[openssl-dev] arch (ARM) capabilities

[Andy Polyakov]

>>> But when I've tested it on AArch64 with openssl-1.1.0-pre5 and
>>> current master (./configure no-shared no-engine) I'm getting
>>> 100524.03k vs 52172.12k/s in favour of the non-EVP version.
>>>
>>> Is that really expected?
>>
>> Depends on your system. Not all AArch64 processors were born equal,
>> some have crypto extensions, some don't, some have mighty pipelines,
>> some don't. The presented numbers suggest that you ended up on APM
>> X-Gene processor which doesn't.
> 
> yes, it is an X-Gene, though I don't know which exact one
> 
> The product briefs I could find, say that X-Genes do support cryptograpic
> extensions:
> https://myapm.apm.com/technical_documents/download/apm883208-product-brief1

I says "APM883208-X1 *can* deliver advanced security capabilities with
the *optional* security engine." Mention of "engine" and then algorithm
list suggests that it's kind of embedded peripheral, which would appear
to user-land as device. While what we are talking about here is AES-NI
equivalent for ARMv8, i.e. ARM *instruction* set extension directly
accessible to user-land. Availability of this extension is normally
reflected in Features: in /proc/cpuinfo.