Commit Graph

651 Commits

Author SHA1 Message Date
(no author)
660b1caaf4 Fixed multiplications being too unbalanced for Toom8h. 2010-02-18 13:15:23 +00:00
(no author)
c4746c30a7 tune for windows nehalem 2010-02-17 10:43:21 +00:00
(no author)
bb96dd4a16 Finished putting too8 mul in. 2010-02-17 00:29:00 +00:00
(no author)
710f109c76 Added missing function and made some corrections. 2010-02-17 00:09:31 +00:00
(no author)
3fad25a740 Added toom8 files. 2010-02-16 23:47:07 +00:00
(no author)
11f57996a4 1. remove C99 out of order declarations in rootrem.c
2. Add latest source code to Windows build
2010-02-15 10:15:43 +00:00
(no author)
deb47bec36 Removed an unnecessary line. 2010-02-14 01:07:54 +00:00
(no author)
2515976d28 Minor cleanup. 2010-02-13 23:36:35 +00:00
(no author)
a39e1da77b Whoops, correct a typo. 2010-02-13 21:55:35 +00:00
(no author)
6d174ba39a Corrected small root case. 2010-02-13 21:54:26 +00:00
(no author)
a2c42da38f Fiddled with tuning value. 2010-02-13 21:20:12 +00:00
(no author)
20ae4de5b1 Ran autoconf, connnected up rootrem code and changed tuning value. 2010-02-13 21:18:11 +00:00
(no author)
afd2f2bca4 Made some minor changes to rootrem code. 2010-02-13 21:11:18 +00:00
(no author)
01e8f2702a Added GMP rootrem code and added crossover from old code, which is faster. 2010-02-13 21:01:24 +00:00
(no author)
ee376cce80 Turned on inv_div_qr code in the tdiv_qr function. 2010-02-13 18:34:48 +00:00
(no author)
397a0689cd Turned on new inv_div_qr division code. 2010-02-13 18:09:07 +00:00
(no author)
d2eac73245 Added configure and prototype infor for new inv_div_qr(_n) functions and
fixed a typo.
2010-02-13 17:58:56 +00:00
(no author)
dcc043e3ab Added code for balanced and unbalanced division with remainder using a
precomputed inverse.
2010-02-13 17:48:01 +00:00
(no author)
b4e21c591f Fixed approximate quotient code. 2010-02-13 17:13:29 +00:00
(no author)
c100247b24 Added a test for inv_divappr_q_n and made it pass. 2010-02-11 20:36:56 +00:00
(no author)
e3b6db3580 Fixed some segfaults. 2010-02-11 18:52:17 +00:00
(no author)
31b09b4302 Fixed some typos. 2010-02-11 16:49:43 +00:00
(no author)
5924f0b6da Added asymptotically fast division code. 2010-02-11 16:39:41 +00:00
(no author)
048820278c Added tdiv_qr.c from GMP. 2010-02-11 14:23:21 +00:00
(no author)
64da22761e Fixed a bug I accidentally introduced into the code. 2010-02-11 13:59:20 +00:00
(no author)
f83a14a538 More fiddling with tuning. 2010-02-11 00:20:44 +00:00
(no author)
1d08ade198 Added sb_div_qr and dc_div_qr from GMP sources. Fixed some
bugs, incorrect #defines and fiddled with tuning.
2010-02-10 22:48:46 +00:00
(no author)
490c26c544 Removed and error. 2010-02-10 11:38:05 +00:00
(no author)
25e405dee5 Restored some of original GMP code. 2010-02-10 11:36:40 +00:00
(no author)
808c57b5b3 Fixed some typos. 2010-02-10 11:22:39 +00:00
(no author)
f4a69eec60 Added sb_div_q, sb_divappr_q from GMP and tdiv_3by2 and invert_1 from GMP
and changed everything to use them.
2010-02-10 11:17:59 +00:00
(no author)
61e882ef48 Added sb_div_q based on dc_div_q code.
Added better tuning values.
Fixed a segfault.
