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.

Fix race condition in blas_server_omp.c

Change -openmp to -fopenmp for ifort entry as well

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

Change-Id: Ic896276cd073d6b41930c7c5a29d66348cd1725d

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.

This avoids an error exit a few lines down as USE_THREAD=0 conflicts with USE_OPENMP=1

Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>

(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

* The DLL is named cygopenblas.dll, not libopenblas.dll
* The import lib (still called libopenblas.dll.a) is installed

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.

* Disable ReLAPACK by default; mention it in final build message if included

* Add files via upload

* Add files via upload

* Add files via upload

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.

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.

Signed-off-by: Paul Osmialowski <pawel.osmialowski@arm.com>

make ARM_SOFTFP_ABI=1

Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>

Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>

Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>

Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>

Signed-off-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>

Seperated mips and mips64 files.
Configurations support for mips 32 bit.

Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>

partial revert of 299cdcdc29
from #696, was not explained why that was needed