Replace spurious AVX512 requirement in the Haswell srot microkernel with an AVX2/FMA3 guard

Optimize s/dasum function for POWER10

This patch makes use of new POWER10 vector pair instructions for
loads and stores.

Support timing Apple M1 in the benchmarks

Fix AMD AOCC compiler detection

Fix unusual stack size requirements of the LAPACK EIG tests (from Reference-LAPACK PR 335)

POWER10: Rename mma builtins

The LLVM and GCC teams agreed to rename the __builtin_mma_assemble_pair and
__builtin_mma_disassemble_pair built-ins to __builtin_vsx_assemble_pair and
__builtin_vsx_disassemble_pair respectively. This patch is to make
corresponding changes in dgemm kernel. Also made changes in
inputs to those builtins to avoid some potential typecasting issues.

Reference gcc commit id:77ef995c1fbcab76a2a69b9f4700bcfd005d8e62

Fix use of undefined CC variable in f_check

rebase

Revert "Fix undefined CC in f_check (again)"

Fix undefined CC in f_check (again)

rebase

use /usr/bin/env perl

Fix dll_callback and p_process_term signatures for USE_TLS on Windows x64

Replace unoptimized OMATCOPY_RT with 4x4 blocked version

as suggested by MigMuc in issue 2532

rebase

Fix race in blas_thread_shutdown.

blas_server_avail was read without holding server_lock. If multiple threads call blas_thread_shutdown simultaneously, for example, by calling fork(), then they can attempt to shut down multiple times. This can lead to a segmentation fault.

Fix get_num_procs()  in the USE_TLS branch for non-glibc systems

Recognize Intel Tiger Lake CPUID as SkylakeX

Fix build issue on POWER8 with DYNAMIC_ARCH

Running make DYNAMIC_ARCH=1 on POWER 8 BE with gcc10.2 version, gives
the following error due to the difference in UNROLL_M/N.
'No rule to make target 'dgemm_incopy_POWER10.o', needed by kernel'