1 2001-11-03 Richard Kreckel <kreckel@ginac.de>
3 * src/float/transcendental/cl_F_sinx.cc (sinx_naive): For small
4 values of x, return square(x) instead of x.
5 * src/float/transcendental/cl_F_sinhx.cc (sinhx_naive): Likewise.
7 2001-07-25 Richard Kreckel <kreckel@ginac.de>
9 * Version 1.1.2 released.
11 2001-07-24 Richard Kreckel <kreckel@ginac.de>
13 * src/base/hash/cl_hash.h: declare _cl_hashtable_iterator<htentry> a
14 friend of cl_heap_hashtable<htentry>.
16 2001-07-22 Richard Kreckel <kreckel@ginac.de>
18 * src/float/base/cl_ieee.cc: try to do magic to the FPU only if
19 _FPU_IEEE is really defined.
20 * include/cln/modules.h: change assembler labels from `label' to
21 `label:' on hppa, needed by Linux (see comment).
22 * autoconf/acinclude.m4: new file (for storing CLN-specific macros).
23 * Makefile.devel: adjusted.
24 * autoconf/aclocal.m4: regenerate.
25 * src/base/low/cl_low_mul.cc: moved POD variables that are declared
26 extern "C" elsewhere out of the namespace.
27 * src/base/low/cl_low_div.cc: Likewise.
29 2001-06-08 Bruno Haible <haible@clisp.cons.org>
31 * autoconf/config.{guess,sub}: Update to GNU version 2001-05-11.
32 * autoconf/aclocal.m4: Upgrade to libtool-1.4.
33 * autoconf/ltmain.sh: Likewise.
34 * autoconf/ltconfig: Remove file.
35 * autoconf/install-sh: New file.
36 * configure.in: Add AC_CONFIG_AUX_DIR call.
38 2001-06-05 Richard Kreckel <kreckel@ginac.de>
40 * tests/tests.cc: resolve namespace ambiguity about strcmp().
42 2001-05-31 Richard Kreckel <kreckel@ginac.de>
44 * Version 1.1.1 released.
46 2001-05-28 Richard Kreckel <kreckel@ginac.de>
48 * cln/cln.tex: documented problems with shared library on Sparc
49 using gcc older than 2.95.3.
50 * configure.in: Fixed typos in versioning docu.
52 2001-05-25 Bruno Haible <haible@clisp.cons.org>
54 * src/base/digitseq/cl_asm_arm_.cc: Use #0x instead of #& to designate
55 hexadecimal constants.
57 2001-05-25 Richard Kreckel <kreckel@ginac.de>
59 * autoconf/floatparam.c (double_wordorder_bigendian_p): new symbol.
60 * src/float/dfloat/cl_DF.h: Check for double_wordorder_bigendian_p.
61 * Removed LiDIA interface since that is now outdated (namespace cln)
62 and maintained elsewhere.
63 * Adjusted dates and final touches for 1.1.1.
65 2001-05-19 Richard Kreckel <kreckel@ginac.de>
67 * INSTALL: Update toolchain info: no egcs, some more platforms.
68 * doc/cln.tex: Likewise.
70 2001-05-18 Richard Kreckel <kreckel@ginac.de>
72 * src/base/cl_low.h: prepended variables declared inside macros
73 with underscore. Fixes equal_hashcode() on various platforms.
75 2001-04-25 Richard Kreckel <kreckel@ginac.de>
77 * src/base/cl_low.h: Added several checks if NO_ASM is defined, so
78 this definition becomes actually useful. This is needed for
79 compilation on Arm until somebody fixes the assembler files for Arm.
80 * src/base/digitseq/cl_asm.h: Likewise.
81 * src/base/digitseq/cl_asm_.cc: Likewise.
82 * */Makefile.in: Added `override' in front of `CPPFLAGS +=' so
83 one can say `make CPPFLAGS=-DFOOBAR'.
85 2001-03-26 Arvid Norberg <c99ang@cs.umu.se>
87 * src/base/random/cl_random_from.cc: ported to beos.
89 2001-03-05 Richard Kreckel <kreckel@ginac.de>
91 * include/cln/modules.h (CL_JUMP_TO): Fix IA64 brokenness.
93 2001-01-28 Richard Kreckel <kreckel@ginac.de>
95 * include/cln/number.h (cl_as_N): Remove bogus comment.
