26d7f0620
Merge pull request #827 from wernsaar/develop by
2016-03-30 12:04:49 +0200
68a69c5b5
(refs/pull/827/head)
added optimized dgemv_n kernel for POWER8 by
2016-03-30 11:10:53 +0200
a571359af
Merge pull request #826 from wernsaar/develop by
2016-03-28 15:09:52 +0200
c2464a7c4
(refs/pull/826/head)
added optimized casum kernel for POWER8 by
2016-03-28 14:12:08 +0200
294f93386
added optimized zasum kernel for POWER8 by
2016-03-28 13:37:32 +0200
f59c9bd6e
added optimized sasum kernel for POWER8 by
2016-03-28 12:44:25 +0200
c53be46d7
added optimized dasum kernel for POWER8 by
2016-03-28 12:17:15 +0200
bbb2d73d7
Merge pull request #825 from wernsaar/develop by
2016-03-27 19:04:06 +0200
659ed1659
(refs/pull/825/head)
added otimized cswap and zswap kernels for POWER8 by
2016-03-27 18:31:37 +0200
35c98a355
added optimized zscal kernel for POWER8 by
2016-03-27 16:31:50 +0200
f1a5dd06c
added optimized sscal kernel for POWER8 by
2016-03-27 11:05:56 +0200
e125a3dc3
Merge pull request #824 from wernsaar/develop by
2016-03-27 10:43:17 +0200
35f1f21a7
(refs/pull/824/head)
added drot- and srot-kernel optimimized for POWER8 by
2016-03-27 08:57:11 +0200
7b4b7179b
Merge pull request #819 from ashwinyes/develop_20160324_fixes_optimizations by
2016-03-27 00:04:20 -0400
7a92c1538
added benchmark test for srot and drot by
2016-03-26 07:14:13 +0100
572726814
Merge pull request #823 from wernsaar/develop by
2016-03-25 18:08:48 +0100
3d9a50e84
(refs/pull/823/head)
added optimized sswap kernel for POWER8 by
2016-03-25 17:34:55 +0100
828c849b4
added optimized ccopy kernel for POWER8 by
2016-03-25 16:54:25 +0100
ecc0bc981
added optimized scopy kernel for POWER8 by
2016-03-25 16:06:56 +0100
12f209b7b
added optimized zswap kernel for POWER8 by
2016-03-25 15:27:34 +0100
7316a8793
added optimized dswap kernel for POWER8 by
2016-03-25 14:35:43 +0100
0bff057a8
added optimized dcopy kernel for POWER8 by
2016-03-25 13:03:02 +0100
7ee1d29dd
Merge pull request #822 from wernsaar/develop by
2016-03-25 10:15:51 +0100
1e6cf9808
(refs/pull/822/head)
added optimized dscal kernel for POWER8 by
2016-03-25 09:42:08 +0100
749e605e9
(refs/pull/820/head)
Affinity shared memory area must be at least a page by
2016-03-24 06:17:06 +0000
278511ad2
(refs/pull/819/head)
Cortex-A57: Fix clang compilation errors by
2016-03-24 10:31:28 +0530
3b5ffb49d
Cortex-A57: Improve DGEMM 8x4 Implementation by
2016-03-17 10:23:51 +0530
8519e4ed9
Merge pull request #817 from wernsaar/develop by
2016-03-23 13:37:04 +0100
55eda3813
(refs/pull/817/head)
added optimized zaxpy kernel for POWER8 by
2016-03-23 11:20:23 +0100
53bfc83c2
Update appveyor version. by
2016-03-22 11:37:35 -0400
13ca89f6f
Merge pull request #813 from theoractice/develop by
2016-03-22 11:31:37 -0400
461cf9ea3
Merge pull request #814 from wernsaar/develop by
2016-03-22 15:24:59 +0100
0664ba4c9
(refs/pull/814/head)
added optimized daxpy kernel for POWER8 by
2016-03-22 14:50:03 +0100
aa744dfa5
(refs/pull/813/head)
Update memory.c by
2016-03-22 20:02:37 +0800
61cf8f74d
