Description:kernelspace musl code

Team:OTHERS
Feature or Bugfix:Feature
Binary Source:NA
PrivateCode(Yes/No):No

Change-Id: I99874a7c570b7d22b4a3d34840260eb48ea3ffa1
Reviewed-on: http://mgit-tm.rnd.huawei.com/10276995Reviewed-by: Nshenwei 00579521 <denny.shenwei@huawei.com>
Tested-by: Npublic jenkins <public_jenkins@notesmail.huawei.com>

Both sqrt and sqrtf shifted the signed exponent as signed int to adjust
the bit representation of the result. There are signed right shifts too
in the code but those are implementation defined and are expected to
compile to arithmetic shift on supported compilers and targets.

zero, one, two, half are replaced by const literals
The policy was to use the f suffix for float consts (1.0f),
but don't use suffix for long double consts (these consts
can be exactly represented as double).

thanks to the hard work of Szabolcs Nagy (nsz), identifying the best
(from correctness and license standpoint) implementations from freebsd
and openbsd and cleaning them up! musl should now fully support c99
float and long double math functions, and has near-complete complex
math support. tgmath should also work (fully on gcc-compatible
compilers, and mostly on any c99 compiler).

based largely on commit 0376d44a890fea261506f1fc63833e7a686dca19 from
nsz's libm git repo, with some additions (dummy versions of a few
missing long double complex functions, etc.) by me.

various cleanups still need to be made, including re-adding (if
they're correct) some asm functions that were dropped.