Do not run any cleanup if the program is exiting anyway

From keno's PR #2350 - this avoids the potential hang in blas_thread_shutdown where we may wait for threads to exit while they are waiting on the loader lock from DllMain

Free Windows thread memory with MEM_RELEASE rather than MEM_DECOMMIT

fix error messages

as suggested by hjmndv in #2370

Improve paralleled SGEMM performance on SKYLAKEX CPUs

BTW some kernels were adjusted to improve performance

Add new file lapack.h from LAPACK 3.9.0 to installable headers

new file in LAPACK 3.9.0, split off from lapacke.h

rebase

Fix SKYLAKEX STRMM issues

Optimize AVX2 SGEMM & STRMM

Apply LAPACKE fix for eigenvector transposition in symmetric eigensolvers

from Reference-LAPACK PR 330

rebase

Test all 7 declared values of N in float and double cblas3 tests

kernel/arm64/dgemm_beta.S: add beta == zero branch

Use arm neon instructions to optimize ncopy operation (and enable 7th column in float+complex cblas3 test drivers)

added beta == zero branch, and no need to load C matrix.

Signed by: Xuqiang Chen <chenxuqiang3@hisilicon.com>

Use arm neon instructions to optimize sgemm_beta operation

[WIP] Use arm neon instructions to optimize tcopy operation

* Update make.inc entries for LAPACK 3.9.0

Reference-LAPACK PR 347 changed some variable names and relative paths

* Update LAPACK to 3.9.0

* Add new functions from LAPACK 3.9.0

* Add new functions from LAPACK 3.9.0

* Restore LOADER command 

as it makes it easier to specify pthread as needed

* Restore LOADER

* Restore EIG/LIN prefixes in cmdbase

* add binary path to lapack_testing.py call

* Restore OpenMP version check

* Restore OpenMP version check

* Restore fix for out-of-bounds array accesses

from #2096

AVX2 ZGEMM3M kernel

prefetching for dgemm_beta

add in runtime cpu detection for zarch

modified:   KERNEL.ARMV8
	modified:   KERNEL.TSV110
	new file:   sgemm_ncopy_4.S

modified:   ctest/sin3