97 2001-01-22 Richard Kreckel <kreckel@ginac.de>
99 * configure.in: Make build in separate builddir possible (again).
101 2001-01-22 Richard Kreckel <kreckel@ginac.de>
103 * include/cln/*.h: Change signatures of all classes' methods
104 cln::cl_foo::operator new(size_t, cl_foo*) to
105 cln::cl_foo::operator new(size_t, void*) so one can declare
106 std::vector<cln::cl_foo>, std::list<cln::cl_foo> etc. for
107 certain STL implementations (like libstdc++-v3).
109 2000-12-14 Richard Kreckel <kreckel@ginac.de>
111 * Version 1.1 released.
113 2000-12-13 Richard Kreckel <kreckel@ginac.de>
115 * */*: cl_istream -> std::istream, cl_ostream -> std::ostream.
117 2000-12-05 Richard Kreckel <kreckel@ginac.de>
119 * Makefile.in, src/Makefile.in, doc/Makefile.in: Use mkdir -p.
120 * include/cln/version.h.in, src/base/verion.cc: New files.
121 * configure.in: Generate include/cln/version.h.
122 * cln.m4: Rewrote it. Check result of cln-config without compiling.
123 Do cross-check library version and header version information.
125 2000-12-02 Christian Bauer <cbauer@ginac.de>
127 * Makefile.in, src/Makefile.in, doc/Makefile.in: Added $DESTDIR.
128 * cln.m4, cln.spec.in: some minor fixes with respect to RPM package
131 2000-11-24 Richard Kreckel <kreckel@ginac.de>
133 * */*: Removed problematic stdin, stdout and stderr definitions.
134 Use std::cin, std::cout, std::cerr instead (obsoletes 2000-10-29).
136 2000-11-20 Bruno Haible
138 * cln-config.1: change title.
140 2000-11-18 Richard Kreckel <kreckel@ginac.de>
143 * doc/cln.tex: Document package tools cln-config and cln.m4.
144 * Makefile.in: Care about cln.m4.
146 2000-11-17 Richard Kreckel <kreckel@ginac.de>
148 * cln-config.1: added manpage, as required by a couple of distros.
149 * Makefile.in, doc/Makefile.in: target install depends on installdirs.
151 2000-11-16 Richard Kreckel <kreckel@ginac.de>
153 * autoconf/aclocal.m4 (CL_GMP_SET_UINTD): New macro...
154 * configure.in: ...used here.
155 * include/cln/config.h.in: Put in macros defined by CL_GMP_SET_UINTD...
156 * include/cln/types.h: ...used here.
157 * autoconf/acgeneral.m4, autoconf/aclocal.m4: Adhere to FHS.
159 2000-11-13 Richard Kreckel <kreckel@ginac.de>
161 * src/base/digitseq/cl_asm.h: Test if (intDsize==32) for MIPS and HPPA,
162 in order to guard against an accidented configuration.
163 * src/integer/conv/cl_I_to_digits.cc (table): member b_hoch_k of
164 struct power_table_entry initialized as ULL instead of as UL, if
165 intDsize==64 (caused misprinting on MIPS w/ GMP).
166 * src/base/cl_macros.h (minus_bitm, bitc): implemented ULL, if
168 * src/integer/bitwise/cl_I_mkfx.cc (mkf_extract): Likewise.
169 * src/integer/conv/cl_I_from_L.cc (cl_I_from_L): Added trival
170 generation of Bignum for intDsize==64 and a notreached-check at end.
171 * autoconf/config.guess, autoconf/config.sub: updated from FSF.
172 * include/cln/config.h.in: Prepared support for IA64.
173 * include/cln/types.h: Likewise.
174 * include/cln/object.h: Likewise.
175 * include/cln/modules.h: Likewise.
176 * src/base/cl_macros.h (nonreturning_function): Likewise (NUE's
177 compiler claims __GNUC_MINOR__==9).
179 2000-11-03 Richard Kreckel <kreckel@ginac.de>
181 * src/base/cl_macros.h (bit, minus_bit): changed criterion for ULL from
182 HAVE_DD to HAVE_FAST_LONGLONG.
183 * src/base/cl_macros.h (bitm): implemented ULL, if HAVE_FAST_LONGLONG.
184 * src/base/cl_low.h: actually no need to include "cln/types.h" here.
