From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christian Borntraeger Date: Mon, 01 Jun 2015 07:35:57 +0000 Subject: Re: [PATCH 0/5] Remove s390 sw-emulated hugepages and cleanup Message-Id: <556C0B5D.40205@de.ibm.com> List-Id: References: <1432813957-46874-1-git-send-email-dingel@linux.vnet.ibm.com> In-Reply-To: <1432813957-46874-1-git-send-email-dingel@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-arm-kernel@lists.infradead.org Am 28.05.2015 um 13:52 schrieb Dominik Dingel: > Hi everyone, > > there is a potential bug with KVM and hugetlbfs if the hardware does not > support hugepages (EDAT1). > We fix this by making EDAT1 a hard requirement for hugepages and > therefore removing and simplifying code. The cleanup itself is nice and probably the right thing to do. Emulating large pages makes the code more complex and asks for trouble (as outlined above) The only downside that I see is that z/VM as of today does not announce EDAT1 for its guests so the "emulated" large pages for hugetlbfs would be useful in that case. The current code allocates the page table only once and shares it for all mappers - which is useful for some big databases that spawn hundreds of processes with shared mappings of several hundred GBs. In these cases we do save a decent amount of page table memory. Not sure if that case is actually important, though. Christian From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christian Borntraeger Subject: Re: [PATCH 0/5] Remove s390 sw-emulated hugepages and cleanup Date: Mon, 01 Jun 2015 09:35:57 +0200 Message-ID: <556C0B5D.40205@de.ibm.com> References: <1432813957-46874-1-git-send-email-dingel@linux.vnet.ibm.com> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1432813957-46874-1-git-send-email-dingel@linux.vnet.ibm.com> Sender: linux-sh-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="us-ascii" To: Dominik Dingel , linux-kernel@vger.kernel.org Cc: Russell King , Catalin Marinas , Will Deacon , Tony Luck , Fenghua Yu , James Hogan , Ralf Baechle , Benjamin Herrenschmidt , Paul Mackerras , Michael Ellerman , Martin Schwidefsky , Heiko Carstens , linux390@de.ibm.com, "David S. Miller" , Chris Metcalf , Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , x86@kernel.org, Andrew Morton , Zhang Zhen , David Rientjes , "Aneesh Kumar K.V" , Nathan Am 28.05.2015 um 13:52 schrieb Dominik Dingel: > Hi everyone, > > there is a potential bug with KVM and hugetlbfs if the hardware does not > support hugepages (EDAT1). > We fix this by making EDAT1 a hard requirement for hugepages and > therefore removing and simplifying code. The cleanup itself is nice and probably the right thing to do. Emulating large pages makes the code more complex and asks for trouble (as outlined above) The only downside that I see is that z/VM as of today does not announce EDAT1 for its guests so the "emulated" large pages for hugetlbfs would be useful in that case. The current code allocates the page table only once and shares it for all mappers - which is useful for some big databases that spawn hundreds of processes with shared mappings of several hundred GBs. In these cases we do save a decent amount of page table memory. Not sure if that case is actually important, though. Christian From mboxrd@z Thu Jan 1 00:00:00 1970 Received: with ECARTIS (v1.0.0; list linux-mips); Mon, 01 Jun 2015 09:36:14 +0200 (CEST) Received: from e06smtp11.uk.ibm.com ([195.75.94.107]:40356 "EHLO e06smtp11.uk.ibm.com" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S27006840AbbFAHgMyVH44 (ORCPT ); Mon, 1 Jun 2015 09:36:12 +0200 Received: from /spool/local by e06smtp11.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Mon, 1 Jun 2015 08:36:05 +0100 Received: from d06dlp03.portsmouth.uk.ibm.com (9.149.20.15) by e06smtp11.uk.ibm.com (192.168.101.141) with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted; Mon, 1 Jun 2015 08:36:01 +0100 Received: from b06cxnps4076.portsmouth.uk.ibm.com (d06relay13.portsmouth.uk.ibm.com [9.149.109.198]) by d06dlp03.portsmouth.uk.ibm.com (Postfix) with ESMTP id 4B9851B0807D; Mon, 1 Jun 2015 08:36:55 +0100 (BST) Received: from d06av05.portsmouth.uk.ibm.com (d06av05.portsmouth.uk.ibm.com [9.149.37.229]) by b06cxnps4076.portsmouth.uk.