Fails on Raspberry Pi 3 (ARM Cortex-A53)

[lamasp]

Hi,

Just tried chacha-opt on a Raspberry Pi 3 and it fails (poly1305-opt works fine). The secondary issue is it doesn't fall back to the generic implementation.

CPU is: Broadcom BCM2387 chipset Quad-Core ARM Cortex-A53 CPU 1.2GHz

Results from bench test are:

$ bin/chacha-util bench
ChaCha

---- 8 rounds ----
time granularity: 17 us, 1000412 us/second

neon: error in implementation!
armv6: error in implementation!
neon: error in implementation!
armv6: error in implementation!
neon: error in implementation!
armv6: error in implementation!
neon: error in implementation!
armv6: error in implementation!

---- 12 rounds ----
neon: error in implementation!
armv6: error in implementation!
neon: error in implementation!
armv6: error in implementation!
neon: error in implementation!
armv6: error in implementation!
neon: error in implementation!
armv6: error in implementation!

---- 20 rounds ----
neon: error in implementation!
armv6: error in implementation!
neon: error in implementation!
armv6: error in implementation!
neon: error in implementation!
armv6: error in implementation!
neon: error in implementation!
armv6: error in implementation!
HChaCha

---- 8 rounds ----
neon: error in implementation!
armv6: error in implementation!

---- 12 rounds ----
neon: error in implementation!
armv6: error in implementation!

---- 20 rounds ----
neon: error in implementation!
armv6: error in implementation!

Info from /proc/cpuinfo is:

$ cat /proc/cpuinfo
processor : 0
model name : ARMv7 Processor rev 4 (v7l)
BogoMIPS : 38.40
Features : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm crc32
CPU implementer : 0x41
CPU architecture: 7
CPU variant : 0x0
CPU part : 0xd03
CPU revision : 4

processor : 1
model name : ARMv7 Processor rev 4 (v7l)
BogoMIPS : 38.40
Features : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm crc32
CPU implementer : 0x41
CPU architecture: 7
CPU variant : 0x0
CPU part : 0xd03
CPU revision : 4

processor : 2
model name : ARMv7 Processor rev 4 (v7l)
BogoMIPS : 38.40
Features : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm crc32
CPU implementer : 0x41
CPU architecture: 7
CPU variant : 0x0
CPU part : 0xd03
CPU revision : 4

processor : 3
model name : ARMv7 Processor rev 4 (v7l)
BogoMIPS : 38.40
Features : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm crc32
CPU implementer : 0x41
CPU architecture: 7
CPU variant : 0x0
CPU part : 0xd03
CPU revision : 4

Hardware : BCM2709
Revision : a02082
Serial : 00000000bedaaca5