Werner Saar
3119def9a7
updated cdot and zdot
11 years ago
Werner Saar
33b332372a
add optimized cdot- and zdot-kernel for sandybridge
11 years ago
Werner Saar
fd838c75bc
add optimized cdot- and zdot-kernel for haswell
11 years ago
Werner Saar
b57a60dac8
updated cdot and zdot for piledriver
11 years ago
Werner Saar
5c51163972
added optimized cdot- and zdot-kernel for steamroller
11 years ago
Werner Saar
9299d8cfd6
added optimized cdot- and zdot-kernels for bulldozer
11 years ago
Zhang Xianyi
0a3d3b945d
Refs #535 . Fix the wrong vector instruction in sgemm sandy bridge kernel.
11 years ago
Zhang Xianyi
4f680a7d61
Merge pull request #534 from wernsaar/develop
Refs #533 . added optimized saxpy- and daxpy-kernel for haswell and sandybridge
11 years ago
Werner Saar
ba926e807c
added cdot- and zdot benchmark
11 years ago
Werner Saar
60c6dec6e6
updated some lines for bulldozer
11 years ago
Werner Saar
47898cca35
added optimized saxpy- and daxpy-kernel for sandybridge
11 years ago
Werner Saar
53bb924287
added optimized saxpy- and daxpy-kernel for haswell
11 years ago
Zhang Xianyi
1e80b8b0d3
Merge pull request #531 from wernsaar/develop
added optimized sdot- and ddot-kernels for Haswell and Sandybridge
11 years ago
Werner Saar
a901b065d3
added optimized ddot-kernel for sandybridge
11 years ago
Werner Saar
3937e2a0a0
add optimized sdot-kernel for sandybridge
11 years ago
Werner Saar
9707d608d5
removed double definition line
11 years ago
Werner Saar
701b9d7556
added optimized sdot- and ddot-kernel for HASWELL
11 years ago
Zhang Xianyi
8977b3f235
Refs #529 . Support Intel Broadwell by Haswell kernels.
11 years ago
Zhang Xianyi
f6426395ea
Merge pull request #527 from xantares/patch-1
fix mingw install
11 years ago
xantares
0ac787eefe
fix mingw install
11 years ago
Zhang Xianyi
e5b96e55a7
Fix build bug for ARM64.
11 years ago
Zhang Xianyi
d0c51c4de9
Merge branch 'develop'
11 years ago
Zhang Xianyi
a3491e1e88
Update the doc for 0.2.14.
11 years ago
Zhang Xianyi
e81a5d61e4
Merge branch 'develop' of github.com:xianyi/OpenBLAS into develop
11 years ago
Zhang Xianyi
c674fa32be
Add ARM targets.
11 years ago
Zhang Xianyi
e34911a73d
Fix compiling bug for ARM with setting BINARY.
11 years ago
Zhang Xianyi
76dcaf2281
Merge pull request #521 from maxlevesque/patch-1
Correct typo /proc/ instead of /pros/
11 years ago
Maximilien Levesque
770fac92eb
Correct typo /proc/ instead of /pros/
11 years ago
Zhang Xianyi
e95d64333a
Refs #519 . Avoid calling strncpy.
11 years ago
Zhang Xianyi
75c40bcc48
Refs #520 . Fixed ONLY_CBLAS=1 compiling bug on OSX.
11 years ago
Zhang Xianyi
b62f9f4120
Merge pull request #518 from ton/issue-508
Fix issue #508
11 years ago
Ton van den Heuvel
b6438dedea
Fix issue #508
Fix race condition during shutdown causing a crash in
gotoblas_set_affinity().
11 years ago
Hank Anderson
1d183dcda8
Added lapacke sources.
11 years ago
Zhang Xianyi
cdefdb21cd
Refs #492 . Fixed c/zsyr bug with negative incx.
11 years ago
Hank Anderson
e19bf3a28b
Removed MSVC cpuid func when using clang.
11 years ago
Hank Anderson
3649cfbd7b
Fixed EPILOGUE for clang.
11 years ago
Hank Anderson
5ae8993752
Added intrinsics for MSVC.
11 years ago
Hank Anderson
84d90d6ed8
Fixed some compiler errors/warnings for clang.
11 years ago
Hank Anderson
518e2424a8
Fixed bad filename for cpuid.S compile.
11 years ago
Zhang Xianyi
ea7f9dacf4
Refs #509 . Fixed geadd building bug with DYNAMIC_ARCH=1.
11 years ago
Zhang Xianyi
bf5dbb7e2a
Refs#509. Merge branch 'grisuthedragon-develop' into develop
11 years ago
Hank Anderson
00e373aea6
Added LAPACK sources directly to add_library call instead of OBJECT.
11 years ago
Hank Anderson
9eaea02f33
Added additional gemm defines for complex types.
11 years ago
Hank Anderson
ab7043373f
Fixed bug generating trmv complex source names.
11 years ago
Hank Anderson
504cdb10ed
Added check for MSVC before enabling fortran.
Currently forcing gfortran, instead of assuming ifort.
11 years ago
Hank Anderson
a8002b0c5f
Separated getarch ASM file when using MSVC.
11 years ago
Hank Anderson
0553476fba
Added TRANS defines for complex sources in lapack.
11 years ago
Hank Anderson
2416d9dbac
Fixed TRANSA defines for complex sources in driver/level2.
11 years ago
Hank Anderson
0d8e227ea7
Changed strategy for setting preprocessor definitions.
Instead of generating separate object files for each permutation of
defines for a source file, GenerateNamedObjects now writes an entirely
new source file and inserts the defines as #define c statements.
This solves a problem I ran into with ar.exe where it was refusing to
link objects that had the same filename despite having different paths.
11 years ago
Hank Anderson
12d1fb2e40
Fixed incorrect object name in kernel CMakeLists.txt
11 years ago