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id t517a1Y821627092; Mon, 1 Jun 2015 07:36:01 GMT Received: from d06av05.portsmouth.uk.ibm.com (localhost [127.0.0.1]) by d06av05.portsmouth.uk.ibm.com (8.14.4/8.14.4/NCO v10.0 AVout) with ESMTP id t517Zw7k028390; Mon, 1 Jun 2015 01:36:01 -0600 Received: from oc1450873852.ibm.com (dyn-9-152-224-118.boeblingen.de.ibm.com [9.152.224.118]) by d06av05.portsmouth.uk.ibm.com (8.14.4/8.14.4/NCO v10.0 AVin) with ESMTP id t517ZvRx028372; Mon, 1 Jun 2015 01:35:57 -0600 Message-ID: <556C0B5D.40205@de.ibm.com> Date: Mon, 01 Jun 2015 09:35:57 +0200 From: Christian Borntraeger User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.6.0 MIME-Version: 1.0 To: Dominik Dingel , linux-kernel@vger.kernel.org CC: Russell King , Catalin Marinas , Will Deacon , Tony Luck , Fenghua Yu , James Hogan , Ralf Baechle , Benjamin Herrenschmidt , Paul Mackerras , Michael Ellerman , Martin Schwidefsky , Heiko Carstens , linux390@de.ibm.com, "David S. Miller" , Chris Metcalf , Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , x86@kernel.org, Andrew Morton , Zhang Zhen , David Rientjes , "Aneesh Kumar K.V" , Nathan Lynch , Andy Lutomirski , Michael Holzheu , Hugh Dickins , Naoya Horiguchi , Paolo Bonzini , "Jason J. Herne" , Davidlohr Bueso , "Kirill A. Shutemov" , Luiz Capitulino , Mike Kravetz , linux-arm-kernel@lists.infradead.org, linux-ia64@vger.kernel.org, linux-metag@vger.kernel.org, linux-mips@linux-mips.org, linuxppc-dev@lists.ozlabs.org, linux-s390@vger.kernel.org, linux-sh@vger.kernel.org, sparclinux@vger.kernel.org, linux-mm@kvack.org Subject: Re: [PATCH 0/5] Remove s390 sw-emulated hugepages and cleanup References: <1432813957-46874-1-git-send-email-dingel@linux.vnet.ibm.com> In-Reply-To: <1432813957-46874-1-git-send-email-dingel@linux.vnet.ibm.com> Content-Type: text/plain; charset=iso-8859-15 Content-Transfer-Encoding: 7bit X-TM-AS-MML: disable X-Content-Scanned: Fidelis XPS MAILER x-cbid: 15060107-0041-0000-0000-0000049E36A0 Return-Path: X-Envelope-To: <"|/home/ecartis/ecartis -s linux-mips"> (uid 0) X-Orcpt: rfc822;linux-mips@linux-mips.org Original-Recipient: rfc822;linux-mips@linux-mips.org X-archive-position: 47759 X-ecartis-version: Ecartis v1.0.0 Sender: linux-mips-bounce@linux-mips.org Errors-to: linux-mips-bounce@linux-mips.org X-original-sender: borntraeger@de.ibm.com Precedence: bulk List-help: List-unsubscribe: List-software: Ecartis version 1.0.0 List-Id: linux-mips X-List-ID: linux-mips List-subscribe: List-owner: List-post: List-archive: X-list: linux-mips Am 28.05.2015 um 13:52 schrieb Dominik Dingel: > Hi everyone, > > there is a potential bug with KVM and hugetlbfs if the hardware does not > support hugepages (EDAT1). > We fix this by making EDAT1 a hard requirement for hugepages and > therefore removing and simplifying code. The cleanup itself is nice and probably the right thing to do. Emulating large pages makes the code more complex and asks for trouble (as outlined above) The only downside that I see is that z/VM as of today does not announce EDAT1 for its guests so the "emulated" large pages for hugetlbfs would be useful in that case. The current code allocates the page table only once and shares it for all mappers - which is useful for some big databases that spawn hundreds of processes with shared mappings of several hundred GBs. In these cases we do save a decent amount of page table memory. Not sure if that case is actually important, though. Christian From mboxrd@z Thu Jan 1 00:00:00 1970 From: borntraeger@de.ibm.com (Christian Borntraeger) Date: Mon, 01 Jun 2015 09:35:57 +0200 Subject: [PATCH 0/5] Remove s390 sw-emulated hugepages and cleanup In-Reply-To: <1432813957-46874-1-git-send-email-dingel@linux.vnet.ibm.com> References: <1432813957-46874-1-git-send-email-dingel@linux.vnet.ibm.com> Message-ID: <556C0B5D.40205@de.ibm.