185 * src/base/cl_low.h (logcount_64): always ULL, independent of HAVE_DD.
186 * src/base/random/cl_UL_random.cc (random32): a is always ULL.
188 2000-11-01 Richard Kreckel <kreckel@ginac.de>
190 * include/cln/object.h (cl_combine): define additional signatures, if
191 HAVE_LONGLONG is defined, in order to keep the compiler happy.
192 * src/base/cl_macros.h: include "cln/types.h", since we need HAVE_DD...
193 * src/base/cl_macros.h (bit): ...for this macro...
194 * src/base/cl_macros.h (minus_bit): ...and this one.
195 * src/base/cl_low.h: include "cln/types.h", since we need HAVE_DD...
196 * src/base/cl_low.h (logcount_64): ...for this macro.
197 * src/base/random/cl_UL_random.cc (random32): if HAVE_DD a is an ULL.
198 * src/integer/gcd/cl_I_gcd_aux2.cc (floorDD): fixed algorithmic bug.
199 that turned up when intDsize==32 and cl_word_size==64.
200 * src/float/dfloat/elem/cl_DF_div.cc (operator/): fixed a missing cast
201 to uint64 that turned up when intDsize==32 and cl_word_size==64.
203 2000-10-29 Richard Kreckel <kreckel@ginac.de>
205 * src/real/input/cl_R_read.cc, src/complex/input/cl_N_read.cc:
206 #undef stderr, if it's defined so cln::stderr isn't confused.
207 * src/base/input/cl_read_globals.cc: stdin should not be extern.
209 2000-09-05 Richard Kreckel <kreckel@ginac.de>
211 * include/cln/number.h (As): Fix it in namespace by suffixing `_As'
212 to the appropiate method instead of prefixing `as_'.
213 * src/complex/misc/cl_N_as.cc (cl_N_As): Likewise.
214 * src/real/misc/cl_R_as.cc (cl_R_As): Likewise.
215 * src/rational/misc/cl_RA_as.cc (cl_RA_As): Likewise.
216 * src/integer/misc/cl_I_as.cc (cl_I_As): Likewise.
217 * src/float/misc/cl_F_as.cc (cl_F_As): Likewise.
218 * src/float/sfloat/misc/cl_SF_as.cc (cl_SF_As): Likewise.
219 * src/float/lfloat/misc/cl_LF_as.cc (cl_LF_As): Likewise.
220 * src/float/ffloat/misc/cl_FF_as.cc (cl_FF_As): Likewise.
221 * src/float/dfloat/misc/cl_DF_as.cc (cl_DF_As): Likewise.
223 2000-09-05 Richard Kreckel <kreckel@ginac.de>
225 * src/complex/transcendental/cl_C_expt_C.cc (expt): fix logic for
228 2000-08-30 Richard Kreckel <kreckel@ginac.de>
230 * include/cln/number.h, cl_number::_as_cl_private_thing(): removed.
231 Rearranged for a clearer distinction between macros and remaining
232 identifiers, so Cint can parse it smoothly.
234 2000-08-29 Richard Kreckel <kreckel@ginac.de>
236 * include/cln/number.h, the(const cl_number& x): New template
239 2000-08-29 Richard Kreckel <kreckel@ginac.de>
241 * */*: Pushed CLN into a namespace `cln'. While doing so, the
242 following identifiers got their poor-man's namespace (i.e. the
243 prefix `cl_') stripped off:
244 cl_catalanconst() -> catalanconst()
245 cl_compare() -> compare()
246 cl_cos_sin() -> cos_sin()
247 cl_cos_sin_t -> cos_sin_t
248 cl_cosh_sinh() -> cosh_sinh()
249 cl_cosh_sinh_t -> cosh_sinh_t
250 cl_decoded_dfloat -> decoded_dfloat
251 cl_decoded_ffloat -> decoded_ffloat
252 cl_decoded_float -> decoded_float
253 cl_decoded_lfloat -> decoded_lfloat
254 cl_decoded_sfloat -> decoded_sfloat
255 cl_default_float_format -> default_float_format
256 cl_default_print_flags -> default_print_flags
257 cl_default_random_state -> default_random_state
258 cl_double_approx() -> double_approx()
259 cl_equal() -> equal()
260 cl_equal_hashcode() -> equal_hashcode()
261 cl_eulerconst() -> eulerconst()
262 cl_find_modint_ring() -> find_modint_ring()
263 cl_find_univpoly_ring() -> find_univ_poly_ring()
264 cl_float_approx() -> float_approx
265 cl_float_format() -> float_format()
266 cl_float_format_t -> float_format_t
267 cl_free_hook() -> free_hook()
268 cl_hermite() -> hermite()
269 cl_laguerre() -> laguerre()
270 cl_legendre() -> legandre()
271 cl_malloc_hook() -> malloc_hook()
273 cl_tschebychev() -> tschebychev()
