From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756619AbcA2RdK (ORCPT ); Fri, 29 Jan 2016 12:33:10 -0500 Received: from mail-wm0-f43.google.com ([74.125.82.43]:34052 "EHLO mail-wm0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756595AbcA2RdH convert rfc822-to-8bit (ORCPT ); Fri, 29 Jan 2016 12:33:07 -0500 Subject: Re: [PATCH] of: resolver: Add missing of_node_put Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2104\)) Content-Type: text/plain; charset=utf-8 From: Pantelis Antoniou In-Reply-To: <20160129164531.GB15053@rob-hp-laptop> Date: Fri, 29 Jan 2016 19:33:02 +0200 Cc: Mark Rutland , Amitoj Kaur Chawla , Frank Rowand , Grant Likely , Devicetree List , Linux Kernel Mailing List , julia.lawall@lip6.fr Content-Transfer-Encoding: 8BIT Message-Id: References: <20160127152017.GA16048@amitoj-Inspiron-3542> <20160127160531.GB17123@leverpostej> <20160129164531.GB15053@rob-hp-laptop> To: Rob Herring X-Mailer: Apple Mail (2.2104) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Rob, > On Jan 29, 2016, at 18:45 , Rob Herring wrote: > > On Wed, Jan 27, 2016 at 06:14:00PM +0200, Pantelis Antoniou wrote: >> Hi Mark, >> >>> On Jan 27, 2016, at 18:05 , Mark Rutland wrote: >>> >>> On Wed, Jan 27, 2016 at 08:50:17PM +0530, Amitoj Kaur Chawla wrote: >>>> for_each_child_of_node performs an of_node_get on each iteration, so >>>> to break out of the loop an of_node_put is required. >>>> >>>> Found using Coccinelle. The semantic patch used for this is as follows: >>>> >>>> // >>>> @@ >>>> expression e; >>>> local idexpression n; >>>> @@ >>>> >>>> for_each_child_of_node(..., n) { >>>> ... when != of_node_put(n) >>>> when != e = n >>>> ( >>>> return n; >>>> | >>>> + of_node_put(n); >>>> ? return ...; >>>> ) >>>> ... >>>> } >>>> // >>> >>>> Signed-off-by: Amitoj Kaur Chawla >>>> --- >>>> drivers/of/resolver.c | 4 +++- >>>> 1 file changed, 3 insertions(+), 1 deletion(-) >>>> >>>> diff --git a/drivers/of/resolver.c b/drivers/of/resolver.c >>>> index 640eb4c..e2a0143 100644 >>>> --- a/drivers/of/resolver.c >>>> +++ b/drivers/of/resolver.c >>>> @@ -40,8 +40,10 @@ static struct device_node *__of_find_node_by_full_name(struct device_node *node, >>>> >>>> for_each_child_of_node(node, child) { >>>> found = __of_find_node_by_full_name(child, full_name); >>>> - if (found != NULL) >>>> + if (found != NULL) { >>>> + of_node_put(child); >>>> return found; >>>> + } >>>> } >>>> >>>> return NULL; >>> >>> I don't think this is quite right. When child == found, this change will >>> leave it decremented. >>> >> >> >> This patch is bogus. >> >> __of_find_node_by_full_name() is not taking a reference on the node if found. >> This method relies on keeping the reference taken by the loop. >> >> Taking this into account all of these conccinelle tests are bogus. >> >> The DT internal method are not using the object model in an obvious manner >> and applying these patches without vetting each and everyone is bound to >> break things. > > Things are already broken. But does it matter? > > Our time would be better spent re-designing any refcounting around where > we actually need it rather than trying to fix up the many locations > which are wrong and don't matter. As long as it is callers' > responsibility to get this right, it will never be right. Even the core > code has a hard time getting it right. > Let me pile up. Refcounting for DT is broken. There’s no point trying to fix it as it is. I have a big pile of TODO, one of these is fixing (as in severely cutting down) the areas where refcounting is needed. The idea would be to keep refcounting only in core and provide interfaces that use different semantics for drivers and subsystems. We can discuss things in ELC this April, perhaps on a BoF session again. > Rob Regards — Pantelis