From mboxrd@z Thu Jan 1 00:00:00 1970 From: Joel Fernandes Subject: Re: [RFC] high preempt off latency in vfree path Date: Wed, 23 Mar 2016 12:03:49 -0700 Message-ID: <56F2E895.7080003@lab126.com> References: <56F1F4A6.2060400@lab126.com> <20160323024402.GA27856@tassilo.jf.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Cc: , , , , Nick Piggin To: Andi Kleen Return-path: In-Reply-To: <20160323024402.GA27856@tassilo.jf.intel.com> Sender: owner-linux-mm@kvack.org List-Id: linux-rt-users.vger.kernel.org On 03/22/16 19:44, Andi Kleen wrote: >> (1) >> One is we reduce the number of lazy_max_pages (right now its around 32MB per core worth of pages). >> >> diff --git a/mm/vmalloc.c b/mm/vmalloc.c >> index aa3891e..2720f4f 100644 >> --- a/mm/vmalloc.c >> +++ b/mm/vmalloc.c >> @@ -564,7 +564,7 @@ static unsigned long lazy_max_pages(void) >> >> log = fls(num_online_cpus()); >> >> - return log * (32UL * 1024 * 1024 / PAGE_SIZE); >> + return log * (8UL * 1024 * 1024 / PAGE_SIZE); >> } > > This seems like the right fix to me. Perhaps even make it somewhat smaller. > > Even on larger systems it's probably fine because they have a lot more > cores/threads these days, so it will be still sufficiently large. > Thanks Andi. I'll post a patch then. Regards, Joel > -Andi > -- 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 From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk0-f171.google.com (mail-qk0-f171.google.com [209.85.220.171]) by kanga.kvack.org (Postfix) with ESMTP id 377DB6B0005 for ; Wed, 23 Mar 2016 15:04:41 -0400 (EDT) Received: by mail-qk0-f171.google.com with SMTP id s68so10894069qkh.3 for ; Wed, 23 Mar 2016 12:04:41 -0700 (PDT) Received: from smtp-fw-6001.amazon.com (smtp-fw-6001.amazon.com. [52.95.48.154]) by mx.google.com with ESMTPS id c52si3310844qgc.5.2016.03.23.12.04.40 for (version=TLS1 cipher=AES128-SHA bits=128/128); Wed, 23 Mar 2016 12:04:40 -0700 (PDT) Subject: Re: [RFC] high preempt off latency in vfree path References: <56F1F4A6.2060400@lab126.com> <20160323024402.GA27856@tassilo.jf.intel.com> From: Joel Fernandes Message-ID: <56F2E895.7080003@lab126.com> Date: Wed, 23 Mar 2016 12:03:49 -0700 MIME-Version: 1.0 In-Reply-To: <20160323024402.GA27856@tassilo.jf.intel.com> Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Andi Kleen Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, tj@kernel.org, linux-rt-users@vger.kernel.org, Nick Piggin On 03/22/16 19:44, Andi Kleen wrote: >> (1) >> One is we reduce the number of lazy_max_pages (right now its around 32MB per core worth of pages). >> >> diff --git a/mm/vmalloc.c b/mm/vmalloc.c >> index aa3891e..2720f4f 100644 >> --- a/mm/vmalloc.c >> +++ b/mm/vmalloc.c >> @@ -564,7 +564,7 @@ static unsigned long lazy_max_pages(void) >> >> log = fls(num_online_cpus()); >> >> - return log * (32UL * 1024 * 1024 / PAGE_SIZE); >> + return log * (8UL * 1024 * 1024 / PAGE_SIZE); >> } > > This seems like the right fix to me. Perhaps even make it somewhat smaller. > > Even on larger systems it's probably fine because they have a lot more > cores/threads these days, so it will be still sufficiently large. > Thanks Andi. I'll post a patch then. Regards, Joel > -Andi > -- 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 From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755818AbcCWTEm (ORCPT ); Wed, 23 Mar 2016 15:04:42 -0400 Received: from smtp-fw-6001.amazon.com ([52.95.48.154]:41636 "EHLO smtp-fw-6001.amazon.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753782AbcCWTEl (ORCPT ); Wed, 23 Mar 2016 15:04:41 -0400 X-IronPort-AV: E=Sophos;i="5.24,383,1454976000"; d="scan'208";a="75753705" Subject: Re: [RFC] high preempt off latency in vfree path To: Andi Kleen References: <56F1F4A6.2060400@lab126.com> <20160323024402.GA27856@tassilo.jf.intel.com> CC: , , , , Nick Piggin From: Joel Fernandes Message-ID: <56F2E895.7080003@lab126.com> Date: Wed, 23 Mar 2016 12:03:49 -0700 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.5.1 MIME-Version: 1.0 In-Reply-To: <20160323024402.GA27856@tassilo.jf.intel.com> Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: [10.43.162.125] X-ClientProxiedBy: EX13D11UWB003.ant.amazon.com (10.43.161.206) To EX13D03UWC001.ant.amazon.com (10.43.162.136) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 03/22/16 19:44, Andi Kleen wrote: >> (1) >> One is we reduce the number of lazy_max_pages (right now its around 32MB per core worth of pages). >> >> diff --git a/mm/vmalloc.c b/mm/vmalloc.c >> index aa3891e..2720f4f 100644 >> --- a/mm/vmalloc.c >> +++ b/mm/vmalloc.c >> @@ -564,7 +564,7 @@ static unsigned long lazy_max_pages(void) >> >> log = fls(num_online_cpus()); >> >> - return log * (32UL * 1024 * 1024 / PAGE_SIZE); >> + return log * (8UL * 1024 * 1024 / PAGE_SIZE); >> } > > This seems like the right fix to me. Perhaps even make it somewhat smaller. > > Even on larger systems it's probably fine because they have a lot more > cores/threads these days, so it will be still sufficiently large. > Thanks Andi. I'll post a patch then. Regards, Joel > -Andi >