275 NB: For functions these changes includes all signatures.
276 * include/*: moved to include/cln/*, stripped `cl_' off filenames.
277 * cln-config.in: new file.
279 2000-08-26 Bruno Haible <haible@clisp.cons.org>
281 * autoconf/acgeneral.m4 (AC_OUTPUT): Use braces in exec_prefix default
284 2000-08-18 Bruno Haible <haible@clisp.cons.org>
286 * include/cl_univpoly_modint.h: Fix typo.
288 2000-07-13 Bruno Haible <haible@clisp.cons.org>
290 * src/float/input/cl_F_read_stream.cc (number_char_p): Accept '_',
291 used as precision marker for floats.
292 Reported by Keith Briggs (in 1998) and Thomas Roessler.
293 * src/integer/input/cl_I_read_stream.cc (number_char_p): Likewise.
294 * src/rational/input/cl_RA_read_stream.cc (number_char_p): Likewise.
295 * src/real/input/cl_R_read_stream.cc (number_char_p): Likewise.
296 * src/complex/input/cl_N_read_stream.cc (number_char_p): Likewise.
298 2000-06-22 Bruno Haible <haible@clisp.cons.org>
300 * include/cl_object.h: Rename cl_word_size to cl_pointer_size.
301 * include/cl_types.h (HAVE_FAST_LONGLONG): Also define on Irix6
303 (cl_word_size): New macro.
304 * src/float/sfloat/cl_SF.h: Use cl_pointer_size instead of
307 2000-05-31 Bruno Haible <haible@clisp.cons.org>
309 * tests/exam_I_floor.cc (integer_floor_tests): Add one more entry.
310 From a sample that fails with gcc-2.95.2 on Sparc.
311 * tests/exam_I_gcd.cc (integer_gcd_tests): Likewise.
313 2000-05-30 Richard Kreckel <kreckel@ginac.de>
315 * configure.in, autoconf/aclocal.m4 (CL_GMP_H_VERSION, CL_GMP_CHECK):
317 * configure.in, include/cl_config.h.in (CL_VERSION, CL_VERSION_MINOR,
318 CL_VERSION_PATCHLEVEL): New definitions.
319 * autoconf/config.guess, autoconf/config.sub, autoconf/ltconfig,
320 autoconf/ltmain.sh: updated from FSF (libtool 1.3.5, etc).
321 * src/Makefile.in, configure.in: release-variables renamed from
323 * configure.in: default to build both shared and static library
324 (i.e. default to the most common behaviour).
326 2000-05-29 Richard Kreckel <kreckel@ginac.de>
328 * autoconf/aclocal.m4 (CL_CANONICAL_HOST): Added missing changequote
329 environment around the patch of 2000-05-23.
331 2000-05-29 Bruno Haible <haible@clisp.cons.org>
333 * autoconf/aclocal.m4 (CL_PROG_INSTALL): Fix typo.
334 Reported by Thomas Klausner <wiz@danbala.ifoer.tuwien.ac.at>.
336 2000-05-27 Richard Kreckel <kreckel@ginac.de>
338 * src/float/lfloat/algebraic/cl_LF_sqrt.cc,
339 src/base/digitseq/cl_DS_sqrt.cc: Readjusted break-even points.
341 2000-05-24 Richard Kreckel <kreckel@ginac.de>
343 * autoconf/config.*: Updated to new version from FSF
344 (the new libtool wants this).
345 * src/Makefile.in: added $(LDFLAGS) to link step.
346 * src/base/digitseq/cl_2DS_div.cc, cl_2DS_recip.cc: Readjusted
349 2000-05-23 Bruno Haible <haible@clisp.cons.org>
351 * autoconf/aclocal.m4 (CL_CANONICAL_HOST): Determine host_cpu,
352 host_vendor, host_os correctly if $host has more than two hyphens.
