Re: lib/49990: libc++ fails to build with g++ 5.1 due to the following types of errors:

[Martin Husemann <martin%duskware.de@localhost>]

The following reply was made to PR lib/49990; it has been noted by GNATS.

From: Martin Husemann <martin%duskware.de@localhost>
To: Antti Kantee <pooka%iki.fi@localhost>
Cc: gnats-bugs%NetBSD.org@localhost
Subject: Re: lib/49990: libc++ fails to build with g++ 5.1 due to the following types of errors:
Date: Tue, 23 Jun 2015 14:50:19 +0200

 On Tue, Jun 23, 2015 at 12:38:58PM +0000, Antti Kantee wrote:
 > I seriously wonder what you expected, was wrong with looking at the 
 > build output, and especially with "If there are some specific build 
 > options you're interested in, I can try to dig them out".
 
 The line numbers reported for the errors do not make any sense to me,
 if I look at those lines in exception.cc.
 
 I find this forward declaration:
 
     676 #if __GNUC__ > 3 && __GNUC_MINOR__ > 2 
     677 extern "C" void *__cxa_begin_catch(void *e) throw();
     678 #else
     679 extern "C" void *__cxa_begin_catch(void *e);
     680 #endif  
 
 and this definition:
 
    1192 #if __GNUC__ > 3 && __GNUC_MINOR__ > 2
    1193 extern "C" void *__cxa_begin_catch(void *e) throw()
    1194 #else           
    1195 extern "C" void *__cxa_begin_catch(void *e)
    1196 #endif          
    1197 {       
 
 which seem to be consistent to me. I also don't see any changes to the
 predefined __GNUC__ or __GNUC_MINOR__ macros.
 
 So looking at this code (which is the only I have) and the error messages
 given, something is strange, but I can not spot any error in our tree.
 
 It may be a bug in gcc, or some host system header interfering - or whatever.
 No chance of tracking this down remotely, sorry.
 
 Martin