckel  <kreckel@ginac.de>
584
585         * include/cln/modules.h (CL_CONCATENATE): New macro.
586
587 2002-05-10  Richard Kreckel  <kreckel@ginac.de>
588
589         * doc/cln.tex (Building the library): Update recommendations for 
590         compiling on Tru64 using g++ 3.0 and 3.1.
591         * README: Update homepage.
592
593 2002-05-05  Bruno Haible  <bruno@clisp.org>
594
595         * doc/cln.tex (Building the library): Give some recommendations for
596         g++ 3.0 and 3.1.
597
598         Force link-time references despite optimizations done by g++ 2.95
599         and newer.
600         * include/cln/modules.h (CL_FORCE_LINK): New macro.
601         * Use CL_FORCE_LINK.
602         * include/cln/GV_integer.h (cl_GV_I_debug_dummy): Likewise.
603         * include/cln/GV_number.h (cl_GV_number_debug_dummy): Likewise.
604         * include/cln/SV_number.h (cl_SV_number_debug_dummy): Likewise.
605         * include/cln/SV_ringelt.h (cl_SV_ringelt_debug_dummy): Likewise.
606         * include/cln/dfloat.h (cl_DF_debug_dummy): Likewise.
607         * include/cln/ffloat.h (cl_FF_classes_dummy, cl_FF_debug_dummy):
608         Likewise.
609         * include/cln/integer.h (cl_I_classes_dummy, cl_I_debug_dummy):
610         Likewise.
611         * include/cln/lfloat.h (cl_LF_debug_dummy): Likewise.
612         * include/cln/modinteger.h (cl_MI_debug_dummy): Likewise.
613         * include/cln/rational.h (cl_RA_debug_dummy): Likewise.
614         * include/cln/real.h (cl_R_debug_dummy): Likewise.
615         * include/cln/ring.h (cl_ring_debug_dummy): Likewise.
616         * include/cln/sfloat.h (cl_SF_classes_dummy, cl_SF_classes_dummy):
617         Likewise.
618         * include/cln/string.h (cl_string_debug_dummy): Likewise.
619         * include/cln/univpoly.h (cl_UP_debug_dummy): Likewise.
620         * src/float/base/cl_ieee.h (cl_ieee_dummy_NNN): Likewise.
621
622         Avoid g++ 3.1 warnings.
623         * src/base/cl_offsetof.h (offsetof): Redefine each time.
624         * src/base/digitseq/cl_DS.h: Include "cl_offsetof.h" after <gmp.h>.
625
626         Avoid g++ 3.1 warnings.
627         * src/base/hash/cl_hash.h (struct cl_heap_hashtable): Use typename
628         where needed.
629         * src/base/hash/cl_hash1.h (struct cl_heap_hashtable_1): typedef
630         htxentry as a shortcut.
631         * src/base/hash/cl_hash2.h (struct cl_heap_hashtable_2): Likewise.
632         * src/base/hash/cl_hashset.h (struct cl_heap_hashtable_set): Likewise.
633         * src/base/hash/cl_hashuniq.h (struct cl_heap_hashtable_uniq):
634         Likewise.
635
636 2002-03-15  Bruno Haible  <bruno@clisp.org>
637
638         * cln.tex: Document problem with GNU make 3.77.
639         Reported by Michael Somos <somos@grail.cba.csuohio.edu>.
640
641 2002-02-16  Richard Kreckel  <kreckel@ginac.de>
642
643         * cln.m4: quote macro name.
644           Pointed out by Roberto Bagnara.
645
646 2002-01-20  Richard Kreckel  <kreckel@ginac.de>
647
648         * autoconf/config.{guess,sub}: Update to GNU version 2002-01-02.
649           (the old one was broken on Linux/Mips.)
650
651 2002-01-04  Richard Kreckel  <kreckel@ginac.de>
652
653         * autoconf/autoconf.m4f: get brutal in order to adhere to FHS.
654         * Version 1.1.4 released.
655
656 2002-01-03  Richard Kreckel  <kreckel@ginac.de>
657
658         * autoconf/acinclude.m4: revamp MPN-matcher.
659         * autoconf/aclocal.m4: upgrade to autoconf-2.52 infrastructure,
660           sync with CLisp from CVS.
661         * autoconf/autoconf: Likewise.
662         * autoconf/autoconf.m4f: Likewise (new file).
663         * configure.ac: Likewise (new file, replaces configure.in).
664         * configure.in: Likewise (deleted, replaced by configure.ac).
665         * autoconf/config.{guess,sub}: Update to GNU version 2001-12-13.
666         * src/Makefile.in: made VPATH safe for autoconf-2.52.
667         * include/cln/config.h.in: Add __s390__.
668
669 2001-12-31  Richard Kreckel  <kreckel@ginac.de>
670
671         * src/base/digitseq/cl_DS.h: <gmp.h> is not included extern "C"
672           any more since GMP4 has some C++ support in it.
673
674 2001-12-14  Richard Kreckel  <kreckel@ginac.de>
675
676         * include/cln/modules.h, include/cln/object.h: add support for
677           s390.
678         * src/numtheory/cl_nt_sqrtmodp.cc: workaround for GCC2.x compiler-bug
679           on s390, provided by Gerhard Tonn.
680
681 2001-11-05  Richard Kreckel  <kreckel@ginac.de>
682
683         * autoconf/ltmain.sh: Upgrade to libtool-1.4.2.