354 2000-05-19 Richard Kreckel <kreckel@ginac.de>
356 * src/base/digitseq/cl_DS_mul.cc: Rearranged break-even points to
357 better match present-day CPUs whenever GMP3 is used.
358 * src/base/digitseq/cl_DS_div.cc: dto.
359 * src/TUNING: Added comment about order of tuning.
360 * configure, configure.in: Safer GMP3-detection.
361 * INSTALL.generic: Clarified behaviour of --with-gmp.
362 * autoconf/config.guess: updated from Clisp-2000-03-06 sources.
364 2000-05-04 Richard Kreckel <kreckel@ginac.de>
366 * gmp/: removed completely. From now on we are going to link
367 externally against libgmp3.0 or above!
368 * configure, configure.in, Makefile.in, */Makeflags: removed support
369 of internal gmp 2.0.2, like $GMP_INCLUDES, which should be done by
370 setting $CPPFLAGS instead.
371 * Makefile.in: Added libtool inter-library dependency for -lgmp and
372 conforming interface versioning (-version-info).
373 * autoconf/ltconfig, autoconf/ltmain.sh: Updated to newer versions
376 2000-02-22 Bruno Haible <haible@clisp.cons.org>
378 * src/base/digitseq/cl_asm_mips_.cc: Starting at argument 5 the
379 parameter passing was changed for the MIPS n32 ABI.
381 2000-01-24 Richard Kreckel <kreckel@ginac.de>
383 * gmp/*: Replaced the complete mpn sources with the ones from
384 Debian since they are maintained while the ones from FSF
385 aren't and there were problems on some architectures, PowerPC
386 in particular. See the file gmp/README.CLN. This way the
387 hard links in this directory have vanished, they were causing
388 trouble for people working in AFS. This became necessary for
389 Debian, because there it woudn't compile on PPC.
391 2000-01-13 Richard Kreckel <kreckel@ginac.de>
393 * Version 1.0.3 released.
395 2000-01-13 Richard Kreckel <kreckel@ginac.de>
397 * src/base/cl_macros.h (nonreturning_function): For egcs and newer
398 use __attribute__ ((__noreturn__)) instead of the __volatile__
400 (nonreturning): Remove macro.
401 * include/*: Minor fixes to stop -ansi -pedantic from complaining.
402 * include/cl_integer.h: (doublefactorial): New declaration.
403 * src/integer/misc/combin/cl_I_doublefactorial.cc: New file.
405 1999-12-18 Bruno Haible <haible@clisp.cons.org>
407 * autoconf/acgeneral.m4 (AC_ARG_PROGRAM): Create conftestsed using
409 * autoconf/ltconfig: Improve support for recent FreeBSD 3.
410 * include/cl_GV.h (cl_GV_vectorops): Change return type from 'T' to
411 'const T', to match definition in src/vector/cl_GV_number.cc.
412 Reported by Duncan Simpson <dps@io.stargate.co.uk>.
413 * gmp/mpn/Makefile.in (.S.lo): Use 'if', not '&&', because '&&' may
416 1999-09-07 Bruno Haible <haible@clisp.cons.org>
418 * Version 1.0.2 released.
420 1999-09-06 Bruno Haible <haible@clisp.cons.org>
422 * src/rational/cl_RA.h (integerp, ratiop): Instead of returning a
423 boolean expression, write alternatives ending with either
424 "return cl_true;" or "return cl_false;". This way, g++ does a
425 better job inlining it.
426 * src/float/cl_F.h (longfloatp): Likewise.
427 * src/real/cl_R.h (rationalp, integerp, floatp): Likewise.
428 * src/complex/cl_C.h (realp, complexp): Likewise.
430 1999-09-05 Bruno Haible <haible@clisp.cons.org>
432 * include/cl_integer.h (cl_equal_hashcode): New declaration.
433 * include/cl_rational.h (cl_equal_hashcode): New declaration.
434 * include/cl_sfloat.h (cl_equal_hashcode): New declaration.
435 * include/cl_ffloat.h (cl_equal_hashcode): New declaration.
436 * include/cl_dfloat.h (cl_equal_hashcode): New declaration.
437 * include/cl_lfloat.h (cl_equal_hashcode): New declaration.
438 * include/cl_float.h (cl_equal_hashcode): New declaration.
