Date: Sat, 13 Dec 2008 11:45:32 -0800 From: "Garrett Cooper" <yanefbsd@gmail.com> To: "=?ISO-8859-1?Q?Horst_G=FCnther_Burkhardt_III?=" <horst@sxemacs.org> Cc: FreeBSD PowerPC ML <freebsd-ppc@freebsd.org> Subject: Re: Two issues. One minor, one major. Message-ID: <7d6fde3d0812131145i4560cdb3rbea4c8a072452145@mail.gmail.com> In-Reply-To: <1229075882.1832.85.camel@horst-tla> References: <d19d56910e9f6d3ce0102df7fed9017b@212.203.104.226> <1229075882.1832.85.camel@horst-tla>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Dec 12, 2008 at 1:58 AM, Horst G=FCnther Burkhardt III <horst@sxemacs.org> wrote: > On Thu, 2008-12-11 at 11:32 +0100, Andreas Tobler wrote: >> -mcpu=3D7400 does imply -mppc -maltivec. See the rs6000.h in the >> gcc/config/rs6000 directory. >> I'd suggest to build w/o special cpu flags. > > Thanks. > > OK. Idea: > > -Put patch in devel/gcc and patch in the base system that either > > makes altivec do nothing (NOT a good idea) > makes mcpu not imply maltivec ( a better idea, as it can just be undone > when the kernel gets altivec instruction support ) > > > good idea y/y? > > (also, so it seems the fix is just explicitly supply -mno-altivec in the > commandline) > > --Horst I think that putting some logic in the ports and src makefiles to automatically add -mno-altivec would be wise. -Garrett
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?7d6fde3d0812131145i4560cdb3rbea4c8a072452145>