Werner Saar [Sat, 21 Nov 2015 09:48:37 +0000 (10:48 +0100)]
fixes for cross compile
Werner Saar [Fri, 20 Nov 2015 12:47:22 +0000 (13:47 +0100)]
bugfix for cross compiling
Werner Saar [Fri, 20 Nov 2015 08:45:46 +0000 (09:45 +0100)]
added lapack-3.6.0
Werner Saar [Fri, 20 Nov 2015 08:41:59 +0000 (09:41 +0100)]
removed lapack-3.5.0
Werner Saar [Fri, 20 Nov 2015 08:23:01 +0000 (09:23 +0100)]
increase the stack size limit in the constructor
Ashwin Sekhar T K [Thu, 19 Nov 2015 20:15:35 +0000 (01:45 +0530)]
Fix blas_lock for arm64
Ashwin Sekhar T K [Thu, 19 Nov 2015 19:45:04 +0000 (01:15 +0530)]
lapack-test fixes for Cortex A57
Ashwin Sekhar T K [Thu, 19 Nov 2015 19:42:04 +0000 (01:12 +0530)]
Change BUFFER_SIZE for Cortex A57 to 20 MB
Change the GEMM_P, GEMM_Q, GEMM_R values for Cortex A57
Zhang Xianyi [Tue, 10 Nov 2015 20:30:26 +0000 (04:30 +0800)]
Fix #686. Merge branch 'ashwinyes-develop' into develop
Zhang Xianyi [Tue, 10 Nov 2015 20:22:34 +0000 (04:22 +0800)]
Use 40 MB buffer for ARM Cortex A57.
Zhang Xianyi [Tue, 10 Nov 2015 20:19:43 +0000 (04:19 +0800)]
Delete vi swap file.
Zhang Xianyi [Tue, 10 Nov 2015 20:16:22 +0000 (04:16 +0800)]
Merge branch 'develop' of https://github.com/ashwinyes/OpenBLAS into ashwinyes-develop
Zhang Xianyi [Tue, 10 Nov 2015 20:14:58 +0000 (04:14 +0800)]
Update develop version.
Zhang Xianyi [Mon, 9 Nov 2015 17:39:21 +0000 (11:39 -0600)]
Merge pull request #684 from sebastien-villemot/develop
Fix detection of POWER architecture in c_check.
Sébastien Villemot [Mon, 9 Nov 2015 17:36:04 +0000 (18:36 +0100)]
Fix detection of POWER architecture in c_check.
This is necessary to avoid the false detection of a cross-compiling
environment.
Ashwin Sekhar T K [Fri, 6 Nov 2015 14:45:05 +0000 (20:15 +0530)]
Fix bug in benchmark/gemm.c
Ashwin Sekhar T K [Mon, 2 Nov 2015 14:00:28 +0000 (19:30 +0530)]
Optimized trmm kernels for CORTEXA57
Ashwin Sekhar T K [Mon, 2 Nov 2015 13:28:28 +0000 (18:58 +0530)]
Optimized zgemm kernel for CORTEXA57
Ashwin Sekhar T K [Mon, 2 Nov 2015 13:10:27 +0000 (18:40 +0530)]
Optimized cgemm kernel for CORTEXA57
Also, add a generic ztrmm 4x4 kernel
Ashwin Sekhar T K [Mon, 2 Nov 2015 12:23:28 +0000 (17:53 +0530)]
Optimized dgemm kernel for CORTEXA57
Ashwin Sekhar T K [Mon, 2 Nov 2015 12:15:24 +0000 (17:45 +0530)]
Improve the sgemm kernel for CORTEXA57
Ashwin Sekhar T K [Mon, 2 Nov 2015 11:47:47 +0000 (17:17 +0530)]
Optimized gemv kernels for CORTEXA57
Co-Authored-By: Ralph Campbell <ralph.campbell@broadcom.com>
Ashwin Sekhar T K [Tue, 6 Oct 2015 09:09:02 +0000 (14:39 +0530)]
Optimized swap kernels for CORTEXA57
Co-Authored-By: Ralph Campbell <ralph.campbell@broadcom.com>
Ashwin Sekhar T K [Tue, 6 Oct 2015 09:06:31 +0000 (14:36 +0530)]
Optimized scal kernels for CORTEXA57
Co-Authored-By: Ralph Campbell <ralph.campbell@broadcom.com>
Ashwin Sekhar T K [Tue, 6 Oct 2015 09:03:00 +0000 (14:33 +0530)]
Optimized rot kernels for CORTEXA57
Co-Authored-By: Ralph Campbell <ralph.campbell@broadcom.com>
