Martin Kroeker
15a78d6b66
export NO_AVX512 setting
7 years ago
Martin Kroeker
b7feded85a
Propagate NO_AVX512 via CCOMMON_OPT
7 years ago
Martin Kroeker
a7d0f49cec
Add SKYLAKEX to DYNAMIC_CORE list only if AVX512 is available
7 years ago
Martin Kroeker
f1fb9a4745
Propagate NO_AVX512 if needed
7 years ago
Arjan van de Ven
99c7bba8e4
Initial support for SkylakeX / AVX512
This patch adds the basic infrastructure for adding the SkylakeX (Intel Skylake server)
target. The SkylakeX target will use the AVX512 (AVX512VL level) instruction set,
which brings 2 basic things:
1) 512 bit wide SIMD (2x width of AVX2)
2) 32 SIMD registers (2x the number on AVX2)
This initial patch only contains a trivial transofrmation of the Haswell SGEMM kernel
to AVX512VL; more will follow later but this patch aims to get the infrastructure
in place for this "later".
Full performance tuning has not been done yet; with more registers and wider SIMD
it's in theory possible to retune the kernels but even without that there's an
interesting enough performance increase (30-40% range) with just this change.
7 years ago
Zhang Xianyi
50acc40613
Merge pull request #1536 from WestAlgo/develop
Fix race condition in blas_server_omp.c
7 years ago
Martin Kroeker
12398e53ce
Merge pull request #1553 from martin-frbg/ifort-openmpflag
Change -openmp to -fopenmp for ifort entry as well
7 years ago
Martin Kroeker
193f835662
Change -openmp to -fopenmp for ifort entry as well
7 years ago
Martin Kroeker
e3a069f108
Merge pull request #1550 from martin-frbg/ifort-openmpflag
Update compiler flag for openmp use with ICC
7 years ago
Martin Kroeker
65b8a5c5d8
Update compiler flag for openmp use with ICC
The deprecated -openmp option was finally removed in favor of -qopenmp or -fopenmp, picking the latter to stay compatible with Intel compiler versions before 2015 (when -q options were introduced). Fixes #1546
7 years ago
Martin Kroeker
71051259e0
Restore compiler options for mips P5600 target
7 years ago
Martin Kroeker
d94d7baf7e
Add mips32r2 api target
7 years ago
Zhiyong Dang
1b83341d19
Fix race condition in blas_server_omp.c
Change-Id: Ic896276cd073d6b41930c7c5a29d66348cd1725d
7 years ago
Jerry Zhao
c167a3d6f4
Added RISCV build
7 years ago
Alex Arslan
8f811a9312
Reinstate macOS logic
7 years ago
Alex Arslan
36a17536ca
Compile with cc rather than gcc whenever possible
7 years ago
Alex Arslan
a41d241a0e
Add support for DragonFly BSD
7 years ago
Alex Arslan
8da6b6ae52
Allow building on OpenBSD
With this change, OpenBLAS builds and all tests pass on OpenBSD 6.2
using Clang. Tested on x86-64 only, with and without DYNAMIC_ARCH=1.
7 years ago
Erik M. Bray
ce2028b425
Rewrite this test to work with ctest and re-enable it on the appropriate platforms (including Cygwin, which has fork())
7 years ago
Martin Kroeker
4a4f6658de
When forcing USE_THREAD=0, override USE_OPENMP as well
This avoids an error exit a few lines down as USE_THREAD=0 conflicts with USE_OPENMP=1
7 years ago
Shivraj Patil
e3d844b062
Added mips I6500 core
Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>
8 years ago
Martin Kroeker
601c71fe54
Merge pull request #1304 from martin-frbg/aix-build-fixes
(Plain make) build system fixes for AIX
8 years ago
Martin Kroeker
3810a6fd99
(Plain make) build system fixes for AIX
- retry fortran compiler test with aix-specific option if generic -m32/-m64 fails
- pass any custom ARFLAGS to lapack
- no addition of -m32/-m64 to the CFLAGS and FFLAGS on AIX
8 years ago
Erik M. Bray
dddedbab5d
More canonical installation on Cygwin:
* The DLL is named cygopenblas.dll, not libopenblas.dll
* The import lib (still called libopenblas.dll.a) is installed
8 years ago
Christoph Conrads
9cc6599056
ARM: do not add linker flag `-lm` unconditionally
On ARM the required math library depends on whether the soft floating
point ABI is used or not but this is already handled in
`Makefile.system`, lines 499-505.
