1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
|
Copyright 1996, 1999-2016, 2018-2020 Free Software Foundation, Inc.
Verbatim copying and distribution of this entire article is permitted in any
medium, provided this notice is preserved.
Changes between GMP version 6.2.0 and 6.2.1
BUGS FIXED
* A possible overflow of type int is avoided for mpz_cmp on huge operands.
* Overflows are more carefully detected and reported for mpz_pow_ui.
* A bug in longlong.h for aarch64 sub_ddmmss, not affecting GMP, was healed.
* mini-gmp: mpz_out_str and mpq_out_str now correctly handle out of
range bases.
FEATURES
* C90 compliance.
* Initial support for Darwin on arm64, and improved portability.
* Support for more processors.
SPEEDUPS
* None, except indirectly through recognition of new CPUs.
MISC
* None.
Changes between GMP version 6.1.* and 6.2.0
BUGS FIXED
* gmp_snprintf now correctly reports an error (returns -1) when snprintf
does.
* Conversion to double is now more robust even if the configuration process
does not recognize the float format.
* mpz_powm could return a not completely normalised value, when the
exponent was 1 and the base < 0.
* mpf_set_str could create invalid mpf_t variables for input strings with
many leading zeros.
FEATURES
* New C++ functions factorial, primorial and fibonacci for mpz_class.
* Functions to detect primality now substitute the first 24 Miller-Rabin
iterations with the BPSW test.
* Mini-GMP: new functions mpz_2fac_ui and mpz_mfac_uiui.
* Mini-GMP: mpz_sizeinbase, mpz_get_str, and mpz_set_str now support bases up
to 62.
* Mini-GMP: added support for the mpq_t layer.
* MIPS r6 cores are now supported.
SPEEDUPS
* Major speedup for AMD Ryzen and Epyc thanks to an extensive set of assembly
code.
* Major speedup for IBM POWER9 thanks to assembly code making use of new madd
instruction.
* Speedup for 64-bit ARM CPUs thanks to new/improved assembly code.
* The n-over-k function mpz_bin_ui has been reimplemented for great speedups
for large operands.
* Speedup for the worst case of mpz_perfect_power_p.
* Speedup for gcd for small and medium size operands.
* Speedup for really huge multiplies thanks to much larger FFT tables.
MISC
* Internal representation of the mpz_t variables now supports lazy
allocation; memory is allocated only when a value is stored.
* Small improvements and better coverage for the test suite.
* The tune/speed program can measure some more functions.
* The low-level function mpn_mul no longer diverts to mpn_sqr, users should
call mpn_sqr directly when applicable.
* New installed files gmp.pc and gmpxx.pc, for use with pkg-config.
Contributed by Hugh McMaster.
Changes between GMP version 6.1.1 and 6.1.2
BUGS FIXED
* Mini-GMP: Fixed a division bug, which on a machine with 64-bit
unsigned long affects approximately 1 out of 2^32 divisors.
* Mini-GMP: Fix mpz_set_str crash on inputs with a large number of
leading zeros. Also stricter input validation, rejecting inputs
with no digits.
FEATURES
* Handle more systems which require PIC code in static libraries (e.g.,
"hardened" Gentoo and Debian 9).
* Configuration for arm (-32 and -64) has been rewritten, fixing poor
code selection for many CPUs.
* Mini-GMP: Updated to the latest development version, including
new functions mpn_com and mpn_neg.
SPEEDUPS
* None, except for arm CPUs affected by the configuration rewrite.
MISC
-
Changes between GMP version 6.1.0 and 6.1.1
BUGS FIXED
* Make Intel Broadwell configurations work on Windows.
FEATURES
* Work around faulty cpuid on some recent Intel chips (this allows GMP to run
on Skylake Pentiums).
* Support thumb-less ARM chips.
Changes between GMP version 6.0.* and 6.1.0
BUGS FIXED
* The public function mpn_com is now correctly declared in gmp.h.
* Healed possible failures of mpn_sec_sqr for non-cryptographic sizes for
some obsolete CPUs.
* The option --disable-assembly now disables all inlined asm.
* Fixed bug affecting mini-gmp's bitwise functions mpz_setbit, mpz_clrbit,
and mpz_combit.
* Various problems related to precision for mpf have been fixed.
* Fixed ABI incompatible stack alignment in calls from assembly code.
* Fixed PIC bug in popcount affecting Intel processors using the 32-bit ABI.
SPEEDUPS
* Speedup for Intel Broadwell and Skylake through assembly code making use of
new ADX instructions.
* Square root is now faster when the remainder is not needed. Also the speed
to compute the k-th root improved, for small sizes.
* Improved arm64 support.
FEATURES
* New C++ functions gcd and lcm for mpz_class.
* New public mpn functions mpn_divexact_1, mpn_zero_p, and mpn_cnd_swap.
* New public mpq_cmp_z function, to efficiently compare rationals with
integers.
