Re: [v3] Handle C++11 <math.h> overloads on Solaris 12

[Jonathan Wakely <jwakely@redhat.com>]

On 24/11/15 13:32 +0100, Rainer Orth wrote:
>Jonathan Wakely <jwakely@redhat.com> writes:
>
>> On 17/11/15 13:33 +0100, Rainer Orth wrote:
>>>Solaris 12 recently introduced the C++11 <math.h> overloads, which
>>>caused bootstrap to be broken on both mainline and the gcc-5 branch:
>>>
>>>In file included from /vol/gcc/src/hg/trunk/local/libstdc++-v3/include/precompiled/stdc++.h:41:0:
>>>/var/gcc/regression/trunk/12-gcc/build/i386-pc-solaris2.12/libstdc++-v3/include/cmath: In function 'constexpr int std::fpclassify(float)':
>>>/var/gcc/regression/trunk/12-gcc/build/i386-pc-solaris2.12/libstdc++-v3/include/cmath:561:3: error: redefinition of 'constexpr int std::fpclassify(float)'
>>>   fpclassify(float __x)
>>>   ^
>>>In file included from /usr/include/math.h:13:0,
>>>                 from /var/gcc/regression/trunk/12-gcc/build/i386-pc-solaris2.12/libstdc++-v3/include/cmath:44,
>>>                 from /vol/gcc/src/hg/trunk/local/libstdc++-v3/include/precompiled/stdc++.h:41:
>>>/usr/include/iso/math_c99.h:647:13: note: 'int std::fpclassify(float)' previously defined here
>>>  inline int fpclassify(float __X) { return __builtin_fpclassify(
>>>             ^
>>>
>>>The following patch fixes this by testing for the problem and wrapping
>>>the overloads in include/c_global/cmath and include/tr1/cmath
>>>appropriately.  The test needs to be dynamic since apparently a backport
>>>to Solaris 11 (and perhaps even Solaris 10) is planned.
>>
>> Why does <tr1/cmath> need to change, did Solaris also define the
>> functions in namespace std::tr1?
>
>Without that part, I get errors like
>
>In file included from /vol/gcc/src/hg/trunk/local/libstdc++-v3/include/precompiled/stdtr1c++.h:37:0:
>/var/gcc/regression/trunk/12-gcc/build/i386-pc-solaris2.12/libstdc++-v3/include/tr1/cmath: In function 'float std::tr1::acosh(float)':
>/var/gcc/regression/trunk/12-gcc/build/i386-pc-solaris2.12/libstdc++-v3/include/tr1/cmath:423:18: error: 'float std::tr1::acosh(float)' conflicts with a previous declaration
>   acosh(float __x)

Ah yes, because there's a using std::acosh, which puts Solaris' new
overloads into namespace std::tr1.

OK, that's correct then, and the patch is OK for trunk and
gcc-5-branch as far as I'm concerned.