Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Thu, 20 Jan 2022 11:28:22 -0700
From:   Mathieu Poirier <mathieu.poirier@linaro.org>
To:     Miaoqian Lin <linmq006@gmail.com>
Cc:     Suzuki K Poulose <suzuki.poulose@arm.com>,
        Mike Leach <mike.leach@linaro.org>,
        Leo Yan <leo.yan@linaro.org>,
        Alexander Shishkin <alexander.shishkin@linux.intel.com>,
        coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH] coresight: syscfg: Fix memleak on registration failure
 in cscfg_create_device
Message-ID: <20220120182822.GA1342128@p14s>
References: <20220110073100.15497-1-linmq006@gmail.com>
 <20220120175642.GB1338735@p14s>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220120175642.GB1338735@p14s>
Precedence: bulk

On Thu, Jan 20, 2022 at 10:56:42AM -0700, Mathieu Poirier wrote:
> Good morning,
> 
> On Mon, Jan 10, 2022 at 07:31:00AM +0000, Miaoqian Lin wrote:
> > device_register() calls device_initialize(),
> > according to doc of device_initialize:
> > 
> >     Use put_device() to give up your reference instead of freeing
> >     * @dev directly once you have called this function.
> 
> That is _if_ device_initialize() is called manually.  In this instance
> @dev is registered with device_register() and unregistered with
> device_unregister().  The latter conforms to the comment you pointed out and
> calls put_device() as expected.

I originally misunderstood the context - you are referring to the failure path in
cscfg_create_device().  You are correct about needing to call put_device() but
your solution will not work when the module is unloaded properly and
device_unregister() is called by way of cscfg_clear_device().  In that case
device_unregister() is already calling put_device().

Here simply calling put_device() instead of cscfg_dev_release() in the error
path should do just fine.  That will call cscfg_dev_release() and the memory
allocated for cscfg_mgr will be release.

> 
> Thanks,
> Mathieu
> 
> > 
> > To prevent potential memleak, use put_device() instead call kfree
> > directly.
> > 
> > Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
> > ---
> >  drivers/hwtracing/coresight/coresight-syscfg.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/drivers/hwtracing/coresight/coresight-syscfg.c b/drivers/hwtracing/coresight/coresight-syscfg.c
> > index 43054568430f..007fa1c761a7 100644
> > --- a/drivers/hwtracing/coresight/coresight-syscfg.c
> > +++ b/drivers/hwtracing/coresight/coresight-syscfg.c
> > @@ -764,7 +764,7 @@ struct device *cscfg_device(void)
> >  /* Must have a release function or the kernel will complain on module unload */
> >  static void cscfg_dev_release(struct device *dev)
> >  {
> > -	kfree(cscfg_mgr);
> > +	put_device(dev);
> >  	cscfg_mgr = NULL;
> >  }
> >  
> > -- 
> > 2.17.1
> >