439 * include/cl_real.h (cl_equal_hashcode): New declaration.
440 * include/cl_complex.h (cl_equal_hashcode): New declaration.
441 * src/base/cl_N.h (equal_hashcode_low, equal_hashcode_one): New macros.
442 * src/integer/misc/cl_I_eqhashcode.cc: New file.
443 * src/rational/misc/cl_RA_eqhashcode.cc: New file.
444 * src/float/sfloat/misc/cl_SF_eqhashcode.cc: New file.
445 * src/float/ffloat/misc/cl_FF_eqhashcode.cc: New file.
446 * src/float/dfloat/misc/cl_DF_eqhashcode.cc: New file.
447 * src/float/lfloat/misc/cl_LF_eqhashcode.cc: New file.
448 * src/float/misc/cl_F_eqhashcode.cc: New file.
449 * src/real/misc/cl_R_eqhashcode.cc: New file.
450 * src/complex/misc/cl_C_eqhashcode.cc: New file.
451 * doc/cln.tex: Document `cl_equal_hashcode'.
453 1999-09-05 Bruno Haible <haible@clisp.cons.org>
455 * include/cl_ring.h (cl_number_ring_ops): Add `contains' member.
456 (cl_number_ring): New class.
457 (cl_specialized_number_ring<T>): Inherit from cl_number_ring.
458 (instanceof): New function.
459 * src/integer/ring/cl_I_ring.cc (cl_I_p): New function.
460 * src/integer/misc/cl_I_as.cc (cl_I_p): Add comment.
461 * src/rational/ring/cl_RA_ring.cc (cl_RA_p): New function.
462 * src/rational/misc/cl_RA_as.cc (cl_RA_p): Add comment.
463 * src/real/ring/cl_R_ring.cc (cl_R_p): New function.
464 * src/real/misc/cl_R_as.cc (cl_R_p): Add comment.
465 * src/complex/ring/cl_C_ring.cc (cl_N_p): New function.
466 * src/complex/misc/cl_N_as.cc (cl_N_p): Add comment.
467 * doc/cln.tex: Document `instanceof'.
469 1999-09-05 Bruno Haible <haible@clisp.cons.org>
471 * include/cl_rational.h (numerator, denominator): New declarations.
472 * src/rational/elem/cl_RA_numerator.cc: New file.
473 * src/rational/elem/cl_RA_denominator.cc: New file.
474 * include/cl_integer.h (numerator, denominator): New inline functions.
475 * doc/cln.tex: Document `numerator' and `denominator'.
477 1999-09-05 Bruno Haible <haible@clisp.cons.org>
479 * src/rational/algebraic/cl_RA_rootp.cc (rootp): Fix endless loop
481 * src/rational/algebraic/cl_RA_rootp_I.cc (rootp): Likewise.
483 1999-09-05 Bruno Haible <haible@clisp.cons.org>
485 * include/cl_config.h.in: Support for sparc64 CPU.
486 * include/cl_modules.h: Likewise.
487 * include/cl_types.h: Likewise.
488 * include/cl_object.h: Likewise.
489 * include/cl_GV.h: Likewise.
490 * src/Makefile.in: Likewise.
491 * src/base/cl_alloca.h: Likewise.
492 * src/base/cl_macros.h: Likewise.
493 * src/base/cl_sysdep.h: Likewise.
494 * src/base/cl_low.h: Likewise.
495 * src/base/digitseq/cl_asm.h: Likewise.
496 * src/base/digitseq/cl_asm_.cc: Likewise.
497 * src/base/digitseq/cl_asm_sparc64.h: New file.
498 * src/base/digitseq/cl_asm_sparc64_.cc: New file.
499 * src/modinteger/cl_MI_int.h: Support for sparc64 CPU.
500 * src/polynomial/elem/cl_UP_no_ring.cc: Likewise.
501 * src/polynomial/elem/cl_UP_GF2.h: Likewise.
502 * src/polynomial/elem/cl_asm_GF2.cc: Likewise.
504 1999-09-04 Bruno Haible <haible@clisp.cons.org>
506 * src/base/digitseq/cl_asm_sparc_.cc (orc2_loop_up, orc2_loop_down):
507 Use the `orn' instruction.
509 1999-08-14 Bruno Haible <haible@clisp.cons.org>
511 Assume all platforms have <stdlib.h> and clock_t.
512 * configure.in: Don't call CL_STDLIB_H and CL_CLOCK_T.
