b752858d6
added dgemm-, dtrmm-, zgemm- and ztrmm-kernel for power8 by
2016-03-01 07:33:56 +0100
4fc8c937d
Refs #695 add testcase. by
2016-03-01 01:05:56 -0500
efa4f5c93
Refs #695 #783. Replace default x86_64 cgemv_t asm kernel by C kernel. by
2016-03-01 11:18:56 +0800
17d655fa6
Merge pull request #784 from peterph/develop by
2016-02-27 11:24:20 -0500
f68141cf1
(refs/pull/784/head)
collected usage notes by
2016-02-27 16:57:22 +0100
aa9051820
Update Changelog for 0.2.16.rc1. by
2016-02-24 15:21:22 -0500
6b85dbb6d
Refs #696. Turn off stack limit setting on Linux. by
2016-02-24 14:18:39 -0500
a0debd429
Refs #696. Turn off stack limit setting on Linux. by
2016-02-24 14:18:39 -0500
937493bfe
(tag: v0.2.16.rc1)
Release 0.2.16 rc1 by
2016-02-23 18:29:21 -0500
74b067222
Fix c/zaxpyc kernel bug on Cortex-A57. by
2016-02-23 22:47:53 +0000
6e7be06e0
Refs JuliaLang/julia#5728. Fix gemv performance bug on Haswell Mac OSX. by
2016-02-19 17:56:07 -0500
a04d0555b
[av skip] Fix utest makefile bug on travis ci. by
2016-02-20 00:21:43 +0800
3761c30ba
Fix makefile bug for utest. by
2016-02-18 17:01:48 -0500
38593cd3a
Fix compiling bug on ARM Cortex-A57. by
2016-02-13 15:38:52 +0000
e3b7781c2
Update readme. by
2016-02-13 00:33:53 +0800
5e6965ea4
Run utest when building. by
2016-02-13 00:33:31 +0800
5cc0301fc
Enable utest for appveyor. by
2016-02-12 01:50:20 -0500
19a6dedfd
Add utest for CMake. by
2016-02-12 05:38:13 +0800
0e2b92e21
Added mising lapacke files for CMake. by
2016-02-12 05:28:16 +0800
d06b92906
Add gemm3m building for CMake. by
2016-02-12 05:02:51 +0800
8e98478ff
Update ctest.h from github.com:xianyi/ctest.git. by
2016-02-12 05:01:57 +0800
fb8968fb8
Refs #707. Bugfix for previous commit. by
2016-02-11 05:14:53 +0800
dae6b82a7
Refs #707. Add BUILD_LAPACK_DEPRECATED flag in Makefile.rule. by
2016-02-11 04:22:53 +0800
d73244b82
Refs #727. Align stack buffer address on 32-bytes. by
2016-02-11 03:51:26 +0800
233c6b959
Merge pull request #780 from jeromerobert/bug727 by
2016-02-08 13:24:40 -0500
16ec5323c
(refs/pull/780/head)
Fix zgemv.c compilation when stack allocation is disabled by
2016-02-08 12:05:02 +0100
0ad02ef2d
update CONTRIBUTORS.md by
2016-01-18 18:54:51 +0100
73397faf6
Add benchmark/smallscaling.c by
2016-01-03 14:04:33 +0100
5fc2203d8
zgemv: Add a workaround for #746 by
2016-01-24 10:14:41 +0100
78dcf5c3d
Improve performances of ztrmv on small matrices by
2016-01-14 22:12:57 +0100
32f793195
Use stack allocation in zgemv and zger by
2016-01-03 14:01:12 +0100
d87db1d24
(refs/pull/769/merge)
Merge 48599f0a3b into 52eba814ce by
2016-02-02 18:44:17 +0000
48599f0a3
(refs/pull/769/head)
Update dynamic.c by
2016-02-02 12:33:14 +0100
1a1f3245d
Update dynamic.c by
2016-02-02 11:59:00 +0100
edae5b930
Update dynamic.c by
2016-02-02 09:00:18 +0100
d4adc7140