684         * autoconf/config.{guess,sub}: Update to GNU version 2001-09-07.
685         * Version 1.1.3 released.
686
687 2001-11-04  Bruno Haible  <haible@clisp.cons.org>
688
689         Interoperability with gcc-3.0 -fuse-cxa-atexit.
690         * autoconf/aclocal.m4 (CL_GLOBAL_CONSTRUCTORS): Add test whether
691           global destructors actually exist.
692         * include/cln/modules.h (CL_PROVIDE, CL_PROVIDE_END, CL_PROVIDE_END):
693           Don't hack the global destructors if there is no global destructors
694           function.
695
696 2001-11-03  Richard Kreckel  <kreckel@ginac.de>
697
698         * src/float/transcendental/cl_F_sinx.cc (sinx_naive): For small
699           values of x, return square(x) instead of x.
700         * src/float/transcendental/cl_F_sinhx.cc (sinhx_naive): Likewise.
701
702 2001-07-25  Richard Kreckel  <kreckel@ginac.de>
703
704         * Version 1.1.2 released.
705
706 2001-07-24  Richard Kreckel  <kreckel@ginac.de>
707
708         * src/base/hash/cl_hash.h: declare _cl_hashtable_iterator<htentry> a
709           friend of cl_heap_hashtable<htentry>.
710
711 2001-07-22  Richard Kreckel  <kreckel@ginac.de>
712
713         * src/float/base/cl_ieee.cc: try to do magic to the FPU only if
714           _FPU_IEEE is really defined.
715         * include/cln/modules.h: change assembler labels from `label' to
716           `label:' on hppa, needed by Linux (see comment).
717         * autoconf/acinclude.m4: new file (for storing CLN-specific macros).
718         * Makefile.devel: adjusted.
719         * autoconf/aclocal.m4: regenerate.
720         * src/base/low/cl_low_mul.cc: moved POD variables that are declared
721           extern "C" elsewhere out of the namespace.
722         * src/base/low/cl_low_div.cc: Likewise.
723
724 2001-06-08  Bruno Haible  <haible@clisp.cons.org>
725
726         * autoconf/config.{guess,sub}: Update to GNU version 2001-05-11.
727         * autoconf/aclocal.m4: Upgrade to libtool-1.4.
728         * autoconf/ltmain.sh: Likewise.
729         * autoconf/ltconfig: Remove file.
730         * autoconf/install-sh: New file.
731         * configure.in: Add AC_CONFIG_AUX_DIR call.
732
733 2001-06-05  Richard Kreckel  <kreckel@ginac.de>
734
735         * tests/tests.cc: resolve namespace ambiguity about strcmp().
736
737 2001-05-31  Richard Kreckel  <kreckel@ginac.de>
738
739         * Version 1.1.1 released.
740
741 2001-05-28  Richard Kreckel  <kreckel@ginac.de>
742
743         * cln/cln.tex: documented problems with shared library on Sparc
744           using gcc older than 2.95.3.
745         * configure.in: Fixed typos in versioning docu.
746
747 2001-05-25  Bruno Haible  <haible@clisp.cons.org>
748
749         * src/base/digitseq/cl_asm_arm_.cc: Use #0x instead of #& to designate
750           hexadecimal constants.
751
752 2001-05-25  Richard Kreckel  <kreckel@ginac.de>
753
754         * autoconf/floatparam.c (double_wordorder_bigendian_p): new symbol.
755         * src/float/dfloat/cl_DF.h: Check for double_wordorder_bigendian_p.
756         * Removed LiDIA interface since that is now outdated (namespace cln)
757           and maintained elsewhere.
758         * Adjusted dates and final touches for 1.1.1.
759
760 2001-05-19  Richard Kreckel  <kreckel@ginac.de>
761
762         * INSTALL: Update toolchain info: no egcs, some more platforms.
763         * doc/cln.tex: Likewise.
764
765 2001-05-18  Richard Kreckel  <kreckel@ginac.de>
766
767         * src/base/cl_low.h: prepended variables declared inside macros
768           with underscore.  Fixes equal_hashcode() on various platforms.
769
770 2001-04-25  Richard Kreckel  <kreckel@ginac.de>
771
772         * src/base/cl_low.h: Added several checks if NO_ASM is defined, so
773           this definition becomes actually useful.  This is needed for 
774           compilation on Arm until somebody fixes the assembler files for Arm.
775         * src/base/digitseq/cl_asm.h: Likewise.
776         * src/base/digitseq/cl_asm_.cc: Likewise.
777         * */Makefile.in: Added `override' in front of `CPPFLAGS +=' so
778            one can say `make CPPFLAGS=-DFOOBAR'.
779
780 2001-03-26  Arvid Norberg <c99ang@cs.umu.se>
781
782         * src/base/random/cl_random_from.cc: ported to beos.
783
784 2001-03-05  Richard Kreckel  <kreckel@ginac.de>
785
786         * include/cln/modules.h (CL_JUMP_TO): Fix IA64 brokenness.
787
788 2001-01-28  Richard Kreckel  <kreckel@ginac.de>
789
790         * include/cln/number.h (cl_as_N): Remove bogus comment.
791
792 2001-01-22  Richard Kreckel  <kreckel@ginac.de>
793
794         * configure.in: Make build in separate builddir possible (again).
