On Mon, Mar 19, 2018 at 10:32 AM, Thomas Munro
<thomas.munro@enterprisedb.com> wrote:
> On Sun, Mar 18, 2018 at 7:33 AM, Andres Freund <andres@anarazel.de> wrote:
>> On 2018-03-17 14:20:26 -0400, Tom Lane wrote:
>>> It might be worth studying the icc manual to see if it has an
>>> equivalent of -fwrapv.
>>
>> Yes.
>>
>> A *quick* look through https://software.intel.com/en-us/node/522795
>> unfortunately didn't show anything.
>
> Apparently it does support -fno-strict-overflow. Is that useful here?
Hmm. This was already discussed here:
https://www.postgresql.org/message-id/51016409.2020808%40gmail.com
Noah seemed to be at least slightly in favour of considering turning
it on despite doubt about its precise meaning, but also pointed out
that even -fwrapv doesn't mean exactly the same thing in GCC and
Clang.
Curiously -fno-strict-overflow doesn't seem to appear in the
documentation that Andres posted (well I couldn't find it quickly,
anyway), but we can see that it exists from this interaction between
Xi Wang and Intel compiler engineering:
https://software.intel.com/en-us/forums/intel-c-compiler/topic/358200
And we can see a reference here:
https://www.spec.org/cpu2017/flags/Intel-ic18.0-official-linux64.html
--
Thomas Munro
http://www.enterprisedb.com