these optimizing routing/cjdns examples for armv7

the cjdns/Makefile should be using these[1] performance optimizations. How many other target systems can follow suit here??

• https://github.com/phillymesh/cjdns-optimizations/issues/3

Orange Pi Zero

$ CFLAGS="-s -static -Wall -march=armv7ve -mcpu=cortex-a7 -mfpu=neon-vfpv4 -mfloat-abi=hard -mtune=cortex-a7 -fomit-frame-pointer -marm" ./do

Pine A64

$ Seccomp_NO=1 CFLAGS="-s -static -Wall -march=armv8-a+crc+crypto -mcpu=cortex-a53 -ftree-vectorize -mtune=cortex-a53 -fomit-frame-pointer" ./do

BeagleBone Black

$ CFLAGS="-s -static -Wall -march=armv7-a -mcpu=cortex-a8 -mtune=cortex-a8 -marm -mfloat-abi=hard -mfpu=neon -fomit-frame-pointer" ./do

Raspberry Pi 3

$ Seccomp_NO=1 CFLAGS="-s -static -Wall -mfpu=neon-vfpv4 -mfloat-abi=hard -mcpu=cortex-a7 -mtune=cortex-a7 -fomit-frame-pointer -marm" ./do

Raspberry Pi 2

$ Seccomp_NO=1 CFLAGS="-s -static -Wall -mfpu=neon -mcpu=cortex-a7 -mtune=cortex-a7 -fomit-frame-pointer -marm" ./do

ODROID C2

$ Seccomp_NO=1 CFLAGS="-s -static -Wall -mfpu=neon-vfpv4 -mfloat-abi=hard -mcpu=cortex-a7 -mtune=cortex-a7 -fomit-frame-pointer -marm" ./do

Start with cjdns/Makefile and the ./do invocation examples in the issue, then compare the CFLAGS for each listed ARM board. Review the linked cjdns-optimizations issue for context. Done means the Makefile applies appropriate optimizations to the supported target systems.