* Support for Darwin in all x86 code, thereby enabling fat builds on Darwin.
* Support for more 32-bit arm processors.
* Support for compilation with clang/llvm on more platforms. Caution: GMP
triggers mis-compilation bugs in clang for many platforms, such as arm, x86
(32-bit and 64-bit), powerpc, mips.
* Support for AVX-less modern x86 CPUs. (Such support might be missing either
because the CPU vendor chose to disable AVX, or because the running kernel
lacks AVX context switch support.)
* Stack usage trimmed; we believe 512 KiB is now sufficient for any GMP
call, irrespective of operand size.
* Support for NetBSD under Xen; we switch off AVX unconditionally under
NetBSD since a bug in NetBSD makes AVX fail under Xen.
MISC
* We now use manufacturers' code names for x86 CPUs, e.g., "haswell" instead
of names derived from the commercial brands.
* Small improvements and better coverage for the test suite.
* The various FreeBSD problems listed for 6.0.0 affect this release too.
* Tuned values for FFT multiplications are provided for larger number on
many platforms.
Changes between GMP version 5.1.* and 6.0.0
BUGS FIXED
* The function mpz_invert now considers any number invertible in Z/1Z.
* The mpn multiply code now handles operands of more than 2^31 limbs
correctly. (Note however that the mpz code is limited to 2^32 bits on
32-bit hosts and 2^37 bits on 64-bit hosts.)
* Contains all fixes from release 5.1.3.
SPEEDUPS
* Plain division of large operands is faster and more monotonous in operand
size.
* Major speedup for ARM, in particular ARM Cortex-A15, thanks to improved
assembly.
* Major speedup for SPARC T4/T5 and speedup also for T3, thanks to a lot of
new assembly.
* Speedup for Intel Sandy Bridge, Ivy Bridge, Haswell, thanks to rewritten
and vastly expanded assembly support. Speedup also for the older Core 2
and Nehalem.
* Faster mixed arithmetic between mpq_class and double.
* With g++, optimise more operations when one argument is a simple constant.
FEATURES
* Support for new Intel and AMD CPUs.
* Support for ARM64 alias Aarch64 alias ARMv8.
* New public functions mpn_sec_mul and mpn_sec_sqr, implementing side-channel
silent multiplication and squaring.
* New public functions mpn_sec_div_qr and mpn_sec_div_r, implementing
side-channel silent division.
* New public functions mpn_cnd_add_n and mpn_cnd_sub_n. Side-channel silent
conditional addition and subtraction.
* New public function mpn_sec_powm, implementing side-channel silent modexp.
* New public function mpn_sec_invert, implementing side-channel silent
modular inversion.
* Better support for applications which use the mpz_t type, but nevertheless
need to call some of the lower-level mpn functions. See the documentation
for mpz_limbs_read and related functions.
MISC
* This release will not work on NetBSD 5.x, FreeBSD 7.x, 8.x or 9 series
before 9.3. The reason is that the m4 command is not correctly
implemented. (Workaround: Use an older GMP release, or install GNU m4 from
/usr/ports and tell GMP to use it.)
* This release will not build properly on FreeBSD/amd64 before version 10
using the 32-bit ABI (once a working m4 is installed). The reason is
broken limits.h. (Workaround: Use an older GMP release if using the 32-bit
ABI on these FreeBSD releases is important.)
* This release will not work reliably on FreeBSD 10.0 for i386 or amd64 using
the 32-bit ABI. The reason is bugs in the compiler 'clang'. Depending on
CPU-dependent compiler flags, GMP may or may not be miscompiled in a
particular build. (Workaround: Compiling gcc from /usr/ports should work,
except that gcc circularly depends on GMP; we have not been able to test
that workaround due to FreeBSD 10.0 bugs affecting its ability to run under
KVM and Xen.)
* This release will not compile on FreeBSD before version 10 for i386,
targeting any modern AMD processor. The reason is bugs in the old gcc
bundled with FreeBSD. (Workaround: install a less obsolete gcc from
/usr/ports and tell GMP to use it, or override the -march=amdfam10
GMP configure command line argument.)
Changes between GMP version 5.1.2 and 5.1.3
BUGS FIXED
* The internal functions mpn_sbpi1_div_qr_sec mpn_sbpi1_div_r_sec could
compute garbage with a low probability. They are now rewritten, and the
test code has been improved.
* A bug in the ia64 implementation of mpn_divrem_2, clobbering some
callee-save registers, has been fixed. This is an internal
function, with the bug manifesting itself as miscomputation in,
e.g., mpn_sqrtrem.
* The documentation now correctly says 'const' for input arguments.
SPEEDUPS
* None.
FEATURES
* None.
MISC
* None.
Changes between GMP version 5.1.1 and 5.1.2
BUGS FIXED
* A bug in mpz_powm_ui triggered by base arguments of at least 15000 decimal
digits or mod arguments of at least 7500 decimal digits has been fixed.