Ashwin Sekhar T K [Tue, 6 Oct 2015 08:59:27 +0000 (14:29 +0530)]
Optimized nrm2 kernels for CORTEXA57
Ashwin Sekhar T K [Tue, 6 Oct 2015 08:46:04 +0000 (14:16 +0530)]
Optimized dot kernels for CORTEXA57
Co-Authored-By: Ralph Campbell <ralph.campbell@broadcom.com>
Ashwin Sekhar T K [Tue, 6 Oct 2015 06:49:05 +0000 (12:19 +0530)]
Optimized copy kernels for CORTEXA57
Co-Authored-By: Ralph Campbell <ralph.campbell@broadcom.com>
Ashwin Sekhar T K [Tue, 6 Oct 2015 06:42:08 +0000 (12:12 +0530)]
Optimized axpy kernels for CORTEXA57
Co-Authored-By: Ralph Campbell <ralph.campbell@broadcom.com>
Ashwin Sekhar T K [Tue, 6 Oct 2015 06:22:15 +0000 (11:52 +0530)]
Optimized asum kernels for CORTEXA57
Co-Authored-By: Ralph Campbell <ralph.campbell@broadcom.com>
Ashwin Sekhar T K [Tue, 6 Oct 2015 06:11:15 +0000 (11:41 +0530)]
Optimized iamax kernels for CORTEXA57
Co-Authored-By: Ralph Campbell <ralph.campbell@broadcom.com>
Ashwin Sekhar T K [Mon, 5 Oct 2015 14:19:44 +0000 (19:49 +0530)]
Optimized amax kernels for CORTEXA57
Co-Authored-By: Ralph Campbell <ralph.campbell@broadcom.com>
Ashwin Sekhar T K [Mon, 5 Oct 2015 12:16:11 +0000 (17:46 +0530)]
Fix compiler errors in common.h
Ashwin Sekhar T K [Fri, 4 Sep 2015 07:56:52 +0000 (13:26 +0530)]
Adding arm64 target CORTEXA57
Co-Authored-By: Ralph Campbell <ralph.campbell@broadcom.com>
Ralph Campbell [Thu, 3 Sep 2015 12:30:12 +0000 (18:00 +0530)]
Minor C code fixes in interface/
Ralph Campbell [Thu, 3 Sep 2015 12:27:06 +0000 (17:57 +0530)]
Minor C code fixes in driver/
Ralph Campbell [Fri, 6 Nov 2015 14:34:56 +0000 (20:04 +0530)]
Minor C code fixes in kernel/arm
Ralph Campbell [Fri, 6 Nov 2015 14:34:17 +0000 (20:04 +0530)]
Remove duplicate -D args in kernel/Makefile.L1
Zhang Xianyi [Sat, 7 Nov 2015 05:46:20 +0000 (23:46 -0600)]
Refs #682. Enable LAPACK_COMPLEX_STRUCTURE when __ANDROID_API_ < 21.
Zhang Xianyi [Wed, 28 Oct 2015 18:53:29 +0000 (02:53 +0800)]
Detect AMD Trinity and Richland.
Zhang Xianyi [Wed, 28 Oct 2015 14:44:25 +0000 (09:44 -0500)]
Merge pull request #677 from j-bo/develop
Refs #676. Fixed ONLY_CBLAS=1 compiling bug on windows.
j-bo [Wed, 28 Oct 2015 14:10:42 +0000 (15:10 +0100)]
Refs #676. Fixed ONLY_CBLAS=1 compiling bug on windows.
Zhang Xianyi [Tue, 27 Oct 2015 20:44:50 +0000 (15:44 -0500)]
Merge branch 'develop'
Zhang Xianyi [Tue, 27 Oct 2015 20:44:35 +0000 (15:44 -0500)]
Update doc for OpenBLAS 0.2.15 version. [CI skipped]
Zhang Xianyi [Tue, 27 Oct 2015 15:49:22 +0000 (10:49 -0500)]
Merge pull request #674 from j-bo/develop
Fix #673
Zhang Xianyi [Tue, 27 Oct 2015 15:47:55 +0000 (10:47 -0500)]
Only include complex.h since Android 5.0
j-bo [Tue, 27 Oct 2015 12:56:45 +0000 (13:56 +0100)]
Merge pull request #1 from j-bo/j-bo-patch-673
Fix #673
j-bo [Tue, 27 Oct 2015 12:55:24 +0000 (13:55 +0100)]
Fix #673
Add lacking headers declarations when compiling for Android ARM7
Zhang Xianyi [Tue, 27 Oct 2015 00:02:51 +0000 (19:02 -0500)]
Refs #668. Raise the signal when pthread_create fails.
Thank James K. Lowden for the patch.
