From:   Ondrej Zary <linux@zary.sk>
To:     Sergey Shtylyov <s.shtylyov@omp.ru>
Subject: Re: [PATCH] pata_parport: fix possible memory leak
Date:   Sat, 11 Mar 2023 22:39:21 +0100
User-Agent: KMail/1.9.10
Cc:     Sergei Shtylyov <sergei.shtylyov@gmail.com>,
        Damien Le Moal <damien.lemoal@opensource.wdc.com>,
        Dan Carpenter <error27@gmail.com>,
        Christoph Hellwig <hch@lst.de>, linux-ide@vger.kernel.org,
        linux-kernel@vger.kernel.org
References: <3ab89ddc-cb60-47c4-86ad-cdad94a8a3d7@kili.mountain> <2223c1dc-1869-a307-e5da-772cb27b34e5@omp.ru> <202303112211.57761.linux@zary.sk>
In-Reply-To: <202303112211.57761.linux@zary.sk>
MIME-Version: 1.0
Content-Type: Text/Plain;
  charset="utf-8"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Message-Id: <202303112239.21234.linux@zary.sk>
Precedence: bulk

On Saturday 11 March 2023 22:11:57 Ondrej Zary wrote:
> On Saturday 11 March 2023 21:23:25 Sergey Shtylyov wrote:
> > On 3/11/23 11:19 PM, Sergei Shtylyov wrote:
> > 
> > >> When ida_alloc() fails, "pi" is not freed although the misleading
> > >> comment says otherwise.
> > >> Move the ida_alloc() call up so we really don't have to free it.
> > 
> >    Wait, but don't we still need to call kfree() in pi_init_one()?
> 
> If it fails at device_register, the dev.release is already set to
> pata_parport_dev_release which does the kfree(). put_device() should call
> it. If it fails later, device_unregister() should do it.  

But I see that the ida_free() at the end of pi_init_one() is wrong. It uses pi->dev.id but pi is either uninitialized or already freed.
  
-- 
Ondrej Zary