513 * src/base/cl_base_config.h.in (CLOCK_T): Remove definition.
514 * src/base/random/cl_random_from.cc: Use clock_t instead of CLOCK_T.
515 * src/timing/cl_t_config.h.in (CLOCK_T): Remove definition.
516 * src/timing/cl_t_current2.cc: Use clock_t instead of CLOCK_T.
518 1999-07-18 Bruno Haible <haible@clisp.cons.org>
520 * gmp/config.guess: Link to autoconf/config.guess.
521 gmp/config.sub: Link to autoconf/config.sub.
522 Needed for Win32 platforms.
524 1999-07-17 Bruno Haible <haible@clisp.cons.org>
526 * autoconf/aclocal.m4 (CL_CANONICAL_HOST_CPU): Distinguish "sparc" and
527 "sparc64" according to the C compiler, not the uname result.
529 1999-06-17 Bruno Haible <haible@clisp.cons.org>
531 * src/base/digitseq/cl_asm_sparc_.cc (compare_loop_up): Fix
532 COUNTER_LOOPS version, fortunately not used yet.
534 * include/cl_modules.h: Prepare for gcc version 3.
536 1999-06-12 Bruno Haible <haible@clisp.cons.org>
538 * src/rational/elem/cl_RA_plus.cc, src/rational/elem/cl_RA_minus.cc:
539 Change the last call from I_I_to_RT to I_I_to_RA.
541 1999-06-09 Bruno Haible <haible@clisp.cons.org>
543 * Version 1.0.1 released.
545 1999-06-09 Bruno Haible <haible@clisp.cons.org>
547 * src/integer/cl_I.h (pFN_maxlength_digits_at): Define also when
550 1999-06-08 Bruno Haible <haible@clisp.cons.org>
552 * autoconf/intparam.c (printf_underscored): Change argument type to
554 * include/cl_modules.h (CL_OUTPUT_LABEL, CL_JUMP_TO): New macros.
555 (CL_PROVIDE, CL_PROVIDE_END): Use them.
556 * include/cl_string.h (cl_heap_string::operator new): Return 1, not 0.
557 * include/cl_GV.h (cl_GV_inner<T>::operator new): Likewise.
558 * src/base/ring/cl_no_ring.cc (dummy_canonhom, dummy_expt_pos): Don't
559 cast a cl_I to void here. Works around a bug in g++-2.95.
560 * src/complex/misc/cl_C_class.cc: Include "cl_C.h".
561 * src/polynomial/elem/cl_UP_no_ring.cc (dummy_canonhom,
562 dummy_expt_pos): Don't cast a cl_I to void here. Works around a bug
564 * src/polynomial/elem/cl_asm_sparc_GF2.cc (DECLARE_FUNCTION): New
566 * src/rational/misc/cl_RA_class.cc: Include "cl_RA.h".
567 * src/vector/cl_GV_I.cc (cl_heap_GV_I_general::operator new,
568 DEFINE_cl_heap_GV_I_bits): Return 1, not 0.
569 * src/vector/cl_GV_number.cc (cl_heap_GV_number_general::operator new):
572 1999-06-01 Bruno Haible <haible@clisp.cons.org>
574 * autoconf/aclocal.m4 (CL_CANONICAL_HOST_CPU): Canonicalize alpha
577 1999-05-29 Bruno Haible <haible@clisp.cons.org>
579 * src/base/digitseq/cl_asm_i386_.cc (DECLARE_FUNCTION): Treat
581 * src/base/digitseq/cl_asm_sparc_.cc (DECLARE_FUNCTION): Likewise.
582 * src/base/digitseq/cl_asm_m68k_.cc (DECLARE_FUNCTION): Treat
583 OpenBSD like NetBSD, and Linux/ELF like SVR4.
585 1999-05-16 Bruno Haible <haible@clisp.cons.org>
587 * src/base/cl_low.h (integerlength32) [__rs6000__]: Use old assembler
588 syntax on AIX systems and new assembler syntax on non-AIX systems.
590 1999-05-01 Bruno Haible <haible@clisp.cons.org>
592 * autoconf/config.guess, autoconf/config.sub: Upgrade to newest
593 version from GNU CVS.
