Zhang Xianyi
|
a71e8c82f6
|
Fix change log typo.
|
9 years ago |
Zhang Xianyi
|
1619b2f3c8
|
Merge branch 'release-0.2.17'
|
9 years ago |
Zhang Xianyi
|
4f3153395a
|
Update doc for 0.2.17.
|
9 years ago |
Zhang Xianyi
|
308e6195b7
|
Refs #807. Enable BUILD_LAPACK_DEPRECATED=1 by default.
|
9 years ago |
Zhang Xianyi
|
fced5744fb
|
Merge branch 'release-0.2.16'
|
9 years ago |
Zhang Xianyi
|
8c0fb1258d
|
Update 0.2.16 doc
|
9 years ago |
Zhang Xianyi
|
aae581d004
|
Merge branch 'develop' into release-0.2.16
|
9 years ago |
Zhang Xianyi
|
e17303933a
|
Merge pull request #802 from ashwinyes/develop_20160314_dgemm_optimization
DGEMM Optimizations for Cortex-A57
|
9 years ago |
Zhang Xianyi
|
f9226275f4
|
Merge pull request #801 from Keno/patch-3
Don't pass REALNAME to `.end`
|
9 years ago |
Ashwin Sekhar T K
|
cf8c7e28b3
|
Update CONTRIBUTORS.md
|
9 years ago |
Ashwin Sekhar T K
|
5ac02f6dc7
|
Optimize Dgemm 4x4 for Cortex A57
|
9 years ago |
Ashwin Sekhar T K
|
7aa1ad4923
|
Functional Assembly Kernels for CortexA57
Adding functional (non-optimized) kernels for Cortex-A57
with the following layouts.
SGEMM - 16x4, 8x8
CGEMM - 8x4
DGEMM - 8x4, 4x8
|
9 years ago |
Keno Fischer
|
d5e1255ca7
|
Don't pass REALNAME to `.end`
Putting the procedure there is an MSVC-ism, where it is optional. GCC silently ignores and Clang errors, so it is best to remove this.
|
9 years ago |
Zhang Xianyi
|
587455868e
|
Merge pull request #800 from jeromerobert/smallscaling
Fix smallscaling compilation
|
9 years ago |
Jerome Robert
|
323c237e7b
|
Fix smallscaling compilation
Also revert 0bbca5e
|
9 years ago |
Werner Saar
|
faa5e2e5e3
|
FIX: forgot the add the files cgemv_n_4.c and cgemv_t_4.c
|
9 years ago |
wernsaar
|
551fdf53e8
|
Merge pull request #799 from wernsaar/develop
Added optimized cgemv_n and cgemv_t kernels for bulldozer, piledriver…
|
9 years ago |
Werner Saar
|
fdf291be30
|
Added optimized cgemv_n and cgemv_t kernels for bulldozer, piledriver and steamroller
|
9 years ago |
Zhang Xianyi
|
68eb4fa329
|
Add missing openblas_env makefile.
|
9 years ago |
Zhang Xianyi
|
05196a8497
|
Refs #716. Only call getenv at init function.
|
9 years ago |
wernsaar
|
db9b611b12
|
Merge pull request #798 from wernsaar/develop
Optimized zgemv_n kernel for bulldozer, piledriver and steamroller
|
9 years ago |
Werner Saar
|
2e6333f74e
|
modified common.h for piledriver
|
9 years ago |
Werner Saar
|
c99cc41cbd
|
Added optimized zgemv_n kernel for bulldozer, piledriver and steamroller
|
9 years ago |
wernsaar
|
711ecb8bd5
|
Merge pull request #797 from wernsaar/develop
bugfixes for lapack and lapacke
|
9 years ago |
Werner Saar
|
10c2ebdfc5
|
BUGFIX: removed fixes for bugs #148 and #149, because info for xerbla is wrong
|
9 years ago |
Werner Saar
|
26b3b3a3e6
|
bugfixes form lapack svn for bugs #142 - #155
|
9 years ago |
Werner Saar
|
acdff55a6a
|
Bugfix for ztrmv
|
9 years ago |
Zhang Xianyi
|
7d6b68eb4a
|
Refs #786. Revert to default assembly kernel.
|
9 years ago |
Werner Saar
|
0bbca5e803
|
removed build of smallscaling, because build on arm, arm64 and power fails
|
9 years ago |
Werner Saar
|
cd5241d0cf
|
modified KERNEL for power, to use the generic DSDOT-KERNEL
|
9 years ago |
Werner Saar
|
8d652f11e7
|
updated smallscaling.c to build without C99 or C11
increased the threshold value of nep.in to 40
|
9 years ago |
Zhang Xianyi
|
6c86570e1f
|
Merge pull request #790 from jeromerobert/bug786
ztrmv_L.c: no longer need a 4kB buffer
|
9 years ago |
Jerome Robert
|
53ba1a77c8
|
ztrmv_L.c: no longer need a 4kB buffer
Fix #786
|
9 years ago |
Zhang Xianyi
|
d23c7c713c
|
Fixed #789 Fix utest/ctest.h on Mingw.
|
9 years ago |
Zhang Xianyi
|
8c43d7fa5f
|
Merge remote-tracking branch 'origin/power8' into develop
Refs #774
|
9 years ago |
Werner Saar
|
085f215257
|
Modified assembly label name, so that they are hidden.
Added license informations.
|
9 years ago |
Zhang Xianyi
|
8f758eeff9
|
Refs #786. avoid old assembly c/zgemv kernels.
|
9 years ago |
Werner Saar
|
0afc76fd65
|
enabled gemm_beta assembly kernels
|
9 years ago |
Werner Saar
|
91e1c5080c
|
modified configuration, to use power6 sgemm kernel for power8
|
9 years ago |
Werner Saar
|
73f04c2c72
|
enabled hemv assemly function for power8
|
9 years ago |
Werner Saar
|
3e633152c6
|
enabled symv assembly kernels on power8
|
9 years ago |
Werner Saar
|
d5130ce7e3
|
enabled gemv assembly on power8
|
9 years ago |
Werner Saar
|
4824b88fcb
|
enabled all level1 assembly kernels for power8
|
9 years ago |
Werner Saar
|
cc26d888b8
|
BUGFIX: increased BUFFER_SIZE for POWER8
|
9 years ago |
Zhang Xianyi
|
8577be2a95
|
Modify travis script.
|
9 years ago |
Zhang Xianyi
|
1edf30b790
|
Change Opteron(SSE3) to Opteron_SSE3 at dyanmaic core name.
|
9 years ago |
Werner Saar
|
b752858d6c
|
added dgemm-, dtrmm-, zgemm- and ztrmm-kernel for power8
|
9 years ago |
Zhang Xianyi
|
4fc8c937d4
|
Refs #695 add testcase.
|
9 years ago |
Zhang Xianyi
|
efa4f5c936
|
Refs #695 #783. Replace default x86_64 cgemv_t
asm kernel by C kernel.
|
9 years ago |
Zhang Xianyi
|
17d655fa64
|
Merge pull request #784 from peterph/develop
collected usage notes
|
9 years ago |