Date: Sat, 13 Sep 2014 00:49:11 +0400 From: Andrey Chernov <ache@freebsd.org> To: Tijl Coosemans <tijl@FreeBSD.org> Cc: ports@FreeBSD.org, Gerald Pfeifer <gerald@pfeifer.com> Subject: Re: Can't build lang/gcc port on i386: segmentation fault Message-ID: <54135C47.4010804@freebsd.org> In-Reply-To: <20140912164918.31e008d5@kalimero.tijl.coosemans.org> References: <5412D743.70005@freebsd.org> <alpine.LSU.2.11.1409121400460.2669@tuna.site> <5412E4A1.7040101@freebsd.org> <5412E64E.7050701@freebsd.org> <alpine.LSU.2.11.1409121436401.2669@tuna.site> <5412EFED.9000100@freebsd.org> <5412F480.8070100@freebsd.org> <20140912162342.2e6d557f@kalimero.tijl.coosemans.org> <5413067D.5050407@freebsd.org> <20140912164918.31e008d5@kalimero.tijl.coosemans.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 12.09.2014 18:49, Tijl Coosemans wrote: >>> Can you try with NO_CPU_CFLAGS=yes >> >> It works! >> But CPU optimizations are gone... > > Then I suspect clang produces a 64 bit instruction somewhere with > -march=core2. Maybe toolchain@ would be interested in debugging this > further. I contact toolchain@ related to this subj, providing core file, etc. It works on -current, amd64, -march=core2 combination (on better CPU), so it either -stable or i386-specific clang bug. Alas, I don't have -current, i386 machine with this CPU. -- http://ache.vniz.net/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?54135C47.4010804>