diff mbox

libstdc++/60637 Fix C++98 std::signbit<long double>

Message ID 20160118162856.GY15084@redhat.com
State New
Headers show

Commit Message

Jonathan Wakely Jan. 18, 2016, 4:28 p.m. UTC
This fixes PR60637 by using the appropriate built-in for the size of
the argument type. In Bugzilla Marc asked why we don't just use the
same code as for C++11, but I want to make this less intrusive change
on the branches (trunk is already OK anyway).

The new test is only run on x86 because it fails with a -Woverflow
warning on (at least) powerpc.

Tested x86_64-linux and powerpc64le-linux. Committed to gcc-5-branch
and gcc-4_9-branch.

Comments

Marc Glisse Jan. 18, 2016, 8:24 p.m. UTC | #1
On Mon, 18 Jan 2016, Jonathan Wakely wrote:

> This fixes PR60637 by using the appropriate built-in for the size of
> the argument type. In Bugzilla Marc asked why we don't just use the
> same code as for C++11, but I want to make this less intrusive change
> on the branches (trunk is already OK anyway).

That's fine for the branch indeed, but I don't see why we couldn't remove 
that illegal signbit declaration on trunk in C++98 mode. It works but it 
is still wrong to have it. Now that C++14 is the default, explicitly 
asking for C++98 looks like a clear sign to me that people don't want such 
extensions.
Jonathan Wakely Jan. 18, 2016, 9 p.m. UTC | #2
On 18/01/16 21:24 +0100, Marc Glisse wrote:
>On Mon, 18 Jan 2016, Jonathan Wakely wrote:
>
>>This fixes PR60637 by using the appropriate built-in for the size of
>>the argument type. In Bugzilla Marc asked why we don't just use the
>>same code as for C++11, but I want to make this less intrusive change
>>on the branches (trunk is already OK anyway).
>
>That's fine for the branch indeed, but I don't see why we couldn't 
>remove that illegal signbit declaration on trunk in C++98 mode. It 
>works but it is still wrong to have it. Now that C++14 is the default, 
>explicitly asking for C++98 looks like a clear sign to me that people 
>don't want such extensions.

For -std=c++98 rather than -std=gnu++98 I agree, but I'm not going to
make that change in stage 3, let's do it in stage 1.

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69350
diff mbox

Patch

commit e30df471f94c383d742bf4506a794b6505526f77
Author: Jonathan Wakely <jwakely@redhat.com>
Date:   Mon Jan 18 14:39:57 2016 +0000

    Fix C++98 std::signbit<long double>
    
    	PR libstdc++/60637
    	* include/c_global/cmath (signbit) [__cplusplus < 201103L]: Use
    	__builtin_signbitf or __builtin_signbitl as appropriate.
    	* testsuite/26_numerics/headers/cmath/60637.cc: New.

diff --git a/libstdc++-v3/include/c_global/cmath b/libstdc++-v3/include/c_global/cmath
index 4cafe5f..d3fc8b7 100644
--- a/libstdc++-v3/include/c_global/cmath
+++ b/libstdc++-v3/include/c_global/cmath
@@ -880,7 +880,11 @@  _GLIBCXX_BEGIN_NAMESPACE_VERSION
     signbit(_Tp __f)
     {
       typedef typename __gnu_cxx::__promote<_Tp>::__type __type;
-      return __builtin_signbit(__type(__f));
+      return sizeof(__type) == sizeof(float)
+	? __builtin_signbitf(__type(__f))
+	: sizeof(__type) == sizeof(double)
+	? __builtin_signbit(__type(__f))
+	: __builtin_signbitl(__type(__f));
     }
 
   template<typename _Tp>
diff --git a/libstdc++-v3/testsuite/26_numerics/headers/cmath/60637.cc b/libstdc++-v3/testsuite/26_numerics/headers/cmath/60637.cc
new file mode 100644
index 0000000..16a7896
--- /dev/null
+++ b/libstdc++-v3/testsuite/26_numerics/headers/cmath/60637.cc
@@ -0,0 +1,35 @@ 
+// Copyright (C) 2016 Free Software Foundation, Inc.
+//
+// This file is part of the GNU ISO C++ Library.  This library is free
+// software; you can redistribute it and/or modify it under the
+// terms of the GNU General Public License as published by the
+// Free Software Foundation; either version 3, or (at your option)
+// any later version.
+
+// This library is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License for more details.
+
+// You should have received a copy of the GNU General Public License along
+// with this library; see the file COPYING3.  If not see
+// <http://www.gnu.org/licenses/>.
+
+// { dg-options "-std=gnu++98 -ffast-math" }
+// { dg-do run { target i?86-*-* x86_64-*-* } }
+
+#include <cmath>
+#include <testsuite_hooks.h>
+
+void
+test01()
+{
+  long double ld = -5.3165867831218916301793863361917824e-2467L;
+  VERIFY( std::signbit(ld) == 1 );
+}
+
+int
+main()
+{
+  test01();
+}