Collected fixes for PPC64LE

[jzmaddock]

This PR get's things almost green when running the tests on PPC64LE.

I have one remaining failure from autodiff which I don't understand: https://github.com/jzmaddock/math/runs/1856451176?check_suite_focus=true

Other than that most of the changes involve fixing the tests to not test long double when BOOST_MATH_NO_LONG_DOUBLE_FUNCTIONS is defined.

Test suites covered: special_fun, distribution_tests, misc, quadrature, interpolators, autodiff

Not currently tested/covered: mp, float128_tests, long-running-tests, examples, tools. Mostly due to time and space constraints as these are run on the QEMU interpolator. Adding these would be nice to have, but the core of the library is now functional and tested.

This is a partial fix for #510.

Things discovered about this platform:

• long double is the non-contiguous "double double" type which we have never supported as it's basically impossible to reason about. It's probably possible to get quite a bit working with enough work, but the type continues to surprise at every turn as it's very non IEEE conforming.
• long double is not usable in constexpr (even though the software emulated __float128 is).
• Mixing long double and __float128 in the same expression is a big no no that results in compiler errors.

For the benefit of future googlers, here's the math_info output:

Macros from <math.h>
    HUGE_VAL                                =inf
    HUGE_VALF                               =inf
    HUGE_VALL                               =inf
    INFINITY                                =inf
    NAN                                     =(__builtin_nanf (""))
    FP_INFINITE                             =1
    FP_NAN                                  =0
    FP_NORMAL                               =4
    FP_SUBNORMAL                            =3
    FP_ZERO                                 =2
    FP_FAST_FMA                             =1
    FP_FAST_FMAF                            =1
    FP_ILOGB0                               =(-2147483647)
    FP_ILOGBNAN                             =2147483647
    MATH_ERRNO                              =1
    MATH_ERREXCEPT                          =2
    FLT_MIN_10_EXP                          =-37
    FLT_DIG                                 =6
    FLT_MIN_EXP                             =-125
    FLT_EPSILON                             =1.1920929e-07
    FLT_RADIX                               =2
    FLT_MANT_DIG                            =24
    FLT_ROUNDS                              =1
    FLT_MAX                                 =3.4028235e+38
    FLT_MAX_10_EXP                          =38
    FLT_MAX_EXP                             =128
    FLT_MIN                                 =1.1754944e-38
    DBL_DIG                                 =15
    DBL_MIN_EXP                             =-1021
    DBL_EPSILON                             =2.2204460492503131e-16
    DBL_MANT_DIG                            =53
    DBL_MAX                                 =1.7976931348623157e+308
    DBL_MIN                                 =2.2250738585072014e-308
    DBL_MAX_10_EXP                          =308
    DBL_MAX_EXP                             =1024
    DBL_MIN_10_EXP                          =-307
    LDBL_MAX_10_EXP                         =308
    LDBL_MAX_EXP                            =1024
    LDBL_MIN                                =2.00416836000897277799610805135016e-292
    LDBL_MIN_10_EXP                         =-291
    LDBL_DIG                                =31
    LDBL_MIN_EXP                            =-968
    LDBL_EPSILON                            =4.94065645841246544176568792868221e-324
    LDBL_MANT_DIG                           =106
    LDBL_MAX                                =1.79769313486231580793728971405301e+308

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
std::numeric_limits information for type float
    is_specialized       = 1
    min()                = 1.1754944e-38
    max()                = 3.4028235e+38
    digits               = 24
    digits10             = 6
    is_signed            = 1
    is_integer           = 0
    is_exact             = 0
    radix                = 2
    epsilon()            = 1.1920929e-07
    round_error()        = 0.5
    min_exponent         = -125
    min_exponent10       = -37
    max_exponent         = 128
    max_exponent10       = 38
    has_infinity         = 1
    has_quiet_NaN        = 1
    has_signaling_NaN    = 1
    has_denorm           = 1
    has_denorm_loss      = 0
    infinity()           = inf
    quiet_NaN()          = nan
    signaling_NaN()      = nan
    denorm_min()         = 1.4012985e-45
    is_iec559            = 1
    is_bounded           = 1
    is_modulo            = 0
    traps                = 0
    tinyness_before      = 0
    round_style          = 1

