From owner-freebsd-current Tue May 21 00:19:57 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id AAA03904 for current-outgoing; Tue, 21 May 1996 00:19:57 -0700 (PDT) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id AAA03896 for ; Tue, 21 May 1996 00:19:53 -0700 (PDT) Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id RAA26825; Tue, 21 May 1996 17:18:32 +1000 Date: Tue, 21 May 1996 17:18:32 +1000 From: Bruce Evans Message-Id: <199605210718.RAA26825@godzilla.zeta.org.au> To: asami@cs.berkeley.edu, wollman@lcs.mit.edu Subject: Re: some more on fast bcopy Cc: current@FreeBSD.org Sender: owner-current@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk >> Another place that it may help (and may fool lmbench and the likes :) >> is libc, do I need to the complicated FP state save/restore in there >> too? Or can that be a simple fnsave/frstor? >Don't forget that you can't make this the default because libc doesn't >know what CPU it's running on. There should be a sysctl or something to determine the cpu class. I want one for the fpu class. Currently we have hw.floatingpoint, but I want more than a binary value: { FPU_EMULATED_AND_BROKEN, FPU_EMULATED, FPU_COPROCESSOR, FPU_CPU }, where FPU_PROCESSOR might need to be multi-valued, and FPU_CPU says that the fpu behaviour is determined by the cpu class. Bruce