Update cpuid_x86.c by
2016-01-31 15:33:56 +0100
92058a75e
(refs/pull/944/head, optimized_for_deeplearning)
For gemm multi-threading, simply split M. by
2015-11-25 05:14:56 +0800
1367a64d0
Merge branch 'develop' of github.com:xianyi/OpenBLAS into arm_soft_fp_abi by
2015-11-11 19:25:07 +0000
3673c77ff
(refs/pull/686/merge)
Merge 93718dcc67 into d00ada378f by
2015-11-10 14:07:54 +0000
93718dcc6
(refs/pull/686/head)
Fix bug in benchmark/gemm.c by
2015-11-06 20:15:05 +0530
924f4b00b
Optimized trmm kernels for CORTEXA57 by
2015-11-02 19:30:28 +0530
bc4e96311
Optimized zgemm kernel for CORTEXA57 by
2015-11-02 18:58:28 +0530
82b791bf1
Optimized cgemm kernel for CORTEXA57 by
2015-11-02 18:40:27 +0530
262c1479a
Optimized dgemm kernel for CORTEXA57 by
2015-11-02 17:53:28 +0530
3d8be7b6a
Improve the sgemm kernel for CORTEXA57 by
2015-11-02 17:45:24 +0530
8055add70
Optimized gemv kernels for CORTEXA57 by
2015-11-02 17:17:47 +0530
f801abd58
Optimized swap kernels for CORTEXA57 by
2015-10-06 14:39:02 +0530
5ee916d1f
Optimized scal kernels for CORTEXA57 by
2015-10-06 14:36:31 +0530
f4cadff03
Optimized rot kernels for CORTEXA57 by
2015-10-06 14:33:00 +0530
c29ea30dc
Optimized nrm2 kernels for CORTEXA57 by
2015-10-06 14:29:27 +0530
d1b2ec5eb
Optimized dot kernels for CORTEXA57 by
2015-10-06 14:16:04 +0530
183b2e6cd
Optimized copy kernels for CORTEXA57 by
2015-10-06 12:19:05 +0530
b5143de00
Optimized axpy kernels for CORTEXA57 by
2015-10-06 12:12:08 +0530
9af52f1af
Optimized asum kernels for CORTEXA57 by
2015-10-06 11:52:15 +0530
25a9a1da4
Optimized iamax kernels for CORTEXA57 by
2015-10-06 11:41:15 +0530
64df449fe
Optimized amax kernels for CORTEXA57 by
2015-10-05 19:49:44 +0530
45b275761
Fix compiler errors in common.h by
2015-10-05 17:46:11 +0530
c425f99e3
Adding arm64 target CORTEXA57 by
2015-09-04 13:26:52 +0530
86333efdb
Minor C code fixes in interface/ by
2015-09-03 18:00:12 +0530
03faa3c06
Minor C code fixes in driver/ by
2015-09-03 17:57:06 +0530
3e8d6ea74
Init POWER8 kernels by POWER6. by
2015-11-03 12:25:05 +0800
be4e5fcd2
Fixed #778. Merge branch 'buffer51-develop' into develop by
2016-02-05 08:39:08 +0800
855e0cb70
(refs/pull/778/head)
Restored LAPACK_COMPLEX_STRUCTURE for Android prior to 21. Refs #682. by
2016-02-04 17:20:07 -0500
7f7d04dcd
Fixed linking error when compiling ARMv7 for Android (disabled -lpthread and added -Wl,--no-warn-mismatch). by
2016-02-04 17:05:31 -0500
4e1b521e2
Fix lapack complex implementation of lauu2 and potf2 for Android (use FLOAT instead of FLOAT[2] as imaginary part is not used). by
2015-11-07 19:31:13 -0500
a1a96589a
Fixed #773 blas_quickdivide bug on CMake and Visual Studio x86 32-bit. by
2016-02-04 15:23:32 -0500
0e68beb89
Fixed #711, #698. Merge branch 'byzhang-develop' into develop by
2016-02-03 02:56:27 +0800
926ba8b7c
