The provided -march flags, especially for ARMv5 and ARMv6 may not
necessarily match the needed ones: for ARMv5, it might be armv5,
armv5te, armv5t, etc. If the wrong one is used, the incorrect toolchain
sysroot can be used in a multilib toolchain.

Therefore, let the user building OpenBLAS pass the appropriate -march
flag.

The other flags, such as -mfpu=vfp or -mfloat-abi=hard are kept, as they
are actually required for the build to proceed (OpenBLAS uses VFP
instructions, and assume an EABIhf ABI).

[Peter: update for v0.2.20]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
[Retrieved from:
https://git.buildroot.net/buildroot/tree/package/openblas/0001-Makefile.arm-remove-march-flags.patch]
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>

Disable the SkyLakeX DGEMMIxCOPY kernels as well

#1955

as a stopgap measure for https://github.com/numpy/numpy/issues/13401 as mentioned in #1955

Misc. typo fixes in comments and documentation

sgemm/strmm kernel for power9

Correct argument of CPU_ISSET for glibc <2.5

fixes #2104

Found via `codespell -q 3 -w -L ith,als,dum,nd,amin,nto,wis,ba -S ./relapack,./kernel,./lapack-netlib`

Merge develop in preparation of 0.3.6 release

Disable repeated recursion on Ab_BR in ReLAPACK xGBTRF

due to crashes in LAPACK tests

Disable reallocation of work array in ReLAPACK xSYTRF

Correct INFO=4 condition in ReLAPACK xGETRF

as it appears to cause memory management problems (seen in the LAPACK tests)

Avoid out-of-bounds accesses in LAPACK EIG tests

Correct length of name string in xerbla call

see https://github.com/Reference-LAPACK/lapack/issues/333

Fix ReLAPACK integration problems

1. Replaced all instances of "int" with "blasint"
2. Added string length as "hidden" third parameter in calls to fortran XERBLA

snprintf define consolidated to common.h

Add (C)BLAS extension ?sum

Add in runtime CPU detection for POWER.

Add -lm and disable EXPRECISION support on *BSD

fixes #2075

Detect 32bit environment on 64bit ARM hardware

for #2056, using same approach as #2058

as trivial copies of the respective ?asum kernels with the ABS and vflpsb calls removed

as trivial copy of ?asum with the fabs calls removed

as trivial copy of ?asum with the fabs calls removed