From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christian Borntraeger Subject: Re: [RFC]Two ideas to optimize updating irq routing table Date: Wed, 26 Mar 2014 12:14:34 +0100 Message-ID: <5332B69A.6060102@de.ibm.com> References: <33183CC9F5247A488A2544077AF19020815DA9BD@SZXEMA503-MBS.china.huawei.com> <53317892.2050506@redhat.com> <33183CC9F5247A488A2544077AF19020815DAEBA@SZXEMA503-MBS.china.huawei.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: "gleb@redhat.com" , "mst@redhat.com" , "avi.kivity@gmail.com" , "Herongguang (Stephen)" , "Huangweidong (C)" To: "Gonglei (Arei)" , Paolo Bonzini , "qemu-devel@nongnu.org" , "kvm@vger.kernel.org" Return-path: Received: from e06smtp14.uk.ibm.com ([195.75.94.110]:46238 "EHLO e06smtp14.uk.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751432AbaCZLOk (ORCPT ); Wed, 26 Mar 2014 07:14:40 -0400 Received: from /spool/local by e06smtp14.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Wed, 26 Mar 2014 11:14:38 -0000 Received: from b06cxnps4076.portsmouth.uk.ibm.com (d06relay13.portsmouth.uk.ibm.com [9.149.109.198]) by d06dlp02.portsmouth.uk.ibm.com (Postfix) with ESMTP id B0CA72190041 for ; Wed, 26 Mar 2014 11:14:30 +0000 (GMT) Received: from d06av03.portsmouth.uk.ibm.com (d06av03.portsmouth.uk.ibm.com [9.149.37.213]) by b06cxnps4076.portsmouth.uk.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id s2QBEOsI48431212 for ; Wed, 26 Mar 2014 11:14:24 GMT Received: from d06av03.portsmouth.uk.ibm.com (localhost [127.0.0.1]) by d06av03.portsmouth.uk.ibm.com (8.14.4/8.14.4/NCO v10.0 AVout) with ESMTP id s2QBEZVI010850 for ; Wed, 26 Mar 2014 05:14:35 -0600 In-Reply-To: <33183CC9F5247A488A2544077AF19020815DAEBA@SZXEMA503-MBS.china.huawei.com> Sender: kvm-owner@vger.kernel.org List-ID: On 26/03/14 09:22, Gonglei (Arei) wrote: > Without patches, ping time can jump from 0.3ms to 2ms-30ms. With synchronize_srcu > patch, ping time is worse. With synchronize_srcu_expedited patch, ping time is > overall good, though sometimes ping time jump to 1ms-3ms. Just to understand whats going on, does something like diff --git a/kernel/rcu/srcu.c b/kernel/rcu/srcu.c index 3318d82..432c2a2 100644 --- a/kernel/rcu/srcu.c +++ b/kernel/rcu/srcu.c @@ -331,7 +331,7 @@ EXPORT_SYMBOL_GPL(__srcu_read_unlock); */ #define SRCU_RETRY_CHECK_DELAY 5 #define SYNCHRONIZE_SRCU_TRYCOUNT 2 -#define SYNCHRONIZE_SRCU_EXP_TRYCOUNT 12 +#define SYNCHRONIZE_SRCU_EXP_TRYCOUNT 50 /* * @@@ Wait until all pre-existing readers complete. Such readers make the problem go away? From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:39144) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WSlnI-0006Hi-Dj for qemu-devel@nongnu.org; Wed, 26 Mar 2014 07:15:01 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WSln6-0004UT-OJ for qemu-devel@nongnu.org; Wed, 26 Mar 2014 07:14:52 -0400 Received: from e06smtp12.uk.ibm.com ([195.75.94.108]:57425) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WSln6-0004UN-EP for qemu-devel@nongnu.org; Wed, 26 Mar 2014 07:14:40 -0400 Received: from /spool/local by e06smtp12.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Wed, 26 Mar 2014 11:14:38 -0000 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 0A8411B08061 for ; Wed, 26 Mar 2014 11:14:29 +0000 (GMT) Received: from d06av03.portsmouth.uk.ibm.com (d06av03.portsmouth.uk.ibm.com [9.149.37.213]) by b06cxnps4076.portsmouth.uk.ibm.com (8.13.8/8.13.8/NCO v10.0) with ESMTP id s2QBEOFT50856166 for ; Wed, 26 Mar 2014 11:14:24 GMT Received: from d06av03.portsmouth.uk.ibm.com (localhost [127.0.0.1]) by d06av03.portsmouth.uk.ibm.com (8.14.4/8.14.4/NCO v10.0 AVout) with ESMTP id s2QBEZVE010850 for ; Wed, 26 Mar 2014 05:14:35 -0600 Message-ID: <5332B69A.6060102@de.ibm.com> Date: Wed, 26 Mar 2014 12:14:34 +0100 From: Christian Borntraeger MIME-Version: 1.0 References: <33183CC9F5247A488A2544077AF19020815DA9BD@SZXEMA503-MBS.china.huawei.com> <53317892.2050506@redhat.com> <33183CC9F5247A488A2544077AF19020815DAEBA@SZXEMA503-MBS.china.huawei.com> In-Reply-To: <33183CC9F5247A488A2544077AF19020815DAEBA@SZXEMA503-MBS.china.huawei.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [RFC]Two ideas to optimize updating irq routing table List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Gonglei (Arei)" , Paolo Bonzini , "qemu-devel@nongnu.org" , "kvm@vger.kernel.org" Cc: "Herongguang (Stephen)" , "Huangweidong (C)" , "avi.kivity@gmail.com" , "gleb@redhat.com" , "mst@redhat.com" On 26/03/14 09:22, Gonglei (Arei) wrote: > Without patches, ping time can jump from 0.3ms to 2ms-30ms. With synchronize_srcu > patch, ping time is worse. With synchronize_srcu_expedited patch, ping time is > overall good, though sometimes ping time jump to 1ms-3ms. Just to understand whats going on, does something like diff --git a/kernel/rcu/srcu.c b/kernel/rcu/srcu.c index 3318d82..432c2a2 100644 --- a/kernel/rcu/srcu.c +++ b/kernel/rcu/srcu.c @@ -331,7 +331,7 @@ EXPORT_SYMBOL_GPL(__srcu_read_unlock); */ #define SRCU_RETRY_CHECK_DELAY 5 #define SYNCHRONIZE_SRCU_TRYCOUNT 2 -#define SYNCHRONIZE_SRCU_EXP_TRYCOUNT 12 +#define SYNCHRONIZE_SRCU_EXP_TRYCOUNT 50 /* * @@@ Wait until all pre-existing readers complete. Such readers make the problem go away?