795
796 2001-01-22  Richard Kreckel  <kreckel@ginac.de>
797
798         * include/cln/*.h: Change signatures of all classes' methods
799           cln::cl_foo::operator new(size_t, cl_foo*) to
800           cln::cl_foo::operator new(size_t, void*) so one can declare
801           std::vector<cln::cl_foo>, std::list<cln::cl_foo> etc. for
802           certain STL implementations (like libstdc++-v3).
803
804 2000-12-14  Richard Kreckel  <kreckel@ginac.de>
805
806         * Version 1.1 released.
807
808 2000-12-13  Richard Kreckel  <kreckel@ginac.de>
809
810         * */*: cl_istream -> std::istream, cl_ostream -> std::ostream.
811
812 2000-12-05  Richard Kreckel  <kreckel@ginac.de>
813
814         * Makefile.in, src/Makefile.in, doc/Makefile.in: Use mkdir -p.
815         * include/cln/version.h.in, src/base/verion.cc: New files.
816         * configure.in: Generate include/cln/version.h.
817         * cln.m4: Rewrote it. Check result of cln-config without compiling.
818           Do cross-check library version and header version information.
819
820 2000-12-02  Christian Bauer  <cbauer@ginac.de>
821
822         * Makefile.in, src/Makefile.in, doc/Makefile.in: Added $DESTDIR.
823         * cln.m4, cln.spec.in: some minor fixes with respect to RPM package
824           building.
825
826 2000-11-24  Richard Kreckel  <kreckel@ginac.de>
827
828         * */*: Removed problematic stdin, stdout and stderr definitions.  
829           Use std::cin, std::cout, std::cerr instead (obsoletes 2000-10-29).
830
831 2000-11-20  Bruno Haible
832
833         * cln-config.1: change title.
834
835 2000-11-18  Richard Kreckel  <kreckel@ginac.de>
836
837         * cln.m4: New file.
838         * doc/cln.tex: Document package tools cln-config and cln.m4.
839         * Makefile.in: Care about cln.m4.
840
841 2000-11-17  Richard Kreckel  <kreckel@ginac.de>
842
843         * cln-config.1: added manpage, as required by a couple of distros.
844         * Makefile.in, doc/Makefile.in: target install depends on installdirs.
845
846 2000-11-16  Richard Kreckel  <kreckel@ginac.de>
847
848         * autoconf/aclocal.m4 (CL_GMP_SET_UINTD): New macro...
849         * configure.in: ...used here.
850         * include/cln/config.h.in: Put in macros defined by CL_GMP_SET_UINTD...
851         * include/cln/types.h: ...used here.
852         * autoconf/acgeneral.m4, autoconf/aclocal.m4: Adhere to FHS.
853
854 2000-11-13  Richard Kreckel  <kreckel@ginac.de>
855
856         * src/base/digitseq/cl_asm.h: Test if (intDsize==32) for MIPS and HPPA,
857           in order to guard against an accidented configuration.
858         * src/integer/conv/cl_I_to_digits.cc (table): member b_hoch_k of 
859           struct power_table_entry initialized as ULL instead of as UL, if 
860           intDsize==64 (caused misprinting on MIPS w/ GMP).
861         * src/base/cl_macros.h (minus_bitm, bitc): implemented ULL, if 
862           HAVE_FAST_LONGLONG.
863         * src/integer/bitwise/cl_I_mkfx.cc (mkf_extract): Likewise.
864         * src/integer/conv/cl_I_from_L.cc (cl_I_from_L): Added trival 
865           generation of Bignum for intDsize==64 and a notreached-check at end.
866         * autoconf/config.guess, autoconf/config.sub: updated from FSF.
867         * include/cln/config.h.in: Prepared support for IA64.
868         * include/cln/types.h: Likewise.
869         * include/cln/object.h: Likewise.
870         * include/cln/modules.h: Likewise.
871         * src/base/cl_macros.h (nonreturning_function): Likewise (NUE's 
872           compiler claims __GNUC_MINOR__==9).
873
874 2000-11-03  Richard Kreckel  <kreckel@ginac.de>
875
876         * src/base/cl_macros.h (bit, minus_bit): changed criterion for ULL from
877           HAVE_DD to HAVE_FAST_LONGLONG.
878         * src/base/cl_macros.h (bitm): implemented ULL, if HAVE_FAST_LONGLONG.
879         * src/base/cl_low.h: actually no need to include "cln/types.h" here.
880         * src/base/cl_low.h (logcount_64): always ULL, independent of HAVE_DD.
881         * src/base/random/cl_UL_random.cc (random32): a is always ULL.
882
883 2000-11-01  Richard Kreckel  <kreckel@ginac.de>
884
885         * include/cln/object.h (cl_combine): define additional signatures, if
886           HAVE_LONGLONG is defined, in order to keep the compiler happy.
887         * src/base/cl_macros.h: include "cln/types.h", since we need HAVE_DD...
888         * src/base/cl_macros.h (bit): ...for this macro...
889         * src/base/cl_macros.h (minus_bit): ...and this one.
890         * src/base/cl_low.h: include "cln/types.h", since we need HAVE_DD...
891         * src/base/cl_low.h (logcount_64): ...for this macro.
892         * src/base/random/cl_UL_random.cc (random32): if HAVE_DD a is an ULL.
893         * src/integer/gcd/cl_I_gcd_aux2.cc (floorDD): fixed algorithmic bug.
894           that turned up when intDsize==32 and cl_word_size==64.
