jasonmoxham
|
bc2fb3d67e
|
New core2/penryn mul_1 asm function
|
2010-12-17 00:54:02 +00:00 |
|
jasonmoxham
|
f4cad36a24
|
change calling convention on the asm code in x86_64 for redc_1
|
2010-12-16 22:39:03 +00:00 |
|
jasonmoxham
|
e34cec1068
|
change name of redc_basecase to redc_1 and export
|
2010-12-16 02:32:35 +00:00 |
|
jasonmoxham
|
874e62cbf0
|
New K8 asm function mpn_store
|
2010-12-05 10:35:49 +00:00 |
|
jasonmoxham
|
a87caeb1f8
|
faster core2/penryn mpn_hamdist by using the K8 version
|
2010-12-05 07:49:17 +00:00 |
|
jasonmoxham
|
77e6892a3c
|
windows conversion script doestn't like sums in code
|
2010-12-05 01:34:06 +00:00 |
|
jasonmoxham
|
b820af3933
|
New K8 asm logic functions mpn_nand_n mpn_nior_n
|
2010-12-05 00:55:20 +00:00 |
|
jasonmoxham
|
3fbe565821
|
New nehalem asm function popcount
|
2010-12-04 14:29:18 +00:00 |
|
jasonmoxham
|
014ff79403
|
New asm function mpn_hamdist for nehalem
|
2010-12-04 10:08:24 +00:00 |
|
jasonmoxham
|
08f9f8b572
|
New core2/penryn asm function mpn_popcount
|
2010-12-04 00:10:53 +00:00 |
|
jasonmoxham
|
8e1e363c9e
|
update asm def file
|
2010-12-03 23:49:07 +00:00 |
|
jasonmoxham
|
b6c1ca003c
|
new K8 asm function popcount hamdist
|
2010-12-03 23:22:36 +00:00 |
|
jasonmoxham
|
d9e61dbf34
|
new asm code for K10 popcount hamdist
|
2010-12-03 23:00:04 +00:00 |
|
jasonmoxham
|
72316c1513
|
add sandybridge to cpu's
|
2010-12-02 23:05:35 +00:00 |
|
jasonmoxham
|
97e2c4c0ef
|
x86_64 cpuid always availible , simplify detection code
|
2010-10-26 15:53:26 +00:00 |
|
jasonmoxham
|
5fe96fe1ec
|
replace multifunction aors_err1,2 with add,sub separate src code
|
2010-09-04 20:26:53 +00:00 |
|
jasonmoxham
|
42eb646f27
|
replace pre_divrem_1.* with preinv_divrem_1.*
|
2010-08-13 14:28:46 +00:00 |
|
jasonmoxham
|
7a678e4d9d
|
replace pre_mod_1.* with preinv_mod_1.*
|
2010-08-13 14:21:10 +00:00 |
|
jasonmoxham
|
4a8f88119c
|
replace mode1o.* with modexact_1c_odd.*
|
2010-08-13 14:09:23 +00:00 |
|
jasonmoxham
|
2708d3f949
|
replace divebyfobm1.* with divexact_byfobm1.* and rename consistently
|
2010-08-13 13:53:34 +00:00 |
|
jasonmoxham
|
411107f047
|
replace divebyff.* with divexact_byff.*
|
2010-08-13 13:23:52 +00:00 |
|
jasonmoxham
|
ff3a95453c
|
replace diveby3.* with divexact_by3c.*
|
2010-08-13 13:18:45 +00:00 |
|
jasonmoxham
|
5dad96491b
|
change dive_1.* to divexact_1.*
|
2010-08-13 12:54:25 +00:00 |
|
jasonmoxham
|
e0aae614a4
|
add missing asm functions for nano,netburst as fat requires asm (not C) functions for any function in the fat structure
|
2010-07-22 15:42:21 +00:00 |
|
jasonmoxham
|
b474f41608
|
correct fat entry , got wrong case
|
2010-07-22 11:29:38 +00:00 |
|
gladman
|
700ead092c
|
Change "Bm1of" to "fobm1" in text files (hopefully fairly safe)
|
2010-07-09 15:09:09 +00:00 |
|
gladman
|
