Return-Path: Received: from bombadil.infradead.org ([198.137.202.9]:50456 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753928AbbHJI0X (ORCPT ); Mon, 10 Aug 2015 04:26:23 -0400 Date: Mon, 10 Aug 2015 01:26:22 -0700 From: Christoph Hellwig To: Jeff Layton Cc: Christoph Hellwig , bfields@fieldses.org, linux-nfs@vger.kernel.org Subject: Re: [PATCH v2 03/18] nfsd: convert laundry_wq to something less nfsd4 specific Message-ID: <20150810082622.GA18624@infradead.org> References: <1438264341-18048-1-git-send-email-jeff.layton@primarydata.com> <1438809216-4846-1-git-send-email-jeff.layton@primarydata.com> <1438809216-4846-4-git-send-email-jeff.layton@primarydata.com> <20150809071438.GC9772@infradead.org> <20150809071137.28ca8f19@tlielax.poochiereds.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20150809071137.28ca8f19@tlielax.poochiereds.net> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Sun, Aug 09, 2015 at 07:11:37AM -0400, Jeff Layton wrote: > create_singlethread_workqueue already makes an unbound workqueue. This > patch just lifts the "max_active" value to the default, and removes the > WQ_MEM_RECLAIM flag. > > We certainly could turn this into a bound workqueue, but given the sort > of job that the laundromat runs I'm not sure we'd benefit much from the > locality. > > ...and sure, I can turn this into two patches if you'd prefer. The patch was just rather confusing to me. Do you want the existing laundromat to scale better with lots of namespaces? Sounds reasonable, but I don't really see the use case. Looking at the later patches I now see you're overloading a totally different job to it. I don't think there's a point given how cheap workqueues are these days. Even more it seems like you really should use the mm/list_lru.c infrastructure and a shrinker for a your file cache.