Date: Tue, 26 Aug 2008 17:06:33 +1000 From: Andrew Reilly <andrew-freebsd@areilly.bpc-users.org> To: Rohit Tripathi <rohit.trip@gmail.com> Cc: freebsd-current@freebsd.org Subject: Re: gnome2/x11 slow after optimization! Message-ID: <20080826070633.GA85357@duncan.reilly.home> In-Reply-To: <33615c8e0808211038l7b9ab665l67f0439d3f40e6a6@mail.gmail.com> References: <33615c8e0808211038l7b9ab665l67f0439d3f40e6a6@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Aug 21, 2008 at 01:38:09PM -0400, Rohit Tripathi wrote: > CFLAGS= -O3 -pipe -mtune=core2 -m128bit-long-double -mfpmath=sse Just on top of what everyone else has said, (i.e., "don't do that"), the -m128bit-long-double option is almost certainly going to break something if you're operating in 32-bit mode (it changes the ABI for some functions and struct alignment), with the caveat that the "long double" type is almost never used, so you might get away with it just by default. It certainly isn't going to help when you're using sse math, which doesn't do long doubles at all. I'd be very surprised if either Gnome2 or x11 did anything with long double, so this comment really defaults to the initial statement: "don't do that". Cheers, -- Andrew
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080826070633.GA85357>