wernsaar
|
793175be3a
|
added experimental support for big numa machines
|
11 years ago |
wernsaar
|
271af406f3
|
bugfix for linux affinity code
|
11 years ago |
wernsaar
|
f5f50b3563
|
added benchmarks for lapack potrf, potrs and potri functions
|
11 years ago |
wernsaar
|
651dd22d7d
|
added benchmark program for lapack ?getri functions
|
11 years ago |
wernsaar
|
7c611a2f95
|
bugfix for zgeev
|
11 years ago |
wernsaar
|
296564e369
|
added lapack geev benchmark
|
11 years ago |
wernsaar
|
799a0eabbd
|
bugfix in cholesky.c
|
11 years ago |
wernsaar
|
ca63503e61
|
extented plot-filter.sh for linpack and cholesky benchmarks
|
11 years ago |
wernsaar
|
5087096711
|
optimization of sandybridge cgemm-kernel
|
11 years ago |
Zhang Xianyi
|
21f7768b26
|
Merge pull request #421 from wernsaar/develop
optimized sgemm- and cgemm-kernel for haswell
|
11 years ago |
wernsaar
|
46bc4fd50c
|
optimized cgemm kernel for haswell
|
11 years ago |
wernsaar
|
1cc02b4337
|
optimized sgemm kernel for haswell
|
11 years ago |
Zhang Xianyi
|
6e223db7fc
|
Merge pull request #420 from wernsaar/develop
Optimizations for HASWELL
|
11 years ago |
wernsaar
|
1d33547222
|
optimized zgemm kernel for haswell
|
11 years ago |
wernsaar
|
3ea4dadd30
|
optimizations for trsm
|
11 years ago |
wernsaar
|
1b10ff129a
|
optimizations for trmm
|
11 years ago |
wernsaar
|
125610d23b
|
allow to set custom value for ?GEMM_DEFAULT_UNROLL_MN, optimizations for syrk
|
11 years ago |
wernsaar
|
e213a42cde
|
added a sample plot-filter scripts and a header file for gnuplot
|
11 years ago |
wernsaar
|
e4663be46a
|
added symv benchmark
|
11 years ago |
wernsaar
|
11637b6926
|
add benchmark for ger
|
11 years ago |
Zhang Xianyi
|
80bf3e6a35
|
Merge pull request #419 from wernsaar/develop
added optimized sgemv kernels for Sandy Bridge, Haswell, Bullldozer, and Piledriver.
|
11 years ago |
wernsaar
|
6acbafe45b
|
added sgemv_n microkernel for haswell
|
11 years ago |
wernsaar
|
5392d11b04
|
optimized sgemv_n_microk_sandy.c
|
11 years ago |
wernsaar
|
c0fe95fb72
|
added sgemv_n microkernel for sandybridge
|
11 years ago |
wernsaar
|
d9d4077c93
|
added sgemv_t microkernel for haswell
|
11 years ago |
wernsaar
|
02eb72ac42
|
bugfix in sgemv_t_microk_sandy.c
|
11 years ago |
wernsaar
|
c06f9986d4
|
added sgemv_t microkernel for sandybridge
|
11 years ago |
wernsaar
|
2cce125c79
|
added optimized sgemv_t for bulldozer and piledriver
|
11 years ago |
wernsaar
|
b3938fe371
|
don't use this sgemv_n on Windows
|
11 years ago |
Zhang Xianyi
|
e6668dd83b
|
Merge pull request #414 from staticfloat/sf/symlinkfix
Don't create an absolute symlink when installing on Darwin
|
11 years ago |
wernsaar
|
c8a4a56177
|
performance optimizations for sgemv_n
|
11 years ago |
wernsaar
|
3c5732615d
|
added blocked sgemv_n and microkernel for bulldozer and piledriver
|
11 years ago |
Zhang Xianyi
|
134fa320e6
|
Refs #415. Fixed the x86/i386 compiling bug with DYNAMIC_ARCH=1.
|
11 years ago |
Elliot Saba
|
a79df1ff49
|
Don't create an absolute symlink when installing on Darwin
|
11 years ago |
wernsaar
|
7ceb25d7b3
|
changed string GFORTRAN to lowercase
|
11 years ago |
Zhang Xianyi
|
f2eb480738
|
OpenBLAS 0.2.10 version.
|
11 years ago |
Zhang Xianyi
|
c94762bb56
|
Refs #401. Added NO_AVX2 flag for old binutils (e.g. RHEL6)
|
11 years ago |
wernsaar
|
51413925bd
|
adjust number of threads for small size in cgemv and zgemv
|
11 years ago |
wernsaar
|
b985cea65d
|
adjust number of threads for sgemv and dgemv
|
11 years ago |
wernsaar
|
d286daa2ba
|
adjusted number of threads for small size
|
11 years ago |
wernsaar
|
bcb115b55b
|
added benchmark for gemv
|
11 years ago |
Zhang Xianyi
|
3dd094f17a
|
Merge pull request #413 from wernsaar/develop
additional benchmarks
|
11 years ago |
wernsaar
|
339ab34c4c
|
added additional test value to dstest.in
|
11 years ago |
wernsaar
|
7424e2b609
|
added additional test value
|
11 years ago |
wernsaar
|
73594cff73
|
segment violation in x86_64 sgemv kernels
|
11 years ago |
wernsaar
|
880597b301
|
segment violation in sgemv kernels
|
11 years ago |
wernsaar
|
9c835431d0
|
modified pathes to atlas, mkl and acml
|
11 years ago |
wernsaar
|
1d4ffddf69
|
added conf option for number of loops
|
11 years ago |
wernsaar
|
b0e7810a6b
|
added her2k benchmark
|
11 years ago |
wernsaar
|
2b92a8c499
|
added herk benchmark
|
11 years ago |