floraachy
/
OpenBLAS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
3,414 Commits 47 Branches 64 Tags 71 MiB
Commit Graph

3414 Commits

Author SHA1 Message Date
Zhang Xianyi fb8968fb83 Refs #707. Bugfix for previous commit. 2016-02-11 05:14:53 +08:00
Zhang Xianyi dae6b82a71 Refs #707. Add BUILD_LAPACK_DEPRECATED flag in Makefile.rule. 
If you want to build LAPACK deprecated functions since LAPACK 3.6.0

make BUILD_LAPACK_DEPRECATED=1
 2016-02-11 04:22:53 +08:00
Zhang Xianyi d73244b825 Refs #727. Align stack buffer address on 32-bytes. 2016-02-11 03:52:02 +08:00
Zhang Xianyi 233c6b959f Merge pull request #780 from jeromerobert/bug727 
Bug727
 2016-02-08 13:24:40 -05:00
Jerome Robert 16ec5323c9 Fix zgemv.c compilation when stack allocation is disabled 2016-02-08 12:05:02 +01:00
Jerome Robert 0ad02ef2d6 update CONTRIBUTORS.md 2016-02-08 11:26:51 +01:00
Jerome Robert 73397faf68 Add benchmark/smallscaling.c 
* Bench small matrices with multi-threading
* Close #727
 2016-02-08 11:25:27 +01:00
Jerome Robert 5fc2203d8a zgemv: Add a workaround for #746 2016-02-08 11:25:15 +01:00
Jerome Robert 78dcf5c3d5 Improve performances of ztrmv on small matrices 
* Use stack allocation
* Disable multi-threading
* Ref #727
 2016-02-08 11:25:02 +01:00
Jerome Robert 32f793195f Use stack allocation in zgemv and zger 
For better performance with small matrices
Ref #727
 2016-02-08 11:24:21 +01:00
Zhang Xianyi be4e5fcd20 Fixed #778. Merge branch 'buffer51-develop' into develop 2016-02-05 08:39:08 +08:00
buffer51 855e0cb700 Restored LAPACK_COMPLEX_STRUCTURE for Android prior to 21. Refs #682. 2016-02-04 17:20:07 -05:00
buffer51 7f7d04dcd2 Fixed linking error when compiling ARMv7 for Android (disabled -lpthread and added -Wl,--no-warn-mismatch). 2016-02-04 17:05:31 -05:00
buffer51 4e1b521e27 Fix lapack complex implementation of lauu2 and potf2 for Android (use FLOAT instead of FLOAT[2] as imaginary part is not used). 2016-02-04 16:59:56 -05:00
Zhang Xianyi a1a96589aa Fixed #773 blas_quickdivide bug on CMake and Visual Studio x86 32-bit. 2016-02-04 15:23:32 -05:00
Zhang Xianyi 0e68beb89f Fixed #711, #698. Merge branch 'byzhang-develop' into develop 2016-02-03 02:56:27 +08:00
Zhang Xianyi 926ba8b7ca Merge branch 'develop' of https://github.com/byzhang/OpenBLAS into byzhang-develop 2016-02-03 02:48:32 +08:00
Zhang Xianyi 9f080c47e1 Merge pull request #743 from tkelman/patch-1 
re enable Fortran optimization flag on windows
 2016-02-02 13:46:12 -05:00
Zhang Xianyi 52eba814ce Fixed #769. Merge branch 'martin-frbg-develop' into develop 2016-02-02 13:43:51 -05:00
Martin Kroeker 935356c34f Update dynamic.c and cpuid_x86.c for Intel Avoton. 
Second part of "support Intel Avoton via Nehalem kernel"
 2016-02-02 13:42:55 -05:00
Zhang Xianyi ff9388d625 Refs #768. Swap the result of zdot x87 fp kernel. 2016-02-02 13:38:01 -05:00
Martin Kroeker 4f05c23673 Update cpuid_x86.c 
Add recognition of Intel Atom C27xx (Avoton, model code 4D)
 2016-02-02 13:38:01 -05:00
Benyu Zhang 4a1263f609 Fix the source paths 2016-02-01 18:32:42 -08:00
Zhang Xianyi 962376664d Refs #768. Swap the result of zdot x87 fp kernel. 2016-02-02 09:15:02 +08:00
Tony Kelman 5fef0d1b75 re enable Fortran optimization flag on windows 
partial revert of 299cdcdc29
from #696, was not explained why that was needed
 2016-01-30 01:13:51 -08:00
Zhang Xianyi 578f471808 Fix utest bug when INTERFACE64=1. 2016-01-28 22:18:38 -06:00
Zhang Xianyi 5a8447e97e Use ctest.h for unit test. Enable unit test on travis CI. 2016-01-29 11:35:31 +08:00
Zhang Xianyi be95bdaf47 Detect ARMV8 on 32-bit mode by using ARMV7 kernels. 2016-01-28 17:30:26 +00:00
Zhang Xianyi c44ff4d648 Refs #714. avoid compiling warnings. 2016-01-28 04:38:07 +08:00
Zhang Xianyi e003a1294c Merge pull request #764 from martin-frbg/develop 
Update Makefile.system to fix awk/nawk issue #763
 2016-01-26 14:03:27 -06:00
Martin Kroeker 44062517eb Update Makefile.system 
Define AWK as "nawk" for SunOS (actually Illumos) only - fixes #763
 2016-01-26 20:35:25 +01:00
Zhang Xianyi 13f0f8c10e Refs #723. Avoid out of boundary for getf2. 2016-01-26 09:14:57 -06:00
Zhang Xianyi f5df444ceb Merge pull request #762 from jeromerobert/bug760 
Let openblas_get_num_threads return the number of active threads
 2016-01-26 08:45:16 -06:00
Zhang Xianyi e382713423 Merge pull request #759 from jeromerobert/bug742 
Bug742
 2016-01-26 08:43:32 -06:00
Zhang Xianyi aaa8551c57 Merge pull request #749 from lotheac/illumos_fixes 
illumos fixes
 2016-01-26 08:42:20 -06:00
Jerome Robert 0d87c1ffb6 Let openblas_get_num_threads return the number of active threads 
... not the number of allocated threads.

