Zhang Xianyi
c937090121
Added gfortran dependency for LSB/lsbcc.
12 years ago
wernsaar
fe8c5666f9
optimized dgemm_kernel for HASWELL
12 years ago
wernsaar
f6b50057e2
corrected and testet FMA3 Code
12 years ago
wernsaar
2840d56aeb
added dgemm_kernel for Piledriver
12 years ago
wernsaar
2d49db2f5b
moved compiler flags from Makefile.rule to Makefile.arm
12 years ago
wernsaar
04391e6d9c
optimized param.h
12 years ago
wernsaar
85484a42df
added kernels for cgemm, ctrmm, zgemm and ztrmm
12 years ago
wernsaar
3983011f0b
added sgemm- and strmm_kernel
12 years ago
wernsaar
2a1515c9dd
added dgemm_ncopy_4_vfpv3.S
12 years ago
wernsaar
31f51e78bc
minor optimizations on dgemm_kernel
12 years ago
wangqian
beffee7d91
Fixed buffer overflow bug in kernel/x86_64/dgemv_t.S file.
12 years ago
Zhang Xianyi
a35f4343fa
Merge pull request #301 from yieldthought/develop
Remove -Wl,--retain-symbols-file from dynamic link line to fix tool support
12 years ago
yieldthought
ce5626a384
Remove -Wl,--retain-symbols-file from dynamic library linking to fix tool support
The aim is to restrict the symbols exported from openBLAS, but for dynamic libraries --retain-symbols-file has the opposite effect intended. It removes symbols from the .symtab section of the library, which is used by tools such as debuggers, profilers, objdump etc. but does not remove them from .dynsym, which is the section used by the runtime when loading the binary.
In short, using --retain-symbols-file like this doesn't prevent symbol leakage but does prevent tools from analyzing applications linked with openBLAS.
This patch improves tool support for openBLAS applications but does not address symbol leakage through .dynsym.
12 years ago
wernsaar
e0b968c3a7
Changed kernels for dgemm and dtrmm
12 years ago
wernsaar
93f1074dd4
changed some values for arm
12 years ago
wernsaar
1c63180bb6
updated dgemm_kernel_8x2_vfpv3.S
12 years ago
wernsaar
22a8fcc4b7
add modified c_check perl program
12 years ago
wernsaar
9965d48005
added Makefile.arm
12 years ago
wernsaar
4a474ea7dc
changed dgemm_kernel to use fused multiply add
12 years ago
wernsaar
69ce737cc5
modified Makefile.L3 for ARM
12 years ago
wernsaar
d13788d1b4
common files modified for ARM
12 years ago
wernsaar
70411af888
initial checkin of kernel/arm
12 years ago
Zhang Xianyi
16eb780e13
Refs #262 . Fixed compatibility issues of GNU stack markings with PathScale EKOPath(tm) Compiler Suite: Version 4.0.12.1
12 years ago
Zhang Xianyi
a746724e84
Added backers.
12 years ago
Lars Buitinck
3f7b0cd994
Merge pull request #290 from larsmans/missing-threshold
check if GEMM_MULTITHREAD_THRESHOLD defined in gemm.c
Set a fallback value.
12 years ago
Zhang Xianyi
cc6db2ecfe
Merge pull request #291 from larsmans/fix-makefile-prefix
fix default prefix handling in makefiles
12 years ago
Zhang Xianyi
3175be4b3d
Merge pull request #289 from larsmans/no-noconst
get rid of the generated cblas_noconst.h file
12 years ago
Lars Buitinck
a29e6592da
fix default prefix handling in makefiles
PREFIX wasn't communicated to Makefile.install (where it matters)
by Makefile. The result is that the default PREFIX is empty and
OpenBLAS was being installed in /lib.
12 years ago
Lars Buitinck
212463dce9
get rid of the generated cblas_noconst.h file
12 years ago
Zhang Xianyi
037bd82bef
Merge pull request #288 from sebastien-villemot/develop
Avoid failure on qemu guests declaring an Athlon CPU without 3dnow!
12 years ago
Sébastien Villemot
eae4cfa3f6
Avoid failure on qemu guests declaring an Athlon CPU without 3dnow!
The present patch verifies that, on machines declaring an Athlon CPU model and
family, the 3dnow and 3dnowext feature flags are indeed present. If they are
not, it fallbacks on the most generic x86 kernel. This prevents crashes due to
illegal instruction on qemu guests with a weird configuration.
Closes #272
12 years ago
Zhang Xianyi
6c4a7d0828
Import AMD Piledriver DGEMM kernel generated by AUGEM.
So far, this kernel doesn't deal with edge.
AUGEM: Automatically Generate High Performance Dense Linear Algebra
Kernels on x86 CPUs.
Qian Wang, Xianyi Zhang, Yunquan Zhang, and Qing Yi. In the
International Conference for High Performance Computing, Networking,
Storage and Analysis (SC'13). Denver, CO. Nov, 2013.
12 years ago
Zhang Xianyi
fe98de2f68
Merge branch 'bulldozer' into develop
12 years ago
Zhang Xianyi
db389b5915
Refs #281 . Detect __CYGWIN__ macro for Cygwin x86_64.
Signed-off-by: Zhang Xianyi <traits.zhang@gmail.com>
12 years ago
Zhang Xianyi
52f587db7f
Refs #281 . Detect _WIN32 macro for Windows API.
http://www.mail-archive.com/bug-gnulib@gnu.org/msg05722.html
12 years ago
wernsaar
067e8417fd
removed unnessesary instructions from zgemm_kernel_2x2_bulldozer.S
12 years ago
wernsaar
a82da3d069
removed unnessesary instructions
12 years ago
Zhang Xianyi
1569bf14f8
Refs #282 . Fixed zgemv_n typo bug on Win64.
12 years ago
Zhang Xianyi
df554aebd2
Merge pull request #280 from ViralBShah/develop
Patch LAPACK XLASD4.f as discussed in JuliaLang/julia#2340
12 years ago
Viral B. Shah
eae6920f2d
Patch LAPACK XLASD4.f as discussed in JuliaLang/julia#2340
12 years ago
Zhang Xianyi
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.
12 years ago
Zhang Xianyi
f51a849d91
Merge pull request #278 from wernsaar/haswell
Merge wernsaar's Haswell gemm kernels.
12 years ago
wernsaar
44ef70420c
added cgemm_kernel_8x2_haswell.S
12 years ago
wernsaar
d488b1b1aa
added zgemm_kernel_4x2_haswell.S
12 years ago
wernsaar
4070d9a123
added dgemm_kernel_16x2_haswell.S
12 years ago
wernsaar
0b90c0ec64
added sgemm_kernel_16x4_haswell.S
12 years ago
wernsaar
2b8ab8f55b
sgemm_kernel_16x4_haswell.S minor changes
12 years ago
wernsaar
1cb9579cd0
added zgemm_kernel_4x2_haswell.S and fixed a bug in sgemm_kernel_16x4_haswell.S
12 years ago
Zhang Xianyi
2638370844
Init code base for Intel Haswell.
12 years ago
wernsaar
89637f87c8
added sgemm- and dgemm-kernel for HASWELL processor
12 years ago