[RFC, VECTOR ABI] Allow __attribute__((vector)) in GCC by default.

[Jeff Law]

On 10/07/2015 11:22 AM, Joseph Myers wrote:
> On Wed, 7 Oct 2015, Jeff Law wrote:
>
>> I'm not sure why this attribute isn't documented, but clearly that should be
>> fixed.
>
> I assume the reasoning was: we document support for Cilk+ (and what's
> included in Cilk+ is externally documented).  When it becomes a generic
> GNU C attribute, that reasoning no longer applies.
Seems reasonable.  I had a suspicion it was something of this nature.


>
>> I thought the ABI for this stuff was consistent across the implementations
>> (that was certainly the goal).  So aside from an example of how to use the
>> attribute to get calls into the vector math library, I'm not sure what's
>> needed.  Essentially the attribute is just another way to ensure we exploit
>> the vector library when possible.
>
> Well, the ABI should at least say that the attribute has the same
> implications as the pragma regarding what vector function versions are
> available.  (So that a future compiler building with -mavx1024 won't
> interpret old glibc headers as meaning that AVX1024 function versions are
> available in old glibc - that will instead require a further attribute /
> pragma to declare availability of those versions.)
Also seems reasonable.  Thanks for clarifying the longer term 
implications you're concerned about.

jeff