Fix access violation on Windows while static linking by
2016-03-22 19:14:54 +0800
de202fa37
Merge pull request #1 from xianyi/develop by
2016-03-22 05:33:20 -0500
6f93b5359
Merge pull request #812 from wernsaar/develop by
2016-03-21 13:59:44 +0100
11c44dede
(refs/pull/812/head)
added optimized sdot kernel for POWER8 by
2016-03-21 13:18:23 +0100
f00d64259
Merge pull request #811 from wernsaar/develop by
2016-03-21 10:48:41 +0100
9e4584d06
(refs/pull/811/head)
added optimized zdot kernel for POWER8 by
2016-03-21 10:12:07 +0100
2a5679da5
Merge branch 'release-0.2.17' into develop by
2016-03-20 20:52:43 -0400
a71e8c82f
(tag: v0.2.17)
Fix change log typo. by
2016-03-20 20:52:15 -0400
9b987badb
Merge branch 'master' into develop Bump to 0.2.18.dev by
2016-03-20 20:48:21 -0400
1619b2f3c
Merge branch 'release-0.2.17' by
2016-03-20 20:44:01 -0400
4f3153395
Update doc for 0.2.17. by
2016-03-20 20:43:42 -0400
d7a1a7ff2
Merge branch 'release-0.2.17' into develop by
2016-03-20 09:24:28 -0400
308e6195b
Refs #807. Enable BUILD_LAPACK_DEPRECATED=1 by default. by
2016-03-20 09:22:56 -0400
7a3d7b1f5
Merge pull request #808 from theoractice/develop by
2016-03-20 09:07:47 -0400
74cc2d662
Merge pull request #809 from wernsaar/develop by
2016-03-20 13:16:41 +0100
fc3a55851
(refs/pull/808/head)
Fix a minor compiler error in VisualStudio with CMake by
2016-03-20 18:58:18 +0800
cd9fafc05
(refs/pull/809/head)
ddot for POWER8: updated licence information by
2016-03-20 11:19:27 +0100
84b92e637
added optimized ddot kernel for POWER8 by
2016-03-20 11:06:06 +0100
c279a53ed
Merge pull request #806 from wernsaar/develop by
2016-03-18 12:46:16 +0100
e1df5a6e2
(refs/pull/806/head)
fixed sgemm- and strmm-kernel by
2016-03-18 12:12:03 +0100
5c658f874
add optimized cgemm- and ctrmm-kernel for POWER8 by
2016-03-18 08:17:25 +0100
ec4390a96
Bump devlop version to 0.2.17.dev. by
2016-03-15 14:52:01 -0400
fced5744f
(tag: v0.2.16)
Merge branch 'release-0.2.16' by
2016-03-15 14:49:10 -0400
8c0fb1258
Update 0.2.16 doc by
2016-03-15 14:48:41 -0400
aae581d00
Merge branch 'develop' into release-0.2.16 by
2016-03-15 13:56:01 -0400
e17303933
Merge pull request #802 from ashwinyes/develop_20160314_dgemm_optimization by
2016-03-14 20:31:03 -0400
f9226275f
Merge pull request #801 from Keno/patch-3 by
2016-03-14 15:42:31 -0400
cf8c7e28b
(refs/reviewable/pr802/r1, refs/pull/802/head)
Update CONTRIBUTORS.md by
2016-03-14 19:59:41 +0530
5ac02f6dc
Optimize Dgemm 4x4 for Cortex A57 by
2016-03-14 19:35:23 +0530
7aa1ad492
Functional Assembly Kernels for CortexA57 by
2016-03-14 19:33:21 +0530
dcd15b546
BUGFIX: KERNEL.POWER8 by
2016-03-14 14:36:59 +0100
96284ab29
added sgemm- and strmm-kernel for POWER8 by
2016-03-14 13:52:44 +0100
d5e1255ca
(refs/reviewable/pr801/r1, refs/pull/801/head)
Don't pass REALNAME to `.end` by
2016-03-13 18:56:21 -0400
587455868
Merge pull request #800 from jeromerobert/smallscaling by
2016-03-10 15:45:33 -0500
323c237e7
(refs/reviewable/pr800/r1, refs/pull/800/head)
Fix smallscaling compilation by
