From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932630AbcAIASa (ORCPT ); Fri, 8 Jan 2016 19:18:30 -0500 Received: from mail-ob0-f172.google.com ([209.85.214.172]:35390 "EHLO mail-ob0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753535AbcAIAS2 (ORCPT ); Fri, 8 Jan 2016 19:18:28 -0500 MIME-Version: 1.0 In-Reply-To: References: From: Andy Lutomirski Date: Fri, 8 Jan 2016 16:18:08 -0800 Message-ID: Subject: Re: [RFC 09/13] x86/mm: Disable interrupts when flushing the TLB using CR3 To: Linus Torvalds Cc: Oleg Nesterov , X86 ML , Dave Hansen , Borislav Petkov , Linux Kernel Mailing List , "linux-mm@kvack.org" , Brian Gerst Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Jan 8, 2016 3:41 PM, "Linus Torvalds" wrote: > > On Fri, Jan 8, 2016 at 3:15 PM, Andy Lutomirski wrote: > > + /* > > + * We mustn't be preempted or handle an IPI while reading and > > + * writing CR3. Preemption could switch mms and switch back, and > > + * an IPI could call leave_mm. Either of those could cause our > > + * PCID to change asynchronously. > > + */ > > + raw_local_irq_save(flags); > > native_write_cr3(native_read_cr3()); > > + raw_local_irq_restore(flags); > > This seems sad for two reasons: > > - it adds unnecessary overhead on non-pcid setups (32-bit being an > example of that) I can certainly skip the flag saving on !PCID. > > - on pcid setups, wouldn't invpcid_flush_single_context() be better? > I played with that and it was slower. I don't pretend that makes any sense. > So on the whole I hate it. > > Why isn't this something like > > if (static_cpu_has_safe(X86_FEATURE_INVPCID)) { > invpcid_flush_single_context(); > return; > } > native_write_cr3(native_read_cr3()); > > *without* any flag saving crud? > > And yes, that means that we'd require X86_FEATURE_INVPCID in order to > use X86_FEATURE_PCID, but that seems fine. I have an SNB "Extreme" with PCID but not INVPCID, and there could be a whole generation of servers like that. I think we should fully support them. We might be able to get away with just disabling preemption instead of IRQs, at least if mm == active_mm. > > Or is there some reason you wanted the odd flags version? If so, that > should be documented. What do you mean "odd"? --Andy