* An AMD Bulldozer specific bug affecting the 64-bit Windows ABI has been
fixed. This bug was in a key function (mpn_mul_1) and made both Bulldozer
specific builds and fat builds run on Bulldozer completely non-functional.
SPEEDUPS
* None.
FEATURES
* None.
MISC
* Fixes and generalisations to the test suite.
* Minor portability enhancements.
Changes between GMP version 5.1.0 and 5.1.1
BUGS FIXED
* On Windows 64-bit, an error causing link errors about
__gmp_binvert_limb_table has been fixed.
* Aarch64 alias ARM64 support now works.
* A possible buffer overrun in mpz_ior has been fixed.
* A rare sign flip in mpz_remove has been fixed.
* A bug causing problems with mpf numbers with absolute value >= 2^31 has
been fixed.
* Several bugs in mini-gmp have been fixed.
* A bug caused by automake, related to the 'distcheck' target, has been fixed
by upgrading the automake used for GMP release engineering.
SPEEDUPS
* None.
FEATURES
* Preliminary support for the x32 ABI under x86-64.
MISC
* The mini-gmp testsuite now tests the entire set of functions.
* Various improvements of the GMP testsuite.
Changes between GMP version 5.0.* and 5.1.0
BUGS FIXED
* When reading a C++ number (like mpz_class) in an istream reaches the end
of the stream, the eofbit is now set.
* The result sign of mpz_rootrem's remainder is now always correct.
* The mpz_remove function now handles negative divisors.
* Contains all fixes from release 5.0.5.
SPEEDUPS
* The n-factorial and n-over-k functions have been reimplemented for great
speedups for small and large operands.
* New subquadratic algorithm for the Kronecker/Jacobi/Legendre symbol.
* Major speedup for ARM, in particular ARM Cortex-A9 and A15, thanks to broad
assembly support.
* Significant speedup for POWER6 and POWER7 thanks to improved assembly.
* The performance under M$ Windows' 64-bit ABI has been greatly improved
thanks to complete assembly support.
* Minor speed improvements of many functions and for many platforms.
FEATURES
* Many new CPUs recognised.
* New functions for multi-factorials, and primorial: mpz_2fac_ui,
mpz_mfac_uiui and mpz_primorial_ui.
* The mpz_powm_sec function now uses side-channel silent division for
converting into Montgomery residues.
* The fat binary mechanism is now more robust in its CPU recognition.
MISC
* Inclusion of assembly code is now controlled by the configure options
--enable-assembly and --disable-assembly. The "none" CPU target is gone.
* In C++, the conversions mpq_class->mpz_class, mpf_class->mpz_class and
mpf_class->mpq_class are now explicit.
* Includes "mini-gmp", a small, portable, but less efficient, implementation
of a subset of GMP's mpn and mpz interfaces. Used in GMP bootstrap, but it
can also be bundled with applications as a fallback when the real GMP
library is unavailable.
* The ABIs under AIX are no longer called aix32 and aix64, but mode64 and 32.
This is more consistent with other powerpc systems.
* The coverage of the testsuite has been improved, using the lcov tool. See
also https://gmplib.org/devel/lcov/.
* It is now possible to compile GMP using a C++ compiler.
* K&R C compilers are no longer supported.
* The BSD MP compatibility functions have been removed.
Changes between GMP version 5.0.4 and 5.0.5
BUGS FIXED
* A bug causing AMD 11h processors to be treated like AMD 10h has been fixed.
The 11h processors do not correctly handle all 10h (aka K10) instructions,
and GMP's use of these instructions results in major miscomputations (not
as one would have hoped CPU traps of some 'illegal instruction' sort).
* A bug affecting recent Intel Sandy Bridge CPUs resulting in configuration
failures has been fixed.
SPEEDUPS
* None.
FEATURES
* A couple of tests added to the self-check suite.
MISC
* None.
Changes between GMP version 5.0.3 and 5.0.4
BUGS FIXED
* Thresholds in mpn_powm_sec for both fat and non-fat builds are now used
safely, plugging a one-word buffer overrun introduced in the 5.0.3 release
(for non-fat) and a multi-word buffer overrun that existed since 5.0 (for
fat). (We have not been able to provoke malign stack smashing in any of
the ~100 configurations explored by the GMP nightly builds, but the bug
should be assumed to be exploitable.)
* Two bugs in multiplication code causing incorrect computation with
extremely low probability have been fixed.
* A bug in the test suite causing buffer overruns during "make check",
sometimes leading to subsequent malloc crashes, has been fixed.
* Two bugs in the gcd code have been fixed. They could lead to incorrect
results, but for uniformly distributed random operands, the likelihood for
that is infinitesimally small. (There was also a third bug, but that was
an incorrect ASSERT, which furthermore was not enabled by default.)
* A bug affecting 32-bit PowerPC division has been fixed. The bug caused
miscomputation for certain divisors in the range 2^32 ... 2^64-1 (about 1
in 2^30 of these).