Epsilon has sane value of std::pow(std::numeric_limits<T>::radix, 1-std::numeric_limits<T>::digits).

    sizeof(float) = 4
    alignment_of<float> = 4

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
std::numeric_limits information for type double
    is_specialized       = 1
    min()                = 2.2250738585072014e-308
    max()                = 1.7976931348623157e+308
    digits               = 53
    digits10             = 15
    is_signed            = 1
    is_integer           = 0
    is_exact             = 0
    radix                = 2
    epsilon()            = 2.2204460492503131e-16
    round_error()        = 0.5
    min_exponent         = -1021
    min_exponent10       = -307
    max_exponent         = 1024
    max_exponent10       = 308
    has_infinity         = 1
    has_quiet_NaN        = 1
    has_signaling_NaN    = 1
    has_denorm           = 1
    has_denorm_loss      = 0
    infinity()           = inf
    quiet_NaN()          = nan
    signaling_NaN()      = nan
    denorm_min()         = 4.9406564584124654e-324
    is_iec559            = 1
    is_bounded           = 1
    is_modulo            = 0
    traps                = 0
    tinyness_before      = 0
    round_style          = 1

Epsilon has sane value of std::pow(std::numeric_limits<T>::radix, 1-std::numeric_limits<T>::digits).

    sizeof(double) = 8
    alignment_of<double> = 8

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
std::numeric_limits information for type long double
    is_specialized       = 1
    min()                = 2.00416836000897277799610805135016e-292
    max()                = 1.79769313486231580793728971405301e+308
    digits               = 106
    digits10             = 31
    is_signed            = 1
    is_integer           = 0
    is_exact             = 0
    radix                = 2
    epsilon()            = 4.94065645841246544176568792868221e-324
    round_error()        = 0.5
    min_exponent         = -968
    min_exponent10       = -291
    max_exponent         = 1024
    max_exponent10       = 308
    has_infinity         = 1
    has_quiet_NaN        = 1
    has_signaling_NaN    = 1
    has_denorm           = 1
    has_denorm_loss      = 0
    infinity()           = inf
    quiet_NaN()          = nan
    signaling_NaN()      = nan
    denorm_min()         = 4.94065645841246544176568792868221e-324
    is_iec559            = 1
    is_bounded           = 1
    is_modulo            = 0
    traps                = 0
    tinyness_before      = 0
    round_style          = 1

CAUTION: epsilon does not have a sane value.

    sizeof(long double) = 16
    alignment_of<long double> = 16

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Math function overload information for type float
The Math functions are overloaded for type float
std::fabs looks OK for type float
std::abs looks OK for type float
std::sqrt looks OK for type float
std::atan2 looks OK for type float
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Math function overload information for type double
The Math functions are overloaded for type double
std::fabs looks OK for type double
std::abs looks OK for type double
std::sqrt looks OK for type double
std::atan2 looks OK for type double
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Math function overload information for type long double
The Math functions are overloaded for type long double
std::fabs looks OK for type long double
std::abs looks OK for type long double
std::sqrt looks OK for type long double
std::atan2 looks OK for type long double

[jwakely]

It seems that for boost 1.76.0 libboost_math_tr1l.so.1.75.0 and libboost_math_c99l.so.1.75.0 are no longer built on ppc64le, which breaks the Fedora build. Is that expected? Was it caused by this change?

[jzmaddock]

It seems that for boost 1.76.0 libboost_math_tr1l.so.1.75.0 and libboost_math_c99l.so.1.75.0 are no longer built on ppc64le, which breaks the Fedora build. Is that expected? Was it caused by this change?

No, there are no changes to the build system here, and none anywhere I can think of that would cause this.

If you cd into libs/math/build and run ../../../b2 what output do you see?

[rodgert]

For now I'm going to bypass expecting them to be present in ppc64le builds of boost. It sounds like it's dubious whether they are even correct on this target. After I finish with the Boost 1.76.0 upgrade for Fedora 35 I'll try to dig deeper into why they aren't being built.