8 years ago
Martin Kroeker
31e086d6a6
Disable ReLAPACK by default ( #1238 )
* Disable ReLAPACK by default; mention it in final build message if included
* Add files via upload
* Add files via upload
* Add files via upload
8 years ago
Zhang Xianyi
4239dd65ce
Merge branch 'develop' into develop_arm_softfp
8 years ago
Zhang Xianyi
fa6a920caa
Link -lm or -lm_hard for Android ARMv7.
8 years ago
Ashwin Sekhar T K
37efb5bc1d
arm: Remove unnecessary files/code
Since softfp code has been added to all required vfp kernels,
the code for auto detection of abi is no longer required.
The option to force softfp ABI on make command line by giving
ARM_SOFTFP_ABI=1 is retained. But there is no need to give this option
anymore.
Also the newly added C versions of 4x4/4x2 gemm/trmm kernels are removed.
These are longer required. Moreover these kernels has bugs.
8 years ago
Ashwin Sekhar T K
0d5c8e5386
arm: Determine the abi from compiler if not specified on command line
If ARM abi is not explicitly mentioned on the command line, then set the
arm abi to softfp or hard according to the compiler environment.
This assumes that compiler sets the defines __ARM_PCS and __ARM_PCS_VFP
accordingly.
8 years ago
Zhang Xianyi
482015f8d6
Merge branch 'arm_soft_fp_abi' into develop
8 years ago
Paul Osmialowski
42bbe74791
build: LLVM: Add Flang compiler support and enable OpenMP for Clang
Signed-off-by: Paul Osmialowski <pawel.osmialowski@arm.com>
8 years ago
Elliot Saba
6e972994b2
Force `DYNAMIC_ARCH` to empty when `DYNAMIC_CORE` is not set
8 years ago
Denis Steckelmacher
c9ff735da6
Add ZEN support (tested for auto-detected static backend)
8 years ago
Zhang Xianyi
90e02ccf68
Support ARM softfp ABI for sgemm on ARMV7.
make ARM_SOFTFP_ABI=1
8 years ago
Zhang Xianyi
503dcbfde6
Merge branch 'develop' into arm_soft_fp_abi
8 years ago
Martin Kroeker
d0035b857d
Apply COMMON_OPT to default FFLAGS to avoid building non-optimized LAPACK by mistake
8 years ago
François Bissey
32ca9a9f68
Do not override MACOSX_DEPLOYMENT_TARGET if it is already defined.
8 years ago
Shivraj Patil
9687437928
MIPS n32 ABI and build time mips simd support check
Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>
9 years ago
Shivraj Patil
d1c6469283
MIPS n32 ABI support, MSA support detection and rename ARCH, ARCHFLAGS
Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>
9 years ago
Shivraj Patil
22b9c2747d
P6600/I6400 Build fix. Reverted the changes which was done to support for MIPS n32 ABI
Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>
9 years ago
Shivraj Patil
beb1d076a4
Added MSA optimization for GEMV_N, GEMV_T, ASUM, DOT functions
Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>
9 years ago
Vicente Olivert Riera
7f28cd1f88
Complete support for MIPS n32 ABI
Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
9 years ago
Vicente Olivert Riera
e12cff87b8
Makefile.system: P5600 and I6400 cores need -mmsa
Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
9 years ago
Shivraj Patil
2c3dfe2bf3
MIPS P5600(32 bit) and I6400(64 bit) cores support added.
Seperated mips and mips64 files.
Configurations support for mips 32 bit.
Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>
9 years ago
buffer51
7f7d04dcd2
Fixed linking error when compiling ARMv7 for Android (disabled -lpthread and added -Wl,--no-warn-mismatch).
9 years ago
Tony Kelman
5fef0d1b75
re enable Fortran optimization flag on windows
partial revert of 299cdcdc29
from #696 , was not explained why that was needed
9 years ago
Martin Kroeker
44062517eb
Update Makefile.system
Define AWK as "nawk" for SunOS (actually Illumos) only - fixes #763
9 years ago
Zhang Xianyi
aaa8551c57
Merge pull request #749 from lotheac/illumos_fixes
illumos fixes
9 years ago
Lauri Tirkkonen
31aff441ce
use $(AWK) in Makefile.install and switch it to nawk
9 years ago