Zhang Xianyi [Mon, 26 Oct 2015 23:14:41 +0000 (18:14 -0500)]
Add AppVeyor badge.
Zhang Xianyi [Mon, 26 Oct 2015 23:08:54 +0000 (18:08 -0500)]
[ci skip] Build Visual Studio 12 Win64 on Appveyor
Zhang Xianyi [Mon, 26 Oct 2015 21:11:07 +0000 (05:11 +0800)]
Only test x64 Windows CI.
Zhang Xianyi [Mon, 26 Oct 2015 21:10:40 +0000 (05:10 +0800)]
Fix DYNAMIC_ARCH=1 bug.
Zhang Xianyi [Mon, 26 Oct 2015 20:08:17 +0000 (15:08 -0500)]
Use AppVeyor Windows CI.
Zhang Xianyi [Mon, 26 Oct 2015 19:54:59 +0000 (14:54 -0500)]
Merge branch 'cmake' into develop
Zhang Xianyi [Mon, 26 Oct 2015 19:54:34 +0000 (14:54 -0500)]
Merge branch 'develop' into cmake
Conflicts:
driver/others/memory.c
Zhang Xianyi [Mon, 26 Oct 2015 19:52:13 +0000 (14:52 -0500)]
Fix cmake bug on MSVC 32-bit.
Zhang Xianyi [Mon, 26 Oct 2015 18:54:53 +0000 (02:54 +0800)]
Fix cmake bug on x86 32-bit.
e.g. Build 32-bit on 64-bit Linux.
cmake -DBINARY=32
Zhang Xianyi [Mon, 26 Oct 2015 15:42:21 +0000 (23:42 +0800)]
Add CBLAS test for CMAKE.
Zhang Xianyi [Fri, 23 Oct 2015 17:16:34 +0000 (01:16 +0800)]
Refs #671. the return of i?max cannot larger than N.
Zhang Xianyi [Thu, 22 Oct 2015 16:07:35 +0000 (11:07 -0500)]
Refs #669. Fixed the build bug with gcc on Mac OS X.
Zhang Xianyi [Tue, 20 Oct 2015 19:37:22 +0000 (14:37 -0500)]
Fixed cmake bug on Visual Studio.
Zhang Xianyi [Tue, 20 Oct 2015 18:24:54 +0000 (02:24 +0800)]
Fixed cmake bug on haswell.
Zhang Xianyi [Mon, 19 Oct 2015 20:30:55 +0000 (04:30 +0800)]
Fixe cmake config bugs.
Zhang Xianyi [Mon, 19 Oct 2015 19:35:25 +0000 (03:35 +0800)]
Detect cmake test result.
Zhang Xianyi [Mon, 12 Oct 2015 20:46:08 +0000 (04:46 +0800)]
Merge branch 'develop' into cmake
Conflicts:
driver/others/memory.c
Zhang Xianyi [Thu, 8 Oct 2015 15:07:24 +0000 (15:07 +0000)]
Include time.h.
Zhang Xianyi [Tue, 6 Oct 2015 18:31:51 +0000 (02:31 +0800)]
Refs #615. Import bug fixes for LAPACKE dormlq.
Zhang Xianyi [Mon, 5 Oct 2015 19:14:32 +0000 (14:14 -0500)]
Fixed #654. Make sure the gotoblas_init function is run before all other static initializations.
Zhang Xianyi [Mon, 5 Oct 2015 15:25:15 +0000 (10:25 -0500)]
Merge pull request #656 from stevengj/libname
default to lib$(SYMBOLPREFIX)openblas$(SYMBOLSUFFIX)
Zhang Xianyi [Mon, 5 Oct 2015 15:23:52 +0000 (10:23 -0500)]
Merge pull request #659 from Keno/patch-2
Fix cross compilation suffix detection
Keno Fischer [Mon, 5 Oct 2015 04:58:07 +0000 (00:58 -0400)]
Fix cross compilation suffix detection
If the path involves `-`, this would have otherwise detected this as a cross compile suffix.
Steven G. Johnson [Thu, 1 Oct 2015 19:07:04 +0000 (15:07 -0400)]
default to lib$(SYMBOLPREFIX)openblas$(SYMBOLSUFFIX), as discussed in #646: if you rename the symbols, it is best to rename the library
Zhang Xianyi [Sat, 26 Sep 2015 14:42:44 +0000 (14:42 +0000)]
Fixed make TARGET=CORTEXA9 and CORTEXA15 bug.
