[cfe-dev] [PROPOSAL] Reintroduce guards for Intel intrinsic headers

[David Blaikie]

On Thu, Jul 30, 2015 at 11:04 AM, Reid Kleckner <rnk at google.com> wrote:

> On Thu, Jul 30, 2015 at 10:33 AM, Eric Christopher <echristo at gmail.com>
> wrote:
>>
>> What kind of file is this? Keep in mind that things in the global
>> namespace prefixed with an underscore is a reserved name for implementers
>> as well. That would make this code not standards compliant as well.
>>
>
> The actual C++ rules are that any name with double underscores is
> reserved, and names beginning with an underscore followed by a capital
> letter. So, the Intel intrinsics are *not* in the implementer's namespace,
> but that was probably a mistake.
>

There's a second clause in 17.6.4.3.2\1:

"Each name that begins with an underscore is reserved to the implementation
for use as a name in the global namespace" (the other names you mentioned
are reserved for any use - so they can be used for macros, etc too - these
one's cannot)


>
> I'm sympathetic to users who are probably implementing a compatibility
> layer here and don't want to write their own intrinsic wrappers, but I
> think the right tradeoff is probably to fix the code.
>
> _______________________________________________
> cfe-dev mailing list
> cfe-dev at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20150730/842a1e80/attachment.html>