595 1999-04-24 Bruno Haible <haible@clisp.cons.org>
597 * src/integer/bitwise/cl_I_logand.cc (logand): Optimize the case when
598 either operand is a positive fixnum, O(1) instead of O(N).
599 * src/integer/bitwise/cl_I_lognand.cc (lognand): Likewise.
600 * src/integer/bitwise/cl_I_logandc2.cc (logandc2): Likewise for the
603 1999-04-14 Bruno Haible <haible@clisp.cons.org>
605 * autoconf/aclocal.m4 (CL_GLOBAL_CONSTRUCTORS): Add check whether
606 ctor/dtor needs to be exported.
607 * include/cl_config.h.in (CL_NEED_GLOBALIZE_CTORDTOR): New macro.
608 * include/cl_modules.h (CL_GLOBALIZE_JUMP_LABEL): Renamed from
610 (CL_GLOBALIZE_LABEL): New macro.
611 (CL_GLOBALIZE_CTORDTOR_LABEL): Renamed from CL_GLOBALIZE_ASM_LABEL.
612 (CL_PROVIDE): Update.
614 1999-04-12 Bruno Haible <haible@clisp.cons.org>
616 * src/Makefile.in ($(ASMFILES_S)): On HPPA, ignore preprocessing
617 errors ("unterminated string or character constant").
618 ($(ASMFILES_LO)): On HPPA, try with various settings of
619 COMPILER_PATH, in order to try /usr/ccs/bin/as and /bin/as.
621 1999-04-11 Bruno Haible <haible@clisp.cons.org>
623 * INSTALL: Mention gmp problems on MIPS.
624 * doc/cln.tex: Likewise.
626 1999-03-24 Mumit Khan <khan@xraylith.wisc.edu>
628 * src/Makefile.in (SUBDIRS): Filter out CVS and RCS directories from
630 * include/cl_modules.h (CL_GLOBALIZE_LABEL): Define for Win32.
631 (CL_GLOBALIZE_ASM_LABEL): New macro.
632 (CL_PROVIDE): Use it.
633 * src/base/random/cl_random_from.cc: Handle WIN32.
634 * src/timing/cl_t_current.cc: Likewise.
636 1999-03-15 Bruno Haible <haible@clisp.cons.org>
638 * autoconf/intparam.c (main7): Use %lX instead of %X for a `long'.
639 (main8): Adapt for C++.
641 1999-03-09 Bruno Haible <haible@clisp.cons.org>
643 * INSTALL: Mention egcs-1.1 problems on Sparc.
644 * doc/cln.tex: Likewise.
646 1999-03-08 Bruno Haible <haible@clisp.cons.org>
648 * autoconf/aclocal.m4 (CL_FPU_CONTROL): Fix the "checking for"
651 1999-02-25 Bruno Haible <haible@clisp.cons.org>
653 * autoconf/aclocal.m4: In test programs, declare `int main()', not
656 * lidia-interface/src/interfaces/integers/cln/bigint.c
657 (bigint_to_string): Fix for negative arguments.
659 * src/base/cl_low.h: Check for `__sparc__', not `SPARC'.
660 * src/base/cl_alloca.h: Likewise.
662 * src/base/cl_low.h: Eliminate CLISP style "# " comments.
663 * src/base/digitseq/cl_asm_arm_.cc,
664 src/base/digitseq/cl_asm_mips_.cc,
665 src/float/dfloat/elem/cl_DF_mul.cc: Likewise.
667 * src/modinteger/cl_MI_pow2.h,
668 src/modinteger/cl_MI_pow2m1.h,
669 src/modinteger/cl_MI_pow2p1.h: Workaround g++-2.7.2 inlining bug.
671 1999-01-18 Bruno Haible <haible@clisp.cons.org>
673 * autoconf/acgeneral.m4,
674 autoconf/acspecific.m4: Upgrade to autoconf-2.13.
675 * autoconf/config.guess, autoconf/config.sub: Likewise.
676 * autoconf/aclocal.m4 (CL_ALLOCA): Test for _MSC_VER and alloca being
677 a macro. Use ${ac_objext}.
678 * src/base/cl_macros.h (alloca): Put _MSC_VER test before the others,
679 conforming with CL_ALLOCA.
681 1999-01-12 Bruno Haible <haible@clisp.cons.org>
683 * Version 1.0 released.