Re: [Boost-bugs] [Boost C++ Libraries] #10925: fialed to build math library with studo 12.4 c++ on Solaris 11.2

Subject: Re: [Boost-bugs] [Boost C++ Libraries] #10925: fialed to build math library with studo 12.4 c++ on Solaris 11.2
From: Boost C++ Libraries (noreply_at_[hidden])
Date: 2015-04-02 17:10:33

#10925: fialed to build math library with studo 12.4 c++ on Solaris 11.2
-------------------------------+-------------------------
  Reporter:  angela.xie@…      |      Owner:  johnmaddock
      Type:  Bugs              |     Status:  new
 Milestone:  To Be Determined  |  Component:  math
   Version:  Boost 1.56.0      |   Severity:  Showstopper
Resolution:                    |   Keywords:
-------------------------------+-------------------------

Comment (by johnmaddock):

 I think the fundamental issue with flushing-to-zero is that this is
 potentially a runtime setting (certainly it is if you're using the x87
 FPU, not so sure about the SSE instructions).  So there's no way that the
 compile time constant numeric_limits::has_denorm can accurately reflect
 reality.   IEC559 would give the definitive answer as to whether flush-to-
 zero is acceptible or not.

 BTW the issue here is when flush-to-zero ''may'' occur, but isn't
 absolutely guarenteed to ''always'' occur.  That does genuinely make it
 hard to reason about some floating-point logic, though I agree it's a
 reasonable optimization.

-- 
Ticket URL: <https://svn.boost.org/trac/boost/ticket/10925#comment:22>
Boost C++ Libraries <http://www.boost.org/>
Boost provides free peer-reviewed portable C++ source libraries.