RE: [PATCH v2, middle end]: Fix PR 37908, thinko with atomic NAND operation

["Lu, Hongjiu" <hongjiu dot lu at intel dot com>]

>
>Richard Guenther wrote:
>> On Wed, Oct 29, 2008 at 8:34 PM, David Daney <ddaney@caviumnetworks.com>
>wrote:
>
>
>>> That GCC behavior diverges from a 'published standard' is lamentable,
>you
>>> think we should change GCC to match the standard immediately, I wonder
>if
>>> there are better options.
>>
>> Another option is to remove support for this function entirely and
>> re-introduce it
>> using a different name.
>
>That's a much better idea than silently changing it at the last moment.
>

Maybe we should disable it with a warning by default and

1. Enable the old semantics with a switch, like -fbroken-nand.
2. Enable the correct semantics with a switch, like -fnand.

Both -fbroken-and and -fnand should be off by default.

H.J.