895         * src/float/dfloat/elem/cl_DF_div.cc (operator/): fixed a missing cast
896           to uint64 that turned up when intDsize==32 and cl_word_size==64.
897
898 2000-10-29  Richard Kreckel  <kreckel@ginac.de>
899
900         * src/real/input/cl_R_read.cc, src/complex/input/cl_N_read.cc:
901           #undef stderr, if it's defined so cln::stderr isn't confused.
902         * src/base/input/cl_read_globals.cc: stdin should not be extern.
903
904 2000-09-05  Richard Kreckel  <kreckel@ginac.de>
905
906         * include/cln/number.h (As): Fix it in namespace by suffixing `_As'
907           to the appropiate method instead of prefixing `as_'.
908         * src/complex/misc/cl_N_as.cc (cl_N_As): Likewise.
909         * src/real/misc/cl_R_as.cc (cl_R_As): Likewise.
910         * src/rational/misc/cl_RA_as.cc (cl_RA_As): Likewise.
911         * src/integer/misc/cl_I_as.cc (cl_I_As): Likewise.
912         * src/float/misc/cl_F_as.cc (cl_F_As): Likewise.
913         * src/float/sfloat/misc/cl_SF_as.cc (cl_SF_As): Likewise.
914         * src/float/lfloat/misc/cl_LF_as.cc (cl_LF_As): Likewise.
915         * src/float/ffloat/misc/cl_FF_as.cc (cl_FF_As): Likewise.
916         * src/float/dfloat/misc/cl_DF_as.cc (cl_DF_As): Likewise.
917
918 2000-09-05  Richard Kreckel  <kreckel@ginac.de>
919
920         * src/complex/transcendental/cl_C_expt_C.cc (expt): fix logic for
921           the 0^y cases.
922
923 2000-08-30  Richard Kreckel  <kreckel@ginac.de>
924
925         * include/cln/number.h, cl_number::_as_cl_private_thing(): removed.
926           Rearranged for a clearer distinction between macros and remaining
927           identifiers, so Cint can parse it smoothly.
928
929 2000-08-29  Richard Kreckel  <kreckel@ginac.de>
930
931         * include/cln/number.h, the(const cl_number& x): New template
932           function.
933
934 2000-08-29  Richard Kreckel  <kreckel@ginac.de>
935
936         * */*: Pushed CLN into a namespace `cln'.  While doing so, the
937           following identifiers got their poor-man's namespace (i.e. the 
938           prefix `cl_') stripped off:
939           cl_catalanconst()        ->  catalanconst()
940           cl_compare()             ->  compare()
941           cl_cos_sin()             ->  cos_sin()
942           cl_cos_sin_t             ->  cos_sin_t
943           cl_cosh_sinh()           ->  cosh_sinh()
944           cl_cosh_sinh_t           ->  cosh_sinh_t
945           cl_decoded_dfloat        ->  decoded_dfloat
946           cl_decoded_ffloat        ->  decoded_ffloat
947           cl_decoded_float         ->  decoded_float
948           cl_decoded_lfloat        ->  decoded_lfloat
949           cl_decoded_sfloat        ->  decoded_sfloat
950           cl_default_float_format  ->  default_float_format
951           cl_default_print_flags   ->  default_print_flags
952           cl_default_random_state  ->  default_random_state
953           cl_double_approx()       ->  double_approx()
954           cl_equal()               ->  equal()
955           cl_equal_hashcode()      ->  equal_hashcode()
956           cl_eulerconst()          ->  eulerconst()
957           cl_find_modint_ring()    ->  find_modint_ring()
958           cl_find_univpoly_ring()  ->  find_univ_poly_ring()
959           cl_float_approx()        ->  float_approx
960           cl_float_format()        ->  float_format()
961           cl_float_format_t        ->  float_format_t
962           cl_free_hook()           ->  free_hook()
963           cl_hermite()             ->  hermite()
964           cl_laguerre()            ->  laguerre()
965           cl_legendre()            ->  legandre()
966           cl_malloc_hook()         ->  malloc_hook()
967           cl_pi()                  ->  pi()
968           cl_tschebychev()         ->  tschebychev()
969           cl_zeta()                ->  zeta()
970           NB: For functions these changes includes all signatures.
971         * include/*: moved to include/cln/*, stripped `cl_' off filenames.
972         * cln-config.in: new file.
973
974 2000-08-26  Bruno Haible  <haible@clisp.cons.org>
975
976         * autoconf/acgeneral.m4 (AC_OUTPUT): Use braces in exec_prefix default
977           value, not parens.
978
979 2000-08-18  Bruno Haible  <haible@clisp.cons.org>
980
981         * include/cl_univpoly_modint.h: Fix typo.
982
983 2000-07-13  Bruno Haible  <haible@clisp.cons.org>
984
985         * src/float/input/cl_F_read_stream.cc (number_char_p): Accept '_',
986           used as precision marker for floats.
987           Reported by Keith Briggs (in 1998) and Thomas Roessler.
988         * src/integer/input/cl_I_read_stream.cc (number_char_p): Likewise.
989         * src/rational/input/cl_RA_read_stream.cc (number_char_p): Likewise.
990         * src/real/input/cl_R_read_stream.cc (number_char_p): Likewise.
991         * src/complex/input/cl_N_read_stream.cc (number_char_p): Likewise.
992
993 2000-06-22  Bruno Haible  <haible@clisp.cons.org>
994
995         * include/cl_object.h: Rename cl_word_size to cl_pointer_size.
