[wwwdocs] Note for pr70024 for gcc-5

[Richard Henderson]

Is there anything else we should say?

I thought about recommending that distributions not install the libffi shared 
library from gcc and instead use upstream source.  But that doesn't really help 
one way or the other, so I dropped that language.


r~
-------------- next part --------------
Index: changes.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-5/changes.html,v
retrieving revision 1.137
diff -u -r1.137 changes.html
--- changes.html	20 Jan 2016 17:42:34 -0000	1.137
+++ changes.html	3 Mar 2016 02:04:38 -0000
@@ -674,6 +674,10 @@
        have been removed. The AArch64 backend now uses the local register
        allocator (LRA) only.
      </li>
+     <li>The shared library generated for libffi has an accidental ABI
+       change in the size and layout of <code>ffi_closure</code>,
+       <code>ffi_raw_closure</code>, and <code>ffi_java_raw_closure</code>.
+       It is now too late to correct this for GCC 5.
    </ul>
 
 <h3 id="arm">ARM</h3>