SPEEDUPS
* None, except indirectly through recognition of new CPUs, and through better
tuning parameters.
FEATURES
* Some more tests added to the self-check suite.
* The AMD "Bulldozer" CPU is now recognised.
MISC
* None.
Changes between GMP version 5.0.2 and 5.0.3
BUGS FIXED
* A few minor bugs related to portability fixed.
* A slight timing leak of the powm_sec functions have been sealed. (This
leak could possibly be used to extract the most significant few bits of the
exponent. "Few" here means at most 10.)
* The mpz_nextprime function now runs a safer number of pseudo-random prime
tests.
* A bug in division code possibly causing incorrect computation was fixed.
SPEEDUPS
* None, except indirectly through recognition of new CPUs, and through better
tuning parameters.
FEATURES
* New CPUs recognised.
* IBM S/390 are now supported in both 31/32-bit and 64-bit mode. (We have
not been able to fully test this on any multilib machine, since IBM expired
our guest account a few days before our release.)
MISC
* None.
Changes between GMP version 5.0.1 and 5.0.2
BUGS FIXED
* Many minor bugs related to portability fixed.
* The support for HPPA 2.0N now works, after an assembly bug fix.
* A test case type error has been fixed. The symptom of this bug was
spurious 'make check' failures.
SPEEDUPS
* None, except indirectly through recognition of new CPUs.
FEATURES
* Fat builds are now supported for 64-bit x86 processors also under Darwin.
MISC
* None.
Changes between GMP version 5.0.0 and 5.0.1
BUGS FIXED
* Fat builds fixed.
* Fixed crash for huge multiplies when old FFT_TABLE2 type of parameter
selection tables' sentinel was smaller than multiplied operands.
* The solib numbers now reflect the removal of the documented but preliminary
mpn_bdivmod function; we correctly flag incompatibility with GMP 4.3. GMP
5.0.0 has this wrong, and should perhaps be uninstalled to avoid confusion.
SPEEDUPS
* Multiplication of large numbers has indirectly been sped up through better
FFT tuning and processor recognition. Since many operations depend on
multiplication, there will be a general speedup.
FEATURES
* More Core i3, i5 an Core i7 processor models are recognised.
* Fixes and workarounds for Mac OS quirks should make this GMP version build
using many of the different versions of "Xcode".
MISC
* The amount of scratch memory needed for multiplication of huge numbers has
been reduced substantially (but is still larger than in GMP 4.3.)
* Likewise, the amount of scratch memory needed for division of large numbers
has been reduced substantially.
* The FFT tuning code of tune/tuneup.c has been completely rewritten, and
new, large FFT parameter selection tables are provided for many machines.
* Upgraded to the latest autoconf, automake, libtool.
Changes between GMP version 4.3.X and 5.0.0
BUGS FIXED
* None (contains the same fixes as release 4.3.2).
SPEEDUPS
* Multiplication has been overhauled:
(1) Multiplication of larger same size operands has been improved with
the addition of two new Toom functions and a new internal function
mpn_mulmod_bnm1 (computing U * V mod (B^n-1), B being the word base.
This latter function is used for the largest products, waiting for a
better Schoenhage-Strassen U * V mod (B^n+1) implementation.
(2) Likewise for squaring.
(3) Multiplication of different size operands has been improved with the
addition of many new Toom function, and by selecting underlying
functions better from the main multiply functions.
* Division and mod have been overhauled:
(1) Plain "schoolbook" division is reimplemented using faster quotient
approximation.
(2) Division Q = N/D, R = N mod D where both the quotient and remainder
are needed now runs in time O(M(log(N))). This is an improvement of
a factor log(log(N))
(3) Division where just the quotient is needed is now O(M(log(Q))) on
average.
(4) Modulo operations using Montgomery REDC form now take time O(M(n)).
(5) Exact division Q = N/D by means of mpz_divexact has been improved
for all sizes, and now runs in time O(M(log(N))).
* The function mpz_powm is now faster for all sizes. Its complexity has
gone from O(M(n)log(n)m) to O(M(n)m) where n is the size of the modulo
argument and m is the size of the exponent. It is also radically
faster for even modulus, since it now partially factors such modulus
and performs two smaller modexp operations, then uses CRT.
* The internal support for multiplication yielding just the lower n limbs
has been improved by using Mulders' algorithm.
* Computation of inverses, both plain 1/N and 1/N mod B^n have been
improved by using well-tuned Newton iterations, and wrap-around
multiplication using mpn_mulmod_bnm1.
* A new algorithm makes mpz_perfect_power_p asymptotically faster.
* The function mpz_remove uses a much faster algorithm, is better tuned,
and also benefits from the division improvements.
* Intel Atom and VIA Nano specific optimisations.
* Plus hundreds of smaller improvements and tweaks!
FEATURES
* New mpz function: mpz_powm_sec for side-channel quiet modexp
computations.
