From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752391AbcBOQbv (ORCPT ); Mon, 15 Feb 2016 11:31:51 -0500 Received: from mx1.redhat.com ([209.132.183.28]:45365 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751235AbcBOQbq (ORCPT ); Mon, 15 Feb 2016 11:31:46 -0500 Date: Mon, 15 Feb 2016 10:31:34 -0600 From: Josh Poimboeuf To: Peter Zijlstra Cc: Jiri Slaby , Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , x86@kernel.org, linux-kernel@vger.kernel.org, live-patching@vger.kernel.org, Michal Marek , Andy Lutomirski , Borislav Petkov , Linus Torvalds , Andi Kleen , Pedro Alves , Namhyung Kim , Bernd Petrovitsch , Chris J Arges , Andrew Morton , Arnaldo Carvalho de Melo , David Vrabel , Borislav Petkov , Konrad Rzeszutek Wilk , Boris Ostrovsky , Jeremy Fitzhardinge , Chris Wright , Alok Kataria , Rusty Russell , Herbert Xu , "David S. Miller" , Pavel Machek , "Rafael J. Wysocki" , Len Brown , Matt Fleming , Alexei Starovoitov , netdev@vger.kernel.org, Ananth N Mavinakayanahalli , Anil S Keshavamurthy , Masami Hiramatsu , Gleb Natapov , Paolo Bonzini , kvm@vger.kernel.org, Wim Van Sebroeck , Guenter Roeck , linux-watchdog@vger.kernel.org, Waiman Long Subject: Re: [PATCH 00/33] Compile-time stack metadata validation Message-ID: <20160215163134.GA20585@treble.redhat.com> References: <56BDB5A8.9030006@suse.cz> <20160212144543.GA29004@treble.redhat.com> <20160212171037.GV6357@twins.programming.kicks-ass.net> <20160212183206.GB29004@treble.redhat.com> <20160212201011.GW6357@twins.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <20160212201011.GW6357@twins.programming.kicks-ass.net> User-Agent: Mutt/1.5.23.1-rc1 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Feb 12, 2016 at 09:10:11PM +0100, Peter Zijlstra wrote: > On Fri, Feb 12, 2016 at 12:32:06PM -0600, Josh Poimboeuf wrote: > > What I actually see in the listing is: > > > > decl __percpu_prefix:__preempt_count > > je 1f: > > .... > > 1: > > call ___preempt_schedule > > > > So it puts the "call ___preempt_schedule" in the slow path. > > Ah yes indeed. Same difference though. > > > I also don't see how that would be related to the use of the asm > > statement in the __preempt_schedule() macro. Doesn't the use of > > unlikely() in preempt_enable() put the call in the slow path? > > Sadly no, unlikely() and asm_goto don't work well together. But the slow > path or not isn't the reason we do the asm call thing. > > > #define preempt_enable() \ > > do { \ > > barrier(); \ > > if (unlikely(preempt_count_dec_and_test())) \ > > preempt_schedule(); \ > > } while (0) > > > > Also, why is the thunk needed? Any reason why preempt_enable() can't be > > called directly from C? > > That would make the call-site save registers and increase the size of > every preempt_enable(). By using the thunk we can do callee saved > registers and avoid blowing up the call site. So is the goal to optimize for size? If I replace the calls to __preempt_schedule[_notrace]() with real C calls and remove the thunks, it only adds about 2k to vmlinux. There are two ways to fix the warnings: 1. get rid of the thunks and call the C functions directly; or 2. add the stack pointer to the asm() statement output operand list to ensure a stack frame gets created in the caller function before the call. (Note this still allows the thunks to do callee saved registers.) I like #1 better, but maybe I'm still missing the point of the thunks. -- Josh