ba024fcfc0
Allow to force the number of parallel make job
...
This is particularly useful when using distcc
2015-12-28 19:45:29 +01:00
299cdcdc29
lapack fixes for Windos
2015-11-21 14:33:27 +01:00
a8516c5b47
fixes for cross compile
2015-11-21 10:48:37 +01:00
ccf41ebf78
Merge branch 'develop' into arm_soft_fp_abi
2015-10-28 12:12:31 +00:00
88bef3bffc
default to lib$(SYMBOLPREFIX)openblas$(SYMBOLSUFFIX), as discussed in #646 : if you rename the symbols, it is best to rename the library
2015-10-01 15:07:04 -04:00
857899526f
ARM soft fp abi branch.
2015-09-26 14:10:18 +00:00
1cf2b10224
Use pure C generic target on x86 and x86_64.
...
make TARGET=GENERIC
?gemm3m is unimplemented on generic target.
2015-08-03 23:55:56 -05:00
4f5691e5c0
Fix test execution when USE_OPENMP=0
...
The standard way to disable OpenMP support is to set USE_OPENMP=0,
as indicated by other checks to see if USE_OPENMP equals 1. The
problem is obviously then that `ifdef USE_OPENMP` is very much not
what we want to test for. This causes tests to fail when no OpenMP
library is installed.
2015-06-12 23:52:07 -07:00
29293160a4
Fix #593 . Change MACOSX_DEPLOYMENT_TARGET to 10.6.
2015-06-08 10:53:50 -05:00
e127fb8fd8
1) Refs #575 . Remove g77 from compiler list.
...
2) If OpenBLAS cannot find Fortran compiler, it will only build BLAS
(without LAPACK).
2015-05-19 00:01:04 -05:00
51ff17d46e
Add AMD Excavator target.
2015-05-13 16:16:30 -05:00
6743beb748
Refs #565 . Fix the bug of generate FEXTRALIB.
2015-05-07 13:06:53 +08:00
cda29f183b
Add vecLib benchmarks
2015-05-06 21:52:34 -04:00
771b18ae9c
Detect the wrong combined flags of USE_OPENMP=1 and USE_THREAD=0.
2015-02-08 01:42:48 -06:00
41aad0407f
Merge pull request #482 from jeromerobert/develop
...
Allow to do gemv and ger buffer allocation on the stack
2015-01-02 02:26:17 +08:00
4319769b79
added target processor STEAMROLLER
2014-12-28 20:16:46 +08:00
e9d9a8eae3
Allow to do gemv and ger buffer allocation on the stack
...
ger and gemv call blas_memory_alloc/free which in their turn
call blas_lock. blas_lock create thread contention when matrices
are small and the number of thread is high enough. We avoid
call blas_memory_alloc by replacing it with stack allocation.
This can be enabled with:
make -DMAX_STACK_ALLOC=2048
The given size (in byte) must be high enough to avoid thread contention
and small enough to avoid stack overflow.
Fix #478
2014-12-27 14:33:12 +01:00
0b4602b753
add SYMBOLPREFIX and SYMBOLSUFFIX makefile options
...
for adding a prefix or suffix to all exported symbol names in the shared library
Useful to avoid conflicts with other BLAS libraries, especially when using
64 bit integer interfaces in OpenBLAS
Note that since OSX does not have the objcopy utility, setting these options
to non-empty values on Mac requires the objconv tool, available (GPL license)
from http://www.agner.org/optimize/#objconv
2014-10-24 22:27:09 -07:00
5fa6158731
renoved flag no-integrated-as, because not working on macosx
2014-08-26 18:29:40 +02:00
84badf8086
EXPERIMENTAL: added the flag -no-integrated-as for clang compiler in Makefile.system
2014-08-26 17:36:32 +02:00
793175be3a
added experimental support for big numa machines
2014-08-02 13:40:16 +02:00
134fa320e6
Refs #415 . Fixed the x86/i386 compiling bug with DYNAMIC_ARCH=1.