* New mpn functions: mpn_sqr, mpn_and_n, mpn_ior_n, mpn_xor_n, mpn_nand_n,
mpn_nior_n, mpn_xnor_n, mpn_andn_n, mpn_iorn_n, mpn_com, mpn_neg,
mpn_copyi, mpn_copyd, mpn_zero.
* The function mpn_tdiv_qr now allows certain argument overlap.
* Support for fat binaries for 64-bit x86 processors has been added.
* A new type, mp_bitcnt_t for bignum bit counts, has been introduced.
* Support for Windows64 through mingw64 has been added.
* The cofactors of mpz_gcdext and mpn_gcdext are now more strictly
normalised, returning to how GMP 4.2 worked. (Note that also release
4.3.2 has this change.)
MISC
* The mpn_mul function should no longer be used for squaring,
instead use the new mpn_sqr.
* The algorithm selection has been improved, the number of thresholds have
more than doubled, and the tuning and use of existing thresholds have
been improved.
* The tune/speed program can measure many of new functions.
* The mpn_bdivmod function has been removed. We do not consider this an
incompatible change, since the function was marked as preliminary.
* The testsuite has been enhanced in various ways.
Changes between GMP version 4.3.1 and 4.3.2
Bugs:
* Fixed bug in mpf_eq.
* Fixed overflow issues in mpz_set_str, mpz_inp_str, mpf_set_str, and
mpf_get_str.
* Avoid unbounded stack allocation for unbalanced multiplication.
* Fixed bug in FFT multiplication.
Speedups:
* None, except that proper processor recognition helps affected processors.
Features:
* Recognise more "Core 2" processor variants.
* The cofactors of mpz_gcdext and mpn_gcdext are now more strictly
normalised, returning to how GMP 4.2 worked.
Changes between GMP version 4.3.0 and 4.3.1
Bugs:
* Fixed bug in mpn_gcdext, affecting also mpz_gcdext and mpz_invert.
The bug could cause a cofactor to have a leading zero limb, which
could lead to crashes or miscomputation later on.
* Fixed some minor documentation issues.
Speedups:
* None.
Features:
* Workarounds for various issues with Mac OS X's build tools.
* Recognise more IBM "POWER" processor variants.
Changes between GMP version 4.2.X and 4.3.0
Bugs:
* Fixed bug in mpz_perfect_power_p with recognition of negative perfect
powers that can be written both as an even and odd power.
* We might accidentally have added bugs since there is a large amount of
new code in this release.
Speedups:
* Vastly improved assembly code for x86-64 processors from AMD and Intel.
* Major improvements also for many other processor families, such as
Alpha, PowerPC, and Itanium.
* New sub-quadratic mpn_gcd and mpn_gcdext, as well as improved basecase
gcd code.
* The multiply FFT code has been slightly improved.
* Balanced multiplication now uses 4-way Toom in addition to schoolbook,
Karatsuba, 3-way Toom, and FFT.
* Unbalanced multiplication has been vastly improved.
* Improved schoolbook division by means of faster quotient approximation.
* Several new algorithms for division and mod by single limbs, giving
many-fold speedups.
* Improved nth root computations.
* The mpz_nextprime function uses sieving and is much faster.
* Countless minor tweaks.
Features:
* Updated support for fat binaries for x86_32 include current processors
* Lots of new mpn internal interfaces. Some of them will become public
in a future GMP release.
* Support for the 32-bit ABI under x86-apple-darwin.
* x86 CPU recognition code should now default better for future
processors.
* The experimental nails feature does not work in this release, but
it might be re-enabled in the future.
Misc:
* The gmp_version variable now always contains three parts. For this
release, it is "4.3.0".
Changes between GMP version 4.2.3 and 4.2.4
Bugs:
* Fix bug with parsing exponent '+' sign in mpf.
* Fix an allocation bug in mpf_set_str, also affecting mpf_init_set_str, and
mpf_inp_str.
Speedups:
* None, except that proper processor recognition helps affected processors.
Features:
* Recognize new AMD processors.
Changes between GMP version 4.2.2 and 4.2.3
Bugs:
* Fix x86 CPU recognition code to properly identify recent AMD and Intel
64-bit processors.
* The >> operator of the C++ wrapper gmpxx.h now does floor rounding, not
truncation.
* Inline semantics now follow the C99 standard, and works with recent GCC
releases.
* C++ bitwise logical operations work for more types.
* For C++, gmp.h now includes cstdio, improving compiler compatibility.
* Bases > 36 now work properly in mpf_set_str.
Speedups:
* None, except that proper processor recognition helps affected processors.
Features:
* The allocation functions now detect overflow of the mpz_t type. This means
that overflow will now cause an abort, except when the allocation
computation itself overflows. (Such overflow can probably only happen in
powering functions; we will detect powering overflow in the future.)
Changes between GMP version 4.2.1 and 4.2.2
* License is now LGPL version 3.
