From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx3-rdu2.redhat.com ([66.187.233.73]:52198 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1726775AbeIKVfB (ORCPT ); Tue, 11 Sep 2018 17:35:01 -0400 Date: Wed, 12 Sep 2018 00:34:44 +0800 From: Ming Lei To: Tejun Heo Cc: linux-kernel@vger.kernel.org, Jianchao Wang , Kent Overstreet , linux-block@vger.kernel.org Subject: Re: [PATCH] percpu-refcount: relax limit on percpu_ref_reinit() Message-ID: <20180911163443.GD10082@ming.t460p> References: <20180909125824.9150-1-ming.lei@redhat.com> <20180910164920.GE1100574@devbig004.ftw2.facebook.com> <20180911000049.GB30977@ming.t460p> <20180911134836.GG1100574@devbig004.ftw2.facebook.com> <20180911154540.GA10082@ming.t460p> <20180911154959.GI1100574@devbig004.ftw2.facebook.com> <20180911160532.GB10082@ming.t460p> <20180911163032.GA2966370@devbig004.ftw2.facebook.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20180911163032.GA2966370@devbig004.ftw2.facebook.com> Sender: linux-block-owner@vger.kernel.org List-Id: linux-block@vger.kernel.org On Tue, Sep 11, 2018 at 09:30:32AM -0700, Tejun Heo wrote: > On Wed, Sep 12, 2018 at 12:05:33AM +0800, Ming Lei wrote: > > On Tue, Sep 11, 2018 at 08:49:59AM -0700, Tejun Heo wrote: > > > On Tue, Sep 11, 2018 at 11:45:41PM +0800, Ming Lei wrote: > > > > > So, this part seems wrong. The function is called percpu_ref_reinit() > > > > > - the refcnt is expected to be in its initial state with just the base > > > > > ref once this function returns. If you're removing the restriction on > > > > > > > > But the comment says that 'Re-initialize @ref so that it's in the same > > > > state as when it finished', and this invariant isn't changed with this > > > > patch. > > > > > > The comment goes "when perpcu_ref_init() finished". The function is > > > called re _init_. It should put the ref in the initial state, right? > > > > OK, I am fine to keep the behaviour for this API, will introduce a new helper > > for NVMe. > > Why aren't switch_to_atomic/percpu enough? The blk-mq's use case is this _reinit is done on one refcount which was killed via percpu_ref_kill(), so the DEAD flag has to be cleared. > > > However, it is doable to switch to percpu mode from atomic mode when it > > doesn't drop to zero, looks like sort of perpcu_ref_reinit(inherit_old_refcount). > > Isn't that way more contorted than just switching operating modes? As explained above. Thanks, Ming