B2 configuration check information can be confusing?

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

B2 configuration check information can be confusing?

Boost - Build mailing list
B2 configuration check information can be confusing?

I got this configuration info below from a b2 toolset=msvc,gcc command

I was puzzled to see contradictory information

    - __float128               : no

But later

    - __float128               : yes

After some confusion, the penny eventually dropped and I concluded that info was first for msvc and then second for gcc, (or was it address-model=32 and 64)?

Could this be easily made clearer?

Even a newline might give clues, but showing the toolset would be even better?

Thanks

Paul


Paul A. Bristow

Prizet Farmhouse

Kendal, Cumbria

LA8 8AB           UK

Performing configuration checks

    - default address-model    : 32-bit

    - default architecture     : x86

    - symlinks supported       : yes

    - has_icu builds           : no

    - BOOST_COMP_GNUC >= 4.3.0 : no

    - Boost.Config Feature Check: cxx11_auto_declarations : yes

    - Boost.Config Feature Check: cxx11_lambdas : yes

    - Boost.Config Feature Check: cxx11_unified_initialization_syntax : yes

    - Boost.Config Feature Check: cxx11_smart_ptr : yes

    - GCC libquadmath and __float128 support : no

    - Boost.Config Feature Check: cxx11_hdr_initializer_list : yes

    - Boost.Config Feature Check: sfinae_expr : yes

    - Boost.Config Feature Check: cxx11_range_based_for : yes

    - libfftw3                 : no

    - __float128               : no

    - Boost.Config Feature Check: cxx11_decltype : yes

    - Boost.Config Feature Check: cxx11_variadic_templates : yes

    - has_ntl_rr builds        : no

    - has_mpfr_class builds    : no

    - has_mpreal builds        : no

    - has_e_float builds       : no

    - gcc visibility           : no

    - long double support      : yes

    - Boost.Config Feature Check: cxx11_hdr_tuple : yes

    - Boost.Config Feature Check: cxx11_defaulted_functions : yes

    - has_eigen builds         : no

    - Boost.Config Feature Check: cxx11_constexpr : yes

    - Boost.Config Feature Check: cxx11_hdr_array : yes

    - Boost.Config Feature Check: cxx17_if_constexpr : no

    - Boost.Config Feature Check: cxx17_std_apply : no

    - Intel _Quad datatype support : no

    - Boost.Config Feature Check: cxx11_hdr_thread : yes

    - Boost.Config Feature Check: cxx11_hdr_atomic : yes

    - Boost.Config Feature Check: cxx11_hdr_future : yes

    - Boost.Config Feature Check: cxx11_hdr_chrono : yes

    - Boost.Config Feature Check: cxx11_hdr_random : yes

    - Boost.Config Feature Check: cxx11_allocator : yes

    - Boost.Config Feature Check: cxx11_inline_namespaces : yes

    - has_mpfr builds          : no

    - 128-bit floatmax_t       : no

    - Boost.Config Feature Check: cxx11_numeric_limits : yes

    - Boost.Config Feature Check: cxx11_explicit_conversion_operators : yes

    - Boost.Config Feature Check: cxx11_function_template_default_args : yes

    - Boost.Config Feature Check: cxx11_hdr_mutex : yes

    - default address-model    : 64-bit

    - default architecture     : x86

    - has_icu builds           : no

    - BOOST_COMP_GNUC >= 4.3.0 : yes

    - Boost.Config Feature Check: cxx11_auto_declarations : yes

    - Boost.Config Feature Check: cxx11_lambdas : yes

    - Boost.Config Feature Check: cxx11_unified_initialization_syntax : yes

    - Boost.Config Feature Check: cxx11_smart_ptr : yes

    - GCC libquadmath and __float128 support : yes

    - Boost.Config Feature Check: cxx11_hdr_initializer_list : yes

    - Boost.Config Feature Check: sfinae_expr : yes

    - Boost.Config Feature Check: cxx11_range_based_for : yes

    - libfftw3                 : no

    - __float128               : yes

    - Boost.Config Feature Check: cxx11_decltype : yes

    - Boost.Config Feature Check: cxx11_variadic_templates : yes

    - has_ntl_rr builds        : no

    - has_mpfr_class builds    : no

    - has_mpreal builds        : no

    - has_e_float builds       : no

    - gcc visibility           : yes

    - long double support      : yes

    - Boost.Config Feature Check: cxx11_hdr_tuple : yes

    - Boost.Config Feature Check: cxx11_defaulted_functions : yes

    - has_eigen builds         : no

    - Boost.Config Feature Check: cxx11_constexpr : yes

    - Boost.Config Feature Check: cxx11_hdr_array : yes

    - Boost.Config Feature Check: cxx17_if_constexpr : no

    - Boost.Config Feature Check: cxx17_std_apply : no

    - Intel _Quad datatype support : no

    - Boost.Config Feature Check: cxx11_hdr_thread : no

    - Boost.Config Feature Check: cxx11_hdr_atomic : yes

    - Boost.Config Feature Check: cxx11_hdr_future : yes

    - Boost.Config Feature Check: cxx11_hdr_chrono : yes

    - Boost.Config Feature Check: cxx11_hdr_random : yes

    - Boost.Config Feature Check: cxx11_allocator : yes

    - Boost.Config Feature Check: cxx11_inline_namespaces : yes

    - has_mpfr builds          : no

    - 128-bit floatmax_t       : yes

    - Boost.Config Feature Check: cxx11_numeric_limits : yes

    - Boost.Config Feature Check: cxx11_explicit_conversion_operators : yes

    - Boost.Config Feature Check: cxx11_function_template_default_args : yes

    - Boost.Config Feature Check: cxx11_hdr_mutex : no

...patience...


_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: B2 configuration check information can be confusing?

Boost - Build mailing list
AMDG

On 8/23/19 5:05 AM, Paul A Bristow via Boost-build wrote:

> I got this configuration info below from a b2 toolset=msvc,gcc command
>
> I was puzzled to see contradictory information
>
>     - __float128               : no
>
> But later
>
>     - __float128               : yes
>
> After some confusion, the penny eventually dropped and I concluded that info was
> first for msvc and then second for gcc, (or was it address-model=32 and 64)?
>
> Could this be easily made clearer?
>
> Even a newline might give clues, but showing the toolset would be even better?
>

I agree that it can be confusing, but I'm concerned that adding
more output by default would turn it into even more of an
info dump than it already is.

In Christ,
Steven Watanabe
_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build