996         * include/cl_types.h (HAVE_FAST_LONGLONG): Also define on Irix6
997           with N32 ABI.
998           (cl_word_size): New macro.
999         * src/float/sfloat/cl_SF.h: Use cl_pointer_size instead of
1000           cl_word_size.
1001
1002 2000-05-31  Bruno Haible  <haible@clisp.cons.org>
1003
1004         * tests/exam_I_floor.cc (integer_floor_tests): Add one more entry.
1005           From a sample that fails with gcc-2.95.2 on Sparc.
1006         * tests/exam_I_gcd.cc (integer_gcd_tests): Likewise.
1007
1008 2000-05-30  Richard Kreckel  <kreckel@ginac.de>
1009
1010         * configure.in, autoconf/aclocal.m4 (CL_GMP_H_VERSION, CL_GMP_CHECK):
1011           New macros.
1012         * configure.in, include/cl_config.h.in (CL_VERSION, CL_VERSION_MINOR,
1013           CL_VERSION_PATCHLEVEL): New definitions.
1014         * autoconf/config.guess, autoconf/config.sub, autoconf/ltconfig,
1015           autoconf/ltmain.sh: updated from FSF (libtool 1.3.5, etc).
1016         * src/Makefile.in, configure.in: release-variables renamed from 
1017           CLN_* to CL_*.
1018         * configure.in: default to build both shared and static library
1019           (i.e. default to the most common behaviour).
1020
1021 2000-05-29  Richard Kreckel  <kreckel@ginac.de>
1022
1023         * autoconf/aclocal.m4 (CL_CANONICAL_HOST): Added missing changequote
1024           environment around the patch of 2000-05-23.
1025
1026 2000-05-29  Bruno Haible  <haible@clisp.cons.org>
1027
1028         * autoconf/aclocal.m4 (CL_PROG_INSTALL): Fix typo.
1029           Reported by Thomas Klausner <wiz@danbala.ifoer.tuwien.ac.at>.
1030
1031 2000-05-27  Richard Kreckel  <kreckel@ginac.de>
1032
1033         * src/float/lfloat/algebraic/cl_LF_sqrt.cc,
1034           src/base/digitseq/cl_DS_sqrt.cc: Readjusted break-even points.
1035
1036 2000-05-24  Richard Kreckel  <kreckel@ginac.de>
1037
1038         * autoconf/config.*: Updated to new version from FSF
1039           (the new libtool wants this).
1040         * src/Makefile.in: added $(LDFLAGS) to link step.
1041         * src/base/digitseq/cl_2DS_div.cc, cl_2DS_recip.cc: Readjusted 
1042           break-even points.
1043
1044 2000-05-23  Bruno Haible  <haible@clisp.cons.org>
1045
1046         * autoconf/aclocal.m4 (CL_CANONICAL_HOST): Determine host_cpu,
1047           host_vendor, host_os correctly if $host has more than two hyphens.
1048
1049 2000-05-19  Richard Kreckel  <kreckel@ginac.de>
1050
1051         * src/base/digitseq/cl_DS_mul.cc: Rearranged break-even points to
1052           better match present-day CPUs whenever GMP3 is used.
1053         * src/base/digitseq/cl_DS_div.cc: dto. 
1054         * src/TUNING: Added comment about order of tuning.
1055         * configure, configure.in: Safer GMP3-detection.
1056         * INSTALL.generic: Clarified behaviour of --with-gmp.
1057         * autoconf/config.guess: updated from Clisp-2000-03-06 sources.
1058
1059 2000-05-04  Richard Kreckel  <kreckel@ginac.de>
1060
1061         * gmp/: removed completely.  From now on we are going to link
1062           externally against libgmp3.0 or above!
1063         * configure, configure.in, Makefile.in, */Makeflags: removed support
1064           of internal gmp 2.0.2, like $GMP_INCLUDES, which should be done by
1065           setting $CPPFLAGS instead.
1066         * Makefile.in: Added libtool inter-library dependency for -lgmp and
1067           conforming interface versioning (-version-info).
1068         * autoconf/ltconfig, autoconf/ltmain.sh: Updated to newer versions
1069           from libtool 1.3.4.
1070
1071 2000-02-22  Bruno Haible  <haible@clisp.cons.org>
1072
1073         * src/base/digitseq/cl_asm_mips_.cc: Starting at argument 5 the
1074           parameter passing was changed for the MIPS n32 ABI.
1075
1076 2000-01-24  Richard Kreckel  <kreckel@ginac.de>
1077
1078         * gmp/*: Replaced the complete mpn sources with the ones from
1079           Debian since they are maintained while the ones from FSF
1080           aren't and there were problems on some architectures, PowerPC
1081           in particular.  See the file gmp/README.CLN.  This way the
1082           hard links in this directory have vanished, they were causing
1083           trouble for people working in AFS.  This became necessary for
1084           Debian, because there it woudn't compile on PPC.
1085
1086 2000-01-13  Richard Kreckel  <kreckel@ginac.de>
1087
1088         * Version 1.0.3 released.
1089
1090 2000-01-13  Richard Kreckel  <kreckel@ginac.de>
1091
1092         * src/base/cl_macros.h (nonreturning_function): For egcs and newer
1093           use __attribute__ ((__noreturn__)) instead of the __volatile__
1094           storage class.
1095           (nonreturning): Remove macro.
