[v3] Fix ext/random/hypergeometric_distribution/operators/values.cc on Solaris 9

[Rainer Orth]

Hi Paolo,

> On 11/27/2013 03:14 PM, Rainer Orth wrote:
>> hypergeometric_pdf is defined in testsuite/util/testsuite_random.h
>> inside _GLIBCXX_USE_C99_MATH_TR1, but used unconditionally.  Fixed which
>> the following patch which allows the test to compile and pass on
>> i386-pc-solaris2.9.  On i386-pc-solaris2.11, it FAILs with an Arithmetic
>> exception, as it did before the patch:
>>
>> Program received signal SIGFPE, Arithmetic exception.
>> [Switching to Thread 1 (LWP 1)]
>> 0x080542ba in std::generate_canonical<unsigned int, 32u,
>> std::mersenne_twister_engine<unsigned int, 32u, 624u, 397u, 31u,
>> 2567483615u, 11u, 4294967295u, 7u, 2636928640u, 15u, 4022730752u, 18u,
>> 1812433253u> > (__urng=...)
>>      at
>> /var/gcc/regression/trunk/11-gcc/build/i386-pc-solaris2.11/libstdc++-v3/include/bits/random.tcc:3480
>> 3480          return __sum / __tmp;
>>
>> __tmp is 0 here.
>>
>> Ok for mainline?
> Please use // { dg-require-cmath "" } instead.

make sense since this renders the test unsupported instead of pass.
Since I'd already committed the previous version based on Jonathan's
approval, I've installed the following on top of it after testing with
the appropriate runtest command.

	Rainer


2013-11-27  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>

	* testsuite/ext/random/hypergeometric_distribution/operators/values.cc:
	Use dg-require-cmath instead.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: v3.patch
Type: text/x-patch
Size: 826 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/libstdc++/attachments/20131127/b63432ff/attachment.bin>
-------------- next part --------------


-- 
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University