From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ns.skekraft.net (ns.skekraft.net [213.199.96.131]) by ozlabs.org (Postfix) with ESMTP id 3F76367B22 for ; Sat, 14 May 2005 11:03:19 +1000 (EST) Received: from icd.localnet (131.net95.skekraft.net [213.199.95.131]) by ns.skekraft.net (Postfix) with ESMTP id F26BFA4009 for ; Sat, 14 May 2005 03:03:16 +0200 (CEST) From: Roger Larsson To: linuxppc-embedded@ozlabs.org Date: Sat, 14 May 2005 04:55:29 +0200 References: <005b01c557bd$c04e2770$0301a8c0@chuck2> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Message-Id: <200505140455.29179.roger.larsson@norran.net> Subject: floating point in kernel (Re: insmod problem) List-Id: Linux on Embedded PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Friday 13 May 2005 15.46, Kumar Gala wrote: > On May 13, 2005, at 8:14 AM, Mark Chambers wrote: > > > You're using floating arithmetic in the kernel. > > > > > > That's not supported. > > > > > > Regards > > > > > > Pantelis > > > > Just for my edification:=A0 Is floating point itself not supported > > (floating > > point register > > corruption or something like that) or just linking to user space > > libraries? > > Use of floating point registers and instructions. Its possible to use > FP in rare cases inside the kernel buts its extremely discouraged. > The kernel is doing lazy saving of FP registers, only saving for user=20 processes that uses them. And especially avoiding to save for every interrupt! So if you like to use any FP (registers) in kernel you have to: =2D disable kernel preemption (YES) =2D save previous value =2D do your stuff =2D restore =2D enable preemption I remember a preemption bug caused by using MMX/SSE to copy memory uncached. /RogerL