Merge branch 'develop' of https://github.com/byzhang/OpenBLAS into byzhang-develop by
2016-02-03 02:48:32 +0800
9f080c47e
Merge pull request #743 from tkelman/patch-1 by
2016-02-02 13:46:12 -0500
52eba814c
Fixed #769. Merge branch 'martin-frbg-develop' into develop by
2016-02-02 13:43:51 -0500
935356c34
Update dynamic.c and cpuid_x86.c for Intel Avoton. by
2016-02-02 09:00:18 +0100
ff9388d62
Refs #768. Swap the result of zdot x87 fp kernel. by
2016-02-02 09:15:02 +0800
4f05c2367
Update cpuid_x86.c by
2016-01-31 15:33:56 +0100
4a1263f60
(refs/pull/771/head)
Fix the source paths by
2016-02-01 18:32:42 -0800
962376664
Refs #768. Swap the result of zdot x87 fp kernel. by
2016-02-02 09:15:02 +0800
5fef0d1b7
(refs/pull/743/head)
re enable Fortran optimization flag on windows by
2016-01-18 08:44:46 -0800
578f47180
Fix utest bug when INTERFACE64=1. by
2016-01-28 22:18:38 -0600
5a8447e97
Use ctest.h for unit test. Enable unit test on travis CI. by
2016-01-29 11:35:31 +0800
be95bdaf4
Detect ARMV8 on 32-bit mode by using ARMV7 kernels. by
2016-01-28 17:30:26 +0000
c44ff4d64
Refs #714. avoid compiling warnings. by
2016-01-28 04:38:07 +0800
e003a1294
Merge pull request #764 from martin-frbg/develop by
2016-01-26 14:03:27 -0600
44062517e
(refs/pull/764/head)
Update Makefile.system by
2016-01-26 20:35:25 +0100
13f0f8c10
Refs #723. Avoid out of boundary for getf2. by
2016-01-26 09:14:57 -0600
f5df444ce
Merge pull request #762 from jeromerobert/bug760 by
2016-01-26 08:45:16 -0600
e38271342
Merge pull request #759 from jeromerobert/bug742 by
2016-01-26 08:43:32 -0600
aaa8551c5
Merge pull request #749 from lotheac/illumos_fixes by
2016-01-26 08:42:20 -0600
0d87c1ffb
(refs/pull/762/head)
Let openblas_get_num_threads return the number of active threads by
2016-01-26 13:04:16 +0100
0b194426f
Merge pull request #761 from wernsaar/develop by
2016-01-26 09:19:14 +0100
63a7d7fb2
(refs/pull/761/head)
updated gemv_n_vfpv3.S for armv7 by
2016-01-25 15:00:13 +0100
b4ede558a
updated nrm2 kernel for armv7 by
2016-01-25 11:55:25 +0100
de3e2d434
updated trmm kernels for armv7 by
2016-01-25 11:08:56 +0100
a0e51e96f
updated gemm kernels for armv7 by
2016-01-25 10:46:10 +0100
d6afac962
(refs/pull/749/head)
don't pass -Y at all to the linker on illumos by
2016-01-22 18:46:27 +0200
c2891330b
updated KERNEL.ARMV6 by
2016-01-24 17:12:07 +0100
ceaa931e4
updated gemv kernel for armv6 by
2016-01-24 16:31:19 +0100
eaa63165d
updated cgemv and zgemv kernels for armv6 by
2016-01-24 14:42:38 +0100
c65357c56
updated trmm_kernels for armv6 by
2016-01-24 13:03:33 +0100
e63e9f9f2
updated gemm_kernels for armv6 by
2016-01-24 11:55:50 +0100
1fe3aab04
(refs/pull/759/head)
Use GEMM_MULTITHREAD_THRESHOLD as a number of ops by
2016-01-24 10:30:50 +0100
aafd3ab60
updated cdot and zdot on arm by
2016-01-24 10:56:49 +0100