Bugs:
* Shared library numbers corrected for libcxx.
* Fixed serious bug in gmpxx.h where a=a+b*c would generate garbage.
Note that this only affects C++ programs.
* Fix crash in mpz_set_d for arguments with large negative exponent.
* Fix 32-bit ABI bug with Itanium assembly for popcount and hamdist.
* Fix assembly syntax problem for powerpc-ibm-aix with AIX native assembler.
* Fix problems with x86 --enable-fat, where the compiler where told to
generate code for the build machine, not plain i386 code as it should.
* Improved recognition of powerpc systems wrt Altivec/VMX capability.
* Misc minor fixes, mainly workarounds for compiler/assembler bugs.
Speedups:
* "Core 2" and Pentium 4 processors, running in 64-bit mode will get a
slight boost as they are now specifically recognized.
Features:
* New support for x86_64-solaris
* New, rudimentary support for x86-apple-darwin and x86_64-apple-darwin.
(Please see https://gmplib.org/macos.html for more information.)
Changes between GMP version 4.2 and 4.2.1
Bugs:
* Shared library numbers corrected.
* Broken support for 32-bit AIX fixed.
* Misc minor fixes.
Speedups:
* Exact division (mpz_divexact) now falls back to plain division for large
operands.
Features:
* Support for some new systems.
Changes between GMP version 4.1.4 and 4.2
Bugs:
* Minor bug fixes and code generalizations.
* Expanded and improved test suite.
Speedups:
* Many minor optimizations, too many to mention here.
* Division now always subquadratic.
* Computation of n-factorial much faster.
* Added basic x86-64 assembly code.
* Floating-point output is now subquadratic for all bases.
* FFT multiply code now about 25% faster.
* Toom3 multiply code faster.
Features:
* Much improved configure.
* Workarounds for many more compiler bugs.
* Temporary allocations are now made on the stack only if small.
* New systems supported: HPPA-2.0 gcc, IA-64 HP-UX, PowerPC-64 Darwin,
Sparc64 GNU/Linux.
* New i386 fat binaries, selecting optimised code at runtime (--enable-fat).
* New build option: --enable-profiling=instrument.
* New memory function: mp_get_memory_functions.
* New Mersenne Twister random numbers: gmp_randinit_mt, also now used for
gmp_randinit_default.
* New random functions: gmp_randinit_set, gmp_urandomb_ui, gmp_urandomm_ui.
* New integer functions: mpz_combit, mpz_rootrem.
* gmp_printf etc new type "M" for mp_limb_t.
* gmp_scanf and friends now accept C99 hex floats.
* Numeric input and output can now be in bases up to 62.
* Comparisons mpz_cmp_d, mpz_cmpabs_d, mpf_cmp_d recognise infinities.
* Conversions mpz_get_d, mpq_get_d, mpf_get_d truncate towards zero,
previously their behaviour was unspecified.
* Fixes for overflow issues with operands >= 2^31 bits.
Caveats:
* mpfr is gone, and will from now on be released only separately. Please see
www.mpfr.org.
Changes between GMP version 4.1.3 and 4.1.4
* Bug fix to FFT multiplication code (crash for huge operands).
* Bug fix to mpf_sub (miscomputation).
* Support for powerpc64-gnu-linux.
* Better support for AMD64 in 32-bit mode.
* Upwardly binary compatible with 4.1.3, 4.1.2, 4.1.1, 4.1, 4.0.1, 4.0,
and 3.x versions.
Changes between GMP version 4.1.2 and 4.1.3
* Bug fix for FFT multiplication code (miscomputation).
* Bug fix to K6 assembly code for gcd.
* Bug fix to IA-64 assembly code for population count.
* Portability improvements, most notably functional AMD64 support.
* mpz_export allows NULL for countp parameter.
* Many minor bug fixes.
* mpz_export allows NULL for countp parameter.
* Upwardly binary compatible with 4.1.2, 4.1.1, 4.1, 4.0.1, 4.0, and 3.x
versions.
Changes between GMP version 4.1.1 and 4.1.2
* Bug fixes.
Changes between GMP version 4.1 and 4.1.1
* Bug fixes.
* New systems supported: NetBSD and OpenBSD sparc64.
Changes between GMP version 4.0.1 and 4.1
* Bug fixes.
* Speed improvements.
* Upwardly binary compatible with 4.0, 4.0.1, and 3.x versions.
* Asymptotically fast conversion to/from strings (mpz, mpq, mpn levels), but
also major speed improvements for tiny operands.
* mpn_get_str parameter restrictions relaxed.
* Major speed improvements for HPPA 2.0 systems.
* Major speed improvements for UltraSPARC systems.
* Major speed improvements for IA-64 systems (but still sub-optimal code).
* Extended test suite.
* mpfr is back, with many bug fixes and portability improvements.
* New function: mpz_ui_sub.
* New functions: mpz_export, mpz_import.
