a15f22a1f
bugfix for piledriver cgemm-, zgemm- and zgemv-kernel by
2014-06-28 11:46:58 +0200
b94ea89f5
bugfix for haswell cgemm- and zgemm-kernel by
2014-06-28 10:22:40 +0200
35f668bb1
bugfix for cgemm_kernel_8x2_sandy.S by
2014-06-28 10:01:56 +0200
4ebbf758f
(refs/pull/399/head)
.gitignore: add some more entries concerned with kernel by
2014-06-27 13:58:42 -0700
8615d6ec8
Build import libs as .dll.a instead of .lib by
2014-06-27 11:58:14 -0700
6c2ead30f
(refs/pull/398/head)
Remove all trailing whitespace except lapack-netlib by
2014-06-27 12:05:18 -0700
f41f03ab8
(refs/pull/397/head)
fix #394. this cleans up some handles after using them, and doesn't disable ALL process privileges upon success by
2014-06-27 12:10:04 -0400
365e8de34
added optimized cgemm-kernel for SANDYBRIDGE by
2014-06-27 13:40:29 +0200
578d1b621
added DSDOT definition and enabled optimized sdot kernel by
2014-06-27 11:30:29 +0200
a6ae079b1
added blas-test from lapack by
2014-06-27 10:12:19 +0200
d10db52ed
Merge pull request #390 from wernsaar/develop by
2014-06-27 14:57:06 +0800
dabab2b5f
(refs/pull/390/head)
added new optimized sgemm kernel for SANDYBRIGE by
2014-06-26 21:42:08 +0200
aa2709c4e
enabled optimized dgemm kernel for NEHALEM by
2014-06-26 12:22:29 +0200
9d6f2b594
Fortran flag -frecursive is disabled by default by
2014-06-25 13:55:19 +0200
a13bcc171
enabled optimized sgemv kernel for barcelona and piledriver by
2014-06-25 13:50:57 +0200
d2c82d754
enabled optimized sgemv kernel for HASWELL by
2014-06-25 12:56:45 +0200
0517672dd
enabled optimized sgemv kernels for nehalem, sandybridge and bulldozer by
2014-06-25 12:38:14 +0200
15d5dfa92
fixed compiler warnings by
2014-06-25 11:32:44 +0200
d83373db6
added parameter for gemm3m kernels by
2014-06-25 10:40:25 +0200
88b6bf251
force fallback for x86 32bit by
2014-06-22 17:27:11 +0200
4a2ab7460
Ref #391: force fallback for x86 32bit by
2014-06-22 13:51:17 +0200
86d8c8978
Ref #391: disabled SMP in ger.c and zger.c by
2014-06-22 12:01:24 +0200
316df0e82
fixed bug for INTERFACE64 by
2014-06-22 09:49:20 +0200
438002204
Ref #393: fix for INTERFACE64=0 and ARCH_X86 in divtable by
2014-06-21 12:29:23 +0200
23203d52c
Ref #380: lowered stack usage for haswell kernels by
2014-06-19 14:31:52 +0200
73545a79c
Ref #380: lowered stack usage for piledriver and bulldozer kernels by
2014-06-19 14:02:14 +0200
a19d20900
Ref #103: enhancement for small matrix dimensions by
2014-06-18 15:04:11 +0200
860281653
Merge pull request #387 from davidanthoff/fixbuilderroronwin by
2014-06-18 07:57:30 +0800
d52863cfd
Merge pull request #386 from wernsaar/develop by
2014-06-18 07:56:08 +0800
c6361d63c
(refs/pull/387/head)
Add -lgfortran flag to gcc call in a makefile by
2014-06-13 21:10:27 -0700
53bfa51ee
(refs/pull/386/head)
Ref #385: fixed warnings in dynamic.c by
2014-06-12 18:17:08 +0200
ff9cfca24
Ref #385: added missing return instruction by
2014-06-12 15:52:14 +0200
a86d349a5