1096         * include/*: Minor fixes to stop -ansi -pedantic from complaining.
1097         * include/cl_integer.h: (doublefactorial): New declaration.
1098         * src/integer/misc/combin/cl_I_doublefactorial.cc: New file.
1099
1100 1999-12-18  Bruno Haible  <haible@clisp.cons.org>
1101
1102         * autoconf/acgeneral.m4 (AC_ARG_PROGRAM): Create conftestsed using
1103           "cat", not "echo".
1104         * autoconf/ltconfig: Improve support for recent FreeBSD 3.
1105         * include/cl_GV.h (cl_GV_vectorops): Change return type from 'T' to
1106           'const T', to match definition in src/vector/cl_GV_number.cc.
1107           Reported by Duncan Simpson <dps@io.stargate.co.uk>.
1108         * gmp/mpn/Makefile.in (.S.lo): Use 'if', not '&&', because '&&' may
1109           yield return code 1.
1110
1111 1999-09-07  Bruno Haible  <haible@clisp.cons.org>
1112
1113         * Version 1.0.2 released.
1114
1115 1999-09-06  Bruno Haible  <haible@clisp.cons.org>
1116
1117         * src/rational/cl_RA.h (integerp, ratiop): Instead of returning a
1118           boolean expression, write alternatives ending with either
1119           "return cl_true;" or "return cl_false;". This way, g++ does a
1120           better job inlining it.
1121         * src/float/cl_F.h (longfloatp): Likewise.
1122         * src/real/cl_R.h (rationalp, integerp, floatp): Likewise.
1123         * src/complex/cl_C.h (realp, complexp): Likewise.
1124
1125 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1126
1127         * include/cl_integer.h (cl_equal_hashcode): New declaration.
1128         * include/cl_rational.h (cl_equal_hashcode): New declaration.
1129         * include/cl_sfloat.h (cl_equal_hashcode): New declaration.
1130         * include/cl_ffloat.h (cl_equal_hashcode): New declaration.
1131         * include/cl_dfloat.h (cl_equal_hashcode): New declaration.
1132         * include/cl_lfloat.h (cl_equal_hashcode): New declaration.
1133         * include/cl_float.h (cl_equal_hashcode): New declaration.
1134         * include/cl_real.h (cl_equal_hashcode): New declaration.
1135         * include/cl_complex.h (cl_equal_hashcode): New declaration.
1136         * src/base/cl_N.h (equal_hashcode_low, equal_hashcode_one): New macros.
1137         * src/integer/misc/cl_I_eqhashcode.cc: New file.
1138         * src/rational/misc/cl_RA_eqhashcode.cc: New file.
1139         * src/float/sfloat/misc/cl_SF_eqhashcode.cc: New file.
1140         * src/float/ffloat/misc/cl_FF_eqhashcode.cc: New file.
1141         * src/float/dfloat/misc/cl_DF_eqhashcode.cc: New file.
1142         * src/float/lfloat/misc/cl_LF_eqhashcode.cc: New file.
1143         * src/float/misc/cl_F_eqhashcode.cc: New file.
1144         * src/real/misc/cl_R_eqhashcode.cc: New file.
1145         * src/complex/misc/cl_C_eqhashcode.cc: New file.
1146         * doc/cln.tex: Document `cl_equal_hashcode'.
1147
1148 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1149
1150         * include/cl_ring.h (cl_number_ring_ops): Add `contains' member.
1151           (cl_number_ring): New class.
1152           (cl_specialized_number_ring<T>): Inherit from cl_number_ring.
1153           (instanceof): New function.
1154         * src/integer/ring/cl_I_ring.cc (cl_I_p): New function.
1155         * src/integer/misc/cl_I_as.cc (cl_I_p): Add comment.
1156         * src/rational/ring/cl_RA_ring.cc (cl_RA_p): New function.
1157         * src/rational/misc/cl_RA_as.cc (cl_RA_p): Add comment.
1158         * src/real/ring/cl_R_ring.cc (cl_R_p): New function.
1159         * src/real/misc/cl_R_as.cc (cl_R_p): Add comment.
1160         * src/complex/ring/cl_C_ring.cc (cl_N_p): New function.
1161         * src/complex/misc/cl_N_as.cc (cl_N_p): Add comment.
1162         * doc/cln.tex: Document `instanceof'.
1163
1164 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1165
1166         * include/cl_rational.h (numerator, denominator): New declarations.
1167         * src/rational/elem/cl_RA_numerator.cc: New file.
1168         * src/rational/elem/cl_RA_denominator.cc: New file.
1169         * include/cl_integer.h (numerator, denominator): New inline functions.
1170         * doc/cln.tex: Document `numerator' and `denominator'.
1171
1172 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1173
1174         * src/rational/algebraic/cl_RA_rootp.cc (rootp): Fix endless loop
1175           in the integer case.
1176         * src/rational/algebraic/cl_RA_rootp_I.cc (rootp): Likewise.
1177
1178 1999-09-05  Bruno Haible  <haible@clisp.cons.org>
1179
1180         * include/cl_config.h.in: Support for sparc64 CPU.
1181         * include/cl_modules.h: Likewise.
1182         * include/cl_types.h: Likewise.
1183         * include/cl_object.h: Likewise.
1184         * include/cl_GV.h: Likewise.
1185         * src/Makefile.in: Likewise.
1186         * src/base/cl_alloca.h: Likewise.