* Optimization for nth root functions (mpz_root, mpz_perfect_power_p).
* Optimization for extended gcd (mpz_gcdext, mpz_invert, mpn_gcdext).
* Generalized low-level number format, reserving a `nails' part of each
limb. (Please note that this is really experimental; some functions
are likely to compute garbage when nails are enabled.)
* Nails-enabled Alpha 21264 assembly code, allowing up to 75% better
performance. (Use --enable-nails=4 to enable it.)
Changes between GMP version 4.0 and 4.0.1
* Bug fixes.
Changes between GMP version 3.1.1 and 4.0
* Bug fixes.
* Speed improvements.
* Upwardly binary compatible with 3.x versions.
* New CPU support: IA-64, Pentium 4.
* Improved CPU support: 21264, Cray vector systems.
* Support for all MIPS ABIs: o32, n32, 64.
* New systems supported: Darwin, SCO, Windows DLLs.
* New divide-and-conquer square root algorithm.
* New algorithms chapter in the manual.
* New malloc reentrant temporary memory method.
* New C++ class interface by Gerardo Ballabio (beta).
* Revamped configure, featuring ABI selection.
* Speed improvements for mpz_powm and mpz_powm_ui (mainly affecting small
operands).
* mpz_perfect_power_p now properly recognizes 0, 1, and negative perfect
powers.
* mpz_hamdist now supports negative operands.
* mpz_jacobi now accepts non-positive denominators.
* mpz_powm now supports negative exponents.
* mpn_mul_1 operand overlap requirements relaxed.
* Float input and output uses locale specific decimal point where available.
* New gmp_printf, gmp_scanf and related functions.
* New division functions: mpz_cdiv_q_2exp, mpz_cdiv_r_2exp, mpz_divexact_ui.
* New divisibility tests: mpz_divisible_p, mpz_divisible_ui_p,
mpz_divisible_2exp_p, mpz_congruent_p, mpz_congruent_ui_p,
mpz_congruent_2exp_p.
* New Fibonacci function: mpz_fib2_ui.
* New Lucas number functions: mpz_lucnum_ui, mpz_lucnum2_ui.
* Other new integer functions: mpz_cmp_d, mpz_cmpabs_d, mpz_get_d_2exp,
mpz_init2, mpz_kronecker, mpz_lcm_ui, mpz_realloc2.
* New rational I/O: mpq_get_str, mpq_inp_str, mpq_out_str, mpq_set_str.
* Other new rational functions: mpq_abs, mpq_cmp_si, mpq_div_2exp,
mpq_mul_2exp, mpq_set_f.
* New float tests: mpf_integer_p, mpf_fits_sint_p, mpf_fits_slong_p,
mpf_fits_sshort_p, mpf_fits_uint_p, mpf_fits_ulong_p, mpf_fits_ushort_p.
* Other new float functions: mpf_cmp_d, mpf_get_default_prec, mpf_get_si,
mpf_get_ui, mpf_get_d_2exp.
* New random functions: gmp_randinit_default, gmp_randinit_lc_2exp_size.
* New demo expression string parser (see demos/expr).
* New preliminary perl interface (see demos/perl).
* Tuned algorithm thresholds for many more CPUs.
Changes between GMP version 3.1 and 3.1.1
* Bug fixes for division (rare), mpf_get_str, FFT, and miscellaneous minor
things.
Changes between GMP version 3.0 and 3.1
* Bug fixes.
* Improved `make check' running more tests.
* Tuned algorithm cutoff points for many machines. This will improve speed for
a lot of operations, in some cases by a large amount.
* Major speed improvements: Alpha 21264.
* Some speed improvements: Cray vector computers, AMD K6 and Athlon, Intel P5
and Pentium Pro/II/III.
* The mpf_get_prec function now works as it did in GMP 2.
* New utilities for auto-tuning and speed measuring.
* Multiplication now optionally uses FFT for very large operands. (To enable
it, pass --enable-fft to configure.)
* Support for new systems: Solaris running on x86, FreeBSD 5, HP-UX 11, Cray
vector computers, Rhapsody, Nextstep/Openstep, MacOS.
* Support for shared libraries on 32-bit HPPA.
* New integer functions: mpz_mul_si, mpz_odd_p, mpz_even_p.
* New Kronecker symbol functions: mpz_kronecker_si, mpz_kronecker_ui,
mpz_si_kronecker, mpz_ui_kronecker.
* New rational functions: mpq_out_str, mpq_swap.
* New float functions: mpf_swap.
* New mpn functions: mpn_divexact_by3c, mpn_tdiv_qr.
* New EXPERIMENTAL function layer for accurate floating-point arithmetic, mpfr.
To try it, pass --enable-mpfr to configure. See the mpfr subdirectory for
more information; it is not documented in the main GMP manual.
Changes between GMP version 3.0 and 3.0.1
* Memory leaks in gmp_randinit and mpz_probab_prime_p fixed.
