Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Tue, 17 Dec 2002 14:41:23 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Tue, 17 Dec 2002 14:41:23 -0500 Received: from chaos.analogic.com ([204.178.40.224]:62602 "EHLO chaos.analogic.com") by vger.kernel.org with ESMTP id ; Tue, 17 Dec 2002 14:41:20 -0500 Date: Tue, 17 Dec 2002 14:52:08 -0500 (EST) From: "Richard B. Johnson" Reply-To: root@chaos.analogic.com To: Alan Cox cc: Ulrich Drepper , Linus Torvalds , Dave Jones , Ingo Molnar , Linux Kernel Mailing List , hpa@transmeta.com Subject: Re: Intel P6 vs P7 system call performance In-Reply-To: <1040154273.20804.13.camel@irongate.swansea.linux.org.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1252 Lines: 43 On 17 Dec 2002, Alan Cox wrote: > On Tue, 2002-12-17 at 18:48, Ulrich Drepper wrote: > > Alan Cox wrote: > > > > > Is there any reason you can't just keep the linker out of the entire > > > mess by generating > > > > > > .byte whatever > > > .dword 0xFFFF0000 > > > > > > instead of call ? > > > > There is no such instruction. Unless you know about some secret > > undocumented opcode... > > No I'd forgotten how broken x86 was > You can call intersegment with a full pointer. I don't know how expensive that is. Since USER_CS is a fixed value in Linux, it can be hard-coded .byte 0x9a .dword 0xfffff000 .word USER_CS No. I didn't try this, I'm just looking at the manual. I don't know what the USER_CS is (didn't look in the kernel) The book says the pointer is 16:32 which means that it's a dword, followed by a word. Cheers, Dick Johnson Penguin : Linux version 2.4.18 on an i686 machine (797.90 BogoMips). Why is the government concerned about the lunatic fringe? Think about it. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/