1187         * src/base/cl_macros.h: Likewise.
1188         * src/base/cl_sysdep.h: Likewise.
1189         * src/base/cl_low.h: Likewise.
1190         * src/base/digitseq/cl_asm.h: Likewise.
1191         * src/base/digitseq/cl_asm_.cc: Likewise.
1192         * src/base/digitseq/cl_asm_sparc64.h: New file.
1193         * src/base/digitseq/cl_asm_sparc64_.cc: New file.
1194         * src/modinteger/cl_MI_int.h: Support for sparc64 CPU.
1195         * src/polynomial/elem/cl_UP_no_ring.cc: Likewise.
1196         * src/polynomial/elem/cl_UP_GF2.h: Likewise.
1197         * src/polynomial/elem/cl_asm_GF2.cc: Likewise.
1198
1199 1999-09-04  Bruno Haible  <haible@clisp.cons.org>
1200
1201         * src/base/digitseq/cl_asm_sparc_.cc (orc2_loop_up, orc2_loop_down):
1202           Use the `orn' instruction.
1203
1204 1999-08-14  Bruno Haible  <haible@clisp.cons.org>
1205
1206         Assume all platforms have <stdlib.h> and clock_t.
1207         * configure.in: Don't call CL_STDLIB_H and CL_CLOCK_T.
1208         * src/base/cl_base_config.h.in (CLOCK_T): Remove definition.
1209         * src/base/random/cl_random_from.cc: Use clock_t instead of CLOCK_T.
1210         * src/timing/cl_t_config.h.in (CLOCK_T): Remove definition.
1211         * src/timing/cl_t_current2.cc: Use clock_t instead of CLOCK_T.
1212
1213 1999-07-18  Bruno Haible  <haible@clisp.cons.org>
1214
1215         * gmp/config.guess: Link to autoconf/config.guess.
1216           gmp/config.sub: Link to autoconf/config.sub.
1217           Needed for Win32 platforms.
1218
1219 1999-07-17  Bruno Haible  <haible@clisp.cons.org>
1220
1221         * autoconf/aclocal.m4 (CL_CANONICAL_HOST_CPU): Distinguish "sparc" and
1222           "sparc64" according to the C compiler, not the uname result.
1223
1224 1999-06-17  Bruno Haible  <haible@clisp.cons.org>
1225
1226         * src/base/digitseq/cl_asm_sparc_.cc (compare_loop_up): Fix
1227           COUNTER_LOOPS version, fortunately not used yet.
1228
1229         * include/cl_modules.h: Prepare for gcc version 3.
1230
1231 1999-06-12  Bruno Haible  <haible@clisp.cons.org>
1232
1233         * src/rational/elem/cl_RA_plus.cc, src/rational/elem/cl_RA_minus.cc:
1234           Change the last call from I_I_to_RT to I_I_to_RA.
1235
1236 1999-06-09  Bruno Haible  <haible@clisp.cons.org>
1237
1238         * Version 1.0.1 released.
1239
1240 1999-06-09  Bruno Haible  <haible@clisp.cons.org>
1241
1242         * src/integer/cl_I.h (pFN_maxlength_digits_at): Define also when
1243           intDsize==64.
1244
1245 1999-06-08  Bruno Haible  <haible@clisp.cons.org>
1246
1247         * autoconf/intparam.c (printf_underscored): Change argument type to
1248           `const char*'.
1249         * include/cl_modules.h (CL_OUTPUT_LABEL, CL_JUMP_TO): New macros.
1250           (CL_PROVIDE, CL_PROVIDE_END): Use them.
1251         * include/cl_string.h (cl_heap_string::operator new): Return 1, not 0.
1252         * include/cl_GV.h (cl_GV_inner<T>::operator new): Likewise.
1253         * src/base/ring/cl_no_ring.cc (dummy_canonhom, dummy_expt_pos): Don't
1254           cast a cl_I to void here. Works around a bug in g++-2.95.
1255         * src/complex/misc/cl_C_class.cc: Include "cl_C.h".
1256         * src/polynomial/elem/cl_UP_no_ring.cc (dummy_canonhom,
1257            dummy_expt_pos): Don't cast a cl_I to void here. Works around a bug
1258           in g++-2.95.
1259         * src/polynomial/elem/cl_asm_sparc_GF2.cc (DECLARE_FUNCTION): New
1260           macro.
1261         * src/rational/misc/cl_RA_class.cc: Include "cl_RA.h".
1262         * src/vector/cl_GV_I.cc (cl_heap_GV_I_general::operator new,
1263            DEFINE_cl_heap_GV_I_bits): Return 1, not 0.
1264         * src/vector/cl_GV_number.cc (cl_heap_GV_number_general::operator new):
1265           Likewise.
1266
1267 1999-06-01  Bruno Haible  <haible@clisp.cons.org>
1268
1269         * autoconf/aclocal.m4 (CL_CANONICAL_HOST_CPU): Canonicalize alpha
1270           variants to alpha.
1271
1272 1999-05-29  Bruno Haible  <haible@clisp.cons.org>
1273
1274         * src/base/digitseq/cl_asm_i386_.cc (DECLARE_FUNCTION): Treat
1275           OpenBSD like NetBSD.
1276         * src/base/digitseq/cl_asm_sparc_.cc (DECLARE_FUNCTION): Likewise.