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Am 28.05.2015 um 13:52 schrieb Dominik Dingel: > Hi everyone, > > there is a potential bug with KVM and hugetlbfs if the hardware does not > support hugepages (EDAT1). > We fix this by making EDAT1 a hard requirement for hugepages and > therefore removing and simplifying code. The cleanup itself is nice and probably the right thing to do. Emulating large pages makes the code more complex and asks for trouble (as outlined above) The only downside that I see is that z/VM as of today does not announce EDAT1 for its guests so the "emulated" large pages for hugetlbfs would be useful in that case. The current code allocates the page table only once and shares it for all mappers - which is useful for some big databases that spawn hundreds of processes with shared mappings of several hundred GBs. In these cases we do save a decent amount of page table memory. Not sure if that case is actually important, though. Christian From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wi0-f175.google.com (mail-wi0-f175.google.com [209.85.212.175]) by kanga.kvack.org (Postfix) with ESMTP id 345326B0038 for ; Mon, 1 Jun 2015 03:36:05 -0400 (EDT) Received: by wicmx19 with SMTP id mx19so62253143wic.0 for ; Mon, 01 Jun 2015 00:36:04 -0700 (PDT) Received: from e06smtp11.uk.ibm.com (e06smtp11.uk.ibm.com. [195.75.94.107]) by mx.google.com with ESMTPS id d14si15221009wik.46.2015.06.01.00.36.02 for (version=TLSv1 cipher=AES128-SHA bits=128/128); Mon, 01 Jun 2015 00:36:03 -0700 (PDT) Received: from /spool/local by e06smtp11.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Mon, 1 Jun 2015 08:36:02 +0100 Received: from b06cxnps4076.portsmouth.uk.ibm.com (d06relay13.portsmouth.uk.ibm.com [9.149.109.198]) by d06dlp01.portsmouth.uk.ibm.com (Postfix) with ESMTP id 4DC1917D8066 for ; Mon, 1 Jun 2015 08:36:58 +0100 (BST) Received: from d06av05.portsmouth.uk.ibm.com (d06av05.portsmouth.uk.ibm.com [9.149.37.229]) by b06cxnps4076.portsmouth.uk.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id t517a0d025362632 for ; Mon, 1 Jun 2015 07:36:00 GMT Received: from d06av05.portsmouth.uk.ibm.com (localhost [127.0.0.1]) by d06av05.portsmouth.uk.ibm.com (8.14.4/8.14.4/NCO v10.0 AVout) with ESMTP id t517Zw7e028390 for ; Mon, 1 Jun 2015 01:36:00 -0600 Message-ID: <556C0B5D.40205@de.ibm.com> Date: Mon, 01 Jun 2015 09:35:57 +0200 From: Christian Borntraeger MIME-Version: 1.0 Subject: Re: [PATCH 0/5] Remove s390 sw-emulated hugepages and cleanup References: <1432813957-46874-1-git-send-email-dingel@linux.vnet.ibm.com> In-Reply-To: <1432813957-46874-1-git-send-email-dingel@linux.vnet.ibm.com> Content-Type: text/plain; charset=iso-8859-15 Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Dominik Dingel , linux-kernel@vger.kernel.org Cc: Russell King , Catalin Marinas , Will Deacon , Tony Luck , Fenghua Yu , James Hogan , Ralf Baechle , Benjamin Herrenschmidt , Paul Mackerras , Michael Ellerman , Martin Schwidefsky , Heiko Carstens , linux390@de.ibm.com, "David S. Miller" , Chris Metcalf , Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , x86@kernel.org, Andrew Morton , Zhang Zhen , David Rientjes , "Aneesh Kumar K.V" , Nathan Lynch , Andy Lutomirski , Michael Holzheu , Hugh Dickins , Naoya Horiguchi , Paolo Bonzini , "Jason J. Herne" , Davidlohr Bueso , "Kirill A. Shutemov" , Luiz Capitulino , Mike Kravetz , linux-arm-kernel@lists.infradead.org, linux-ia64@vger.kernel.org, linux-metag@vger.kernel.org, linux-mips@linux-mips.org, linuxppc-dev@lists.ozlabs.org, linux-s390@vger.kernel.org, linux-sh@vger.kernel.org, sparclinux@vger.kernel.org, linux-mm@kvack.org Am 28.05.2015 um 13:52 schrieb Dominik Dingel: > Hi everyone, > > there is a potential bug with KVM and hugetlbfs if the hardware does not > support hugepages (EDAT1). > We fix this by making EDAT1 a hard requirement for hugepages and > therefore removing and simplifying code. The cleanup itself is nice and probably the right thing to do. Emulating large pages makes the code more complex and asks for trouble (as outlined above) The only downside that I see is that z/VM as of today does not announce EDAT1 for its guests so the "emulated" large pages for hugetlbfs would be useful in that case. The current code allocates the page table only once and shares it for all mappers - which is useful for some big databases that spawn hundreds of processes with shared mappings of several hundred GBs. In these cases we do save a decent amount of page table memory. Not sure if that case is actually important, though. Christian -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org