Boost logo

Boost Users :

From: Edward Diener (eldiener_at_[hidden])
Date: 2021-08-23 12:00:24

On 8/23/2021 4:23 AM, John Emmas via Boost-users wrote:
> On 18/08/2021 14:11, Edward Diener via Boost-users wrote:
>> On 8/18/2021 5:03 AM, John Emmas via Boost-users wrote:
>>> Safari seemed a bit happier so I installed ver 1.74 and eventually
>>> ver 1.77 - but they both give me the same error list :-(
>>> So I'm guessing this might be a Visual Studio issue.  Edward - do you
>>> happen to know which version of VS2019 you're running?
>> I am running VS2019 16.10.3.
> Hi Edward, there's been a development here after I flagged this up to
> Microsoft....
> I sent them a small test project and it seems that for VS2019 16.11.0
> the problem only happens for an x86 build (if I build as x64 it works
> okay).  Going back to the command line you sent me last week, is there
> some way I can adapt it to force an x86 build and alternatively, an x64
> build?  I could then swap between them and see if there's any difference
> here.  And it looks like John Maddock might be onto something when he
> suggested modifying 'boost/integer/common_factor_rt.hpp'.  I'd be happy
> to try modifying it temporarily if John can suggest something...  Thanks,
> John

It is a matter of the command prompt. The vcvars32.bat command prompt is
for an x86 build and the vcvars64.bat is for an x64 build. You are
correct that I tried it for an x64 build. However trying the code for an
x86 build was also successful for me.

> On 16/08/2021 18:57, John Maddock via Boost-users wrote:
>> #if ((defined(BOOST_MSVC) && (BOOST_MSVC >= 1600)) ||
>> (defined(__clang__) && defined(__c2__)) || (defined(BOOST_INTEL) &&
>> defined(_MSC_VER))) && (defined(_M_IX86) || defined(_M_X64))
>> #include <intrin.h>
>> #endif
>> I'm wondering if the check for x86 needs to be applied to the
>> BOOST_MSVC check as well as the intel one?

Boost-users list run by williamkempf at, kalb at, bjorn.karlsson at, gregod at, wekempf at