From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755863Ab0CVUli (ORCPT ); Mon, 22 Mar 2010 16:41:38 -0400 Received: from mx1.redhat.com ([209.132.183.28]:58892 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754306Ab0CVUlg (ORCPT ); Mon, 22 Mar 2010 16:41:36 -0400 Date: Mon, 22 Mar 2010 16:40:48 -0400 From: Jason Baron To: Steven Rostedt Cc: linux-kernel@vger.kernel.org, mingo@elte.hu, mathieu.desnoyers@polymtl.ca, hpa@zytor.com, tglx@linutronix.de, andi@firstfloor.org, roland@redhat.com, rth@redhat.com, mhiramat@redhat.com, fweisbec@gmail.com Subject: Re: [PATCH 3/5] jump label: x86 support Message-ID: <20100322204048.GD2278@redhat.com> References: <5492858ead53ab5c67d11acc620001abe8d36bcb.1269272444.git.jbaron@redhat.com> <1269276025.2957.5.camel@gandalf.stny.rr.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1269276025.2957.5.camel@gandalf.stny.rr.com> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Mar 22, 2010 at 12:40:25PM -0400, Steven Rostedt wrote: > > add x86 support for jump label. I'm keeping this patch separate so its clear to > > arch maintainers what was required for x86 support this new feature. hopefully, > > it wouldn't be too painful for other arches. > > > > Signed-off-by: Jason Baron > > > > +#ifdef __HAVE_ARCH_JUMP_LABEL > > + > > +# ifdef CONFIG_X86_64 > > +# define JUMP_LABEL_NOP P6_NOP5 > > +# else > > +# define JUMP_LABEL_NOP ".byte 0xe9 \n\t .long 0\n\t" > > +# endif > > Are you sure P6_NOP5 can't happen on non 64bit? Just because it is not > configured does not mean that the CPU can not handle it. Look at the > code I did in arch/x86/kernel/ftrace.c to determine what nop to use. > Maybe we can make that generic and have at boot up, the kernel determine > a proper 5byte nop. > indeed, i've looked at the ftrace nop code...I think was concerned that I would need a baseline nop that would work for all boxes. But I guess a jmp + 5 would be it? -Jason