2016-03-10 20:24:41 +0100
faa5e2e5e
FIX: forgot the add the files cgemv_n_4.c and cgemv_t_4.c by
2016-03-10 11:10:38 +0100
551fdf53e
Merge pull request #799 from wernsaar/develop by
2016-03-10 10:22:08 +0100
fdf291be3
(refs/reviewable/pr799/r1, refs/pull/799/head)
Added optimized cgemv_n and cgemv_t kernels for bulldozer, piledriver and steamroller by
2016-03-10 09:42:07 +0100
68eb4fa32
Add missing openblas_env makefile. by
2016-03-09 14:52:47 -0500
05196a849
Refs #716. Only call getenv at init function. by
2016-03-09 12:50:07 -0500
db9b611b1
Merge pull request #798 from wernsaar/develop by
2016-03-09 15:55:56 +0100
2e6333f74
(refs/reviewable/pr798/r1, refs/pull/798/head)
modified common.h for piledriver by
2016-03-09 15:48:29 +0100
c99cc41cb
Added optimized zgemv_n kernel for bulldozer, piledriver and steamroller by
2016-03-09 14:02:03 +0100
711ecb8bd
Merge pull request #797 from wernsaar/develop by
2016-03-07 16:44:17 +0100
10c2ebdfc
(refs/reviewable/pr797/r1, refs/pull/797/head)
BUGFIX: removed fixes for bugs #148 and #149, because info for xerbla is wrong by
2016-03-07 10:34:04 +0100
26b3b3a3e
bugfixes form lapack svn for bugs #142 - #155 by
2016-03-07 10:10:00 +0100
acdff55a6
Bugfix for ztrmv by
2016-03-07 09:39:34 +0100
7d6b68eb4
Refs #786. Revert to default assembly kernel. by
2016-03-07 11:34:58 +0800
0bbca5e80
removed build of smallscaling, because build on arm, arm64 and power fails by
2016-03-06 11:54:41 +0100
cd5241d0c
modified KERNEL for power, to use the generic DSDOT-KERNEL by
2016-03-06 09:07:24 +0100
8d652f11e
updated smallscaling.c to build without C99 or C11 increased the threshold value of nep.in to 40 by
2016-03-06 08:40:51 +0100
6c86570e1
Merge pull request #790 from jeromerobert/bug786 by
2016-03-05 15:25:27 -0500
53ba1a77c
(refs/reviewable/pr790/r1, refs/pull/790/head)
ztrmv_L.c: no longer need a 4kB buffer by
2016-03-05 19:07:03 +0100
d23c7c713
Fixed #789 Fix utest/ctest.h on Mingw. by
2016-03-05 09:34:37 -0500
8c43d7fa5
Merge remote-tracking branch 'origin/power8' into develop by
2016-03-05 06:03:19 -0500
085f21525
(power8)
Modified assembly label name, so that they are hidden. Added license informations. by
2016-03-05 10:27:27 +0100
8f758eeff
Refs #786. avoid old assembly c/zgemv kernels. by
2016-03-05 08:32:03 +0800
0afc76fd6
enabled gemm_beta assembly kernels by
2016-03-04 15:01:15 +0100
91e1c5080
modified configuration, to use power6 sgemm kernel for power8 by
2016-03-04 13:38:57 +0100
73f04c2c7
enabled hemv assemly function for power8 by
2016-03-04 13:20:50 +0100
3e633152c
enabled symv assembly kernels on power8 by
2016-03-04 13:08:18 +0100
d5130ce7e
enabled gemv assembly on power8 by
2016-03-04 12:53:31 +0100
4824b88fc
enabled all level1 assembly kernels for power8 by
2016-03-04 12:35:25 +0100
cc26d888b
BUGFIX: increased BUFFER_SIZE for POWER8 by
2016-03-04 10:26:53 +0100
8577be2a9
Modify travis script. by
2016-03-04 04:24:43 +0800
1edf30b79
Change Opteron(SSE3) to Opteron_SSE3 at dyanmaic core name. by
2016-03-01 20:13:08 +0800