From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751716AbbGKWGd (ORCPT ); Sat, 11 Jul 2015 18:06:33 -0400 Received: from mail-lb0-f181.google.com ([209.85.217.181]:33239 "EHLO mail-lb0-f181.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751501AbbGKWGc (ORCPT ); Sat, 11 Jul 2015 18:06:32 -0400 Message-ID: <55A19363.1050506@cogentembedded.com> Date: Sun, 12 Jul 2015 01:06:27 +0300 From: Sergei Shtylyov Organization: Cogent Embedded User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.7.0 MIME-Version: 1.0 To: David Daney , linux-arm-kernel@lists.infradead.org, Catalin Marinas , Will Deacon CC: Robert Richter , Andrew Morton , linux-kernel@vger.kernel.org, David Daney Subject: Re: [PATCH 3/3] arm64, mm: Use IPIs for TLB invalidation. References: <1436646323-10527-1-git-send-email-ddaney.cavm@gmail.com> <1436646323-10527-4-git-send-email-ddaney.cavm@gmail.com> In-Reply-To: <1436646323-10527-4-git-send-email-ddaney.cavm@gmail.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hello. On 07/11/2015 11:25 PM, David Daney wrote: > From: David Daney > Most broadcast TLB invalidations are unnecessary. So when > invalidating for a given mm/vma target the only the needed CPUs via The only the needed? > and IPI. > For global TLB invalidations, also use IPI. > Tested on Cavium ThunderX. > This change reduces 'time make -j48' on kernel from 139s to 116s (83% > as long). > The patch is needed because of a ThunderX Pass1 erratum: Exclusive > store operations unreliable in the presence of broadcast TLB > invalidations. The performance improvements shown make it compelling > even without the erratum workaround need. > Signed-off-by: David Daney WBR, Sergei From mboxrd@z Thu Jan 1 00:00:00 1970 From: sergei.shtylyov@cogentembedded.com (Sergei Shtylyov) Date: Sun, 12 Jul 2015 01:06:27 +0300 Subject: [PATCH 3/3] arm64, mm: Use IPIs for TLB invalidation. In-Reply-To: <1436646323-10527-4-git-send-email-ddaney.cavm@gmail.com> References: <1436646323-10527-1-git-send-email-ddaney.cavm@gmail.com> <1436646323-10527-4-git-send-email-ddaney.cavm@gmail.com> Message-ID: <55A19363.1050506@cogentembedded.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hello. On 07/11/2015 11:25 PM, David Daney wrote: > From: David Daney > Most broadcast TLB invalidations are unnecessary. So when > invalidating for a given mm/vma target the only the needed CPUs via The only the needed? > and IPI. > For global TLB invalidations, also use IPI. > Tested on Cavium ThunderX. > This change reduces 'time make -j48' on kernel from 139s to 116s (83% > as long). > The patch is needed because of a ThunderX Pass1 erratum: Exclusive > store operations unreliable in the presence of broadcast TLB > invalidations. The performance improvements shown make it compelling > even without the erratum workaround need. > Signed-off-by: David Daney WBR, Sergei