2010-02-10 02:09:24 +00:00
(no author)
068f94dd5c Fixed a segfault. 2010-02-09 23:25:48 +00:00
(no author)
9086a12d6c Moved dcpi1_dc_div_q.c from GMP to dc_div_q.c (modified trivially).
Moved dcpi1_divappr_q.c from GMP to dc_divappr_q.c (modified heavily to use 
only MPIR functions), removing dcpi1_divappr_q_n function. Moved div_q.c 
from GMP to tdiv_q.c (modified trivially). 

Currently tests do not pass. More work required!
2010-02-09 21:19:08 +00:00
(no author)
bf50a2d72e Add another missed mp_bitcnt_t change 2010-02-09 10:22:47 +00:00
(no author)
1758461822 Another minor change to core2 tuning values. Seems to slightly improve
timings (though almost imperceptibly).
2010-02-09 00:43:06 +00:00
(no author)
88c8b3e0f3 Add mp_bitcnt_t type to MPIR 2010-02-08 17:06:12 +00:00
(no author)
00f292cc48 adjust windows tune slightly 2010-02-08 10:02:55 +00:00
(no author)
6b59c6b0ba Fixed some bugs in mpn_sqr_n to do with SQR_THRESHOLDS. 2010-02-07 20:58:21 +00:00
(no author)
a909902fe4 Fixed a bug in toom7_sqr_n. 2010-02-07 20:46:54 +00:00
(no author)
015ce32bf7 correct bug in new assembler code 2010-02-07 17:59:56 +00:00
(no author)
3306eae2a2 Add HAVE_NATIVE defines for Windows code 2010-02-07 17:33:56 +00:00
(no author)
9639773959 Hand tuned SQR thresholds for core2/penryn. 2010-02-07 15:43:19 +00:00
(no author)
3c16b2f566 1. Update g2y.py to cope with GMP assembler
2. add new assembler to Windows build
2010-02-07 14:48:02 +00:00
(no author)
2edb3c830a Added a clarification to the copyright. 2010-02-07 14:07:42 +00:00
(no author)
0c7b48b1fd Convert divrem_2.asm to yasm format. 2010-02-07 14:02:39 +00:00
(no author)
a54d23d652 Convert divem_euclidean_qr_1.asm to yasm format. 2010-02-07 13:32:16 +00:00
(no author)
0f5bb75287 New tuning values for core2/penryn. 2010-02-07 03:28:55 +00:00
(no author)
c7ade0be6f Replaced nhgcd2 with the GMP hgcd2.c file, removing the matrix1_vector fns and
replacing them with the original.
2010-02-06 15:51:49 +00:00
(no author)
06d4a32943 Added divrem_1 and divrem_2 x86_64 assembly code from GMP, replaced
divrem_euclidean_qr_1 with divrem_1.
2010-02-06 13:28:23 +00:00
(no author)
0bd0a87f7e Temporarily committing these so they can be converted to yasm format. 2010-02-06 03:16:22 +00:00
wbhart
1c2f0506f5 Fixed a bug in mpn_gcdext reported by Dan Grayson. 2010-01-30 22:27:37 +00:00
wbhart
c8d33128b5 Made a change to hopefully support unaligned memory allocation as requested by
Dan Grayson. Passes make check and try mpn_lshift.