2014-07-17 15:02:01 +08:00
c94762bb56
Refs #401 . Added NO_AVX2 flag for old binutils (e.g. RHEL6)
2014-07-16 08:38:25 +08:00
6c2ead30f0
Remove all trailing whitespace except lapack-netlib
...
Signed-off-by: Timothy Gu <timothygu99@gmail.com>
2014-06-27 12:05:18 -07:00
88b6bf251a
force fallback for x86 32bit
2014-06-22 17:27:11 +02:00
4a2ab7460b
Ref #391 : force fallback for x86 32bit
2014-06-22 13:51:17 +02:00
316df0e821
fixed bug for INTERFACE64
2014-06-22 09:49:20 +02:00
438002204d
Ref #393 : fix for INTERFACE64=0 and ARCH_X86 in divtable
2014-06-21 12:29:23 +02:00
409b52255c
changed default optimization flag from O3 to O2 for ARM
2014-05-16 14:36:24 +02:00
a35a1a9ae7
changed makefiles for lapack development
2014-05-07 11:33:02 +02:00
75acf96d94
Refs #329 #287 . Only disable -fopenmp for LAPACK Fortran codes on Windows.
2014-01-24 15:39:46 +08:00
2594728eb7
Merge remote branch 'origin/develop' into haswell
2013-12-01 16:53:39 +01:00
65ebab0688
modified Makefile.system
2013-12-01 16:46:32 +01:00
0b6e13b689
Merge remote branch 'origin/develop' into haswell
2013-12-01 13:38:11 +01:00
5c648a8984
Merge remote branch 'origin/develop' into haswell
2013-12-01 11:25:33 +01:00
5048a80032
Refs #283 . Fixed the incorrect usage of long data type for Windows 64.
2013-11-14 13:46:42 +08:00
dfd1064d7b
refs #287 . Don't enable OpenMP for netlib LAPACK sequential Fortran codes.
2013-11-02 15:09:33 +08:00
c937090121
Added gfortran dependency for LSB/lsbcc.
2013-10-22 13:24:47 +08:00
c92ae012a6
Refs #279 . Provide ONLY_CBLAS flag. If you only need CBLAS without
...
a fortran compiler, please try make ONLY_CBLAS=1.
This mode only compiler CBLAS without BLAS fortran interface and LAPACK.
2013-08-21 00:03:25 +08:00
2638370844
Init code base for Intel Haswell.
2013-08-13 00:54:59 +08:00
673e453b3f
Enable bulldozer kernels.
2013-08-05 16:07:54 +08:00
a07cc39571
Refs #266 . Fixed the compiling bug with Open64 5.0.
2013-07-31 14:41:39 +08:00
5b504d6c23
Refs #263 . Rollback bulldozer and piledriver kernels to barcelona kernels.
2013-07-28 17:39:24 +08:00
77b572fa0b
Merge branch 'loongson3a' into develop
...
Conflicts:
Makefile.system
2013-07-20 22:33:17 +08:00
b67252c2e4
Ensure the correct stack alignment on Win32.
2013-07-17 15:19:07 +08:00
e80e285928
Update build matrix for Travis CI.
2013-07-11 23:49:29 +08:00
6df39ad9e7
Refs #248 . Support LAPACK and LAPACKE with lsbcc.
...
For LAPACKE, use LAPACK_COMPLEX_STRUCTURE.
The reson is lsbcc didn't define complex I in complex.h.
2013-07-10 16:02:27 +08:00
3eb5af1955
Refs #247 . Included lapack source codes. Avoid downloading tar.gz from netlib.org
...
Based on 3.4.2 version, apply patch.for_lapack-3.4.2.
2013-07-09 18:13:48 +08:00
f54f5bac9e
Refs #248 . Fixed the LSB compatiable issue for BLAS only.
...
For example, make CC=lsbcc NO_LAPACK=1.
2013-07-09 15:38:03 +08:00
886cbaf4e4
Support AMD Piledriver by bulldozer kernels.
2013-07-06 12:06:43 -03:00
cc522aa21d
Use quiet make for Travis CI.
2013-07-05 14:52:57 +08:00
cd1d473ba0
Merge pull request #230 from wernsaar/develop
...
