Date: Wed, 03 Mar 2004 10:59:21 -0600 From: Jon Noack <noackjr@alumni.rice.edu> To: current@FreeBSD.ORG Cc: Remi <MrL0L@charter.net> Subject: Re: libxine failure under -CURRENT Message-ID: <40460EE9.6070202@alumni.rice.edu> In-Reply-To: <20040303100548.GA2636@dragon.nuxi.com> References: <404510EF.8040300@charter.net> <1078268844.707.2.camel@falter> <20040303100548.GA2636@dragon.nuxi.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 3/3/2004 4:05 AM, David O'Brien wrote: > On Tue, Mar 02, 2004 at 11:07:25PM +0000, Alistair Sutton wrote: >> On Tue, 2004-03-02 at 22:55, Remi wrote: >>> # uname -a >>> FreeBSD 5.2-CURRENT FreeBSD 5.2-CURRENT #1: Sun Feb 29 19:30:50 PST >>> 2004 root@:/usr/obj/usr/src/sys/BSD i386 >>> >>> Has anyone seen this or know how to fix it? >> >> *snip* >> >> You need to change your CPUTYPE variable in /etc/make.conf to something >> other than P4 and then clean and remake the port. > > Why? GCC 3.3.3 doesn't have the p4 code generation bugs of earlier > GCC's. There are still edge-cases where things don't quite work correctly, especially when you start dealing with assembly. For example, trying to compile mplayer with -funroll-loops will also fail. Perhaps we should make a note somewhere that multimedia software (or any hand-optimized software) may not like some compiler flags. In any case, it's best practice when compilation fails to try again without extra optimizations (-O -pipe at most). If it still fails, there's a problem somewhere. If not, let us know which which compiler flag causes it to fail -- there might be a fix for it... Jon Noack
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?40460EE9.6070202>