From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753122AbdK1Mnd (ORCPT ); Tue, 28 Nov 2017 07:43:33 -0500 Received: from bombadil.infradead.org ([65.50.211.133]:35282 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752577AbdK1Mna (ORCPT ); Tue, 28 Nov 2017 07:43:30 -0500 Date: Tue, 28 Nov 2017 13:43:13 +0100 From: Peter Zijlstra To: Josh Poimboeuf Cc: linux-kernel@vger.kernel.org, Thomas Gleixner , Dave Hansen , Andy Lutomirski , Ingo Molnar , Borislav Petkov , Brian Gerst , Denys Vlasenko , "H . Peter Anvin" , Linus Torvalds , Rik van Riel , daniel.gruss@iaik.tugraz.at, hughd@google.com, keescook@google.com, linux-mm@kvack.org, michael.schwarz@iaik.tugraz.at, moritz.lipp@iaik.tugraz.at, richard.fellner@student.tugraz.at Subject: Re: [PATCH 1/2] x86/mm/kaiser: Remove unused user-mapped page-aligned section Message-ID: <20171128124313.vezo4vkfucuezutl@hirez.programming.kicks-ass.net> References: <666935452d5eef100464b7314be90fccd65e795c.1511842148.git.jpoimboe@redhat.com> <20171128092025.ggbnkx2j4uglbdax@hirez.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20171128092025.ggbnkx2j4uglbdax@hirez.programming.kicks-ass.net> User-Agent: NeoMutt/20170609 (1.8.3) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Nov 28, 2017 at 10:20:25AM +0100, Peter Zijlstra wrote: > On Mon, Nov 27, 2017 at 10:10:12PM -0600, Josh Poimboeuf wrote: > > The '.data..percpu..user_mapped..page_aligned' section isn't used > > anywhere. Remove it and its related macros. > > With my patches: > > arch/x86/events/intel/ds.c:DEFINE_PER_CPU_SHARED_ALIGNED_USER_MAPPED(struct debug_store, cpu_debug_store); > > is the only user left of any of that. > > I suppose we could just allocate a whole page for that and use > kaiser_add_mapping() for it. Then we can remove all of > DEFINE_.*_USER_MAPPED(). Or we could field it a spot in the cpu_entry_area I suppose.