Date: Sat, 06 Aug 2011 10:37:10 -0500 From: Nathan Whitehorn <nwhitehorn@freebsd.org> To: freebsd-ppc@freebsd.org Subject: Re: Lags and setjmp Message-ID: <4E3D5FA6.80207@freebsd.org> In-Reply-To: <CA%2BWntOvpqj0FSE%2BcYA2UqsefvhaxPhaaKdaz3DC%2BPyFe01iRDA@mail.gmail.com> References: <CA%2BWntOvpqj0FSE%2BcYA2UqsefvhaxPhaaKdaz3DC%2BPyFe01iRDA@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 08/05/11 10:47, Super Bisquit wrote: > The file /usr/src/sys/powerpc/powerpc/setjmp.S has > > #ifdef __powerpc64__ > > at line nine. Maybe this is affecting the system because: > the lag seems to occur on Xorg/Xserver with the cursor having to be moved > before there is a response. > I'm wondering if setting the vale to powerpc instead of powerpc64 would stop > this action. > It wouldn't for the reason Justin said, and this file only applies to the kernel besides. I've also never experienced this bug you're referring to. Is X using longjmp/setjmp() internally? One thing we do not do is save/restore FP registers for long jumps. It's not required by the spec, but Linux does do it. -Nathan
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4E3D5FA6.80207>