Ref #380: enhancements for dynamic_arch by
2014-06-12 14:20:03 +0200
7b277f011
Merge pull request #384 from wernsaar/develop by
2014-06-11 09:49:27 +0800
faeab93df
(refs/pull/384/head)
Ref #51: added blas extensions simatcopy, dimatcopy, cimatcopy, zimatcopy by
2014-06-10 16:14:34 +0200
f773f492f
(tag: v0.2.9)
Merge branch 'develop' by
2014-06-10 21:55:47 +0800
21a6b5f79
OpenBLAS 0.2.9 Version. by
2014-06-10 21:55:19 +0800
cee257f38
Ref #51: added blas extensions zomatcopy and comatcopy by
2014-06-10 10:34:54 +0200
7bfb3011e
Ref #51: added blas extension somatcopy by
2014-06-09 20:21:13 +0200
8c8f59623
Ref #51: added blas extension domatcopy as not opimized reference by
2014-06-09 17:11:07 +0200
bff575d0b
Ref #375: added workaround for small sizes to scal.c and zscal.c by
2014-06-08 13:49:19 +0200
faf3ac0aa
Ref #285: added axpby kernels by
2014-06-08 11:54:24 +0200
a40116de2
Fixed generating DLL bug. by
2014-06-06 16:13:08 +0800
b31ec9937
Fixed #374. by
2014-06-05 17:01:12 +0800
0ac073fa9
Merge pull request #376 from wernsaar/develop by
2014-05-26 04:46:06 -0500
25e899b60
(refs/pull/376/head)
fixed function profile in zpotri.c by
2014-05-25 09:15:22 +0200
219bcb119
added lapack and lapacke timing libs by default by
2014-05-24 15:53:25 +0200
566444554
changed threshold value for sep.in from 50.0 to 60.0 by
2014-05-23 17:26:50 +0200
89da45080
enabled and tested optimized potri lapack functions by
2014-05-23 12:14:30 +0200
c26bbee48
enabled abd tested optimized trtri lapack functions by
2014-05-23 10:55:39 +0200
445b6f2a9
(refs/pull/374/merge)
Merge 81e5f0c37e into 10a16bd690 by
2014-05-23 03:27:39 +0000
81e5f0c37
(refs/pull/374/head)
Remove all trailing whitespace by
2014-05-22 20:26:48 -0700
ced13574a
Random "walk (a)round" --> "work-around" typo fixes by
2014-05-22 18:10:33 -0700
fe858873a
Add NO_STATIC variable which disables static lib installation by
2014-05-22 18:05:19 -0700
a8d4d1c4d
Build import library for mingw by
2014-05-17 16:21:21 -0700
c4ccb3fbb
removed lapack/getri because it was never used by
2014-05-21 14:21:19 +0200
a748d3a75
enabled optimized trti2 lapack functions again by
2014-05-21 11:02:07 +0200
a5ab231ad
enabled optimized complex lauum lapack functions again by
2014-05-21 10:35:28 +0200
dbaeea7b5
enabled lauu2 and lauum lapack functions again by
2014-05-21 09:49:18 +0200
10a16bd69
Refs #372. Fixed a lot of bugs about LAPACK testing. As a walk round solution, we rolled back some kernels. by
2014-05-21 11:25:11 +0800
406f5bd22
Merge branch 'develop' of https://github.com/wernsaar/OpenBLAS into wernsaar-develop by
2014-05-21 11:24:39 +0800
a0ae53966
(refs/pull/372/head)
removed debug flag from Makefile.rule by
2014-05-19 15:57:18 +0200
0d75f3b6a
enabled and tested optimized gesv lapack functions by
2014-05-19 14:44:53 +0200
abad6f66d
marked trti2.c and ztrti2.c as bad by
2014-05-19 13:50:02 +0200
2ff66e661
enabled and tested optimized laswp lapack function by