3d80645be9
|
1. Change symbol mpn_divexact_bybM1of to mpn_divexact_fobm1 in *.c, *.h, *.in and config.* files
2. Change filenames divebybm1of.<c|asm|as> to divebyfobm1.<c|asm|as>
3. Change filename t-divebybm1of.c to t-divebyfobm1.c
4. Update Windows builds accordingly
|
2010-07-09 11:56:57 +00:00 |
|
jasonmoxham
|
aefcb57915
|
add in westmere cpu properly
|
2010-06-26 09:58:50 +00:00 |
|
jasonmoxham
|
484cceadc4
|
new K8only dir for asm that is faster on K8 but NOT K10
|
2010-06-26 09:00:48 +00:00 |
|
jasonmoxham
|
edc4f423ba
|
remove unused/tested x86_64 asm
|
2010-06-26 08:21:25 +00:00 |
|
mvngu
|
fa9329788c
|
nehalem: taurus.skynet Linux: tuning values for GCD_THRESHOLD and GCDEXT_THRESHOLD
|
2010-05-20 04:39:21 +00:00 |
|
mvngu
|
e6f15143f9
|
netburst: sextus.skynet Linux: tuning values for GCD_THRESHOLD and GCDEXT_THRESHOLD
|
2010-05-20 04:37:14 +00:00 |
|
mvngu
|
1d0a85ca7a
|
k8: flavius.skynet Linux: tuning values for GCD_THRESHOLD and GCDEXT_THRESHOLD
|
2010-05-20 04:24:32 +00:00 |
|
mvngu
|
089d53ce38
|
penryn: sage.math Linux: tuning values for GCD_THRESHOLD and GCDEXT_THRESHOLD
|
2010-05-20 04:09:19 +00:00 |
|
mvngu
|
e153633e9e
|
core2: tuning values for GCD_THRESHOLD and GCDEXT_THRESHOLD, averaged out for Linux and OS X
|
2010-05-20 04:03:05 +00:00 |
|
jasonmoxham
|
53edce5a20
|
removed all old gcdext_threshold that were set to zero
|
2010-05-12 20:56:21 +00:00 |
|
jasonmoxham
|
44a7c7d514
|
atom 64bit params
|
2010-05-12 16:26:27 +00:00 |
|
wbhart
|
73c82fbe5c
|
Added tuning for k102.
|
2010-05-11 23:36:25 +00:00 |
|
wbhart
|
0291aa8fa5
|
Fixed the gcd tuning code and added core2/penryn tuning.
|
2010-05-11 23:27:43 +00:00 |
|
wbhart
|
236fcac5c7
|
Fixed the xgcd normalisation issue and redid the tuning code for gcd and
xgcd.
|
2010-05-11 18:37:34 +00:00 |
|
(no author)
|
d41166e589
|
retune atom params
|
2010-04-08 21:49:34 +00:00 |
|
(no author)
|
eec9eac26f
|
Added tuning for i7 (thanks to Gonzalo Tornaria).
|
2010-04-08 00:19:26 +00:00 |
|
(no author)
|
fde0379365
|
Tuning for netburst.
|
2010-04-02 04:48:42 +00:00 |
|
(no author)
|
2f7dfb1c01
|
copy old fft tables into the updated gmp-mparam.h
|
2010-03-25 00:03:23 +00:00 |
|
(no author)
|
68ddf76fae
|
mpn/x86_64/k8/gmp-mparam.h from box2
|
2010-03-24 22:44:56 +00:00 |
|
(no author)
|
4edafb7ebc
|
mpn/x86_64/k8/k10/k102/gmp-mparam.h from lena
|
2010-03-24 19:35:05 +00:00 |
|
(no author)
|
5a9c6aa3eb
|
mpn/x86_64/core2/penryn/gmp-mparam.h from mod
|
2010-03-24 19:32:19 +00:00 |
|
(no author)
|
6c57fd310f
|
mpn/x86_64/atom/gmp-mparam.h from atom
|
2010-03-24 18:11:22 +00:00 |
|
(no author)
|
1555039bd3
|
mpn/x86_64/core2/gmp-mparam.h from eno
|
2010-03-24 17:43:47 +00:00 |
|
(no author)
|
627e796d79
|
mpn/x86_64/k8/k10/gmp-mparam.h from cuda1
|
2010-03-24 17:39:26 +00:00 |
|