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