2010-01-10 23:54:55 +00:00
gladman
0ac38fd71c 1. repair Windows build script
2. Add tuning for nehalem on Windows
2010-01-06 15:07:32 +00:00
wbhart
0200f63f3b Added tuning params for K102. Added new fft code to main directory. 2010-01-01 14:06:41 +00:00
gladman
73dccb229b correct bug I introduced into invert.c 2009-12-09 20:30:32 +00:00
gladman
aaf7949430 1. update run-test.py for Windows build
2. remove test code in invert.c in Windows build
2009-12-09 20:02:31 +00:00
wbhart
52a2c8b839 Added my copyright to the file mpn/generic/invert.c. 2009-12-09 18:56:38 +00:00
wbhart
7db71713ef Removed a mid block declaration. 2009-12-09 18:40:54 +00:00
gladman
8d7347c60c 1. Add mpn.invert test to Windows
2. remove test subroutine from mpn/invert.c
2009-12-09 09:06:31 +00:00
gladman
7e5051b5d6 1. Add a new test to Windows build
2. Remove C99 features from two new files
2009-12-07 09:18:30 +00:00
wbhart
d252f5b22e Fixed the test code for mpz_likely_prime_p. 2009-12-07 01:06:47 +00:00
wbhart
ce52d5515b Removed a small bit of unnecessary code. 2009-12-06 21:55:07 +00:00
wbhart
90f2cf62f0 Rewrte dc_divappr giving proof sketch and making it slightly different. 2009-12-06 21:14:36 +00:00
gladman
11c4a69572 Add Windows build for Intel nehalem 2009-12-03 17:32:55 +00:00
gladman
e3f6129a64 2009-12-02 17:09:35 +00:00
gladman
1485e3dc1b 1. Add Jason's new assembler code to the Windows builds
2. Tidy up assembler to prepare for Windows nehalem build
2009-12-02 16:24:00 +00:00
gladman
b086474ddf Bring Windows naming into line with that on Linux
Rename 'amd64' elemnts to 'k8'
2009-12-02 12:29:38 +00:00
jasonmoxham
829dacd87a New core2/penryn asm functions popcount hamdist 2009-11-19 10:53:45 +00:00
jasonmoxham
afc620f493 new K8 asm functions mpn_popcount mpn_hamdist 2009-11-19 10:38:40 +00:00
jasonmoxham
367f00f4fe New K8 asm functions mpn_and,ior,xor 2009-11-19 10:35:17 +00:00
jasonmoxham
beb4d5b735 New K8 asm funtions mpn_lshift2,3,4,5,6 2009-11-19 10:32:03 +00:00
jasonmoxham
7bd0558974 New atom asm funtion mpn-copyi 2009-11-19 09:09:50 +00:00
jasonmoxham
ef2b3db6ef remove # comments 2009-11-19 08:49:14 +00:00
jasonmoxham
8bcfe2975a New nehalem asm mpn_copyi mpn_copyd 2009-11-19 08:45:27 +00:00
jasonmoxham
579e36f2b1 New nehalem asm mpn_com 2009-11-19 08:44:50 +00:00
jasonmoxham
59bf8d86e4 new nehalem asm logic mpn fn's 2009-11-19 08:37:54 +00:00
jasonmoxham
dcf3afa567 convert addlsh from gas to yasm format 2009-11-18 17:43:25 +00:00
wbhart
969b44ec5a attempt to push windows tuning update
Signed-off-by:  <Brian Gladman>
2009-11-17 13:34:35 +00:00
wbhart
dc983a5503 Added some clarifying comments to the mulmid code. 2009-11-17 13:34:30 +00:00
wbhart
9cb7c5cf1f Switch over to exact quotient when divisor gets large in mpn_tdiv_q. 2009-11-17 13:15:23 +00:00
jasonmoxham
1757343142 new directory for core2-apple-darwin 32bit with no PIC code in it 2009-10-24 17:31:05 +00:00
wbhart
be0a749738 Oops, wrong author added to license info. 2009-10-19 05:59:49 +00:00