Zhang Xianyi [Tue, 22 Sep 2015 15:01:59 +0000 (10:01 -0500)]
Merge pull request #652 from larsmans/fixes
Tiny fixes
Lars Buitinck [Tue, 22 Sep 2015 10:01:09 +0000 (12:01 +0200)]
git ignore versioned .so files
Lars Buitinck [Tue, 22 Sep 2015 10:00:30 +0000 (12:00 +0200)]
actually remove cblas_noconst.h
This file hasn't been used since
212463dce961827421a9c54f109a430c1599732c.
Zhang Xianyi [Thu, 10 Sep 2015 15:36:57 +0000 (10:36 -0500)]
Merge pull request #640 from kortschak/dlansy-fix
Fix LAPACK_*lansy routines
Zhang Xianyi [Thu, 10 Sep 2015 15:32:07 +0000 (10:32 -0500)]
Refs #638. Fixed compiling bug with clang on Mac OS X.
kortschak [Thu, 10 Sep 2015 06:02:50 +0000 (15:32 +0930)]
Fix LAPACK_*lansy routines
Fixes #639.
Zhang Xianyi [Wed, 9 Sep 2015 15:48:15 +0000 (10:48 -0500)]
Merge branch 'yuyichao-skylake-id' into develop
Zhang Xianyi [Wed, 9 Sep 2015 15:47:17 +0000 (10:47 -0500)]
Detect other Intel Skylake cores.
http://users.atw.hu/instlatx64/
Yichao Yu [Wed, 9 Sep 2015 15:00:23 +0000 (11:00 -0400)]
Ref #632. Support Intel Skylake by Haswell kernels.
Zhang Xianyi [Wed, 9 Sep 2015 14:56:07 +0000 (09:56 -0500)]
Merge pull request #634 from kortschak/lantr-trans-prep
Fix lantr preparation for row major matrices
kortschak [Tue, 8 Sep 2015 23:55:48 +0000 (09:25 +0930)]
Fix lantr preparation for row major matrices
Zhang Xianyi [Tue, 8 Sep 2015 18:59:08 +0000 (13:59 -0500)]
Merge pull request #633 from grisuthedragon/tune_imatcopy
Improved Ximatcopy when lda==ldb.
Martin Koehler [Mon, 7 Sep 2015 12:33:26 +0000 (14:33 +0200)]
Improved Ximatcopy when lda==ldb.
The Ximatcopy functions create a copy of the input matrix
although they seem to work inplace. The new routines
XIMATCOPY_K_YY perform the operations inplace if the leading
dimension does not change.
Zhang Xianyi [Fri, 4 Sep 2015 18:01:01 +0000 (13:01 -0500)]
Merge pull request #630 from buffer51/develop
Fixed error in common.h for Android compilation introduced by e12cf11
buffer51 [Fri, 4 Sep 2015 00:54:21 +0000 (20:54 -0400)]
Fixed error in common.h for Android compilation introduced by
e12cf1123e8784ce6fe9d2ac14526331fbe2c555
Zhang Xianyi [Thu, 20 Aug 2015 03:50:25 +0000 (22:50 -0500)]
Add notification.
Zhang Xianyi [Thu, 20 Aug 2015 03:48:55 +0000 (22:48 -0500)]
Merge branch 'develop' of github.com:xianyi/OpenBLAS into develop
Zhang Xianyi [Thu, 20 Aug 2015 03:26:20 +0000 (22:26 -0500)]
Merge pull request #619 from gitter-badger/gitter-badge
Add a Gitter chat badge to README.md
The Gitter Badger [Thu, 20 Aug 2015 03:21:09 +0000 (03:21 +0000)]
Added Gitter badge
Zhang Xianyi [Wed, 19 Aug 2015 13:07:47 +0000 (08:07 -0500)]
Use C kernels for s/dgemv on x86.
Zhang Xianyi [Wed, 19 Aug 2015 03:43:42 +0000 (22:43 -0500)]
Fixed cmake bug with NO_LAPACK=1
Zhang Xianyi [Mon, 17 Aug 2015 20:22:37 +0000 (15:22 -0500)]
Merge pull request #617 from notaz/arm_fixes
really fix ARM64 locking
Grazvydas Ignotas [Sun, 16 Aug 2015 23:27:45 +0000 (01:27 +0200)]
really fix ARM64 locking
Zhang Xianyi [Sun, 16 Aug 2015 22:16:18 +0000 (17:16 -0500)]
Merge pull request #616 from notaz/arm_fixes
ARM fixes
Grazvydas Ignotas [Sun, 16 Aug 2015 18:11:13 +0000 (20:11 +0200)]
correct a minor mistake
Grazvydas Ignotas [Sun, 16 Aug 2015 16:13:30 +0000 (18:13 +0200)]
use real armv5 support
there is no more requirement for ARMv6 instructions,
and VFP on ARMv5 is uncommon
projects / platform / upstream / openblas.git / log