Re: pkg/45278: bootstrap fails in libfetch NetBSD -current (gcc4.5)

[christos%zoulas.com@localhost (Christos Zoulas)]

The following reply was made to PR pkg/45278; it has been noted by GNATS.

From: christos%zoulas.com@localhost (Christos Zoulas)
To: gnats-bugs%NetBSD.org@localhost, pkg-manager%netbsd.org@localhost, 
gnats-admin%netbsd.org@localhost, 
        pkgsrc-bugs%netbsd.org@localhost, kuehro%gmx.de@localhost
Cc: 
Subject: Re: pkg/45278: bootstrap fails in libfetch NetBSD -current (gcc4.5)
Date: Sun, 21 Aug 2011 11:47:59 -0400

 On Aug 21,  2:40pm, joerg%britannica.bec.de@localhost (Joerg Sonnenberger) 
wrote:
 -- Subject: Re: pkg/45278: bootstrap fails in libfetch NetBSD -current (gcc4.
 
 | The following reply was made to PR pkg/45278; it has been noted by GNATS.
 | 
 | From: Joerg Sonnenberger <joerg%britannica.bec.de@localhost>
 | To: gnats-bugs%NetBSD.org@localhost
 | Cc: pkg-manager%netbsd.org@localhost, gnats-admin%netbsd.org@localhost, 
pkgsrc-bugs%netbsd.org@localhost,
 |      kuehro%gmx.de@localhost
 | Subject: Re: pkg/45278: bootstrap fails in libfetch NetBSD -current (gcc4.5)
 | Date: Sun, 21 Aug 2011 16:38:00 +0200
 | 
 |  On Sun, Aug 21, 2011 at 08:30:05AM +0000, David Holland wrote:
 |  >  > >  half of the network code. So let us try with -fno-strict-aliasing 
on 
 |  >  > >  ftp.c then. Other suggestions besides turning to another compiler ? 
;-)
 |  >  > 
 |  >  > *sigh* What I am telling you is that this is not about strict aliasing.
 |  >  
 |  >  Ok, but our default compiler thinks so and fails on an important part
 |  >  of pkgsrc infrastructure. But this works on NetBSD current i386:
 |  
 |  Did you read the quoted error message? It is the broken array boundary 
check,
 |  not about aliasing rules.
 
 Which is a verbatim copy of the ftp code... Yes, the compiler is clearly
 broken, but the new code is easier to read anyway and does not require
 the ugly UC() macro.
 
 christos