wbhart
ecdbd4a162 Added copyright info to GCD and XGCD code. 2009-10-19 05:49:29 +00:00
jasonmoxham
c6af9fbfc2 some more masm? movq/movd mixups 2009-10-16 00:45:14 +00:00
jasonmoxham
4ba747128d change movq to movd for old masm assembler 2009-10-15 18:21:27 +00:00
jasonmoxham
be135c7347 change asm #comment to C comment 2009-10-15 18:13:19 +00:00
gladman
0965fc4532 update VC++ readme.txt and Core2 tuning 2009-10-15 16:24:35 +00:00
wbhart
19c4d52566 Fixed a carry bug in mpn_tdiv_q and added the failing cases to tests/mpz/t-tdiv.c. 2009-10-13 04:52:00 +00:00
wbhart
07bbd31e1d Fixed more bugs in sb_divappr_q. 2009-10-12 02:05:16 +00:00
jasonmoxham
d0067e2f3b fix broken sparc32/v9 add/sub code 2009-10-11 16:32:54 +00:00
wbhart
380b80372a Fixed some bugs in mpn_tdiv_q and added a try test for mpn_tdiv_q. 2009-10-11 06:46:10 +00:00
jasonmoxham
6edccbf39a adjust sparc32/v9=t2,ABI=32 params for mulhigh 2009-10-09 20:25:51 +00:00
wbhart
0f39b0f20c Clarified some comments. 2009-10-09 04:17:32 +00:00
wbhart
5e38fe6d8b Added some rough justifications for doing what we do in sb_divappr_q. 2009-10-09 03:48:31 +00:00
jasonmoxham
ce832e54a5 varro powerpc64/mode64 64bit params 2009-10-09 01:27:36 +00:00
jasonmoxham
b8c10614e7 ia64 params 2009-10-09 00:33:52 +00:00
jasonmoxham
45ec139248 pentium4 32bit params 2009-10-09 00:33:19 +00:00
jasonmoxham
b49482cf47 core2 32bit params 2009-10-09 00:10:27 +00:00
jasonmoxham
c9f16233b8 add back in old fft tuning values , better than nothing 2009-10-08 22:50:40 +00:00
jasonmoxham
297e277fe6 add mising fat entry for 32bit nano 2009-10-08 22:39:37 +00:00
jasonmoxham
518226d914 atom params 2009-10-08 22:06:01 +00:00
jasonmoxham
2488ed55e8 k7 params 2009-10-08 22:04:10 +00:00
jasonmoxham
93688a18b6 k10 params 2009-10-08 18:01:18 +00:00
jasonmoxham
1a7c40bb2b k8 32bit params 2009-10-08 17:52:27 +00:00
jasonmoxham
265bd0a989 sparc64 params's 2009-10-08 17:49:13 +00:00
jasonmoxham
f09d7f9df1 sparc 32 bit param 2009-10-08 17:05:41 +00:00
jasonmoxham
2f138f7a16 core2 params 2009-10-08 15:48:27 +00:00
jasonmoxham
e49eccab57 k8 params 2009-10-08 15:30:59 +00:00
jasonmoxham
4053a62930 k102 params 2009-10-08 15:28:51 +00:00
wbhart
8258200ecb Fixed a major bug in mpn_tdiv_q and extended some tests to run longer. 2009-10-08 11:54:06 +00:00
wbhart
a4d22db74e Extended the mpz/t-tdiv test and fixed a bug in mpn_tdiv_q. 2009-10-08 02:54:11 +00:00
jasonmoxham
9fb5a0cb90 nehalem 32bit params 2009-10-07 12:21:08 +00:00
jasonmoxham
2821267426 nehalem 64 params 2009-10-07 12:00:35 +00:00
wbhart
6d16ef649e Fixed unusual bug in sb_divappr_q (I think). 2009-10-07 09:58:10 +00:00
gladman
448c60fd77 Add latest division code to the Windows builds 2009-10-07 09:21:06 +00:00
wbhart
938a8de8ed Added fast mpn_tdiv_q and test code and used it in mpz_tdiv_q. 2009-10-07 05:27:59 +00:00
gladman
f42829cd8e Core2 assembler (copies) for Windows 2009-10-05 15:56:30 +00:00
jasonmoxham
ff4fc75bb8 New core2/penryn asm fns mod_1_? divrem_hensel_qr_1_2 rsh_divrem_hensel_qr_1_2 2009-10-05 15:02:50 +00:00