Refs #230 . New dgemm and sgemm Kernel for BULLDOZER
2013-06-13 07:29:27 -07:00
56f160134d
Refs #231 . Change the default C compiler to clang on Mac OSX.
2013-06-13 22:15:19 +08:00
d854b30ae6
Added UNROLL values for 3M to getarch_2nd.c, Makefile.system and Makefile.L3
2013-06-09 17:26:42 +02:00
960b0c88a7
Refs #227 . Detected LLVM/Clang compiler.
2013-06-06 23:43:40 +08:00
f2fb8c7035
Change LIBSUFFIX from .lib to .a on windows.
2013-06-04 16:05:28 +08:00
357078b93e
Refs #216 . Revert the default value of GEMM_MULTITHREAD_THRESHOLD to 4.
2013-05-03 09:08:54 +08:00
48bdc1ad3b
Added NO_PARALLEL_MAKE flag to disable parallel make.
2013-04-15 21:37:30 +08:00
990efcab6e
Merge branch 'loongson3b' into loongson3a
2013-04-11 16:11:03 +00:00
75a5dc3975
Added the configure for the host loongcc compiling on Loongson3.
2013-04-11 16:10:47 +00:00
6958c1a1aa
Fixed the SEGFAULT bug with Loongcc and Loongson3.
2013-04-11 15:33:43 +08:00
1a57717b1a
Added the configuration of Loongcc compiler for Loongson 3 CPU.
2013-04-07 15:42:07 +08:00
5c8bf6ae0e
Merge branch 'bulldozer' into develop
2013-02-10 01:19:42 +08:00
4db6660de4
Refs #185 . Add missing 'const' to declarations in <cblas.h>. Thanks to Dan Povey!
...
The 'const' modifications were done automatically using this scripts:
https://kaldi.svn.sourceforge.net/svnroot/kaldi/sandbox/dan/tools/for_openblas
2013-01-20 22:52:51 +01:00
b7c0fa6bd2
Init AMD Bulldozer codebase.
2012-12-06 07:29:54 -05:00
e85549ee11
Fix NetBSD build.
2012-11-10 23:20:44 +00:00
08c177ca36
Refs #145 . Update LAPACK to 3.4.2 version.
2012-09-29 23:14:39 +08:00
2573311308
refs #140 . Fixed zdot incompatibility ABI issue with GCC 4.7 on Win 32.
...
GCC 4.7 uses MSVC ABI on Win 32. This means the caller pops the hidden pointer for returning
aggregate structures larger than 8 bytes.
2012-09-24 20:34:33 +08:00
758e34efbb
Fixed the detection bug on Loongson 3A server.
2012-09-21 10:14:07 +00:00
f76a384841
Refs #139 . Added NO_AVX flag to use old Nehalem kernels on Sandy Bridge.
...
For example, make NO_AVX=1 or make DYNAMIC_ARCH=1 NO_AVX=1
2012-09-17 23:25:46 +08:00
d0e731e8b8
provide support for passing CFLAGS, FFLAGS, PFLAGS, FPFLAGS to make on the command line
2012-08-21 00:31:12 -04:00
068861a927
Refs #133 . Users can set COMMON_OPT flag to control CFLAGS and FFLAGS.
2012-08-10 14:36:26 +08:00
e8306f623a
Refs #127 . Generate DLL without a version suffix on Windows.
2012-07-30 19:46:30 +02:00
25f1a573fd
Fixed the build bug when DYNAMIC_ARCH=0.
2012-07-07 12:12:24 +08:00
34fd3b85a8
Refs #113 . Fixed BOBCATE typo in dynamic arch building.
2012-06-19 17:05:16 +08:00
d6cab3f37e
Refs #113 . Support AMD Bobcate using Barcelona kernel codes. Replace 3DNow! with MMX.
2012-05-31 18:17:45 +08:00
a53c6e2440
Merge branch 'develop' into sandybridge
2012-05-25 23:16:44 +08:00
14c3511e92
Respect C compiler set on the command line or inherited from the environment
2012-05-20 18:09:35 +02:00
06e208c5c3
Refs #106 . Fixed wget and md5 bug on FreeBSD and NetBSD.
