From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753925AbcCWCoI (ORCPT ); Tue, 22 Mar 2016 22:44:08 -0400 Received: from mga03.intel.com ([134.134.136.65]:6740 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753525AbcCWCoF (ORCPT ); Tue, 22 Mar 2016 22:44:05 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.24,380,1455004800"; d="scan'208";a="929857085" Date: Tue, 22 Mar 2016 19:44:02 -0700 From: Andi Kleen To: Joel Fernandes Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, tj@kernel.org, linux-rt-users@vger.kernel.org, Nick Piggin Subject: Re: [RFC] high preempt off latency in vfree path Message-ID: <20160323024402.GA27856@tassilo.jf.intel.com> References: <56F1F4A6.2060400@lab126.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <56F1F4A6.2060400@lab126.com> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > (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. -Andi