jasonmoxham
a10fd0d649 New atom asm functions mod_1_? rsh_divrem_hensel_qr_1_1 divrem_hensel_qr_1_1 2009-10-05 14:50:05 +00:00
jasonmoxham
dc1fc39381 New nehalem asm functions mod_1_? rsh_divrem_hensel_qr_1_2 divrem_hensel_qr_1_2 2009-10-05 14:19:08 +00:00
gladman
343fe6ce88 Add mod_1_3 assembler to Windows 2009-10-05 08:29:15 +00:00
jasonmoxham
f8c338ad86 remove old copy,zero from fft 2009-10-05 02:14:27 +00:00
jasonmoxham
4d25d12c67 remove old dead addsub 2009-10-05 02:05:18 +00:00
jasonmoxham
ad4a181469 New AMD asm function mpn_mod_1_3 2009-10-04 23:49:02 +00:00
gladman
01525ea9dd Add new k8 assembler to Windows builds 2009-10-04 11:36:18 +00:00
jasonmoxham
d10523b3f6 Add threshold for euclid/hensel divrem_1 switchover 2009-10-04 02:56:20 +00:00
jasonmoxham
bf0f5c4e6f add carry limb into the existing rsh_divrem_hensel 2009-10-04 02:16:25 +00:00
jasonmoxham
e17eb68ad9 add generic mpn_rsh_divrem_hensel_qr_1_? 2009-10-03 23:13:13 +00:00
jasonmoxham
144deaa17e autotools 2009-10-03 22:23:58 +00:00
jasonmoxham
cdaad5bffc New AMD asm function mpn_rsh_divrem_hensel_qr_1_2 2009-10-03 22:21:44 +00:00
gladman
fc7f72765e Add mod_1_<n> assembler to Windows AMD64 builds 2009-10-03 11:06:24 +00:00
jasonmoxham
90d8b76405 New amd asm function mpn_mod_1_2 2009-10-03 00:04:27 +00:00
gladman
6ca7f4f425 Update Windows timing code 2009-10-02 15:21:04 +00:00
gladman
8c2a0b5882 1. Add generic C mod_1_<n> files to the Windows build
2. Add Windows specific timing code (win_timing.h & .c)
2009-10-02 14:01:13 +00:00
jasonmoxham
520fec686d tweek mod_1_1 amd asm to full speed 2009-10-02 02:57:31 +00:00
jasonmoxham
a7eee56d24 autotools and a few tweeks 2009-10-02 00:41:53 +00:00
wbhart
7edce5c5c6 Fixed a segfault in t-sb_divappr_q. 2009-10-02 00:28:36 +00:00
jasonmoxham
adfded6fe5 split out mpn_mod_1_? from divrem_euclidean_r and add New asm function for AMD for mod_1_1 2009-10-01 22:04:24 +00:00
gladman
1fabd4dbbf 2009-10-01 09:26:35 +00:00
jasonmoxham
07f97a0963 new AMD asm function mpn_divrem_hensel_qr_1_2 2009-09-30 23:41:09 +00:00
gladman
f700c5c84c Add new assembler, C and test files to Windows builds 2009-09-30 16:21:46 +00:00
wbhart
5fcad2a384 Added files for divide conquer division. 2009-09-30 03:07:18 +00:00
jasonmoxham
24d1b6c39b rename divrem_hensel amd asm to match 2009-09-30 02:52:41 +00:00
jasonmoxham
9bbfba8e5e New generic functions for two limb inverse of divrem_hensel_qr_1 and tunes,tests 2009-09-30 02:24:29 +00:00
jasonmoxham
56801786a7 New asm functions for AMD divrem_hensel_qr_1 divrem_hensel_r_1 2009-09-29 23:52:09 +00:00
wbhart
1601f69b05 Fixed a missing mpir.h. 2009-09-29 23:19:37 +00:00
wbhart
5308d1e8fd Got invett and sb_divappr_q working (modulo some linker issue). 2009-09-29 22:55:10 +00:00
jasonmoxham
4783c9dc4e New AMD rsh_divrem_hensel_qr_1 asm fn 2009-09-29 21:53:03 +00:00
jasonmoxham
61c42e5ae1 New generic divrem_hensel's functions and tests,tunes 2009-09-29 16:41:28 +00:00