1277         * src/base/digitseq/cl_asm_m68k_.cc (DECLARE_FUNCTION): Treat
1278           OpenBSD like NetBSD, and Linux/ELF like SVR4.
1279
1280 1999-05-16  Bruno Haible  <haible@clisp.cons.org>
1281
1282         * src/base/cl_low.h (integerlength32) [__rs6000__]: Use old assembler
1283           syntax on AIX systems and new assembler syntax on non-AIX systems.
1284
1285 1999-05-01  Bruno Haible  <haible@clisp.cons.org>
1286
1287         * autoconf/config.guess, autoconf/config.sub: Upgrade to newest
1288           version from GNU CVS.
1289
1290 1999-04-24  Bruno Haible  <haible@clisp.cons.org>
1291
1292         * src/integer/bitwise/cl_I_logand.cc (logand): Optimize the case when
1293           either operand is a positive fixnum, O(1) instead of O(N).
1294         * src/integer/bitwise/cl_I_lognand.cc (lognand): Likewise.
1295         * src/integer/bitwise/cl_I_logandc2.cc (logandc2): Likewise for the
1296           first operand.
1297
1298 1999-04-14  Bruno Haible  <haible@clisp.cons.org>
1299
1300         * autoconf/aclocal.m4 (CL_GLOBAL_CONSTRUCTORS): Add check whether
1301           ctor/dtor needs to be exported.
1302         * include/cl_config.h.in (CL_NEED_GLOBALIZE_CTORDTOR): New macro.
1303         * include/cl_modules.h (CL_GLOBALIZE_JUMP_LABEL): Renamed from
1304           CL_GLOBALIZE_LABEL.
1305           (CL_GLOBALIZE_LABEL): New macro.
1306           (CL_GLOBALIZE_CTORDTOR_LABEL): Renamed from CL_GLOBALIZE_ASM_LABEL.
1307           (CL_PROVIDE): Update.
1308
1309 1999-04-12  Bruno Haible  <haible@clisp.cons.org>
1310
1311         * src/Makefile.in ($(ASMFILES_S)): On HPPA, ignore preprocessing
1312           errors ("unterminated string or character constant").
1313           ($(ASMFILES_LO)): On HPPA, try with various settings of
1314           COMPILER_PATH, in order to try /usr/ccs/bin/as and /bin/as.
1315
1316 1999-04-11  Bruno Haible  <haible@clisp.cons.org>
1317
1318         * INSTALL: Mention gmp problems on MIPS.
1319         * doc/cln.tex: Likewise.
1320
1321 1999-03-24  Mumit Khan  <khan@xraylith.wisc.edu>
1322
1323         * src/Makefile.in (SUBDIRS): Filter out CVS and RCS directories from
1324           the source tree.
1325         * include/cl_modules.h (CL_GLOBALIZE_LABEL): Define for Win32.
1326           (CL_GLOBALIZE_ASM_LABEL): New macro.
1327           (CL_PROVIDE): Use it.
1328         * src/base/random/cl_random_from.cc: Handle WIN32.
1329         * src/timing/cl_t_current.cc: Likewise.
1330
1331 1999-03-15  Bruno Haible  <haible@clisp.cons.org>
1332
1333         * autoconf/intparam.c (main7): Use %lX instead of %X for a `long'.
1334           (main8): Adapt for C++.
1335
1336 1999-03-09  Bruno Haible  <haible@clisp.cons.org>
1337
1338         * INSTALL: Mention egcs-1.1 problems on Sparc.
1339         * doc/cln.tex: Likewise.
1340
1341 1999-03-08  Bruno Haible  <haible@clisp.cons.org>
1342
1343         * autoconf/aclocal.m4 (CL_FPU_CONTROL): Fix the "checking for"
1344           messages.
1345
1346 1999-02-25  Bruno Haible  <haible@clisp.cons.org>
1347
1348         * autoconf/aclocal.m4: In test programs, declare `int main()', not
1349           `main()'.
1350
1351         * lidia-interface/src/interfaces/integers/cln/bigint.c
1352           (bigint_to_string): Fix for negative arguments.
1353
1354         * src/base/cl_low.h: Check for `__sparc__', not `SPARC'.
1355         * src/base/cl_alloca.h: Likewise.
1356
1357         * src/base/cl_low.h: Eliminate CLISP style "# " comments.
1358         * src/base/digitseq/cl_asm_arm_.cc,
1359           src/base/digitseq/cl_asm_mips_.cc,
1360           src/float/dfloat/elem/cl_DF_mul.cc: Likewise.
1361
1362         * src/modinteger/cl_MI_pow2.h,
1363           src/modinteger/cl_MI_pow2m1.h,
1364           src/modinteger/cl_MI_pow2p1.h: Workaround g++-2.7.2 inlining bug.
1365
1366 1999-01-18  Bruno Haible  <haible@clisp.cons.org>
1367
1368         * autoconf/acgeneral.m4,
1369           autoconf/acspecific.m4: Upgrade to autoconf-2.13.
1370         * autoconf/config.guess, autoconf/config.sub: Likewise.
1371         * autoconf/aclocal.m4 (CL_ALLOCA): Test for _MSC_VER and alloca being
1372           a macro. Use ${ac_objext}.
1373         * src/base/cl_macros.h (alloca): Put _MSC_VER test before the others,
1374           conforming with CL_ALLOCA.
1375
1376 1999-01-12  Bruno Haible  <haible@clisp.cons.org>
1377
1378         * Version 1.0 released.
1379