Date: Sat, 18 Aug 2001 23:10:17 -0700 From: Peter Wemm <peter@wemm.org> To: Mike Smith <msmith@freebsd.org> Cc: freebsd-stable@freebsd.org Subject: Re: New kernel option CPU_ENABLE_SSE Message-ID: <20010819061017.92A3438FD@overcee.netplex.com.au>
next in thread | raw e-mail | index | archive | help
Mike Smith wrote: > > >> Because not all i686'es support SSE. > > > > > >So detect it automatically based on the CPU feature bits. > > > > > >Needing a kernel compile option for this is unforgivably lame. If you > > >want to be able to disable it, use a tunable. > > > Perhaps; the "gist" I get is that the compile option is for > > some "field-testing." Maybe similarly appropriate would be > > something similar to "NO_F00F_HACK"; for example, > > "CPU_DISABLE_SSE" or "CPU_NO_ENABLE_SSE" (?). > > All of this stuff is unforgivably lame. Tunables. Tunables. Dammit. 8) Tunables cant compile out the support code. Yes, there should be a tunable to disable the activation of it, but being able to compile out the support code is still worth while in some places. And yes, it is dependent on the cpu_features SSE and FXSR bits. The option is probably misnamed.. It would better be: CPU_SSE_SUPPORT, if we had to have an option. But the original patches had it as CPU_ENABLE_SSE. Cheers, -Peter -- Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010819061017.92A3438FD>