2014-05-19 13:35:32 +0200
5e5503492
marked zlauu2.c and zlauum.c as bad by
2014-05-19 12:53:22 +0200
9a9e81023
marked trtri.c and ztrtri as bad by
2014-05-19 12:42:52 +0200
45be9ac11
moved trtri.c and ztrtri.c to the directory lapack by
2014-05-19 12:29:29 +0200
9f201558c
marked lauu2.c and lauum.c as bad by
2014-05-19 12:00:16 +0200
d4237cb7f
marked larf.c as obsolete by
2014-05-19 11:23:17 +0200
d2a8ff4b0
Merge branch 'TimothyGu-develop' into develop by
2014-05-19 10:37:20 +0800
6c53f15c0
(refs/pull/371/merge)
Merge f331cb1a76 into c38379c9dd by
2014-05-19 02:12:03 +0000
f331cb1a7
(refs/pull/371/head)
Remove code for downloading lapack tarball and the patches themselves by
2014-05-18 19:09:26 -0700
9ed981c5d
Remove unused dll2 target by
2014-05-18 18:54:38 -0700
aaa9d7fbf
marked potri functions as bad because a lot of errors by
2014-05-18 23:41:13 +0200
ebc95e6f1
enabled and tested optimized potf2 lapack functions by
2014-05-18 22:41:43 +0200
61a2c50e8
enabled and tested optimized getf2 lapack functions by
2014-05-18 22:21:16 +0200
4f98f8c9b
enabled and tested optimized potrf lapack functions by
2014-05-18 21:42:37 +0200
536875d46
enabled and tested optimized getrs lapack functions by
2014-05-18 21:13:56 +0200
65f2fba4c
enabled and tested optimized cgetrf lapack function by
2014-05-18 20:32:27 +0200
eea6f51df
enabled and tested optimized sgetrf lapack function by
2014-05-18 20:01:23 +0200
6fc464670
enabled and tested optimized zgetrf lapack function by
2014-05-18 19:36:32 +0200
ac029f81b
enabled and tested optimized dgetrf function by
2014-05-18 19:07:51 +0200
c0cf875a8
added optimized lapack files from OpenBLAS by
2014-05-18 14:09:22 +0200
b6d904838
Remove routines for generating exports/symbol.S by
2014-05-17 16:02:36 -0700
5379eff02
Remove routines for making exports/linux.def by
2014-05-17 16:01:30 -0700
aaddb0541
bugfix for ARMV6 by
2014-05-17 13:00:36 +0200
e52532a9f
enable debug for lapack testing by
2014-05-17 11:18:26 +0200
e826a5a6a
some modifications regarding lapack test by
2014-05-16 20:37:41 +0200
165d5436b
changed threshold to 50.0 by
2014-05-16 20:34:48 +0200
409b52255
changed default optimization flag from O3 to O2 for ARM by
2014-05-16 14:36:24 +0200
5953972a5
changed threshold for 50.0 to 54.0 in svd.in by
2014-05-16 14:32:10 +0200
d751224ea
changed YIELDING for BULLDOZER by
2014-05-15 11:37:38 +0200
4a5938b5c
Modified lapack-test, using lapack_testing.py to run tests by
2014-05-14 15:16:21 +0200
d18bc5468
added FCOMMON_OPT for lapack by
2014-05-14 15:01:03 +0200
8877c6db5
changed label lapack-test by
2014-05-14 13:08:05 +0200
c38379c9d
bugfixes for ARM regarding lapack tests by
2014-05-14 13:03:45 +0200
a0b07c144
bugfixs for ARM regarding lapack tests by
2014-05-14 12:59:20 +0200
43fbdb7a5
added ARMV5 as reference platform by
2014-05-13 17:25:19 +0200
777cebc8c
added ZERO check to zscal.c because bug in lapack-testing by
2014-05-13 16:31:00 +0200