Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH] net: via-rhine: net: Fix a resource leak in rhine_init()
To:     David Miller <davem@davemloft.net>
Cc:     netdev@vger.kernel.org, linux-kernel@vger.kernel.org
References: <20190504030813.17684-1-baijiaju1990@gmail.com>
 <20190503.231308.1440125282445011089.davem@davemloft.net>
 <f2e7c284-aa70-2b67-b9dd-461db102cbc5@gmail.com>
 <20190503.234131.1909607319842703379.davem@davemloft.net>
From:   Jia-Ju Bai <baijiaju1990@gmail.com>
Message-ID: <477140c2-841c-616f-fa25-faa428381111@gmail.com>
Date:   Sat, 4 May 2019 11:57:39 +0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101
 Thunderbird/52.2.0
MIME-Version: 1.0
In-Reply-To: <20190503.234131.1909607319842703379.davem@davemloft.net>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit
Content-Language: en-US
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On 2019/5/4 11:41, David Miller wrote:
> From: Jia-Ju Bai <baijiaju1990@gmail.com>
> Date: Sat, 4 May 2019 11:23:06 +0800
>
>>
>> On 2019/5/4 11:13, David Miller wrote:
>>> From: Jia-Ju Bai <baijiaju1990@gmail.com>
>>> Date: Sat,  4 May 2019 11:08:13 +0800
>>>
>>>> When platform_driver_register() fails, pci_unregister_driver() is not
>>>> called to release the resource allocated by pci_register_driver().
>>>>
>>>> To fix this bug, error handling code for platform_driver_register()
>>>> and
>>>> pci_register_driver() is separately implemented.
>>>>
>>>> This bug is found by a runtime fuzzing tool named FIZZER written by
>>>> us.
>>>>
>>>> Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
>>> I think the idea here is that PCI is not enabled in the kernel, it is
>>> fine for the pci register to fail and only the platform register to
>>> succeed.
>>>
>>> You are breaking that.
>> Okay, I can understand it.
>> If so, I think that platform_driver_register() should be called before
>> pci_register_driver(), and it is still necessary to separately handle
>> their errors.
>> If you agree, I will send a v2 patch.
> It is only a failure if both fail.
>
> If at least one succeeds, the driver can potentially probe properly.

Thanks for the explanation, I understand now :)
The code logic seems a little strange, but it should be correct in 
rhine_init()...
If so, if one fails, I wonder whether it is correct for rhine_cleanup() 
to call both platform_driver_unregister() and pci_unregister_driver()?


Best wishes,
Jia-Ju Bai