2012-05-13 11:43:29 +08:00
7f89edee3e
refs #103 Increase GEMM_MULTITHREAD_THRESHOLD to 50.
2012-05-03 20:05:34 +08:00
f4eee224d8
Refs #93 . Upgraded LAPACK to 3.4.1 version.
2012-04-23 17:15:22 +08:00
fd2ee0c9e2
Refs #88 . Fixed the build bug about LAPACKE C Interface to LAPACKE.
2012-04-13 23:12:06 +08:00
78914475ae
Fixed #88 . Build LAPACKE: C Interface to LAPACK.
2012-04-10 16:19:55 +08:00
910338f071
Merge branch 'develop' into sandybridge
2012-04-05 16:25:57 +08:00
fad089ffff
Fixed #84 the MD5 command line bug on Mac OSX.
2012-04-05 16:21:40 +08:00
19a48b82cf
Init Sandybridge codes based on Nehalem.
2012-03-30 20:01:03 +08:00
3871b6a86d
Merge branch 'loongson3b' into release-0.1.0
2012-03-23 01:26:44 +08:00
31c836ac25
Ref #79 Added GEMM_MULTITHREAD_THRESHOLD flag to use single thread in gemm function with small matrices.
2012-03-23 01:17:41 +08:00
1f15bee02a
Refs #81 . Added LIBNAMESUFFIX flag in Makefile.rule. The user can use this flag to control the library name, e.g. libopenblas.a, libopenblas_ifort.a or libopenblas_omp.a.
2012-03-21 23:57:09 +08:00
d047afe615
Set shared library soname in Linux.
2012-03-15 01:07:34 +08:00
0bbf955d4c
Refs #74 . Added -lgfortran into generating shared library.
2012-03-12 18:20:37 +08:00
0d76196a09
Fix #68 : don't require SystemStubs on OS X.
2011-12-28 23:53:20 -05:00
285e69e2d1
Disable using simple thread level3 to fix a bug on Loongson 3B.
2011-11-17 16:46:26 +00:00
d1baf14a64
Enable thread affinity on Loongson 3B. Fixed the bug of reading cycle counter.
...
In Loongson 3A and 3B, the CPU core increases the counter in every 2 cycles by default.
2011-11-11 17:49:41 +00:00
d0152ec8ca
Fixed #61 a building bug about setting TARGET and DYNAMIC_ARCH at the same time.
2011-09-17 02:27:56 +08:00
30947ea2d5
Fixed #44 a makefile bug when DYNAMIC_ARCH=1 and INTERFACE64=1.
2011-07-14 00:54:23 +08:00
c642b61d4d
Merge branch 'master' of github.com:xianyi/OpenBLAS into develop
2011-06-19 11:59:38 +08:00
49742cb2d3
Make USE_OPENMP=0 disable openmp
2011-06-11 14:38:10 +02:00
8d50a9fd1a
Fixed #35 a build bug with NO_LAPACK=1 & DYNAMIC_ARCH=1.
2011-06-09 11:38:59 +08:00
552f31dbbd
Fixed #13 . Fixed blasint undefined bug in <cblas.h> file.
2011-04-07 14:48:10 +08:00
fbf95688d6
Added utest frame using CUnit( http://cunit.sourceforge.net/ ).
2011-02-16 17:33:06 +08:00
0597c1076f
Added the configures of loongson 3a. refs #1
2011-01-24 22:45:35 +00:00
e6c13e2b3c
changed library name to openblas and modified environment variable.
2011-01-24 17:58:05 +00:00
342bbc3871
Import GotoBLAS2 1.13 BSD version codes.
2011-01-24 14:54:24 +00:00
Jerome Robert
Werner Saar
Zhang Xianyi
Steven G. Johnson
Thomas Anderson
Andreas Noack
Jerome Robert
Tony Kelman
Timothy Gu
Xianyi Zhang
Zaheer Chothia
Alexander Nasonov
Jameson Nash
Stefan Karpinski
Elias Pipping