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;
Subject: Re: [PATCH] proc: Escape more characters in /proc/mounts output
To:     Al Viro <viro@zeniv.linux.org.uk>
Cc:     linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
        Florian Weimer <fweimer@redhat.com>
References: <20201215042454.998361-1-siddhesh@gotplt.org>
 <20201215061005.GF3579531@ZenIV.linux.org.uk>
From:   Siddhesh Poyarekar <siddhesh@gotplt.org>
Message-ID: <85e8114b-31a5-ccd6-f1e4-5b23a67987d6@gotplt.org>
Date:   Tue, 15 Dec 2020 13:00:27 +0530
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.5.0
MIME-Version: 1.0
In-Reply-To: <20201215061005.GF3579531@ZenIV.linux.org.uk>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 12/15/20 11:40 AM, Al Viro wrote:
> On Tue, Dec 15, 2020 at 09:54:54AM +0530, Siddhesh Poyarekar wrote:
> 
>> +	get_user(byte, (const char __user *)data);
>> +
>> +	return byte ? strndup_user(data, PATH_MAX) : NULL;
>>   }
> 
> No.  Not to mention anything else, you
> 	* fetch the same data twice
> 	* fail to check the get_user() results
> 

Ahh sorry, I could inline the strndup_user call and put an additional 
check for length == 1 to return NULL.  Would that be acceptable?

The other alternative would be to not touch copy_mount_string and 
instead, check after fetching the string and if it is blank, free it and 
set to NULL.  That seems more expensive though.

Siddhesh