sw64: optimize ip checksum calculation

Sunway inclusion
category: performance
bugzilla: https://gitee.com/openeuler/kernel/issues/I56W9F



--------------------------------

Function csum_partial_copy_from_user() copy data from userspace and
return checksum result. It used to read 8 bytes data from userspace
every time, calculate checksum and store it to the destination, which
slowed down the copying process. This patch moves the copying process
to the beginning. This patch also rewrites do_csum and move some common
codes to header file to avoid duplication.

Signed-off-by: Cui Mingrui <cuimingrui@wxiat.com>

Signed-off-by: Gu Zitao <guzitao@wxiat.com>
Acked-by: Xie XiuQi <xiexiuqi@huawei.com>
Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>