From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pa0-f70.google.com (mail-pa0-f70.google.com [209.85.220.70]) by kanga.kvack.org (Postfix) with ESMTP id 8698F6B025E for ; Tue, 31 May 2016 17:05:39 -0400 (EDT) Received: by mail-pa0-f70.google.com with SMTP id fg1so337552213pad.1 for ; Tue, 31 May 2016 14:05:39 -0700 (PDT) Received: from mga04.intel.com (mga04.intel.com. [192.55.52.120]) by mx.google.com with ESMTP id s75si33098523pfa.111.2016.05.31.14.05.38 for ; Tue, 31 May 2016 14:05:38 -0700 (PDT) Date: Tue, 31 May 2016 17:13:08 -0400 From: Keith Busch Subject: Re: [PATCH] mm/swap: lru drain on memory reclaim workqueue Message-ID: <20160531211308.GE24107@localhost.localdomain> References: <1464727815-13073-1-git-send-email-keith.busch@intel.com> <20160531210116.GA14868@mtj.duckdns.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160531210116.GA14868@mtj.duckdns.org> Sender: owner-linux-mm@kvack.org List-ID: To: Tejun Heo Cc: linux-mm@kvack.org, LKML On Tue, May 31, 2016 at 05:01:16PM -0400, Tejun Heo wrote: > So, WQ_MEM_RECLAIM on a shared workqueue doesn't make much sense. > That flag guarantees single concurrency level to the workqueue. How > would multiple users of a shared workqueue coordinate around that? > What prevents one events_mem_unbound user from depending on, say, > draining lru? If lru draining requires a rescuer to guarantee forward > progress under memory pressure, that rescuer worker must be dedicated > for that purpose and can't be shared. Gotchya, that fixes my understanding on the rescuer thread operation. In this case, could we revive your previous proposal for consideration? -- 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