Close #760
 2016-01-26 13:04:16 +01:00
wernsaar 0b194426f8 Merge pull request #761 from wernsaar/develop 
Ref #740: all assembly codes now clear floating point register correctly
 2016-01-26 09:19:14 +01:00
Werner Saar 63a7d7fb24 updated gemv_n_vfpv3.S for armv7 2016-01-25 15:00:13 +01:00
Werner Saar b4ede558a5 updated nrm2 kernel for armv7 2016-01-25 11:55:25 +01:00
Werner Saar de3e2d4349 updated trmm kernels for armv7 2016-01-25 11:08:56 +01:00
Werner Saar a0e51e96f1 updated gemm kernels for armv7 2016-01-25 10:46:10 +01:00
Lauri Tirkkonen d6afac9624 don't pass -Y at all to the linker on illumos 
the illumos linker can't understand the "-Y/lib"... form that f_check
generates, and -Wl cannot pass options that include commas
 2016-01-25 11:09:34 +02:00
Werner Saar c2891330bc updated KERNEL.ARMV6 2016-01-24 17:12:07 +01:00
Werner Saar ceaa931e48 updated gemv kernel for armv6 2016-01-24 16:31:19 +01:00
Werner Saar eaa63165df updated cgemv and zgemv kernels for armv6 2016-01-24 14:42:38 +01:00
Werner Saar c65357c566 updated trmm_kernels for armv6 2016-01-24 13:03:33 +01:00
Werner Saar e63e9f9f26 updated gemm_kernels for armv6 2016-01-24 11:55:50 +01:00
Jerome Robert 1fe3aab047 Use GEMM_MULTITHREAD_THRESHOLD as a number of ops 
...not a matrix size. For GEMM_MULTITHREAD_THRESHOLD=4
(the default value) this does not change anything but
for other values it make the GEMM and GEMV thresholds
changing in the same way.

Close #742
 2016-01-24 11:31:40 +01:00
Werner Saar aafd3ab60e updated cdot and zdot on arm 2016-01-24 10:56:49 +01:00
Jerome Robert 1a1935507b [z]ger: increase multithread threshold 
The ones given in 3ae30cd was by far to low because I
mixed m and m*n in my measures. Note that the new ones
are closed to the [z]gemv ones which is comforting
that both are right.
 2016-01-24 10:46:35 +01:00