Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752088AbaFJKDd (ORCPT ); Tue, 10 Jun 2014 06:03:33 -0400 Received: from mx2.parallels.com ([199.115.105.18]:35600 "EHLO mx2.parallels.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750801AbaFJKDb (ORCPT ); Tue, 10 Jun 2014 06:03:31 -0400 Date: Tue, 10 Jun 2014 14:03:15 +0400 From: Vladimir Davydov To: Joonsoo Kim CC: , , , , , , , Subject: Re: [PATCH -mm v2 8/8] slab: make dead memcg caches discard free slabs immediately Message-ID: <20140610100313.GA6293@esperanza> References: <27a202c6084d6bb19cc3e417793f05104b908ded.1402060096.git.vdavydov@parallels.com> <20140610074317.GE19036@js1304-P5Q-DELUXE> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Disposition: inline In-Reply-To: <20140610074317.GE19036@js1304-P5Q-DELUXE> X-Originating-IP: [81.5.110.170] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi, On Tue, Jun 10, 2014 at 04:43:17PM +0900, Joonsoo Kim wrote: > You mentioned that disabling per cpu arrays would degrade performance. > But, this patch is implemented to disable per cpu arrays. Is there any > reason to do like this? How about not disabling per cpu arrays and > others? Leaving it as is makes the patch less intrusive and has low > impact on performance. I guess that amount of reclaimed memory has no > big difference between both approaches. Frankly, I incline to shrinking dead SLAB caches periodically from cache_reap too, because it looks neater and less intrusive to me. Also it has zero performance impact, which is nice. However, Christoph proposed to disable per cpu arrays for dead caches, similarly to SLUB, and I decided to give it a try, just to see the end code we'd have with it. I'm still not quite sure which way we should choose though... Thanks. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/