* Documentation for gmp_randinit fixed. Misc documentation errors fixed.
Changes between GMP version 2.0 and 3.0
* Source level compatibility with past releases (except mpn_gcd).
* Bug fixes.
* Much improved speed thanks to both host independent and host dependent
optimizations.
* Switch to autoconf/automake/libtool.
* Support for building libgmp as a shared library.
* Multiplication and squaring using 3-way Toom-Cook.
* Division using the Burnikel-Ziegler method.
* New functions computing binomial coefficients: mpz_bin_ui, mpz_bin_uiui.
* New function computing Fibonacci numbers: mpz_fib_ui.
* New random number generators: mpf_urandomb, mpz_rrandomb, mpz_urandomb,
mpz_urandomm, gmp_randclear, gmp_randinit, gmp_randinit_lc_2exp,
gmp_randseed, gmp_randseed_ui.
* New function for quickly extracting limbs: mpz_getlimbn.
* New functions performing integer size tests: mpz_fits_sint_p,
mpz_fits_slong_p, mpz_fits_sshort_p, mpz_fits_uint_p, mpz_fits_ulong_p,
mpz_fits_ushort_p.
* New mpf functions: mpf_ceil, mpf_floor, mpf_pow_ui, mpf_trunc.
* New mpq function: mpq_set_d.
* New mpz functions: mpz_addmul_ui, mpz_cmpabs, mpz_cmpabs_ui, mpz_lcm,
mpz_nextprime, mpz_perfect_power_p, mpz_remove, mpz_root, mpz_swap,
mpz_tdiv_ui, mpz_tstbit, mpz_xor.
* New mpn function: mpn_divexact_by3.
* New CPU support: DEC Alpha 21264, AMD K6 and Athlon, HPPA 2.0 and 64,
Intel Pentium Pro and Pentium-II/III, Sparc 64, PowerPC 64.
* Almost 10 times faster mpz_invert and mpn_gcdext.
* The interface of mpn_gcd has changed.
* Better support for MIPS R4x000 and R5000 under Irix 6.
* Improved support for SPARCv8 and SPARCv9 processors.
Changes between GMP version 2.0 and 2.0.2
* Many bug fixes.
Changes between GMP version 1.3.2 and 2.0
* Division routines in the mpz class have changed. There are three classes of
functions, that rounds the quotient to -infinity, 0, and +infinity,
respectively. The first class of functions have names that begin with
mpz_fdiv (f is short for floor), the second class' names begin with mpz_tdiv
(t is short for trunc), and the third class' names begin with mpz_cdiv (c is
short for ceil).
The old division routines beginning with mpz_m are similar to the new
mpz_fdiv, with the exception that some of the new functions return useful
values.
The old function names can still be used. All the old functions names will
now do floor division, not trunc division as some of them used to. This was
changed to make the functions more compatible with common mathematical
practice.
The mpz_mod and mpz_mod_ui functions now compute the mathematical mod
function. I.e., the sign of the 2nd argument is ignored.
* The mpq assignment functions do not canonicalize their results. A new
function, mpq_canonicalize must be called by the user if the result is not
known to be canonical.
* The mpn functions are now documented. These functions are intended for
very time critical applications, or applications that need full control over
memory allocation. Note that the mpn interface is irregular and hard to
use.
* New functions for arbitrary precision floating point arithmetic. Names
begin with `mpf_'. Associated type mpf_t.
* New and improved mpz functions, including much faster GCD, fast exact
division (mpz_divexact), bit scan (mpz_scan0 and mpz_scan1), and number
theoretical functions like Jacobi (mpz_jacobi) and multiplicative inverse
(mpz_invert).
* New variable types (mpz_t and mpq_t) are available that makes syntax of
mpz and mpq calls nicer (no need for & before variables). The MP_INT and
MP_RAT types are still available for compatibility.
* Uses GNU configure. This makes it possible to choose target architecture
and CPU variant, and to compile into a separate object directory.
* Carefully optimized assembly for important inner loops. Support for DEC
Alpha, Amd 29000, HPPA 1.0 and 1.1, Intel Pentium and generic x86, Intel
i960, Motorola MC68000, MC68020, MC88100, and MC88110, Motorola/IBM
PowerPC, National NS32000, IBM POWER, MIPS R3000, R4000, SPARCv7,
SuperSPARC, generic SPARCv8, and DEC VAX. Some support also for ARM,
Clipper, IBM ROMP (RT), and Pyramid AP/XP.
* Faster. Thanks to the assembler code, new algorithms, and general tuning.
In particular, the speed on machines without GCC is improved.
* Support for machines without alloca.
* Now under the LGPL.
INCOMPATIBILITIES BETWEEN GMP 1 AND GMP 2
* mpq assignment functions do not canonicalize their results.
* mpz division functions round differently.
* mpz mod functions now really compute mod.
* mpz_powm and mpz_powm_ui now really use mod for reduction.
|